Birgit Kellner, Patrick McAllister,
Horst Lasic, Sara McClintock (eds.)
Reverberations
of Dharmakīrti‘s
Philosophy
Reverbe
Reverberatio
tions Dharm
Dharmakīr
Proceedings of the Fifth International
Dharmakīrti Conference Heidelberg
August 26 to 30, 2014
ISBN 978-3-7001-8781-3
BIRGIT KELLNER, PATRICK MCALLISTER,
HORST LASIC, SARA MCCLINTOCK (EDS.)
REVERBERATIONS OF DHARMAKĪRTI‘S PHILOSOPHY
ÖSTERREICHISCHE AKADEMIE DER WISSENSCHAFTEN
PHILOSOPHISCH-HISTORISCHE KLASSE
SITZUNGSBERICHTE, 904. BAND
BEITRÄGE ZUR KULTUR- UND
GEISTESGESCHICHTE ASIENS, NR. 104
HERAUSGEGEBEN VOM INSTITUT
FÜR KULTUR- UND GEISTESGESCHICHTE ASIENS
UNTER DER LEITUNG VON BIRGIT KELLNER
Birgit Kellner, Patrick McAllister,
Horst Lasic, Sara McClintock (eds.)
Reverberations
of Dharmakīrti‘s
Philosophy
Proceedings of the Fifth International
Dharmakīrti Conference Heidelberg
August 26 to 30, 2014
Accepted by the publication committee of the Division of Humanities
and Social Sciences of the Austrian Academy of Sciences by:
Michael Alram, Bert G. Fragner, Andre Gingrich, Hermann Hunger, Sigrid Jalkotzy-Deger,
Renate Pillinger, Franz Rainer, Oliver Jens Schmitt, Danuta Shanzer, Peter Wiesinger,
Waldemar Zacharasiewicz
Published with the support of the Austrian Science Fund (FWF): PUB 806
Open access: Where not otherwise specified, this work is licensed
under a Creative Commons Attribution 4.0 Unported License.
To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/
This publication was subjected to an anonymous international review process.
Peer review is an essential part of the Austrian Academy of Sciences Press evaluation process. Before any book
can be accepted for publication, it is assessed by international specialists and ultimately must be approved by
the Austrian Academy of Sciences Publication Committee.
The paper used in this publication is DIN EN ISO 9706 certified and meets the
requirements for permanent archiving of written cultural property.
Some rights reserved.
ISBN 978-3-7001-8781-3
Copyright © Austrian Academy of Sciences, Vienna 2020
Typesetting: Institut für Kultur- und Geistesgeschichte Asiens, Vienna
Print: Prime Rate, Budapest
https://epub.oeaw.ac.at/8781-3
https://verlag.oeaw.ac.at
Made in Europe
Participants of the Fifth International Dharmakīrti Conference, Heidelberg 2014
Contents
Preface
xi
Opening Speech
Shōryū Katsura
xvii
Account of the Fifth International Dharmakīrti Conference
xxiii
1 Jitāri’s Vijñaptimātratāsiddhi
Junjie Chu
2 Consciousness and Causality: Dharmakīrti Against Physicalism
Christian Coseru
3 Action Theory and Scriptural Exegesis in Early Advaita-Vedānta (2):
Maṇḍana Miśra’s Excursus on the Buddha’s Omniscience
Hugo David
4 On the Determination of Causation by Dharmakīrti
Eli Franco
5 Reinterpretation of the Compound svabhāva-pratibandha in
Dharmakīrti’s Logical Theory
Yoichi Fukuda
1
21
41
77
91
6 Jain Conceptions of Non-Cognition: A Dialogue with Dharmakīrti on
Inferential Evidence
Marie-Hélène Gorisse
111
7 Universals, Demons’ Pots, and Demons’ Permanent Pots: Phya pa Chos
kyi seng ge on Unestablished Subjects in Arguments by Consequence
Pascale Hugon
129
8 Two Kinds of Causal Capacity: sāmānyā śaktiḥ and pratiniyatā śaktiḥ
Masahiro Inami
155
9 Jñānaśrīmitra on viparyayabādhakapramāṇa
Kyo Kano
177
viii
CONTENTS
10 How Does One Cognize a Cow? A Dialogue between Mādhava and
Dignāga
Kei Kataoka
191
11 Śāntarakṣita on Two Kinds of Arguments for Self-Awareness:
sahopalambhaniyama and saṃvedana
Hiroko Matsuoka
203
12 Prajñākaragupta’s Argument for Two Means of Valid Cognition
Patrick McAllister
221
13 Balancing the Scales: Dharmakīrti Inside and Out
Lawrence McCrea
251
14 The Concept of vyava√sthā in Dharmottara’s and Prajñākaragupta’s
Explanation of pramāṇa and pramāṇaphala
Mai Miyo
267
15 Dharmakīrti’s Apoha Theory: Perceptual Judgment and the Lack of
Superimposition (samāropaviveka)
Miyuki Nakasuka
283
16 Dharmakīrti’s Notion of Permanence and Its Impact on the Tibetan
Buddhist Doctrine of Buddhahood
Hiroshi Nemoto
301
17 Dignāga on the View of a Generic Term as Denoting a Relation
Hideyo Ogawa
18 Communication between a Speaker and a Listener as the Seeing of a
Double Moon — In Light of the Apoha Theory of Śāntarakṣita and
Kamalaśīla
Kensho Okada
313
331
19 On pramāṇabhūta — The Change of Its Meaning from Dignāga to
Prajñākaragupta
Motoi Ono
343
20 Dharmakīrti on the Role of Salvific Initiation and the Reception of His
Critique in the Later Śaiva Tradition
Cristina Pecchia
363
21 Dharmottara on the viparyaye bādhakapramāṇa and trairūpya in
Dharmakīrti’s sattvānumāna
Masamichi Sakai
375
CONTENTS
ix
22 Dharmakīrti’s Vādanyāya and the History of Conceptions of Debate in
Indian Logic
Ryo Sasaki
391
23 The Meaning of bāhyārtha in Dignāga’s and Jinendrabuddhi’s Theories
of Inference
Kiyokuni Shiga
411
24 Closing a Gap in the Interpretation of Dharmakīrti’s Logic
Ernst Steinkellner
429
25 Philosophical Reflections on the sahopalambhaniyama Argument
John Taber
441
26 The Truth, the Buddha’s Words, and Inference: Bhāviveka’s Theory of
Two Truths
Masaki Tamura
27 The Concept of sādhana in Chinese Buddhist Logic
Mingjun Tang
28 A Causeless Liberation? Kṣemarāja’s Response to Dharmakīrti’s
Critique of Initiation
Somadeva Vasudeva
463
473
497
29 Dharmakīrti and His Successors on āśrayāsiddha and prasaṅgaviparyaya
Toshikazu Watanabe
513
30 Another Look at avinābhāva and niyama in Kumārila’s Exegetical Works
Kiyotaka Yoshimizu
529
Index
549
Preface
This volume contains papers presented at the Fifth International Dharmakīrti Conference,
held at the Crowne Plaza Hotel in Heidelberg, August 26 to 30, 2014. Professor Katsura,
one of the leading senior scholars in the field, recalled the history of the Dharmakīrti
conferences in his opening speech, which is also published in this volume.
The Heidelberg Conference was organized jointly by three projects which Birgit Kellner
directed at the University of Heidelberg: Project MC 13.2 “Reasoning in Buddhism between
South Asia and Tibet” and Project MC 3.3 “Buddhism between South Asia and Tibet –
Negotiating Religious Boundaries in Doctrine and Practice,” both financially supported
by the Cluster of Excellence “Asia and Europe in a Global Context”; and the project
“Systems of Epistemology in Classical Indian Philosophy,” supported by the German
Research Foundation DFG. We gratefully acknowledge the financial support received from
the sponsors of these projects that made the conference possible, as well as the efficient
and kind organizational support by the Cluster’s administrative staff.
Like the previous volumes of Dharmakīrti Conference proceedings, this collection
testifies to a growing and dynamic field, driven by significant discoveries of new sources,
a growing body of historical knowledge, and a continually refined awareness of the sophisticated nature of the Indian, Tibetan and East Asian intellectual traditions that jointly
constitute the historical reference point for Dharmakīrtian Studies. The editing of the proceedings took longer than expected, and for various reasons not all of the papers presented
at the conference could be included. Contrarily, the papers by Hiroko Matsuoka and Patrick
McAllister could not be presented at the conference, but were included here due to their
topical relevance.
Editorial work was conducted chiefly at the Institute for the Cultural and Intellectual
History of Asia of the Austrian Academy of Sciences, where two of the Heidelberg conference organizers, Kellner and McAllister, had in the meantime relocated (and where
Horst Lasic had been working all along; our fourth editor, Sara McClintock, also spent
time here in 2016). Cynthia Peck-Kubaczek of the Academy institute corrected the English
of a number of the papers, and we thank her for her painstaking efforts. We also gratefully
acknowledge editorial assistance by Liudmila Olalde (Heidelberg) whose sharp eyes let no
missing bibliographical reference escape. Together with McAllister, Olalde also handled
technical aspects in the production of the camera-ready copy.
The shorthand “Dharmakīrtian Studies” refers to the study of philosophical currents
in India, China and Tibet which take the theoretical efforts of Dharmakīrti (between
mid-6th and mid-7th centuries CE) and his predecessor Dignāga (ca. 480-540) in the
fields of epistemology and logic as their inspiration – theoretical efforts that revolve
around the explication, justification and defense of a system of “instruments of trustworthy
awareness” (pramāṇa), and, driven by these concerns, also extend into other areas of vital
xii
PREFACE
interest to Buddhist intellectuals in the context of their respective times. Such areas include
problems in the philosophy of mind pertaining to the analysis of consciousness, subjects in
the philosophy of language, here intertwined with the analysis of concepts and concept
formation. Theoretical aspects of Buddhism as a soteriology, as a set of teachings geared
towards the attainment of liberation from suffering in saṃsāra, also play a central role
in Buddhist logico-epistemological discourse. Buddhist pramāṇa theories were adopted,
adapted and criticized by non-Buddhists primarily in their Indian context. Dharmakīrtian
Studies therefore, as a matter of course, also attend to explorations of this larger intellectual
environment between the late fifth and thirteenth centuries CE, an environment shaped by
mutual influence and cross-fertilization, as well as intense polemics between competing
religio-philosophical currents encompassing Brahmanical traditions as well as Jains and
others.
In the past decades, the history of Dharmakīrtian Studies has been significantly shaped,
if not revolutionized, by the discovery of new sources and improved access to them. Within
the larger area of Indian Buddhist literature, Sanskrit pramāṇa literature has been particularly profoundly affected by improved access to Sanskrit manuscripts which have been
preserved in the territory of today’s Tibetan Autonomous Region (TAR) within the People’s
Republic of China.1 Until the beginning of the 21st century, key works of Dignāga and
Dharmakīrti themselves were unknown in the language of their composition, Sanskrit. An
agreement between the Austrian Academy of Sciences and the China Tibetology Research
Center (CTRC) in Beijing, concluded in 2004 and renewed several times since, laid the
foundation for collaborative research based on photocopies of manuscript photographs
kept in the CTRC’s library. Copies of manuscripts of Dharmakīrti’s Pramāṇaviniścaya,
Hetubindu and Santānāntarasiddhi became accessible, as well as of Jinendrabuddhi’s
Pramāṇasamuccayaṭīkā, a commentary on Dignāga’s main logico-epistemological work,
the Pramāṇasamuccaya and -vṛtti. Research on these new sources had already begun
when the Fourth International Dharmakīrti Conference was held in Vienna (August 23-27,
2005). Ernst Steinkellner’s opening speech “News from the manuscript department” lays
out the specifics of the cooperation and its (now) early history, and summarizes ongoing
work and first results; the edition of the first chapter of Jinendrabuddhi’s Pramāṇasamuccayaṭīkā arrived at the Vienna conference just fresh from the press.2 Steinkellner also
outlined the challenges that lie ahead. Besides the enormous task of scholarly work involved in the analysis, edition and translation of these new materials, there remains the
task of a full descriptive catalogue of all Sanskrit manuscripts in the TAR. The actual
1
2
This is also borne out by the significant pramāṇa content in the by now three panels on Sanskrit
manuscripts in China that have been held at the Beijing Seminar of Tibetan Studies, published in the
three volumes Sanskrit Manuscripts in China I (edited by Ernst Steinkellner in cooperation with Duan
Qing and Helmut Krasser, Beijing 2009), II (edited by Horst Lasic and Xuezhu Li, Beijing 2016), and III
(edited by Birgit Kellner, Jowita Kramer and Xuezhu Li, Beijing, forthcoming). Volume II is available
for download at https://www.oeaw.ac.at/fileadmin/Institute/IKGA/PDF/digitales/Lasic_Li_2016.pdf
(last accessed 15 September 2019).
Cf. Helmut Krasser, Horst Lasic, Eli Franco, Birgit Kellner (ed.): Religion and Logic in Buddhist
Philosophical Analysis: Proceedings of the Fourth International Dharmakīrti Conference, Vienna,
August 23-27, 2005 (Vienna 2011: Verlag der Österreichischen Akademie der Wissenschaften), pp.
xvii-xxi.
xiii
manuscripts are still out of bound for Chinese as well as foreign scholars. The same holds
good for a reported 61-volume set of color reproductions of all these manuscripts, of
which five copies are reported to have been printed. It also has not been possible, since
Steinkellner’s report, to access a bundle of paper manuscripts in the Potala palace in Lhasa,
which among others includes a manuscript of Dignāga’s Nyāyamukha and manuscripts of
Dharmakīrti’s Sambandhaparīkṣākārikā, Santānāntarasiddhiprakaraṇa and Pramāṇaviniścayakārikā.3 Steinkellner’s introduction to the volume Sanskrit Manuscripts in China
III (Beijing, forthcoming) summarizes the current situation and formulates a proposal for
further improvement. In the three years since the keynote lecture on which that introduction
is based was held in Beijing, nothing of substance has happened.
While progress in further improving access of scholars to Sanskrit manuscripts in
China has been slow, editorial activities have yielded significant further results. In 2010,
the monograph series Sanskrit Texts from the Tibetan Autonomous Region, founded as
a joint venture of the China Tibetology Publishing House and the Austrian Academy of
Sciences Press, counted eight volumes. In 2019, volumes 21 and 22 are being submitted: the
diplomatic edition of the third chapter of Dharmottara’s Pramāṇaviniścayaṭīkā by Pascale
Hugon (Vienna) in collaboration with Takashi Iwata (Tokyo) and Toshikazu Watanabe
(Vienna, now Tokyo), as well as the critical edition of the first five chapters of Candrakīrti’s
Madhyamakāvatārabhāṣya by Horst Lasic, Xuezhu Li (Beijing) and Anne MacDonald
(Vienna), based on preparatory work by Helmut Krasser. The sixth chapter is being edited
by Anne MacDonald, while the remaining chapters are being edited by Katsura and Li.
As Katsura also recalled in his opening speech, Dharmakīrti’s Pramāṇaviniścaya and
Hetubindu are now available in critical editions by, respectively, Steinkellner (chapters 1
and 2 of the Pramāṇaviniścaya; Hetubindu), as well as Hugon and Toru Tomabechi (Tokyo)
(chapter 3 of the Pramāṇaviniścaya).4 The second chapter of the Pramāṇasamuccayaṭīkā
has been critically edited by Steinkellner, Helmut Krasser and Horst Lasic. Further chapters
are currently being edited by Katsura, Motoi Ono (Tsukuba), Yasutaka Muroya (Vienna),
and Toshikazu Watanabe, with additional support by a group of younger Japanese scholars.
Sections of the second chapter of Dharmottara’s Pramāṇaviniścayaṭīkā, preserved only in
fragmentary manuscripts, have been edited in two Vienna dissertations by Hisataka Ishida
and Masamichi Sakai, supervised by Helmut Krasser.5
Meanwhile, other institutions have been able to conclude cooperation agreements with
the CTRC similar to the Viennese model, for individual manuscripts and texts. In the area
of pramāṇa literature, mention should here be made of efforts at the University of Leipzig,
where Eli Franco, Junjie Chu, Xuezhu Li and Hiroko Matsuoka are editing Yamāri’s (c.
1000-1060) important commentary on Prajñākaragupta’s (c. 750-810) Pramāṇavārttikālaṅkārabhāṣya, as well as rare works by Jitāri (940-1000).6 It is a promising sign that
Chinese scholars are increasingly involved in these editorial activities, as attested by Li’s
3
4
5
6
For a full list of the contents see Steinkellner, “News from the manuscript department”, p. xxi.
For bibliographical references cf. Katsura’s “opening speech” below on page xvii.
Sakai’s 2010 dissertation (PDF download at http://othes.univie.ac.at/9623/) concerns the proof of
momentariness, while Ishida’s 2011 dissertation (PDF at http://othes.univie.ac.at/13375/) deals with
the subject of the logical nexus.
Cf. Junjie Chu and Eli Franco, “Rare Manuscripts of Works by Jitāri”, in: Horst Lasic and Xuezhu Li
(ed.): Sanskrit Manuscripts in China II. Beijing: China Tibetology Publishing House, 15-48.
xiv
PREFACE
participation in several projects; Luo Hong (formerly CTRC, now at Sichuan University
in Chengdu) is working on an edition of Ratnākāraśānti’s Prajñāpāramitopadeśa. In the
long run research on these manuscripts will only be able to flourish if a new generation of
Sanskritists in China carries it forward.
More recent discoveries that may serve as the basis of future projects belong to the final
period of pramāṇa activities in India. There is a third manuscript of Jitāri’s Vijñaptimātratāsiddhi, in addition to the two manuscripts described by Franco and Chu. There is also a copy
of a valuable manuscript of a lengthy work on the sahopalambhaniyama-inference entitled
Sahopalambhaniyamasamarthana, also ascribed to Jitāri. Based on selected sample passages, this text can be assumed to be the same work referred to as Sahopalambhaprakaraṇa
in the colophon of a manuscript that both Rāhula Sāṅkṛtyāyana and Giuseppe Tucci photographed in Ngor monastery.7 However, approximately one third of the Ngor manuscript
is missing in Tucci’s photographs; the remainder is often out of focus. In Sāṅkṛtyāyana’s
photographs, the text is almost completely illegible.8 A hitherto unknown manuscript of
Jñānaśrīmitra’s Advaitabinduprakaraṇa was also recently discovered; it complements the
codex photographed by Sāṅkṛtyāyana in 1938 that formed the basis of Anantalal Thakur’s
edition first published in 1959 (reprinted in 1987) and allows to substantially improve
the text. These are only a few of the many cases where new manuscripts from the TAR
lend invaluable support to editorial work together with other materials; Śāntarakṣita’s
Vādanyāyaṭīkā and Dharmakīrti’s Vādanyāya are another particularly prominent case in
point.9 Lastly, there also remains the extensive manuscript of 123 folios of an otherwise
unknown commentary on Arcaṭa’s Hetubinduṭīkā, in the colophon ascribed to a certain
Jayabhadra or Bhavabhadra.10
Tibetan developments inspired by Indian pramāṇa have similarly benefited from greater
accessibility of sources, as demonstrated by Pascale Hugon’s extensive work on Phya pa
chos kyi seng ge (1109-1169) and other authors from the early period of Tibetan Buddhist
Scholasticism in the 11th to 13th centuries; her paper in this volume offers an entry point
into this newly opened field of enquiry. A considerable amount of pertinent manuscripts
testifying to hitherto largely unknown works have surfaced recently, especially as part of
the private library of the Fifth Dalai Lama Ngag dbang blo bzang rgya mtsho (1617–1682)
in Drepung monastery. They have been published in facsimile in the “Collected Works
of the Bka’ gdams pas” (Bka’ gdams gsung ’bum), released in several installments which
by now number altogether 120 volumes. Hugon and Kazuo Kano (Tokyo) have set out to
7
8
9
10
Tucci’s photographs from 1939 are published in facsimile in Studia Indologiczne 7 (2000) 425-449, as
“Appendix III” to Francesco Sferra’s paper “Sanskrit Manuscripts and Photos of Sanskrit Manuscript in
the [sic] Giuseppe Tucci’s Collection. A Preliminary Report”.
In the Göttingen collection, they are preserved as COD MS SANSCR RAH Xc14/10b (Jitāri, Sahopalambhasiddhi); cf. Bandurski, Frank: “Untersuchungen zur buddhistischen Literatur” = SanskritWörterbuch der buddhistischen Texte aus den Turfan-Funden, Beiheft 5, Göttingen 1994: Vandenhoeck
& Ruprecht, p. 42.
A diplomatic edition of the Kundeling manuscript of the Vādanyāyaṭīkā is currently being prepared by
Yasutaka Muroya at the Academy institute in Vienna, in the framework of the research project “Debate
and rational argumentation in South Asian Buddhism” (P30827) supported by the Austrian Science
Fund FWF.
Cf. Steinkellner, “News from the manuscript department”, p. xx.
xv
produce a descriptive catalogue of this vast collection, while at the same time studies of
individual texts and their contents are being undertaken.11
Last but not least mention should be made of similarly growing research in Chinese
adaptations of Indian pramāṇa. Efforts in this field are undertaken by a younger generation
of scholars in China (cf. the paper by Tang Mingjun in this volume), as well as by a
team of Japanese scholars comprising chiefly Shigeki Moro (Kyoto), Shinya Moriyama
(Matsumoto), Yasutaka Muroya and Motoi Ono – a development facilitated by improved
access to rare manuscripts of commentaries in Japanese temple libraries.12
The contributions to this volume demonstrate that the process of exploring new sources,
of utilizing them in research endeavors and reflecting on how they motivate revisions of
received knowledge, is in full swing. These new discoveries have contributed to a stronger
focus on manuscript research – including problems of paleography and codicology –, and
they have also given precedence to philologically oriented studies. As new texts are to be
edited, new translations are to be produced, and a variety of textual and fundamental historical problems need to be solved. Yet, Dharmakīrtian Studies have at the same time preserved,
even strengthened, their disciplinary openness and methodological pluralism. Philological
and historical studies chiefly concerned with placing texts and thinkers, theories and arguments in the context of intellectual histories that in many respects still remain to be written,
dominate especially in continental Europe and Japan where such methodologies have a
longer academic tradition within Asian Studies at large. But a philosophical engagement
with pramāṇa ideas, an analysis and critical examination of these ideas in terms of their
philosophical significance and substance – more at home in the Angloamerican sphere –,
has also had a place at Dharmakīrti conferences in the past and can by now be considered
an integral part of the world of Dharmakīrtian Studies.13
To take philosophical texts seriously requires reading them as works of philosophy,
just as serious studies of ancient legal literature must take this literature seriously in its
legal dimensions. Many have also argued, convincingly, that a proper history of philosophy
cannot be merely a descriptive account of which philosopher lived when and where and did
what (as notoriously difficult such accounts may be for a field like Indian philosophy where
precise external data is hard to come by). In order to be illuminating, it is to be written
as a history of philosophical thought, with close attention to ideas and content, and by
making plausible why it is that philosophers argue the way they do – in due consideration
11
12
13
The current state of their work is accessible at https://www.oeaw.ac.at/ikga/forschung/tibetologie/mate
rialien/a-gateway-to-early-tibetan-scholasticism/.
Results of these research endeavours were among others presented at the panel “Pramāṇa across Asia:
India, China, Korea, Japan”, held at the XVIIth Congress of the International Association of Buddhist
Studies at the University of Vienna, August 18-23, 2014, published in Wiener Zeitschrift für die Kunde
Südasiens 56-57 (2015-2018). The papers presented at the panel “Transmission and Transformation of
Buddhist Logic and Epistemology in East Asia” (XVIIIth Congress of the International Association of
Buddhist Studies, University of Toronto, August 20–25, 2017) will be published in a volume bearing
the panel title and edited by Shinya Moriyama (Vienna: Arbeitskreis für Tibetische und Buddhistische
Studien, Universität Wien, forthcoming).
John Taber insightfully discusses these different, sometimes divergent, sometimes complementary
approaches to Indian philosophy and their background in disciplinary histories in his paper “On
Engaging Philosophically with Indian Philosophical Texts”, Asiatische Studien / Études Asiatiques 67/1
(2013) 125-163.
PREFACE
xvi
of other forces that drive intellectual history. Philologically oriented historical approaches
and philosophical approaches – each of which may again come in different forms – may
produce tensions, of course, and they often do. The philosopher may find the philologist’s
reticence at wanting to know whether Dharmakīrti is right frustrating (“how could you not
want to know this?”), while the philologist will in turn find the philosopher’s conviction
that they have understood where Dharmakīrti is wrong hyperbolic (“shouldn’t you first
examine his words more carefully before you jump to conclusions?”). The International
Dharmakīrti Conferences have seen a number of discussions along these lines, as the
individual proceedings volumes demonstrate. The field of Dharmakīrtian Studies is best
served by keeping both parties in dialogue, by focusing on what they stand to learn from
each other, and by striving to turn whatever tensions may arise into constructive critical
discourse. It is only then that the seeds which the wealth of our new sources represent will
be able to mature and develop.
September 2019
Birgit Kellner, Vienna
Horst Lasic, Vienna
Sara McClintock, Atlanta
Patrick McAllister, Vienna
Opening Speech
by
Shōryū Katsura
First of all, I would like to thank all of you for coming to participate in the Fifth International
Dharmakīrti Conference and thereby demonstrating the continued vigor and interest in
the study of Dharmakīrti’s works and thought, as well as those of his predecessors and
successors. Special thanks are due to the organizer of this conference, Prof. Birgit Kellner
of the University of Heidelberg, and her assistant, Ms. Ina Chebbi [Buchholz], for their
painstaking efforts. I would also like to thank Dr. Patrick McAllister for his technical
support of all sorts.
I am delighted to see again the faces of many I have met at earlier Dharmakīrti Conferences. But I see many new faces as well, and so this may be a good occasion for me
to give a brief history of the International Dharmakīrti Conferences. It was the late Prof.
Yūichi Kajiyama (1925–2004) who hit upon the idea of holding such a conference. In
1982 Prof. Kajiyama invited Prof. Ernst Steinkellner to Kyōto University as a visiting
professor for one semester to have him read the Vādanyāya with students in Kyōto. Just
before Prof. Steinkellner returned to Vienna, Prof. Kajiyama decided to hold a one-day
workshop on Buddhist logic and epistemology. He invited several Japanese scholars and
students, including Prof. Hiromasa Tosaki, to present papers. Prof. Steinkellner gave a
lecture on the development of the idea of viparyaye bādhakapramāṇam in Dharmakīrti’s
works. Prof. Kajiyama called the event “International Dharmakīrti Conference,” despite the
fact that apart from Prof. Steinkellner there was perhaps no other participant from abroad.
In 1989 Prof. Steinkellner then held the Second International Dharmakīrti Conference
in Vienna. This one was truly “international” in terms of the participants. He called it “the
second conference” as a mark of respect for the first one organized by Prof. Kajiyama. It was
amazing to see that both Prof. Tilmann Vetter (1937–2012) from Leiden and Prof. Lambert
Schmithausen from Hamburg attended the conference. Together with Prof. Steinkellner,
they had both been students of Prof. Erich Frauwallner (1898–1974) at the same time as
Prof. Kajiyama spent a few months in Vienna in the early 1960s. One afternoon in the
middle of the conference we took a Frauwallner Memorial Walk into the Vienna Woods. I
then organized the Third International Dharmakīrti Conference in Hiroshima in 1997; the
fourth one was held again in Vienna in 2005. The proceedings of these three conferences
were published by the Austrian Academy of Sciences Press.14
14
Ernst Steinkellner (ed.), Studies in the Buddhist Epistemological Tradition. Proceedings of the Second
International Dharmakīrti Conference, Vienna, June 11–16, 1989. Vienna 1991. This volume also
includes a “Report on the First International Dharmakīrti Conference at Kyōto, June 16 and 17, 1982”
by Yūichi Kajiyama on p. xi. The proceedings of the third conference were edited by Shōryū Katsura,
Dharmakīrti’s Thought and its Impact on Indian and Tibetan Philosophy. Proceedings of the Third
International Dharmakīrti Conference, Hiroshima, November 4–6, 1997. Vienna 1999. The proceedings
xviii
OPENING SPEECH
In this connection I would like to emphasize the significance of guru-śiṣya relationships
in our field. I happened to be a student of Prof. Kajiyama in the mid-1960s, just after he
came back from his stays in London and Vienna. When I was preparing my MA thesis on
Dharmakīrti’s theory of svasaṃvedana, he read the entire svasaṃvedana portion of the
Pramāṇavārttika Chapter 3 (vv. 320-539) together with Manorathanandin’s commentary
with me almost every Saturday afternoon of 1966–67. And the convener of the present
conference, Prof. Kellner, studied in Vienna and Hiroshima in the 1990s, when both Prof.
Steinkellner and I were teaching at our respective institutes in those cities’ universities.
Today I am glad to see that such guru-śiṣya relationships have developed further and
further in various parts of the world. Here I must remind you that there is neither a formal
association of Dharmakīrti studies, nor any rules that govern us. These conferences have
been held spontaneously and irregularly. Therefore, we do not know when and where the
next Dharmakīrti conference will take place. But that does not bother me at all. As long as
guru-śiṣya relationships continue, I believe that there will be a next one.
It is very sad that I do not see the face of our dear friend Dr. Helmut Krasser (1956–2014)
among you. As you all know, Dr. Krasser passed away last March. We all miss him greatly.
His untimely death is a great loss, not only for those who were immediately associated
with him at the Austrian Academy of Sciences and the University of Vienna, but also for
everyone who is engaged in the field of Buddhist logic and epistemology. He did such a great
service to the development of our field, not only through his own academic contributions
but also by organizing various academic projects, such as the deciphering and editing of
the Sanskrit manuscripts of Jinendrabuddhi’s Ṭīkā on Dignāga’s Pramāṇasamuccayavṛtti
as well as other works. He will be remembered by the younger generation in our field as a
most sympathetic teacher and guide, a person willing to help them with bodhisattva-like
efforts.
Dr. Krasser published an edition and translation of Dharmottara’s Laghuprāmāṇyaparīkṣā (his PhD thesis) and Śaṅkaranandana’s Īśvarāpākaraṇasaṅkṣepa (his Habilitation
thesis). He also edited a number of proceedings of academic conferences, including those of
the Second and the Fourth International Dharmakīrti conferences, as well as the two-volume
Festschrift for Prof. Steinkellner entitled Pramāṇakīrti.15 From the very beginning of Prof.
Steinkellner’s endeavor to open the door to the treasures of Buddhist Sanskrit manuscripts
found in Buddhist monasteries of the Tibetan Autonomous Region, Dr. Krasser helped him,
later succeeding him as the Viennese representative for the cooperation with the China
Tibetology Research Center in Beijing. In that connection, together with Prof. Steinkellner
and Dr. Horst Lasic, he published diplomatic and critical editions of the first two chapters
of Jinendrabuddhi’s Ṭīkā on Dignāga’s Pramāṇasamuccayavṛtti.
Dr. Krasser also published many academic papers on Buddhist epistemology and logic
and related areas. I cannot summarize all that he did in the short period of time of his
active years. Instead I would like to refer to two fundamental hypotheses he left for us: the
15
of the fourth were edited jointly by Helmut Krasser, Eli Franco, Horst Lasic and Birgit Kellner, Religion
and Logic in Buddhist Philosophical Analysis. Proceedings of the Fourth International Dharmakīrti
Conference, Vienna, August 23-27, 2005. Vienna 2011.
Pramāṇakīrti, edited by Birgit Kellner, Helmut Krasser, Horst Lasic, Michael Torsten Wieser-Much
and Helmut Tauscher. Vienna 2007.
xix
dates of Dharmakīrti and the real nature of the Buddhist philosophical treatises attributed
to individual authors.
Regarding the dates of Dharmakīrti, Prof. Frauwallner’s proposal of 600–660 C.E. had
been widely accepted by modern scholars of Dharmakīrti.16 However, Dr. Krasser was
courageous enough to challenge the authority of Prof. Frauwallner, proposing a much
earlier date for Dharmakīrti, chiefly on the grounds that Bhāviveka’s proof of non-eternity
by sattvānumāna was influenced by Dharmakīrti. Dr. Vincent Eltschinger has summarized
the state of affairs in his most recent book as follows:
Kumārila, Dharmakīrti and Candrakīrti have long been considered, ever since
Frauwallner’s influential “Landmarks in the History of Indian Logic” (1961),
roughly contemporary philosophers belonging to the first half of the seventh
century CE. … According to Krasser, however, Bhāviveka, who can be assigned with a fair amount of certainty to 500–570, presupposes both Kumārila
and Dharmakīrti. As a working hypothesis, Krasser proposes “the time of
activity of Kumārila and Dharmakīrti to be the middle of the sixth century.”
Hypothetical (and unpopular) as it may be, Krasser’s chronology relies in
my opinion on much stronger arguments than Frauwallner’s argumentum a
silentio.17
Last week I attended the XVIIth conference of International Association of Buddhist
Studies in Vienna, where I had an opportunity to hear a paper given by Dr. Toshikazu
Watanabe titled “Buddhist Critiques of the Sāṅkhya Theory of Causality, Dharmakīrti and
his Predecessors.” In that paper, referring to Dharmapāla’s commentary on the tenth chapter
of Āryadeva’s Catuḥśataka, he demonstrated that Dharmapāla influenced Dharmakīrti
with regard to the critique of the Sāṅkhya theory of causality. He also demonstrated
that Bhāviveka’s critique of the Sāṅkhya’s logical reason moves along the same lines as
Dharmapāla’s. In his concluding remarks, Dr. Watanabe argued that Dharmapāla and
Bhāviveka must have lived during almost the same period, i.e., the sixth century, and stated
that although he does not accept Dr. Krasser’s proposal that Bhāviveka was influenced by
Dharmakīrti, he sees a close relationship between these two Buddhist philosophers. He thus
concluded that he is inclined to accept Dr. Krasser’s working hypothesis that Dharmakīrti’s
dates could be pushed back into the middle of the sixth century, proposing the dates of
560/570–650.
Also last week in Vienna, Prof. Shinya Moriyama read a paper called “On dharmisvarūpaviparītasādhana,” in which he pointed out the resemblance between Dharmapāla’s idea
of three types of reasons and Dharmakīrti’s idea of three types of śabdārtha in Pramāṇavārttika I.205.18 This gives supporting evidence for Dr. Watanabe’s argument that Dharmapāla
16
17
18
As far as I know, there are at least two exceptions, namely, Christian Lindtner, who proposed 530–600
(“Apropos Dharmakīrti – Two New Works and a New Date,” Acta Orientalia Kobenhavni 41 (1980)
27–37) and Toshihiko Kimura, who proposed 550–620 (“A New Chronology of Dharmakīrti,” Dharmakīrti’s Thought and Its Impact on Indian and Tibetan Philosophy: Proceedings of the Third International Dharmakīrti Conference, Hiroshima, November 4-6, 1997, ed. by Shoryu Katsura. Vienna 1999,
209–214).
Vincent Eltschinger, Buddhist Epistemology as Apologetics: Studies on the History, Self-Understanding
and Dogmatic Foundations of Late Indian Buddhist Philosophy. Vienna 2014, 116, n. 80.
In the meantime published in Wiener Zeitschrift für die Kunde Südasiens LVI–LVII (2015–2018) 37–49.
xx
OPENING SPEECH
influenced Dharmakīrti. Of course, there are other possibilities, such as Dharmakīrti having
influenced Dharmapāla or both having gotten a similar idea from a common source. But
in any case, it is safe to say that Dharmapāla and Dharmakīrti lived at approximately the
same time, as Dr. Watanabe concluded.
Again during last week’s conference, Prof. Eli Franco re-examined and rejected the
earlier date of Dharmakīrti proposed by Dr. Krasser.19 One of his arguments is based on
the silence of Xuanzang, who was in India from 625 to 645, and that of other Indian and
Buddhist philosophers with regard to Dharmakīrti. Against such an argument of silence,
Prof. Steinkellner rightly pointed out that there are other ways to explain this silence,
referring to the well-known legend of Dharmakīrti’s unpopularity with his contemporaries.20
As a student of logic, I do not endorse much power to reasoning based on silence.
Silence does not prove anything; it only raises doubts. In this connection I would like
to refer to my own article “On trairūpya formulae.”21 In that article I demonstrated the
gradual development of the trairūpya formulae in Buddhist logical texts and suggested that
Xuanzang was certainly acquainted with the restriction by the particle eva in the formulation
of the first condition of the valid logical reason (pakṣadharmatā), which was missing in
Dignāga’s formulation of trairūpya but appears in Dharmakīrti’s. It is well known that
the sixth-century Naiyāyika, Uddyotakara, criticized Dignāga’s understanding of the first
condition and Dharmakīrti tried to respond to his criticism by adding the eva-restriction
to the first condition. It is possible that such an eva-restriction was proposed by some
unknown Buddhist logician before Dharmakīrti. But considering Dharmakīrti’s position
in the development of Buddhist logic, I am inclined to believe that it was Dharmakīrti
who initiated this revision in the trairūpya formulae. Therefore, although Xuanzang does
not mention the name of Dharmakīrti, he may well have been acquainted with one of the
important revisions made by Dharmakīrti in Dignāga’s logic. And if this is the case, I
believe that it is possible to refute Prof. Frauwallner’s and Prof. Franco’s argument regarding
Xuanzang’s silence about Dharmakīrti.
Of course, while it is impossible to prove Krasser’s hypothesis of the earlier dates of
Dharmakīrti, it cannot be easily dismissed either. As Eltschinger and Watanabe have both
conjectured, it is quite possible that Dharmakīrti was active in the latter half of the sixth
century, and indeed, perhaps he enjoyed little popularity among his colleagues.
Regarding the second topic left for us by Dr. Krasser, I would like to point out that in
Vasudhararakṣita’s Tibetan translation of the Pramāṇsasamuccayavṛtti on Pramāṇasamuccaya 3.21, there are instructions on how to draw a chart of a hetucakra, which clearly
indicates that at least this portion of PSV is a note recorded by a student during a class given
by, if not Dignāga, some teacher of Buddhist logic lecturing on the Pramāṇasamuccaya.
Moreover, I have recently been reading Avalokitavrata’s Ṭīkā on Bhāviveka’s Prajñāpradīpa, and from time to time I have noticed that Avalokitavrata meticulously points out what
19
20
21
In the meantime published as “Xuanzang’s Silence and Dharmakīrti’s Dates” in Wiener Zeitschrift für
die Kunde Südasiens LVI–LVII (2015–2018) 117–141.
For more about the debate on the dates of Dharmakīrti, please see Dr. Elisa Freschi’s blog: Thinking
about through Sanskrit (and) philosophy, http://elisafreschi.com/2014/08/26/third-day-at-the-iabs-fran
co-on-the-datation-of-dharmakirti-and-some-further-thoughts-on-dharmakirti-dignaga-kumarila/.
Buddhism and Its Relation to Other Religion: Essays in Honour of Dr. Shozen Kumoi on his Seventieth
Birthday. Kyōto 1985, 161–172.
xxi
is the pakṣa, what is the sādhyadharma and what is the hetu of the relevant prayoga. This
also seems in part to support Krasser’s characterization of Buddhist philosophical texts as
being students’ notes taken during monastic lessons. There must be many parallel cases
like this. I would also like to add that among Japanese Buddhists, a tradition existed of
compiling and publishing writings known as kōroku (講録), which are students’ notes of
their teachers’ lectures on certain Buddhist texts or doctrines. From this perspective, too, I
believe that Krasser’s conjecture must be taken more seriously and that we should continue
working on this idea.
As a Japanese Buddhist of the Jōdo-shinshū tradition, I believe that Dr. Krasser, though
invisible, is somewhere among us, having ascended to the Sukhāvatī, Land of the Buddha
Amitābha, and returned from there as a Bodhisattva to watch over and help us. Having
heard what I just said, he would probably say, “Don’t take me too seriously. There are other
nice things to do in the world, like drinking and smoking.”
Before I close my opening speech, I should refer to some of the important achievements
that have been made since the last Dharmakīrti conference. In the opening speech of the
last conference, Prof. Steinkellner presented “News from the manuscript department.”
I would like to report now on some of the further developments in this regard, as far
as I know of them. As I mentioned above, the first two chapters of Jinendrabuddhi’s
Pramāṇasamuccayaṭīkā have been published. I am responsible for editing the third and
the fourth chapter of the same text; I am happy to report that I have finished working
on the third chapter and am now in the middle of the fourth. Regarding the fifth chapter,
as we all know, Mr. Ole Pind finally submitted his work on that apoha chapter to the
University of Vienna as his dissertation and it is now available on line to benefit of us
all;22 since Dr. Krasser is gone, Dr. Lasic and Dr. McAllister have taken over the task
of editing the fifth chapter. Finally, regarding the sixth chapter, Prof. Motoi Ono and his
colleagues have more or less worked out the whole chapter. Meanwhile, Prof. Steinkellner
has published the critical edition of the first two chapters of the Pramāṇaviniścaya and
Dr. Pascale Hugon has published that of the third chapter. Dr. Toshikazu Watanabe and Dr.
Krasser’s students Drs. Masamichi Sakai and Hisataka Ishida have critically edited several
portions of Dharmottara’s Pramāṇaviniścayaṭīkā. And Prof. Steinkellner is now polishing
up the critical edition of the Hetubindu that was prepared by Dr. Krasser.23
Originally I intended to mention some important recent contributions to our field; many
of them come from you. But I decided not to do so because after all, my information is
limited and my impressions of those publications may be biased. So this is the end of my
opening speech. I hope you will all enjoy the forthcoming papers and presentations, and
at the end, I hope we shall have a little better understanding of this marvelous Buddhist
philosopher Dharmakīrti from many different angles. Thank you for your patience.
Heidelberg, 26 August 2014
22
23
In the meantime published in revised form as Dignāga’s Philosophy of Language. Vienna 2016.
In the meantime published as Dharmakīrti’s Hetubindu. Beijing/Vienna 2016.
Account of the Fifth International Dharmakīrti Conference in
Heidelberg, August 26 to 30, 2014
Honorary Chairs
• Shōryū Katsura (Ryūkoku University)
• Ernst Steinkellner (Austrian Academy of Sciences)
Planning Committee
•
•
•
•
Kei Kataoka (Kyūshū University)
Birgit Kellner (University of Heidelberg)
Sara McClintock (Emory University)
Hideyo Ogawa (Hiroshima University)
Organizing Team
• Birgit Kellner, Patrick McAllister, Ina Chebbi (now Buchholz)
xxiv
ACCOUNT OF THE FIFTH INTERNATIONAL DHARMAKĪRTI CONFERENCE
Program
Monday, 25 August: Arrival Day
18:00-20:00 Registration
Tuesday, 26 August
9:00-10:00 Registration
10:00-10:30 Opening Address: Shōryū Katsura
10:30-11:00 Break
Session I chaired by Shōryū Katsura
11:00-11:30 Ernst Steinkellner: Closing a gap in the interpretation of Dharmakīrti’s
logic
11:30-12:00 Eli Franco: The determination of causation and the similarity between
cause and effect
12:00-12:30 Horst Lasic: What is the effect of what and how can one determine it?
12:30-14:30 Lunch Break
Session II chaired by Kei Kataoka
14:30-15:00 Hideyo Ogawa: Dignāga on the view of a generic term as denoting a
relation
15:00-15:30 Akane Saitō: Maṇḍanamiśra’s arguments against Dharmakīrti’s ideas
on language – different definitions of the convention
15:30-16:00 Break
Session III chaired by Hideyo Ogawa
16:00-16:30 Mai Miyo: Dharmottara and Prajñākaragupta on the non-distinction
between pramāṇa and pramāṇaphala
16:30-17:00 Hiroshi Nemoto: Dharmakīrti’s notion of permanence and its impact on
the Tibetan Buddhist doctrine of buddhahood
17:00-17:30 Artur Przybyslawski: The notion of valid cognition (tshad ma) in the
karma bka’ brgyud tradition of Tibetan buddhism
17:30 Birgit Kellner: Buddhist philosophy at the University of Heidelberg – a short
history
18:00 Get-together party
Wednesday, 27 August
Session IV chaired by Sara McClintock
09:00-09:30 Patrick McAllister: Form and content in Ratnakīrti’s writings
PROGRAM
xxv
09:30-10:00 Sara Uckelman: The legitimacy of inference. Argumentation strategy in
Ratnakīrti’s Īśvarasādhanadūṣaṇam
10:00-10:30 Ryō Sasaki: The significance of the Vādanyāya in the historical transition
of the ‘debate’ concept
10:30-11:00 Break
Session V chaired by Pascale Hugon
11:00-11:30 Kyeongjin Choi: The indeterminate role of bādhakapramāṇa in the
Pramāṇaviniścaya
11:30-12:00 Kyō Kanō: On viparyayabādhakapramāṇa
12:00-12:30 Masamichi Sakai: Dharmottara on the viparyaye bādhakapramāṇa and
trairūpya in Dharmakīrti’s sattvānumāna
12:30-14:30 Lunch Break
Session VI chaired by Zhihua Yao
14:30-15:00 Christian Coseru: Consciousness and causal explanation. Śāntarakṣita
against physicalism
15:00-15:30 Zhihua Yao: Non-activity (nirvyāpāra) in Dignāga and Dharmakīrti
15:30-16:00 Break
Session VII chaired by Piotr Balcerowicz
16:00-16:30 Masaki Tamura: The truth, the Buddha’s words, and inference. Bhāviveka’s theory of two truths
16:30-17:00 Motoi Ono: On pramāṇabhūta – the change of its meaning from Dignāga
to Prajñākaragupta
17:00-17:30 Hisataka Ishida: The exclusion of superimposition (samāropavyavaccheda)
17:30 Presentations and discussion of databases and digital resources
Thursday, 28 August
Session VIII chaired by Leonard van der Kuijp
09:00-09:30 Masahiro Inami: Two kinds of causal capacity, sāmānyā śaktiḥ and pratiniyatā śaktiḥ
09:30-10:00 Hisayasu Kobayashi: Dharmottara and Prajñākaragupta on svalakṣaṇa
10:00-10:30 Yōhei Kawajiri: The Pratyabhijñā school’s criticism of the Buddhist
concept of svalakṣaṇa
10:30-11:00 Break
xxvi
ACCOUNT OF THE FIFTH INTERNATIONAL DHARMAKĪRTI CONFERENCE
Session IX chaired by Kei Kataoka
11:00-11:30 Toshikazu Watanabe: Dharmakīrti and his successors on āśrayāsiddha
and prasaṅgaviparyaya
11:30-12:00 Pascale Hugon: Revisiting the prasaṅga-passage of the Pramāṇaviniścaya
12:00-12:30 Lawrence McCrea: Balancing the scales. Dharmakīrti inside and out
12:30-14:30 Lunch Break
Excursion / free afternoon, followed by joint dinner
Friday, 29 August
Session X chaired by Ernst Steinkellner
09:00-09:30 John Taber: The structure of Dharmakīrti’s sahopalambhaniyama argument
09:30-10:00 Serena Saccone: Śubhagupta’s theory of reality in the *Bāhyārthasiddhikārikā
10:00-10:30 Marie-Hélène Gorisse: Jain conceptions of non-apprehension. A criticism
of Dharmakīrti’s theory of inference
10:30-11:00 Break
Session XI chaired by Eli Franco
11:00-11:30 Yoichi Fukuda: Reinterpretation of the compound “svabhāva-pratibandha” in Dharmakīrti’s logical theory
11:30-12:00 Kiyotaka Yoshimizu: Another look at avinābhāva and niyama in Kumārila’s exegetic works
12:00-12:30 Kiyokuni Shiga: On the meaning of bāhyārtha in Dignāga’s and Jinendrabuddhi’s theories of inference
12:30-14:30 Lunch Break
Session XII chaired by Horst Lasic
14:30-15:00 Laura Guerrero: Pramāṇa as conventional truth in the work of Dharmakīrti
15:00-15:30 Kei Kataoka: How does one cognize a cow? A dialogue between Mādhava
and Dignāga
15:30-16:00 Break
Session XIII chaired by Birgit Kellner
16:00-16:30 Vincent Eltschinger: Buddhist epistemology as apologetics
16:30-17:00 Cristina Pecchia: Dharmakīrti on the role of rituals
17:00-17:30 Som Dev Vasudeva: Bhoktṛtva and causation in Kṣemarāja’s response to
Dharmakīrti’s critique of liberation through initiation
17:30 Presentations and discussion of databases and digital resources
PROGRAM
xxvii
Saturday, 30 August
Session XIV chaired by John Taber
09:00-09:30 Junjie Chu: Jitāri’s Vijñaptimātratāsiddhi
09:30-10:00 Hugo David: Maṇḍana Miśra on omniscience (sarvajñatva) and the
perception of yogins (yogipratyakṣa), on the early appropriation of a few Buddhist
concepts in the Mīmāṃsā tradition
10:00-10:30 Brendan Gillon: Ṣaṭkoṭivāda in the Upāyahṛdaya
10:30-11:00 Break
Session XV chaired by Brendan Gillon
11:00-11:30 Huanhuan He / Leonard van der Kuijp: Turning the Wheels: Yet another
look at the *Hetucakra[-ḍamaru]
11:30-12:00 Mingjun Tang: The concept of sādhana in Chinese Buddhist logic
12:00-12:30 Kensho Okada: A way of communication between a speaker and a listener
similar to the way in which two persons with eye disease equally see double moon — in
the light of the apoha theory of Śāntarakṣita and Kamalaśīla
12:30-14:30 Lunch Break
Session XVI chaired by Lawrence McCrea
14:30-15:00 Miyuki Nakasuka: Dharmakīrti’s apoha theory, perceptual judgement,
and lack of superimposition (samāropaviveka)
15:00-15:30 Piotr Balcerowicz: Dharmakīrti and Samantabhadra
15:30-16:00 Break
xxviii ACCOUNT OF THE FIFTH INTERNATIONAL DHARMAKĪRTI CONFERENCE
Participants of the Fifth International Dharmakīrti
Conference
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
Balcerowicz, Piotr (University of Warsaw)
Choi, Kyeongjin (University of Tokyo)
Chu, Junjie (Leipzig University)
Coseru, Christian (College of Charleston)
David, Hugo (University of Cambridge)
Eltschinger, Vincent (Austrian Academy of Sciences)
Franco, Eli (Leipzig University)
Fukuda, Yoichi (Otani University)
Gillon, Brendan (McGill University)
Gorisse, Marie-Hélène (Ghent University)
Guerrero, Laura (Utah Valley University)
Hayashi, Itsuki (Ryukoko University)
He, Huanhuan (Chinese Academy of Social Sciences)
Hugon, Pascale (Austrian Academy of Sciences)
Inami, Masahiro (Tokyo Gakugei University)
Ishida, Hisataka (Tokyo University)
Kano, Kyo (Kobe Women’s University)
Kataoka, Kei (Kyushu University)
Katsura, Shoryu (Ryukoko University)
Kawajiri, Yohei (Chikushi Jogakuen University)
Kellner, Birgit (Heidelberg University)
Kobayashi, Hisayasu (Chikushi Jogakuen University)
Kwon, Soonbeom (Dongguk University)
Lasic, Horst (Austrian Academy of Sciences)
Li, Shenghai (Harvard University)
Lo, King Chung (University Leipzig)
Luo, Hong (Peking University)
McAllister, Patrick (Heidelberg University)
McClintock, Sara (Emory University)
McCrea, Lawrence (Cornell University)
McNamara, Daniel (Emory University)
Miyo, Mai (Waseda University)
Nakasuka, Miyuki (Hiroshima University)
Nemoto, Hiroshi (Hiroshima University)
Ogawa, Hideyo (Hiroshima University)
Okada, Kensho (University of Tsukuba)
O’Leary, Joseph (Sophia University, Tokyo)
Ono, Motoi (University of Tsukuba)
Pecchia, Cristina (Austrian Academy of Sciences)
Prueitt, Catherine (Emory University)
Przbyslawski, Artur (Jagiellonian University)
PARTICIPANTS
42.
43.
44.
45.
46.
47.
48.
49.
50.
51.
52.
53.
54.
55.
56.
57.
58.
59.
60.
Saccone, Margherita Serena (University of Naples)
Saito, Akane (Kyoto University)
Sakai, Masamichi (Kansai University)
Sasaki, Ryo (Waseda University)
Shiga, Kiyokuni (Kyoto Sangyo University)
Siderits, Mark (Illinois State University)
Steinkellner, Ernst (Austrian Academy of Sciences)
Stoltz, Jonathan (University of St. Thomas)
Sung, Chungwhan (Dongguk University)
Taber, John (University of New Mexico)
Tamura, Masaki (Hiroshima University)
Tang, Mingjun (Shanghai Academy of Social Sciences)
Uckelman, Sara (Heidelberg University)
van der Kuijp, Leonard (Harvard University)
Vasudeva, Somadeva (Kyoto University)
Watanabe, Toshikazu (Austrian Academy of Sciences)
Woo, Jeson (Dongguk Univeristy)
Yao, Zhihua (Chinese University of Hong Kong)
Yoshimizu, Kiyotaka (Tohoku University)
xxix
Jitāri’s Vijñaptimātratāsiddhi1
by
Junjie Chu
Introduction
As we have reported elsewhere,2 the newly available manuscripts of Jitāri’s (fl. 940–980)3
contain a number of hitherto unknown works, one of them is the Vijñaptimātratāsiddhi.
This is a short philosophical treatise that become accessible through two manuscripts of
the Sanskrit text.4 In this treatise, Jitāri tries to establish the Yogācāra doctrine of the
cognitive manifestation of the objects and refutes the “externalism” (bahirarthavāda)
around this central topic.5 The basic ideas he presents can be found in Dharmakīrti’s works
and commentaries on them, for instance, PV 3.320–337 and the commentaries ad loc.
At the beginning of the treatise, Jitāri divides the externalists into two groups, according
to their opinions on whether awareness possesses the image of object or not, i.e., the
sākārajñānavādins and the nirākārajñānavādins, with a short outline of the position held by
each group. Jitāri’s refutation, however, focuses mainly on the nirākārajñānavādin-branch.
The discussion is initiated with a formal reasoning (prayoga) aiming to prove his main
thesis that all things that become manifest (pra√kāś) – i.e., that appear in consciousness –
are cognition, which has the nature of self-manifestation. This is followed by the refutation
of various objections that claim the reason in Jitāri’s prayoga is invalid in one of the
three usually recognized ways, that is, by being unestablished (asiddha), inconclusive
(anaikāntika) or contradictory (viruddha).
1
2
3
4
5
I would like to express my sincere thanks to Prof. Ernst Steinkellner for his effort to make it possible
for me to read the manuscripts in Beijing, and to Prof. Eli Franco for his establishing a research project
on Jitāri’s works in the University of Leipzig supported by the German Research Foundation. I am also
indebted to Prof. Franco for many valuable suggestions on an earlier version of this paper. My special
thanks go to Prof. Birgit Kellner who carefully read this paper and made numerous suggestions which
allow for great improvements in both content and language. I am also very grateful to the German
Research Council (DFG) for a substantial grant for the abovementioned research project.
Cf. Chu and Franco 2012.
Cf. Tucci [1930] 1971: 249. Shirasaki (1981: 342) estimates Jitāri’s dates as 960–1040.
In manuscript A the VMS ranges from folios 14b4 to 20ab, in manuscript B from 49a1 to 55b2. A critical
edition of the text based on the two manuscripts is being prepared by the present author and will be
published separately. All quotations of VMS in this paper are based on this forthcoming critical edition.
In the following, two sets of folio-numbers and the line-number separated by forward slash (for instance,
“14b5–6/49a1–2”) refer to manuscript A and manuscript B respectively; however, editorial notes from
the critical apparatus, including the reports of the variant readings in one of the two manuscripts,
are omitted here. I will not describe the physical condition and the philological character of these
manuscripts here, which will be made in the critical edition. For the detailed information of the two
manuscripts of the works attributed to Jitāri, as a whole, cf. Chu and Franco 2012.
I prefer to use the terms “externalism/externalist” (bahirarthavāda/bahirarthavādin) instead of “realism/
realist” to refer to the system/person which/who asserts that object of cognition exists outside of or
independent of cognition, because these terms can easily remind us of its Sanskrit equivalent.
Birgit Kellner et al., eds., Reverberations of Dharmakīrti’s Philosophy: Proceedings of the Fifth International Dharmakīrti
Conference Heidelberg, August 26 to 30, 2014. Vienna 2020, pp. 1–19.
2
Jitāri’s Vijñaptimātratāsiddhi
In the context of countering the attack that the reason is “inconclusive,” Jitāri embarks
on an excursus and refutes the opponent’s thesis that the manifestation of the external object
in cognition is caused by cognition. He does so by negating all four logically possible
propositions with regard to the relationship between manifestation and the object. Jitāri
concludes that the manifestation of the object is merely the cognition’s self-manifestation.
After establishing that the reason is not inconclusive, in the final section, Jitāri replies
to further objections against the thesis of cognition’s self-manifestation raised from the
perspective of non-Buddhist as well as Buddhist systems.
In the following, I will present an analysis of Jitāri’s discussions with the aim of
summarizing the main points.
1. The bahirarthavāda position with regard to the image of the object
After dividing the bahirarthavādins into sākārajñānavādins and nirākārajñānavādins at
the beginning of the VMS, Jitāri describes the position of those bahirarthavādins who
favor sākārajñānavāda with regard to the image of the object as follows:
All this is only the cognition which appears with the image of [object-things
of the five senses respectively, such as] white color, singing voice, fragrant
smell, sweet taste and [tangible] roughness; however, the external (bāhyaḥ)
object-thing having a [physical] body established as entirely different from
[its] product, the cognition, does not appear in its own form.6
The position described here is similar to that of the Sautrāntika. As it is well-known, the
Sautrāntika is a strong proponent of sākārajñānavāda.7 Jitāri then summarizes the position
of those bahirarthavādins who favor nirākārajñānavāda as follows:
Our cognition is devoid of image [of object]; the external object possesses
the image, and is perceptible, for this object is apprehended by perception as
connected to the external world (bahirdeśa).8
6
7
8
VMS 14b5–6/49a1–2: jñānam evedaṃ sitagītasurabhimadhurakarkaśākāram bhāsate, bāhyaḥ* punar
artho jñānakāryavyatirekamātravyavasthāpitaśarīro na svena rūpeṇa cakāsti. *Both manuscripts
clearly read bāhyam. However, since bāhyaḥ appears in related passages quoted below in n. 7 and n. 9,
the text should be emended to bāhyaḥ.
Cf. a frequently quoted statement clarifying the Sautrāntika position, as e.g. in TBh 63,17–18: sautrāntikānāṃ matam — jñānam evedaṃ sarvaṃ nīlādyākāreṇa pratibhāsate, na bāhyo ’rthaḥ, jaḍasya
prakāśāyogāt. Cf. Kajiyama 1965: 428ff.; Kajiyama 1998: 139–140; Dhammajoti 2007: 171ff., 174ff.;
Dhammajoti 2009: 241ff., 269ff., and 274ff.
VMS 14b6–15a1/49a2–3: nirākārā no buddhiḥ, ākāravān bāhyo ’rthaḥ, pratyakṣaś ca, sa hi bahirdeśasambaddhaḥ pratyakṣeṇopalabhyate. In the TSP, we can find a similar outline of the nirākāravāda:
“Cognition is devoid of the image; however, the external thing has the image, and it is perceived distinctly
as connected to the external world. In this manner, the image of cognition is rejected by us.” (TSP
313,4–6 ad TS 980: nirākārā buddhiḥ, ākāravān bāhyo ’rthaḥ, sa ca bahirdeśasambanddho vispaṣṭam
upalabhyata ity evam asmābhir jñānākāro niṣiddhaḥ.)
Junjie Chu
3
This refers evidently to the Mīmāṃsā system,9 which is the major advocate of the nirākāravāda among the non-Buddhist systems, and probably the main opponent functioning as
interlocutor in this treatise. On the other hand, among Buddhist systems, the Vaibhāṣika
is usually regarded as nirākāravādin, with Śubhagupta as the main representative in the
sources available to us. As we will see below, Śubhagupta is quoted and refuted by Jitāri. The
Yogācāra system, as it is attested in many sources, is divided into two sub-systems, the sākāravijñaptimātratāvādin and nirākāravijñaptimātratāvādin. Ratnākaraśānti is a spokesman
of the latter.10 But, needless to say, as a nirākāravijñaptimātratāvādin, Ratnākaraśānti could
not have been the target of Jitāri’s refutation, since he is not a bāhyārthavādin.
2. Formal reasoning proving cognition’s self-manifestation
In the VMS, Jitāri indeed does not pay much attention to the sākārajñānavāda-branch
of the bahirarthavāda, saying that the sākārajñānavādins are not in conformity with the
whole set of the generally established convention and they do not provide anything except
some pieces of false determination.11 So, he concentrates himself on the refutation of the
nirākārajñānavāda-branch of the bahirarthavāda. He starts his refutation by setting forth a
formal reasoning:
What becomes manifest [in cognition] is cognition [itself], just as the conceptual construction of a blue thing; and [a sensory object] like visible matter
becomes manifest [in cognition, therefore, it is cognition with the image of
object]. This is a reason of essential property.12
The main body of the treatise is actually a proof that the reason used in this reasoning
is valid. As Jitāri does in many other works, the proof takes on the form of replies to
various objections that claim the reason to be fallacious according to the Dharmakīrtian
threefold typology of the “pseudo-reason” (hetvābhāsa): (1) unestablished (asiddha); (2)
contradictory (viruddha), and (3) inconclusive (anaikāntika).13
At first, Jitāri declares that, in the formal reasoning, the proving factor (sādhana), i.e.,
“becoming manifest,” is not unestablished, because it is proved by perception as existent in
9
10
11
12
13
Cf. MSBh 28,17–18: nirākārā tu no buddhiḥ, ākāravān bāhyo ’rthaḥ, sa hi bahirdeśasambaddhaḥ
pratyakṣam upalabhyate. Cf. also TSP 101,14–15 ad TS 252: kiṃ ca bhavato mīmāṃsakasya mate yo
bhāsamānaḥ sa ākāro na buddheḥ. kiṃ tv asau bāhyārthasvabhāvo varṇyate, ākāravān bāhyo ’rtho
nirākārā buddhir iti vacanāt. “Further, in your Mīmāṃsaka system, it is explained that it is not the
image of cognition that appears; rather, it is the nature of the external object, since it is said: ‘the external
object possesses the image, the cognition is devoid of the image.’”
Cf. Kajiyama 1965: 421ff.
Cf. VMS 15a1/49a3: parisamāpitasakalavyavahārayogābhāvān mithyābhiniveśaleśād ṛte nātiśerata iti.
VMS 15a2–3/49a4–5: yat prakāśate tat jñānaṃ yathā nīlavikalpaḥ, prakāśate cedaṃ rūpādikam iti
svabhāvahetuḥ.
Dharmakīrti does not directly offer a separate definition of the pseudo-reason (hetvābhāsa). Rather, in
PV 1.1 he defines the threefold valid reason and then adds at the end of the verse that reasons other than
this are “pseudo-reasons” (pakṣadharmas tadaṃśena vyāpto hetus tridhaiva saḥ / avinābhāvaniyamād
dhetvābhāsās tato ’pare //). The verse is translated in Mookerjee and Nagasaki 1964: 6; Hayes and
Gillon 1991: 2f.; and Steinkellner 2008 [2009]: 195, 2013: 4; for further discussion see Frauwallner
1954: 145; Steinkellner 1967: 82f., 2008 [2009]: 195, 2013: 16; and PVin 3.91–131,5 (kārikā 67–68).
4
Jitāri’s Vijñaptimātratāsiddhi
respect to the property-bearer (dharmin), i.e., a “[sensory object] like visible matter.”14
That means, for the reason he uses, the condition of pakṣadharmatā is satisfied. In reply to
the objection that the property-bearer is unestablished either for the internalist who regards
it as internal, or for the externalist who regards it as external, Jitāri argues that the propertybearer, visible matter, is nothing but the exclusion of non-visible-matter. Its properties of
being internal or external that are the subject of disagreement, are characterized with two
further different exclusions (i.e., the exclusion of being internal and the exclusion of being
external, respectively). However, this disagreement does lead to the non-establishment of
the property-bearer, because this property-bearer is established by perception; and in no
inference is the property-bearer regarded as particularized to the certain property that is
subject to controversy, so that it would incur the criticism of non-establishment.15
Then, Jitāri points out further that the reason, “becoming manifest,” is not contradictory,
because it is found among the similar cases. He explains: A proving factor becomes
contradictory when it is pervaded by what is opposed to the factor to be proved (sādhya);16
and a pervaded factor (vyāpya) should never come forth without pervading factor (vyāpaka),
otherwise it would not be pervaded (tadabhāvaprasaṅgāt). So, the reason, insofar as it is
contradictory, does not occur in the similar cases.17
But the opponent does not agree with this, arguing that the proving factor “becoming
manifest” cannot be found in the similar instance, because cognition is not perceptible;
for, when an object is cognized, the cognition is known through inference.18 Here Jitāri
obviously refers to the position of Śabara of the Mīmāṃsā system; in fact, he makes use of
the Śābarabhāṣya verbatim, incorporating literally the passage in question.19 Dharmakīrti
refutes this opinion in PV 3.460ff.20 From his externalist’s point of view, Jitāri’s opponent
says that, in a cognitive event, we perceive merely the object alone that is the externally
existent thing like something blue, which becomes manifest in cognition as blue when
we perceive it. Apart from this no other object can be perceived, for the grasping subject
cannot be perceived in the way “I experience the internal cognition” (āntaraṃ jñānam
14
15
16
17
18
19
20
Cf. VMS 15a3/49a5: atra prayoge na sādhanāsiddhir adhyakṣasiddhatvād dharmiṇi prakāśamānatāyāḥ.
Cf. VMS 15a5–6/49b2–4: ayam evārūpādivyāvṛttiviśiṣṭo dharmī , bāhyatvābāhyatve tasya vivādāspade
vyāvṛttyantare, na ca tayor asiddhiṃ dharmiṇo ākarṣati, tasyādhyakṣasiddhatvenāśakyāpahnavatvāt,
na ca kvacid anumāne vimatyadhikaraṇadharmaviśiṣṭo dṛṣṭo dharmī yenaivam asiddhicodanālāmbī
syāt.
A similar definition of the contradictory reason can be found in RNĀ 33,21–22: tathā hi yo vipakṣa eva
vartate sa khalu sādhyaviparyayavyāpteḥ sādhyaviruddhaṃ sādhayan viruddho ’bhidhīyate.
Cf. VMS 15b1–2/49b4–5: nāpi viruddhatvaṃ sapakṣe bhāvāt. sādhyaviparyayavyāptaṃ hi sādhanaṃ
viruddhaṃ bhavati, na ca vyāpyaṃ vyāpakam antareṇa syāt tadabhāvaprasaṅgād iti sati viruddhatve
na sapakṣe vartate.
Cf. VMS 15b2/49b5–50a1: nanu ca prakāśanaṃ nāma nāsty eva sapakṣe jñānasya sarvasya parokṣatvāt.
na hi kaścid ajñāte ’rthe buddhim upalabhate, jñāte tv anumānād avagacchati.
Cf. MSBh 7,25–29: na hi kaścid ajñāte ’rthe buddhim upalabhate, jñāte tv anumānād avagacchati.
Cf. PV 3.460ab: api cādhyakṣatābhāve dhiyaḥ syāl liṅgato gatiḥ / “Further, cognition should be
known through an inferential sign, insofar as it is not perceptible.” Cf. PV 3.447: etenānātmavitpakṣe
sarvārthādarśanena ye / apratyakṣāṃ dhiyaṃ prāhus te ’pi nirvarṇitottarāḥ // “Through [the unwanted
consequence explained above] that there is no perception of any kind of object in the thesis of nonself-awareness the reply is given to those who say that cognition is non-perceptible.” The opponent
who advocates this theory is at PVV 251,17 labelled as Jaiminīya: ye jaiminīyā apratyakṣāṃ dhiyam
arthāpattigamyām āhuḥ te ’pi nirvarṇitottarā dattottarā boddhavyāḥ.
Junjie Chu
5
anubhavāmīti). That is to say, the opponent does not accept the Buddhist theory that
cognition of an object can be perceived through self-awareness, like the awareness of
sensations such as pleasure; as it is explained by Dignāga.21 Jitāri’s refutation of this
objection is in fact a proof of the existence of the experience of cognition of an objectreferent. He argues that even if you do not have an ascertaining awareness (pratipatti) in the
form “I experienced the cognition,” you cannot prove that the experience of the cognition
does not exist. To support this, he quotes a verse from the Tattvasaṅgraha: The name is
not equal to the characteristics (rūpa) of things, so it is not the case that the things whose
characteristics have been completely apprehended (parijñāta) [through perception] remain
unknown so long their name is unknown.22
Jitāri argues further: If the experience of the cognition of an object-referent would not
exist at all, then the pure conceptual awareness of an object like a pot, perceived as arising
and disappearing, cannot be included in (antar√bhū) or assigned to anything, i.e., it has no
objective substratum at all to be based on; for it can be based neither on the object nor on
the cognition – because the object cannot arise and disappear according to a person’s desire
(īhāvaśena) and it does not have the nature of the subjective conceptualization. Moreover,
the cognition might occur, being imperceptible according to the opponent, but it could not
become manifest, even if it occurs as conceptual awareness; the object is manifest, but it
does not occur like the subjective cognition. To conclude, Jitāri says, if the cognition were
imperceptible, there would be no conceptual awareness consisting in the experience of a
cognition in concordance with the verbal designation of the object at all, but in reality it is
not so.23
Now, Jitāri turns to treat the problem of the reason’s inconclusiveness (anaikāntikatā),
which he defines as the occurrence of reason in the dissimilar cases (vipakṣavṛttikatā). He
argues that this occurrence is impossible in this case, because the proving factor, being
established by valid means of cognition, leaves no room for deviation; he emphasizes also
that this occurrence can never be suspected so long as there is a necessary concomitance
21
22
23
Cf. PS(V) 1.6ab: mānasaṃ cārtharāgādisvasaṃvittir akalpikā / mānasam api rūpādiviṣayālambanam
avikalpakam anubhavākārapravṛttam, rāgādiṣu ca svasaṃvedanam indriyānapekṣatvān mānasaṃ pratyakṣam. “And the mental [perception], [i.e.,] awareness of an object-referent and self[-awareness]
of desire, etc., is free from conceptual construction. Mental [awareness], too, taking the object-field
like a visible matter, etc., as its object-support, [and] occurring with image of direct experience [of
that visible matter], is non-conceptual [and thus can be regarded as perception]; and self-awareness
in respect to the desire, etc., is [also a kind of] mental perception, because it is independent of sense
faculties.” And PS(V) 1.9ab: svasaṃvittiḥ phalaṃ vātra dvyābhāsaṃ hi jñānam utpadyate svābhāsaṃ
viṣayābhāsaṃ ca. tasyobhayābhāsasya yat svasaṃvedanaṃ tat phalam. “Or with regard to this [perception mentioned above as a type of perception] (cf. PSṬ 1 69,6–7: atreti pūrvokte pratyakṣe) the
self-awareness is the result. (9a) Cognition arises actually with two appearances, self-appearance and
object-appearance. The self-awareness of this [cognition] possessing both appearances is the result.”
Cf. TS 1555: na nāma rūpaṃ vastūnāṃ yat tasyāgrahaṇe sati / parijñātātmatattvānām apy avijñātatā
bhavet //
Cf. VMS 15b4–16a1/50a3–6: yadi hi jñānānubhavo nāma nāsty eva, tadā yo ’yaṃ ghaṭādivikalpaḥ
saṃvidita upayann āpayaṃś ca kvāntarbhāvyatām, arthe buddhau vā. na tāvad arthe, tasyehāvaśenodayāstamayāyogāt, asya cānevaṃrūpatvāt. buddhir atra vivarteta, sā cāpratyakṣā vivṛttāpi satī vaḥ
na prakāśeta, na ca prakāśo ’rthas tathāvṛttir ity abhāva evārthābhilāpānukāriṇo ’anubhavātmano
vikalpasya jñānaparokṣapakṣe prasajyeta. na caivam. This is in fact an adaptation of the argument
presented in Dharmakīrti’s PVin 1 14,14–15,7. Cf. the German translation of the Tibetan version in
Vetter 1966: 53.
6
Jitāri’s Vijñaptimātratāsiddhi
between the nature of self-manifestation and the things that become manifest, which he
expresses with a formal reasoning:
Everything that does not possess the nature of self-manifestation does not
become manifest, just like something that never appears. And [a sensory object]
like a blue thing does not have the nature of self-manifestation, if it is not
cognition. This is [the reason of] non-perception of a pervading factor.24
Here, it is worth noting that this is just a repetition in the negative form of the first formal
reasoning given at the beginning of the work when Jitāri starts his refutation.25 The only
difference is that the word “cognition” (jñāna) is here replaced with “possessing the
nature of self-manifestation” (svaprakāśasvabhāva), which Jitāri treats as synonym. This
demonstrates clearly that Jitāri sets forth the discussion from the Yogācāra point of view,
regarding the nature of self-manifestation as a conditio sine qua non for the arising of the
object’s manifestation in the cognition as the cognitive content.
The opponent does not accept the theory of self-manifestation, holding that a thing,
although not attaining (anu√bhū)26 the nature of self-manifestation, can still become
manifest, for instance, a blue thing, etc., not having the nature of manifestation of itself,
when placed (adhīna) near a lamp, etc., can still be manifest, and thus, the nature of selfmanifestation is not a necessary condition for the state of being manifest.27 Jitāri’s reason
consequently still remains inconclusive. Jitāri replies to this by simply pointing out that
manifestation never takes place in any other form than self-manifestation,28 so his reason
cannot be inconclusive.
In the next section, to refute the objection against the Yogācāra’s position of selfmanifestation, he examines the relationship between the cognitive manifestation and the
24
25
26
27
28
Cf. VMS 16a1–3/50a6–b2: anaikāntikatāpy asya na sambhavinī . sā hi bhavantī dṛśyamānavipakṣavṛttikatayā vā sambhavet sambhāvyamānavipakṣavṛttikatayā vā. tatra na tāvad agrimo grāhyaḥ pakṣaḥ,
pramāṇasiddhasya vyabhicāragocarasya kasyacid abhāvāt. nāpi paścimo vipaścitām paritoṣāya. yadi
viparyaye bādhakaṃ pramāṇaṃ na bhavet, asya syād vipakṣapracārāśaṅkā, yāvat tad asti — yad yat
svaprakāśasvabhāvaṃ na bhavati tat tan na prakāśate. yathā kiñcit kadācid apratibhāsamānam. na
bhavati ca svaprakāśasvabhāvam asati jñānatve nīlādikam iti vyāpakānupalabdhiḥ.
Cf. the quotation above in n. 12.
As pointed out by Funayama (2007: 194, n. 35), the subject of the verb anu-√bhū can also be a thing,
and then it does not mean “to experience …” as when it is used with a human being as its subject; he
translates the phrase sattām anubhavati in a passage quoted from TSP 1123,6–8 as “directly manifests
itself” (but, I think, in that case, the phrase could be simply translated as something like “is connected
to/attains its existence”). He gives some other examples of the same expression found in TSP. Actually,
examples for this use of the verb in other contexts can also be found in TSP and other texts, for instance,
cf. TSP 604,12–14: sa vāyur niṣkramaṃs tālvādeḥ saṃyogavibhāgāv anubhavati. gacchaṃś ca na
sa yāvad ākāśam abhigacchati. kiṃ tarhi. yāvad vegam … “It, i.e., wind, when blowing out, attains
connection or disjunction with the palate, etc., and it does not continue to move as long as there is
space, but as long as the impetus (vega) [continues].” (Notably, here, the word anurudhyate in TS
2177 is paraphrased as anubhavati).
Cf. VMS 16a4–6/50b3–4: svaprakāśasvabhāvatām ananubhavann api nīlādiḥ prakāśata iti na kiñcid
anupapannaṃ nāma. ko hy atra niyamo yat svaprakāśasvabhāvenaiva kevalaṃ prakāśitavyam iti,
paraprakāśyatve ’pi prakāśopapatteḥ, svayam aprakāśātmanām api nīlādīnāṃ dīpādisannidhānādhīnaprakāśarūpatopalambhāt.
Cf. VMS 16a6–16b1/50b5: bhaved ayam anaikāntiko hetuḥ, yady anyathāpi prakāśo ghaṭate, kiṃ tu
svaprakāśatām antareṇa prakārāntareṇa prakāśo nopapadyate.
Junjie Chu
7
object that becomes manifest in cognition, with the conclusion that his own initially
presented reason is not inconclusive.
3. Examining the relationship between the “making manifest” (prakāśana) and the object
From the Yogācāra’s point of view, a cognitive event of “manifestation” refers, of course,
to the fact that the cognition is manifest, so the question is raised why it is said that an
object is made manifest by cognition. According to the externalist opponent, however, that
statement means simply the fact that the “making manifest” (prakāśana) of an external
object is performed by its cognition.29 To refute this idea, Jitāri embarks on a rather lengthy
discussion to examine the relationship between the making manifest, which the opponent
alleges to be created by cognition, and the external object (artha). He uses an argument in the
form of “four alternative proposition” (catuṣkoti), a Buddhist favorite dialectical apparatus:
He lists at first exhaustively all conceivable relations between the making manifest and the
object, and then negates them all one by one: the manifestation is a different thing than the
object-referent (1), it is a non-different thing than the object-referent (2), it is both different
and non-different (3), it is neither different nor non-different (4).
Jitāri argues at first that the so-called “making manifest,” which is made by cognition,
cannot be a different entity (padārtha) than the object-referent, because in that case it
cannot be said that it is the object-referent that is made manifest by cognition. He explains:
The object-referent cannot become manifest as itself, since this is contradictory to the
thesis of other-manifestation, i.e., becomes manifest as a different thing. Moreover, the
“other” cannot operate to add a different nature to the object to make it manifest, since in
that case the “other” becomes identical with the object, i.e., one ends up with the second
option.30 Further, in that case, since the object-referent does not undergo any change while
the manifestation arises as a different entity, it would not become manifest; and a future
thing, inasmuch as it is not appearing in the manifestation, cannot be distinguished from
the thing itself (svarūpa) and thus cannot become manifest.31 To avoid this difficulty, the
opponent may argue that the object-referent, without change, becomes manifest with the
same manifestation. However, according to Jitāri, the manifestation is a temporal process,
i.e., the object undergoes the changes from the state of not being manifest to being manifest;
so Jitāri points out, this opponent’s argument does not hold, because in that case also
this unchanging manifestation is useless when the object has already its effecting means
(karaṇa)32 of a different manifestation, i.e., the effecting means of making a change; also
29
30
31
32
Cf. VMS 16b1–3/15a1–2: nanu jñānaprakāśatve ’pi prakāśo yujyata eva, idam eva kim uktaṃ bhavati
jñānenārthaḥ prakāśyata iti. kim atra praṣṭavyam, jñānena tasya prakāśanaṃ kriyata ity ayam artho
’nenābhidhīyata iti.
Cf. VMS 16b3–5/51a2–5: tatra yadi jñānena prakāśānākhyaṃ kim api padārthāntaraṃ kṛtam, kathaṃ
tenārthaḥ prakāśito nāma. na hy asau svayam eva prakāśate, tathātve paraprakāśatvāyogāt. tad api
param asya svarūpaviśeṣādhāne na vyāpriyate, dvitīyavikalpaprasaṅgāt, tasya ca vicārayiṣyamāṇatvāt.
Cf. VMS 16b4–5/51a4–5: tathā ca yathābhūto ’sau prakāśānudaye, tathābhūta eva prakāśodayasamaye
’pīti kathaṃ prakāśeta. aprakāśadṛśā bhāvinaḥ svarūpān na viśiṣyate prakāśate ceti suvyākṛtam.
In the grammatical sense, karaṇa refers to “the most efficient causal factor.” Cf. Pāṇ 1.4.42: sādhakatamaṃ karaṇam. and KV Part I: 557,4–5 (ad loc.): kriyāsiddhau yat prakṛṣṭopakārakaṃ vivakṣitaṃ
8
Jitāri’s Vijñaptimātratāsiddhi
because in that case an infinite regress would result, i.e., an endless arising of manifestation,
when this manifestation can never be (anupapatti) in the state of being currently manifest
because of its not being distinct from its previous form, i.e., the unmanifest form, and
undergoes (upagama) nevertheless the action of producing a manifestation other than that
manifestation.33 That means, in that case, the manifestation would not be a momentary
phenomenon, rather a never-stopping process.
Jitāri then turns to refute the second option, i.e., that the making manifest is nondifferent from the object, i.e., identical with object. He argues that the identity between
them cannot be possible, for in that case, when, through cognition, the object-referent
has the effecting means of making manifest (prakāśanakaraṇa) that is identical to itself,
it itself becomes the effecting means of making manifest, and thus the action (kriyā) of
making manifest becomes impossible, because it is already accomplished before by the
object-referent itself.34 If the action of making manifest takes a form that is different from
what is accomplished by another cause (kāraṇa), then only this different form is created,
since it has not been accomplished, but this form cannot be identical with the object-referent
(tadātmabhūta).35 And thus, if the object-referent ceased to exist in its previous form and
arises in another form, its appeared form (vyaktarūpa) must arise from cognition, and these
two must be momentary, because by accepting a different form, it follows necessarily that
it exists with each action (pratikṛti) [of making manifest].36 According to Jitāri, this is the
idea stated in Dharmakīrti’s PV 3.464–465ab, which he quoted.37
33
34
35
36
37
tat sādhakatamaṃ kārakaṃ karaṇasaṃjñaṃ bhavati. Cf. also Matilal’s discussion on the topic “the
doctrine of karaṇa” in 1985: 372–378.
Cf. VMS 16b5–17a1/51a5–51b2: tenaiva prakāśena prakāśyata iti cet. vārttam etat, tasyāpi prakāśāntarakaraṇe bhāve ’nupayogāt, tasya ca pūrvarūpāviśeṣāt prakāśamānatānupapattau tasyāpi tadaparaprakāśakriyopagame paryavasānaśūnyaprakāśaparaṃparotpādakasya purovasthitapadārthaprakāśamānatāparipanthino ’navasthānasya prasaṅgāt.
Cf. VMS 17a1–2/51b2–3: jñānenārthasyātmabhūtaprakāśanakaraṇe tasyaiva karaṇam āpadyeta. yat
khalv arthasyātmabhūtam asāv artha eva. tac ca kriyate na cārtha iti riktā vāco yuktiḥ. na cāsya kriyā
sambhavinī , prāg eva niṣpannatvāt.
Cf. VMS 17a2–4/51b3–5: yena rūpeṇa sa niṣpanno na tena kriyā, rūpāntareṇa karaṇād iti cet. tad eva
tarhi rūpāntaram aniṣpannatvāt kriyata iti prāptam. tac ca tanniṣpattāv aniṣpannaṃ kāraṇāntarataḥ
paścād upajāyamānaṃ kathaṃ tadātmabhūtaṃ nāma.
Cf. VMS 17a4–5/51b5–52a1: tasmād yady arthasya pūrvarūpavyayo rūpāntarodayaś ca bhavet, bhaved
asya vyaktarūpāntarasya jñānād utpattiḥ, tau cākṣaṇikatve na staḥ, tadupagame pratikṛtīti bhaṅguratvaprasaṅgāt. The phrase pratikṛti (cf. pratijñānaṃ in PV 3.465a quoted below in n. 37) reads in
the manuscript B as prakṛtibhaṅguratva°, “perishable in their nature,” however, in this case the phrase
loses the connection with PV.
Indeed, Dharmakīrti’s statement aims at the refutation of various opponents’ opinions against the
Yogācāra thesis that cognition is cognized by itself (svavedana, cf. PV 3.425ff.). One of them says that
cognition is apprehended by inference, i.e., through inferential sign (liṅga). Dharmakīrti refutes this by
arguing that, the causal forces for arising of a cognition, like sense faculty, object-referent, mental factor,
etc., cannot be established as the inferential sign, because either they could deviate from the causal
connection with cognition or they themselves are the cognition that are to be inferred; also the appearance
(vyakti) of object-referent, being the cognition, cannot be the inferential sign, because the object-referent,
when not being experienced, cannot be ascertained as appearing (cf. PV 3,461’d–463cd: tatra nendriyaṃ
vyabhicārataḥ // tathārtho dhīmanaskārau jñānaṃ tau ca na sidhyataḥ / nāprasiddhasya liṅgatvaṃ
vyaktir arthasya cen matā // saiva nanu jñānaṃ vyakto ’rtho ’nena varṇitaḥ / vyaktāv ananubhūtāyāṃ
tadvyaktatvāviniścayāt //). Dharmakīrti refutes further the objection against this argument in the next
one and a half verses which are quoted by Jitāri: “Now, [the opponent] accepts that the appearance is
Junjie Chu
9
Jitāri points out further that a momentary instance of an object-referent (arthakṣaṇa)
cannot be made manifest by the cognition, for in that case the cognition needs to be either
simultaneously existent or pre-existent, but both cases are impossible: When the cognition
is simultaneous with the object, no causality is possible, just like between the left and right
horns of a cow; again, a verse from Dharmakīrti’s PV is quoted, which argues that all
causes must exist before their results.38 On the other hand, when cognition exists before,
the object-referent arising from cognition with its independent nature of manifestation
(prakāśasvabhāva) would become manifest to all people, since it is common to all. Moreover,
the idea is also incorrect (asaṅgata) that the object-referent has the independent nature of
manifestation only for the person through whose cognition such manifest form (tādṛśa)
is produced, but not for the others; because the object-referents do not have a different
nature (ātmabheda) for each different person (pratipuruṣa), since otherwise it would follow
that they do not possess their own nature (nairātmya) on account of the absence of a fixed
nature (ātmasthiti).39
38
39
an [additional] special property (viśeṣa)* of the very object-referent. [But] the object-referent, since it
does neither arise nor pass away [according to the externalist opponent], cannot have a special property
of any kind. Or, when this [special property] is accepted, it would follow that it decays with each
cognition.” (PV 3.464–465ab: athārthasyaiva kaścit sa viśeṣo vyaktir iṣyate / nānutpādavyayavato
viśeṣo ’rthasya kaścana // tadiṣṭau vā pratijñānaṃ kṣaṇabhaṅgaḥ prasajyate / *In PVV 256,4 viśeṣa is
paraphrased as svabhāvaviśeṣa, so it refers to a form that is different from the object-referent in its own
nature.) It is interesting to observe that, in Jitāri’s text, the expression “with each action [of making
manifest]” (pratikṛti) is actually a referential use of Dharmakīrti’s expression “with each cognition”
(pratijñānam). The point here, in Jitāri’s context, is that the process of making manifest is a cognitive
product, and therefore it arises and disappears with each cognition, consequently it is not independent
of the cognition.
PV 3.246: asataḥ prāg asāmarthyāt paścāc ca na prayogataḥ / prāgbhāvaḥ sarvahetunāṃ nāto ’rthaḥ
svadhiyā saha // “Because [in the case that the object-referent as the cause and cognition as the effect
exist simultaneously, the object-referent], being non-existent before [the arising of the cognition as
result], is not efficient [in producing the result], and after [the arising of result] it is useless [since the
result has been already produced], all causes exist before [their results]; consequently, an object cannot
be simultaneous with its cognition.” The context of this quotation is however not exactly the same as
Jitāri’s argument. In the section of PV 3.245–247, Dharmakīrti discusses the problem of the temporal
relationship between mental awareness and its object, refuting an objection against the theory that
object-referent is an auxiliary factor (sahakārin) for sense faculty in producing mental cognition. The
opponent maintains that the object-referent, being active simultaneously with its cognition and thus
in a different time than sense faculty, cannot serve the function of auxiliary cause (sahakārin) Cf. PV
3.245: tadatulyakriyākālaḥ kathaṃ svajñānakālikaḥ / sahakārī bhaved artha iti ced akṣacetasaḥ // PV
3.246 is the reply to this objection. Jitāri’s aim of quoting this verse seems to be merely to rule out the
possibility of the simultaneous existence of cause and effect. The point here is that, if the object-referent
is external, it cannot be simultaneous with cognition. The Sautrāntika, accepting the externally existent
object-referent, refutes the Vaibhāṣika doctrine of the co-existent causes (sahabhūhetu). However, as an
internalist (antarjñeyavādin), the Yogācāra supports the theory that the object-referent, as a cognitive
aspect of cognition, and therefore existent internally, can be co-existent with cognition. Cf. Kato 1989:
309ff., Dhammajoti 2009: 154ff.
Cf. VMS 17b2–3/52a4–5: samasamayasambhavinā vā jñānenārthakṣaṇaḥ prakāśīkriyeta prāgbhāvinā vā. tatra na tāvad ādyo vikalpaḥ. jñānārthakṣaṇayoḥ sahabhuvoḥ savyetaragoviṣāṇayor iva
hetuhetumattvāyogāt. yad āha — … [quotation of PV 3.246] dvitīye tu vikalpe jñānād utpannaḥ prakāśasvabhāvo ’rthaḥ sādhāraṇatvāt sarvān prati prakāśet. yasyaiva jñānena sa tādṛśo janitas tam eva
prati sa prakāśasvabhāvo nānyaṃ pratīty apy asaṅgatam. na hi pratipuruṣam arthānām ātmabhedaḥ,
10
Jitāri’s Vijñaptimātratāsiddhi
For the third and fourth alternatives, i.e., that the manifestation is both different and nondifferent, as well as neither different nor non-different from the object-referent, Jitāri says
merely that they have already been refuted through the refutation of the first two alternatives,
so it is useless to exert the labor (piṣṭapeṣaṇa) of a separate refutation. He quotes two
verses from the 29th chapter of Arcaṭa’s Hetubinduṭīkā, “Refutation of Non-absolutism”
(dravyaparyāyānekāntavādakhaṇḍanam),40 and concludes that, of things with mutually
contradictory nature, negating the one implies (nāntarīyaka) affirming the other, and one
cannot affirm and negate the same nature in respect to the same thing. The opponents
therefore cannot defend themselves by taking this position.41
After negating all possible alternative interpretations of the relationship between the
cognition’s making manifest and the object, Jitāri says that, since it is not the case that one
thing can be made manifest by the other thing, the conclusion must be as follows: “What
becomes manifest is [cognition] itself, what is not [cognition] itself [, i.e., anything other
than cognition], does not have its manifestation at all.”42 If we compare this conclusion
with his first formal reasoning,43 it becomes clear that here he just repeats the idea of the
pervasion (vyāpti) stated in that reasoning. In the remaining part of the text, Jitāri replies
to several objections against this conclusion.
4. Replies to the objections against the thesis of self-manifestation
The first objection is directly directed against the above-mentioned pervasion, arguing
that a blue thing, etc., can be both not cognition (i.e., external) and self-manifestation in
its nature; so Jitāri’s reason is not established. Jitāri replies to this simply by indicating
the self-contradiction of this argument. He explains: For a cognition, the nature of being
cognition is nothing but the nature of being self-manifestation; so, by admitting the fact
40
41
42
43
nairātmyaprasaṅgād ātmasthiter abhāvāt. (A similar form of the last sentence can be found in PVin 1:
3,11–12.)
In that chapter the Jaina’s non-absolutism (anekāntavāda) is systematically refuted, which holds that
“substance” (dravya) and “mode” (paryāya) are not contradictory or mutually exclusive; rather, they
are in coherence. (For a brief description of this theory, cf. Dasgupta 1975: I 175ff.) Arcaṭa examines
the relationship between the “substance” and its “mode,” also using the analytic method of the “four
alternatives.” After negating the alternatives that the substance and the mode are “different” and “nondifferent” individually, he says in the two verses that are quoted by Jitāri that mutually contradictory
properties, such as difference and non-difference between substance and mode, cannot occur in one and
the same thing, but when they are accepted, the faults mentioned for each case remain, i.e., either “it
cannot be a single thing,” or “it cannot have the double form,” still exist, or the faults that are attached to
each single case come forth also in the case when these two are together. These faults (or “side-effects,”
such as phlegm produced by sugar and bile by ginger in the examples given in HBṬĀ) that come
forth in each single case of difference or non-difference between substance and modes cannot cease
to exist insofar the quality exists. Cf. HBṬ 29,25 (p. 106,11–12): bhedābhedoktadoṣāś ca tayor iṣṭau
kathan na vā / pratyekaṃ ye prasajyante dvayor bhāve kathaṃ na te // and 29,30 (p. 106,21–22): ye
bhedābhedoktimātre hi doṣāḥ sambhavinaḥ kathaṃ / tatsambhave pi te na syur iti brūyād vicakṣaṇaḥ //
Cf. also HBṬĀ 349,8–15, 350,29, 351,3 (ad loc.).
Cf. VMS 18a1/52b4: anyonyapratiṣedharūpāṇāṃ caikapratiṣedhasyāparavidhināntarīyakatvād ekasya
caikatra vidhipratiṣedhayor ayogāt, na tatkalpanānusaraṇaṃ śaraṇaṃ pareṣām.
Cf. VMS 18a2/52b5: tasmād anyasyānyena prakāśyatvānupapatteḥ, yat prakāśate tat svayam eva, yat
tu na svayaṃ tasya nāsyaiva prakāśa iti ekāntaḥ.
Cf. above n. 12.
Junjie Chu
11
that the blue, etc., are self-manifestation, one admits also that they are cognition.44 The
opponent is not satisfied with this reply, asserting that being cognition does not equal
being reflexive manifestation (ātmaprakāśatva), rather it means simply being an agent
of the action of making manifest (prakāśayitṛ). This is refuted by pointing out the fact
that cognition’s being the agent of making manifest is dependent upon the object it makes
manifest (prakāśya), and since that object-referent outside of cognition cannot be proved as
the object to be made manifest by cognition, also cognition cannot be the agent of making
manifest.45
The opponent now introduces the grammatical notion of an “object of action” (karman)
to prove that the object of action is separated from the agent. According to the opponent,
Jitāri’s above argument might be applicable to cases where the object of action is the “object
to be created” (kārya) or the “object to be modified” (vikārya). In the grammatical tradition,
the object of action is divided into three types: The object to be produced, the object to be
modified and the object to be attained.46 The opponent argues that the object made manifest
by cognition could be “the object to be attained” (prāpya), which is neither the object to be
produced or effected (kārya) nor the object to be modified (vikārya), like “village” in the
sentence “he goes to the village” (grāmaṅ gacchatīti); so it is not contradictory to say that
what is attained is made manifest, and is neither produced nor changed.47 This is refuted by
Jitāri based on the following reasons: (1) Other than being cognition no attaining (prāpti) is
possible, and without attaining no object-referent can be attained; (2) the cognition cannot
be characterized as attaining, since [in the opponent’s system] the action (kriyā) and its
causal factor (kāraka) cannot be the same thing;48 (3) In the case that the cognition is an
agent of action and acts with respect to itself (ātmakartṛkakriyārūpa), since the cognition
44
45
46
47
48
Cf. VMS 18a3–4/53a1–2: ajñānaṃ svaprakāśātmakaṃ ca nīlādikam iti kim atrānupapannam. tad ayam
asiddho hetur iti. tad asat, parasparaviruddhatvāt. svaprakāśatvam eva hi jñānasya jñānatvaṃ nānyat
kiñcit. tataḥ svaprakāśatvaṃ nīlāder abhyupannayā jñānatvaṃ nīlāder apy upeti.
Cf. VMS 18a5–6/53a2–4: nātmaprakāśatvaṃ jñānatvam, api tu prakāśayitṛtvam iti cet. prakāśyāpekṣayā hi prakāśayitṛtvam. prakāśyatā cārthasyānupapattimatīti jñānasyāpi prakāśayitṛtvaṃ nopapadyate.
Cf. KV Part II 540,3 (ad Pāṇ 3.2.1: karmaṇy aṇ): trividhaṃ karma, nirvartyam, vikāryaṃ prāpyaṃ
ceti. Cf. also the more detailed explanation in KVP Part II 540,30–541,25: tatra nirvartyaṃ yad asad
evopapapadyate*, yathā kumbhaṃ karoti nagaraṃ karotīti, kumbhādikaṃ hy avidyamānam evotpadyata
iti nirvartyaṃ karma. vikāryam yasya sata eva kaścid vikāro vidhīyate, tad yathā kāṇḍaṃ lunātīti, sata
eva kāṇḍāder lavanena vikrāro vidhīyata iti vikāryaṃ karma. prāpyaṃ nāsata evotpādanaṃ kriyate nāpi
sata eva vikārādhānam, kevalaṃ kriyāsambandhamātraṃ pratīyate, tad yathā vedam adhīte carcāṃ
pārayatīti. atra hy adhyayanādinā vedādeḥ sambandhamātraṃ pratīyate, na tv asata evotpādanam,
nāpi sata evānyathālakṣaṇo vikāra iti prāpyam etat karma. *-upapadyate em.: utpapadyate KVP
Cf. VMS 18b1–2/53a4–53b1: na hi kāryavikārye paraṃ karmaṇī, yenānutpādyam anatiśayaṃ vā
na karma syāt. prāpyam api karma, yat prāpyate param, na kriyate nāpi vikriyate. yathā grāmaṅ
gacchatīti. prāpyakarma vārthaḥ. tad ayam akriyamāṇo ’py avikriyamāno ’pi prāpyamāṇaḥ prakāśyata
ity aviruddham.
The theory of the non-difference between a causal factor of an action (kāraka) or more precisely, the
means of accomplishing the action (karaṇa), and the action (kriyā) as the result is the fundamental
principle of the Yogācāra since Dignāga, which is not accepted by the opponents like the Naiyāyika
who adheres to the grammatical notion of differentiating the action and its various causal factors.
There is evidently a long-lasting debate between Yogācāra and the Naiyāyika on this topic. Cf. PV
3.318–319: kriyākaraṇayor aikyavirodha iti ced asat / dharmabhedābhyupagamād vastv abhinnam
itīṣyate // evaṃprakārā sarvaiva kriyākārakasaṃsthitiḥ / bhāvasya* bhinnābhimateṣv apy āropeṇa
vṛttitaḥ // (*Following PVA and various Tibetan translations, Tosaki reads bhāveṣu. However, the
12
Jitāri’s Vijñaptimātratāsiddhi
does not operate (upayoga) on the [external] object-thing, the latter cannot become the
object of action of that cognition. (4) If the object-thing becomes the object of action
(karmatve) merely due to its presence at the time when the action arises, it would follow
that everything at that moment (tadātana) would be the object. If it becomes the object of
action due to being a cause, then the visual sense, etc., would also come to be the object of
action.49
The opponent argues further for the separation between the agent of making manifest
and the object to be made manifest, saying that a thing that becomes manifest can be an
object of action of making manifest when the agent of making manifest occurs, i.e., when
there is a causal relationship between them. Jitāri replies that an external object-referent
cannot become manifest as itself, nor can it have a different nature, i.e., manifestation
produced by cognition. Thus, it is just nonsense to apply the formula of causality which in
this case could be expressed as: “When that occurs, this becomes manifest.”50
After negating that the relationship between object and cognition is one between object
of action and agent of action, Jitāri adds that, through this negation, also Kumārila’s
statement in ŚVK, Pratyakṣasūtra 54–55, is refuted. There Kumārila explains the word
“birth” in the definition of “perception” in Mīmāṃsāsūtra 1.1.4: “Perception is the birth of
cognition when a person’s sense faculties are connected with an existent object,”51 saying
that the operation of causes is known as additional (atireka) to their birth. The word “birth”
used there is intended to mean that it should not be so also in the case of valid means
of cognition, and the latter cannot last even for a moment, nor is it to be produced as
invalid cognition (apramā), so that the causes should operate later in apprehending of the
49
50
51
genitive form is well supported by Manorathanandin: kriyākaraṇabhāvasya.) “[Opponent:] ‘The action
[of cognizing with the result (phala)] and the means of action [i.e., the means of cognition (pramāṇa)]
are incompatible to be the one and the same thing.’ [Reply:] This is not correct, because they are
understood as a difference in property [which is conceptualized in form of exclusion]. It is held that
there is no real division. All kinds of determination of the action and the causal factor [as separate] takes
that [conceptualized] form, because such relationship occurs also in respect to the existent things which
are considered as different [like wood and axe].” The theory that the object of action is not separated
from the agent of action can be found already in the Sautrāntika system. Cf. the Sautrāntika’s famous
response in the debate on “what perceives?” in AKBh 31,12–15: atra āhuḥ – kim idam ākāśaṃ khādyate.
cakṣur hi pratītya rūpāṇi cotpadyate cakṣurvijñānam. tatra kaḥ paśyati, ko vā dṛśyate. nirvyāpāraṃ
hīdaṃ dharmamātraṃ hetuphalamātraṃ ca. tatra vyavahārārthaṃ cchandata upacārāḥ kriyante –
cakṣuḥ paśyati vijñānaṃ vijānātīti nātrābhiniveṣṭavyam. Cf. Dhammajoti 2007: 87f., Dhammajoti
2009: 263. Here, it is clear, the Sautrāntika tries to desubstantiate the function of perceiving agent and
perceived object, reduce them to an interactive “factor” (dharma) that brings about the cognition, cf.
AKV 14,22: dharmamātram iti svatantrasya kartuḥ pratiṣedhaṃ karoti. “The [phrase] ‘mere factor’
negates an independent agent.”
Cf. VMS 18b2–4/53b1–3: nāviruddham, jñānasattātirekiṇyāḥ prāpter abhāvāt, prāptim antareṇa
cārthasya prāpyamāṇatvāyogāt. jñānasya ca prāptirūpatāyāḥ kriyākārakayor ekatvāyogenāyogāt.
ātmakartṛkakriyārūpatve ’pi jñānasyārthe ’nupayogāt tasya tatkarmatānupapatteḥ, tajjanmakālasattāmātreṇa karmatve sarvasya tadātanasya karmatvaprasaṅgāt, hetutvena tathābhāve cakṣurāder api
tathābhāvāpatteḥ.
Cf. VMS 18b4–5/53b3–4: tasmin sati prakāśamānasya karmatvād ayam adoṣa iti cet. sa tāvad artho
na svayam eva prakāśate, jñānenāpi nātmāntaram āpādyate. tasmiṃś ca sati prakāśata iti ka etad
anunmatto brūyāt.
MS 1.1.4: satsamprayoge puruṣasyendriyāṇāṃ buddhijanma tat pratyakṣam. Cf. Hattori 1968: 161, n.
6.1. For Dignāga’s refutation of this definition, cf. Hattori 1968: 63ff.; Frauwallner 1968: 62ff.; for a
comprehensive discussion of Kumārila’s interpretation of this definition, cf. Taber 2005: 44f.
Junjie Chu
13
object-thing like sense faculty, etc.52 Jitāri explains that this is refuted because with respect
to an external thing, a cognition that does nothing upon that thing cannot be valid, and that
thing cannot be the object of valid cognition; also because, if cognition is valid only upon
its birth, an over-excessive consequence would result (atiprasaṅga).53 This probably means
that if a cognition were a valid cognition merely by arising, then all cognitions, or perhaps
all things, would be valid cognitions merely by arising.
In the final section of the treatise, Jitāri also refutes some opponents from within the
Buddhist tradition. The first opponent is probably the Sautrāntika, whose general opinion on
the topic under discussion is that a cognition cognizes its object without taking any action
upon the object, just by assuming the image of the object that caused it.54 The objection
claims that a cognition can be that which makes its object-referent manifest (prakāśaka),
even without doing anything; and the object-referent can be made manifest (prakāśya) even
without any change made by cognition. That is to say, there can be a relationship between
a manifest-maker and an object even without any action. The idea of “action-less-ness” of
a cognitive event is shared by the Yogācāra, so Jitāri agrees with this completely. However,
he points out that, if two things are determined as having such relationship due to an action
taken by the one upon the other, then this relationship would be broken in the absence
of such action; on the other hand, for a thing that is not causally connected nothing can
be made in the case of the non-existence of this action.55 This means of course that the
theory of non-activity cannot be applied in the case of an external object. According to
Jitāri, in that case, in the absence of any action, the relationship between cognition and
object-referent as the manifest-maker and the object to be made manifest cannot exist; and
further, the external object is not connected with cognition, so without action nothing can
be made. That is to say, the cognition can make the object manifest only in the case that
the object is internal to or inside of cognition – that is, cognition can make only itself
manifest. The opponent argues against this, saying that, if cognition would make itself
manifest, two things would become the manifest-maker and the object to be made manifest
52
53
54
55
ŚVK, Pratyakṣasūtra 54–55 (NR: 109,14–22; Taber 2005: 153; quoted in TS 2923–2924): vyāpāraḥ
kāraṇānāṃ hi dṛṣṭo janmātirekataḥ / pramāṇe pi tathā mā bhūd iti janma vivakṣyate // na hi tat kṣaṇam
apy āste jāyate vā ’pramātmakaṃ / yenārthagrahaṇe paścād vyāpriyetendriyādivat // *kārāṇāṃ appears
also in the quotation in TS 1922a; however, in ŚVK (NR, Taber) it reads: kārakāṇāṃ Cf. the translation
in Taber 2005: 67.
VMS 18b6/54a1 etenaitad api nirastam yad āha kumārilaḥ … [Quotation of ŚVK, Pratyakṣasūtra
54–55] arthe kiṃcid akurvatas tatra prāmāṇyāyogāt, tasya ca prameyatvāyogāt, janmamātre ca prāmāṇye ’tiprasaṅgāt.
On the Sautrāntika’s opinion in this regard which can be found in the Abhidharma sources, cf. Dhammajoti 2007: 87f., Dhammajoti 2009: 263f. (cf. above n. 48). A similar opinion can also be found
in PSV ad PS 9cd: yathā yathā hy arthākāro jñāne pratibhāti śubhāśubhāditvena, tattadrūpaḥ sa
viṣayaḥ pramīyate. evaṃ jñānasaṃvedanaṃ anekākāram upādāya tathā tathā pramāṇaprameyatvam
upacaryate. nirvyāpārās tu sarvadharmāḥ. “The object-field is cognized as this or that form exactly
according to the way in which the image of object (arthākāra) appears in the cognition, as being pleasant
or unpleasant, etc. In this way, on account of (upādāya) the awareness of cognition in multiple forms
[of grasping subject and grasped object] it is metaphorically said that there are means of cognition and
the object of cognition in this or that manner, but [in reality] all phenomena are devoid of the activity [,
since they are instantaneous].”
Cf. VMS 19a2–4/54a3–4: syān matam — kiñcid akurvad api jñānam arthasya prakāśakam, akriyamāṇaviśeṣo ’py arthaḥ prakāśyata iti na kiñcid anucitam. yadi hi kasyacit kriyayā tayos tathābhāvo
vyavasthitaḥ syāt, tadā tadabhāve vighaṭeta. anibandhanasya tv asattve na kiñcit kriyate.
14
Jitāri’s Vijñaptimātratāsiddhi
without separate auxiliary means (upakāra), and everything would be the manifest-maker
and the object to be made manifest with respect to everything else. Jitāri replies to this by
simply pointing out that the object-subject relationship is restrained through their causal
relationship, so the over-excessive consequence mentioned by the opponent would not
occur.56
The topic of discussion then turns to the causal relationship. An opponent from the
Vaibhāṣika system argues that both cognition and object-thing, which arise from their
respective collection (kalāpa) of causes in the preceding moment, will have the form of
the manifest-maker and the object to be made manifest respectively. This is actually the
opinion of Śubhagupta (720–780).57 A verse is quoted from his BSK, which says: “The
causal complex in immediately preceding moment should produce the momentary instance
of object-field together with cognition (sajñāna), just like a visible matter together with
its light; for that reason, they should be perceived together.”58 The conclusion of this
objection is that, since the manifestation of the object-thing occurs even in the absence
of the nature of self-manifestation, so long as the invalidation of this occurrence is not
conclusive, the reason used by Jitāri is not conclusive.59 Jitāri replies: If the object-referent
produced by the collection of its own causes as something with the form to be made manifest
(prakāśyarūpa), then for this object only what is to be made manifest is attained (āpanna),
but not the coexistent cognition; but (ca) it is to be propounded by the opponent that the
object is made manifest by the cognition. Thus, Jitāri says, whoever claims that the object
is produced exclusively from its own cause as being made manifest contradicts the thesis
56
57
58
59
Cf. VMS 19a4–5/54a4–5: asaty upakāre prakāśyaprakāśakabhāve sarvaṃ sarvasya prakāśyaṃ prakāśakaṃ vā bhaved iti cet. na. yadi hīmāv animittāv eva svabhāvāv abhaviṣyatām, aniyamenābhavisyatām.
hetunā tu punar etau niyamyamānau katham atiprasajyeyātām.
Cf. Frauwallner 1961: 147.
Cf. VMS 19a6/54b1: pūrvakaiva tu samāgrī sajñānaṃ viṣayakṣaṇam / sālokarūpavat kuryād yena syāt
sahavedanam // Quotation from BSK 192b2. The verse is also quoted in JNĀ 23,23–24, 351,17–18; TSP
569,15–17; and VMS(R) 308a2. In the TSP, this verse is quoted together with the verse preceding this
one. The preceding verse runs: nānyo ’sti grāhako jñānāc cākṣuṣair viṣayair vinā / ataś ca sahasaṃvittir
nābhedān nīlataddhiyoḥ // “The grasping subject cannot be other than cognition, nor is it without visual
object; for this reason a blue thing and its cognition are apprehended together, not because their being
non-different [as the Yogācāra holds].” Here, in these two verses, as an externalist (Dreyfuss 1997: 363
describes him as Vaibhāṣika) and a Nirākāravādin, Śubhagupta does not accept the Yogācāra’s theory
of non-separation between cognition and its object. In the first verse, he explains sahavedana from the
viewpoint of the subjective aspect; and then in the second verse quoted in our text he explains it again
from the viewpoint of the objective aspect. So, in the second verse the central word must be viṣaya, not
jñāna; consequently, sajñānaṃ viṣayakṣaṇam must be the correct form. However, remarkably, in the
Tibetan translation of the BSK, pāda b appears to be shes pa yul bcas skad cig ste (in the quotation of
VMS(R) 308a2 it has the form: shes dang yul bcas skad cig ma), somewhat like saviṣayaṃ jñānakṣaṇam.
This seems to me not correct. yul bcas or yul dang bcas pa is usually used as the translation of saviṣaya.
Probably the Tibetan translator (or the scribe of the Sanskrit manuscript which the Tibetan translator
used) misread saviṣayaṃ jñānaṃ (cf. the phrase used in in Dignāga’s PSV ad PS 1.9b, cf. Chu 2006
[2008]: 239). In the Tibetan translation of the TSP, the phrase is correctly translated as shes bcas yul
gyi skad cig. Thus, both forms prajñānaṃ and tajjñānaṃ in the editions TSP and TSP (Ś), respectively,
should be corrected to sajñānaṃ, since it is not only supported by the Tibetan translation of the TSP,
but also by the two quotations in JNĀ, and now, additionally by our text.
Cf. VMS 19a6–19b1/54b2: tasmāt svaprakāśasvabhāvābhāve ’pi prakāśopapatter bādhakasyānaikāntikatāyāṃ maulasyāpi hetor anaikāntikateti.
Junjie Chu
15
he proposed.60 Here, “produced from its own cause” refers of course to the external object
that is independent of the cognition.
Now, the opponent argues that the object-referent is produced by its own cause as
having the additional quality (atiśaya) of something whose nature is born from cognition,
so that it could of course be made manifest by the cognition.61 The main points of Jitāri’s
refutation consists in an examination of the relation between the object and the cognition
from the temporal point of view. The opponent’s position that the object possesses an extra
characteristic, i.e., the manifestation, given by cognition, while it arises from its own cause,
implies necessarily the simultaneity between the object and cognition. This simultaneity
is actually completely acceptable for the Yogācāra in the theoretical framework of selfawareness. But for the opponent who is an externalist, as Jitāri points out, the simultaneity
is problematic. Jitāri says that, for two things that arise simultaneously, a relationship as
supporting and supported factor is impossible;62 if they are not related in this manner, but
merely simultaneous, it would follow that also other things born at the same time were
made manifest.63
The opponent has now resource to causality: being connected to the same causal
complex distinguishes the object-reference from other things.64 But, Jitāri replies, the
cognition, too, being dependent on its object, must belong to the same causal complex
(tadbhāva). The opponent still tries to defend himself, saying that the object has to be
made manifest by something else, i.e., the cognition, since it itself is not the manifestmaker. Jitāri then replies, as a coup de grâce, that the cognition, doing nothing, cannot
be the manifest-maker; and if cognition, being simultaneous with and sharing the same
causal complex with the object-referent, could the manifest-maker of the object-referent,
the object-referent would also be manifest-maker of the cognition.65 Jitāri quotes also
statements of Dharmakīrti (PV 3.417b–418a and 3.479’cd) to support his arguments.66
The conclusion is that the fundamental reason (maula), i.e., “becoming manifest,”
cannot be inconclusive, since the manifestation never occurs in any other way; and the
60
61
62
63
64
65
66
Cf. VMS 19b3/54b3–4: yady asāv arthaḥ svakāraṇakalāpena prakāśyarūpo janitaḥ, tasyaiva tarhi
prakāśyo ’yam āpannḥ. na punaḥ sahabhuvo jñānasya. tatprakāśyatā cāsya pareṇa pratipādyā. tad
ayaṃ prastutavastuvirodhinī svahetor evārthasya prakāśyatopapattis varṇayann…
Cf. VMS 19b5/55a1: atha jñānajanyasvabhāvātiśaya evāsau svahetunā janitaḥ, tenāsya tatprakāśyatā
syād eva.
Cf. VMS 19b5–20a1/55a2–3: …upakāryopakārakayoḥ sahotpannayos tadbhāvāyogāt.
Cf. VMS 20a1/55a3: na, samānakālabhāvinām anyeṣām apy aviśeṣeṇa prakāśyatvaprasaṅgāt.
Cf. VMS 20a1/55a4: ekasāmagrīpratibandho niyāmaka iti cet.
Cf. VMS. 20a2f./55a4f.: arthasyāprakāśakasvabhāvatvān naiṣa doṣa iti cet. jñānasyāpy akiñcitkarasya
kim idaṃ prakāśakatvaṃ nāma. tasmād yathārthena samānakālaṃ samānasāmagrīkaṃ vā jñānaṃ
tatprakāśakam, tathārtho ’pi teneti so ’py asya prakāśakaḥ prasajyata eva.
Cf. PV 3.417b–418a: anyasyānupakāriṇaḥ / vyaktau vyajyeta sarvo ’rthaḥ taddhetor niyamo yadi //
naiṣāpi kalpanā jñāne. “When some other thing that does not offer support [in producing cognition]
has appearance [in cognition], [then] all things could be caused to appear. If a restriction [is assumed]
on account of their [simultaneity with cognition], this assumption, too, is not possible with respect to
cognition.” PV 3.479’cd: anyathā tulyadharmā viṣayo ’pi dhiyā saha // “Otherwise, the object-field
sharing the same property with the cognition [would also make the cognition manifest].”
16
Jitāri’s Vijñaptimātratāsiddhi
thesis that is ascertained through the reason free from the three fallacies like the “nonestablishment [of the locus]” should be accepted.67
5. Conclusion
In this short treatise, Jitāri tries to establish the Yogācāra thesis that merely the cognitive
representation exists by means of the reasoning proving the thesis that anything that becomes
manifest is exclusively cognition itself. In doing so, he refutes various objections against
this reasoning and its conclusion. Through these objections, opponents try to demonstrate
that the reasoning is invalid and thereby defend their thesis that cognition and object are
separate things, as, respectively, what makes manifest and what is to be made manifest.
Like many other of Jitāri’s works, the VMS has clear polemical traits. He categorizes his
opponents mainly as belonging to the nirākārajñānavāda-branch of the bāhyārthavāda,
which includes representatives of non-Buddhist systems as well as Buddhists.
Through the presentation in previous sections, I hope that I was able to outline the
most important points of Jitāri’s arguments in this treatise. Although it seems to me that
Jitāri does not offer many innovative ideas, the Vijñaptimātratāsiddhi should still occupy
an important place in the history of Yogācāra philosophical literature, as it summarizes the
most important points of the Yogācāra position with regard to the topic of the cognitive
image of the object, and reports various opponents’ ideas. It thus enriches our knowledge
about later development of the Yogācāra system and its interaction with various Buddhist
and non-Buddhist opponents.
References and abbreviations
AKBh Vasubandhu, Abhidharmakośabhāṣya: Abhidharmakośabhāṣya of Vasubandhu,
ed. P. Pradhan. Revised second edition with introduction and indices etc. by Aruna
Haldar. Patna 1975.
AKV Yaśomittrā, Abhidharmakośavyākhyā: The Abhidharmakośa and Bhāṣya of Ācārya
Vasubandhu with Sphuṭārthā Commentary of Ācārya Yaśomittrā. 2 vols., ed. S. D.
Śāstrī. Varanasi 1998.
BSK Śubhagupta, Bāhyārthasiddhikārikā: Phil rol gyi don grub pa zhes bya ba’i tshig
le’ur byas pa. D 4244, vol. ZHE, 189b3–196b1.
Chu 2006 [2008] J. Chu, On Dignāga’s theory of the object of cognition as presented in
PS(V) 1. Journal of the International Association of Buddhist Studies 29.2 (2006
[2008]) 211–254.
Chu and Franco 2012 J. Chu and E. Franco, Rare Manuscripts of Works by Jitāri. China
Tibetology 1 (2012) 17–32. (Revised and enlarged version in Sanskrit manuscripts
in China III: proceedings of a panel at the 2016 Beijing Seminar on Tibetan Studies,
August 1-4, Beijing, forthcoming.)
67
Cf. VMS 20a5–6/55b1–2: tasmād anyathā prakāśāyogād bādhakasyānaikāntikatāvirahān maulasya
kuto ’naikāntikatā. tad anenāsiddhyādidoṣatrayaviyoginā hetunā yan niścitaṃ vipaścitā tu upādeyam
iti.
Junjie Chu
17
D Derge edition of Tibetan Tripiṭaka: sDe dge Tibetan tripiṭaka, bstan ’gyur – preserved
at the Faculty of Letters, University of Tokyo. Tokyo 1981.
Dasgupta 1975 S. Dasgupta, A History of Indian Philosophy. 5 vols. Delhi 1975.
Dhammajoti 2007 B. K. L. Dhammajoti, Abhidharma Doctrine and Controversy on Perception. Hong Kong 2007.
Dhammajoti 2009 B. K. L. Dhammajoti, Sarvāstivāda Abhidharma. Hong Kong 2009.
Dreyfus 1997 G. B. J. Dreyfus, Recognizing Reality: Dharmakīrti’s Philosophy and Its
Tibetan Interpretations. Albany 1997.
Frauwallner 1954 E. Frauwallner, Die Reihenfolge und Entstehung der Werke Dharmakīrtis. In: Asiatica. Festschrift Friedrich Weller. Leipzig 1954, 142–154. [Reprinted
in Frauwallner 1982: 677–689.]
Frauwallner 1961 E. Frauwallner, Landmarks in the History of Indian Logic. Wiener
Zeitschrift für die Kunde Süd- und Südostasiens 5 (1961) 125–148. [Reprinted in
Frauwallner 1982: 847–870.]
Frauwallner 1968 E. Frauwallner, Materialien zur ältesten Erkenntnislehre der Karmamīmāṃsā. Wien 1968.
Frauwallner 1982 E. Frauwallner, Kleine Schriften, ed. G. Oberhammer and E. Steinkellner. Wiesbaden 1982.
Funayama 2007 T. Funayama, Kamalaśīla’s Distinction between the Two Sub-Schools
of Yogācāra. A Provisional Survey. In: Pramāṇakīrti. Papers Dedicated to Ernst
Steinkellner on the Occasion of his 70th birthday, ed. H. Krasser et al. Wien 2007,
187–202.
Hattori 1968 M. Hattori, Dignāga, On Perception, being the Pratyakṣapariccheda of
Dignāga’s Pramāṇasamuccaya; from the Sanskrit fragments and the Tibetan versions.
Translated and annotated. Cambridge, Massachusetts 1968.
Hayes and Gillon 1991 R. P. Hayes and B. S. Gillon, Introduction to Dharmakīrti’s Theory
of Inference as Presented in Pramāṇavārttika Svopajñavṛtti 1–10. Journal of Indian
Philosophy 19 (1991) 1–73.
HB Dharmakīrti, Hetubindu: Dharmakīrti’s Hetubinduḥ. Teil I: Tibetischer Text und
rekonstruierter Sanskrit Text, ed. E. Steinkellner. Wien 1967.
HBṬ Bhaṭṭa Arcaṭa, Hetubinduṭīkā: Hetubinduṭīkā of Bhaṭṭa Arcaṭa with the SubCommentary Entitled Āloka of Durveka Miśra, ed. Sukhlalji Sanghavi and Muni
Shri Jinavijayaji. Baroda 1949.
HBṬĀ Durvekamiśra, Hetubinduṭīkāloka. See HBṬ.
JNĀ Jñānaśrīmitranibandhāvaliḥ (Buddhist Philosophical Works of Jñānaśrīmitra), ed.
A. Thakur. Patna 1987.
Kajiyama 1965 Y. Kajiyama, Controversy between the sākāra- and nirākāra-vādins of the
yogācāra school – some materials. Journal of Indian and Buddhist Studies XIV.1
(1965) 418–429. [Reprinted in Y. Kajiyama, Studies in Buddhist Philosophy (Selected
Papers). Kyoto 1989, 389–400.]
Kajiyama 1998 Y. Kajiyama, An Introduction to Buddhist Philosophy, an Annotated
Translation of the Tarkabhāṣa of Mokṣākaragupta, Reprint with Corrections in the
Author’s Hand. Wien 1998.
Kato 1989 J. Kato [加藤純章],『経量部の研究』 (Kyōryōbu no kenkyū; *A study of
Sautrāntika). Tokyo 1989.
18
Jitāri’s Vijñaptimātratāsiddhi
KV Jayāditya-Vāmana, Kāśikāvṛtti: Kāśikavṛtti of Jayāditya-Vāmana (Along with Commentaries of Vivaraṇapañcikā—Nyāsa of Jinendrabuddhi and Padamañjarī of Haradatta Miśra), ed. Śrīnārāyaṇa Miśra. Varaṇasi 1985.
KVP Jinendrabuddhi, Kāśikāvivaraṇapañcikā. See KV.
Matilal 1985 B. K. Matilal, Logic, Language & Reality, Indian Philosophy and Contemporary Issues. Delhi 1985.
Mookerjee and Nagasaki 1964 S. Mookerjee and H. Nagasaki, The Pramāṇavārttikam of
Dharmakīrti. An English Translation of the First Chapter with the Autocommentary
and with Elaborate Comments [kārikās I–LI]. Patna 1964.
MS Jaimini, Mīmāṃsāsūtra: The Mīmāṃsā Darśana. 2 vols, ed. Mahamahopadhyaya
Mahesachandra Nyayaratna. Calcutta 1889.
MSBh Śabarasvāmī, Mīmāṃsāsūtrabhāṣyam ad Jaimini’s Mīmāṃsāsūtra 1,1.1–5. See
Frauwallner 1968: 10–61.
NR Pārthasārathimiśra, Nyāyaratnākara: Ślokavārttika of Śrī Kumārila Bhaṭṭa with The
Commentary Nyāyaratnākara of Śrī Pārthasārathi Miśra, ed. and rev. Ganga Sagar
Rai. Varanasi 1993.
Pāṇ Pāṇini, Aṣṭādhyāyī: The Aṣṭādhyāyī of Pāṇini. 2 vols., ed. and trans. Śrīśa Chandra
Vasu. Delhi 1891–1893.
PSṬ 1 Jinendrabuddhi, Viśālāmalavatī Pramāṇasamuccayaṭīkā, Chapter 1, Part 1: Critical Edition, ed. E. Steinkellner, H. Krasser, and H. Lasic. Beijing/Vienna 2005.
PS(V) Dignāga, Pramāṇasamuccaya: Dignāga’s Pramāṇasamuccaya, Chapter 1, A hypothetical reconstruction of the Sanskrit text with the help of the two Tibetan translations
on the basis of the hitherto known Sanskrit fragments and the linguistic materials
gained from Jinendrabuddhis Ṭīkā, ed. E. Steinkellner. 2005. https://www.oeaw.ac.
at/fileadmin/Institute/IKGA/PDF/forschung/buddhismuskunde/dignaga_PS_1.pdf,
last visited 09-08-2019.
PV 1 Dharmakīrti, Pramāṇavārttika, Chapter 1: R. Gnoli, The Pramāṇavārttikam of Dharmakīrti, the First Chapter with the Autocommentary. Text and Critical Notes. Roma
1960.
PV 3 Dharmakīrti, Pramāṇavārttika, Chapter 3. See Tosaki 1979.
PVA Prajñākaragupta, Pramāṇavārttikālaṅkāra: Pramāṇavārtikabhāshyam or Vārtikālaṅkāraḥ of Prajñākaragupta (Being a commentary on Dharmakīrti’s Pramāṇavārtikam), ed. R. Sāṅkṛityāyana. Patna 1953.
PVin 1 Dharmakīrti, Pramāṇaviniścaya: Dharmakīrti’s Pramāṇaviniścayaḥ, Chapter 1
and 2. Critically edited by E. Steinkellner. Beijing/Vienna 2007.
PVV Pramāṇavārttikavṛtti: Dharmakīrti’s Pramāṇavārttika with a commentary by Manorathanandin, ed. R. Sāṅkṛtyāyana. Appendix to Journal of Bihar and Orissa Research
Society, vols. XXIV, XXV, XXVI. Patna 1938–1940.
RNĀ Ratnakīrti, Nibandhāvali: Ratnakīrti-Nibandhāvaliḥ (Buddhist Nyāya Works of
Ratnakīrti), deciphered and edited by Prof. Anantalal Thakur. Patna 1975.
Shirasaki 1981 K. Shirasaki [白崎顕成]: Jitāri,人と思想 (Jitāri, hito to shisō; *Jitāri:
life and thought). In:『僧伝の研究:木村武夫教授古稀記念』(Sōden no kenkyū:
Kimura Takeo kyōjū koki kinen). Kyoto 1981, 1–29.
Steinkellner 1967 E. Steinkellner, Dharmakīrti’s Hetubinduḥ. Teil II. Übersetzung und
Anmerkungen. Wien 1967.
Junjie Chu
19
Steinkellner 2008 [2009] E. Steinkellner, Further Remarks on the Compound avinābhāvaniyama in the Early Dharmakīrti. Wiener Zeitschrift für die Kunde Südasiens 51/52
(2008 [2009]) 193–205.
Steinkellner 2013 E. Steinkellner, Dharmakīrtis frühe Logik, Annotierte Übersetzung der
logischen Teile von Pramāṇavārttika 1 mit der Vṛtti, I. Introduction, Übersetzung,
Analyse, II. Introduction, Anmerkungen, Anhänge etc. Tokyo 2013.
ŚVK Kumārila, Ślokavārttika. See NR.
Taber 2005 J. Taber, A Hindu Critique of Buddhist Epistemology: Kumārila on Perception:
The “Determination of Perception” Chapter of Kumārila Bhaṭṭa’s Ślokavārttika,
Translation and Commentary. London/New York 2005.
TBh Tarkabhāṣā: Tarkabhāṣā and Vādasthāna of Mokṣākaragupta and Jitāripāda, ed. H.
R. Rangaswami Iyengar. Mysore 1952.
Tosaki 1979 H. Tosaki (戸崎宏正),『仏教認識論の研究 – 法称著『ブァマーナ·ヴ
ァールティカ』の現量論 – 上巻』(Bukkyō ninshikiron no kenkyū – hōshōchō
“Pramāṇavārttika” no genryōron, jokan; *Studies in Buddhist epistemology: Dharmakīrti’s Pramāṇavārttika, pratyakṣa chapter, vol. 1). Tokyo 1979.
TS Śāntarakṣita, Tattvasaṅgraha: Tattvasaṅgraha of Śāntarakṣita with the Commentary of
Kamalashīla. 2 vols., ed. E. Krishnamacharya. Baroda 1984.
TSP Kamalaśīla, Tattvasaṅgrahapañjikā. See TS.
TSP (Ś) Kamalaśīla, Tattvasaṅgrahapañjikā: Tattvasaṅgraha of Ācārya Śāntarakṣita
with the Commentary ‘Pañjikā’ of Ācārya (Srī ) Kamalashīla. 2 vols., ed. Swāmī
Dwārikādās Śāstrī. Varanasi 1997.
Tucci [1930] 1971 G. Tucci, The Jātinirākṛti of Jitāri. In: Opera Minora. Part I. Roma
1971, 249–254.
Vetter 1966 T. Vetter, Dharmakīrti’s Pramāṇaviniścayaḥ, 1. Kapitel: Pratyakṣam. Einleitung, Text der tibetischen Übersetzung, Sanskritfragmente, deutsche Übersetzung.
Wien 1966.
VMS Jitāri, Vijñaptimātratāsiddhi: Jitāri’s Vijñaptimātratāsiddhi (Sanskrit Text), critically
edited by J. Chu. Forthcoming.
VMS(R) Ratnākaraśānti, Vijñaptimātratāsiddhi: Rnam par rig pa tsam nyid du grub pa.
In: sDe dge Tibetan tripiṭaka, bstan ’gyur – preserved at the Faculty of Letters,
University of Tokyo. Tokyo 1981, no. 4259, vol. ZHE, folios 306b4–309b3.
Consciousness and Causality: Dharmakīrti Against
Physicalism1
by
Christian Coseru
It is well known that Buddhist conceptions of personal identity entail a version of the
so-called ‘bundle theory of self:’ the human individual comprises five types of aggregates
that serve as the basis for what we ordinarily designate as persons. What is less known
(or least explored) is the extent to which this conception of personal identity informs the
Buddhist epistemological account of cognition. Specifically, the assumption is that with
Dignāga, Dharmakīrti and their successors, the bundle theory of personal identity is either
glossed over in favor of more robust accounts of consciousness and cognition such as that
provided by Yogācāra or challenged for its strict reductionism. I will argue that – rather
than glossing over or challenging it – the Buddhist epistemologists uncover a structural
asymmetry within the bundle theory between the mental and physical domains, and offer an
alternative (if problematic) solution to account for the ineliminable aspects of phenomenal
consciousness.
The following analysis focuses on Dharmakīrti’s arguments against Cārvāka physicalism in the so-called proof of rebirth in the Pramāṇasiddhi chapter of his magnum opus,
the Pramāṇavārttika, with a focus on classical Indian philosophical attempts to address
the mind-body problem. The key issue concerns the relation between cognition and the
body, and the role this relation plays in causal-explanatory accounts of consciousness and
cognition. Here a number of questions arise. Does the central principle of Buddhist Abhidharma reductionism apply to consciousness? Is there a causal criterion for the presence
of consciousness? If there is, can this causal criterion account for the specific features of
consciousness, e.g., its intentionality, phenomenality, and reflexive character (svasaṃvedana)? Can a causal account of phenomena be reconciled with the seeming irreducibility
of consciousness? The Buddhist answer to the challenge of Cārvāka physicalism displays
many of the common features of classical Indian metaphysical debates on personal identity.
My aim is not to trace its exegetical contour and restate its historical significance, but to
propose a philosophical reconstruction that builds on two important features presented
by the Buddhist account: an expanded conception of causality and a robust account of
phenomenal content that, taken together, can help us come to terms with the legacy of
mind-body dualism.
1. Buddhist reductionism
In replacing the subject with a play of momentary cognitive events, the Buddhist account
of personal identity emphasizes what we may call – using the language of contemporary
1
I am grateful to Sara McClintock for her helpful advice and comments, and for suggesting many
conceptual and stylistic improvements.
Birgit Kellner et al., eds., Reverberations of Dharmakīrti’s Philosophy: Proceedings of the Fifth International Dharmakīrti
Conference Heidelberg, August 26 to 30, 2014. Vienna 2020, pp. 21–39.
22
Consciousness and Causality: Dharmakīrti Against Physicalism
philosophy of mind – the dynamic, embodied, and embedded functioning of the five aggregates.2 However, in the schematic analysis of the five aggregates, only “body” or “form”
(rūpa) is a physical aggregate stricto sensu. Sensations, apperception, and volitions can
acquire an objectual aspect, but are not empirical objects proper. Nor are they things in
the generic sense of the Sanskrit term vastu, that is, abstract entities with well-defined
properties and functional characteristics.3 Thus, a sensation of pain is not reducible to
the physical substrate, say a finger, in which it is instantiated (nor presumably to a mere
physiological response). Rather, as object-oriented cognitive aspects (viṣayākāra), sensations and volitions are included in the broader Abhidharma category of mental factors
(caitasika). Feelings may define the quality of the impressions that result from contact
with an object, with the implication that they perhaps stand in a causal relation with these
objects. But as internal mental states, they are also conditioned by habitual tendencies
(vāsanā), which, in turn, they condition: one’s physical condition after strenuous exertion
may feel pleasant or unpleasant depending on one’s level of fitness and degree of exercise
frequency. Likewise, apperception (saṃjñā), the capacity to make intelligible or cause to
be understood, although dependent on a multiplicity of psychological factors, captures the
datum of experience only as fused into a single percept. Volitions too fit the same profile,
with one important difference: rather than attending to the object at hand or providing a
sort of transcendental unity of apperception, they bring forth future states of existence.
As dispositions to act in certain ways, they cleave the mental domain into two classes of
conditioned phenomena: those that are internal to consciousness (saṃprayukta-saṃskāra),
such as, for instance, obsessive dispositions (paryavasthāna) like greed and delusion, and
those that are dissociated from it (viprayukta-saṃskāra), usually taken to refer to latent
dispositions (anuśaya) typically comprising various biological and physical traits.4
This aggregate model of personal identity is not incompatible with the notion that there
are phenomenal primitives (or, in epistemological terms, cognitive universals) – irreducible
features of experience. The experience of vividness (spaṣṭa), for instance, marks perception
apart from mental imagery, thus making it possible to identify visual qualia as irreducibly
perceptual.5 Likewise, the experience of being dragged across the floor as opposed to
2
3
4
5
This conception of embodied cognition finds its roots in Edmund Husserl’s notion of the life-world
(Lebenswelt). The paradigm of embodied (and enactive) cognition is explored at length in Dreyfus
1979, Varela/Thompson/Rosch 1991, Hutchins 1995, Clark 1997, Hurley 1998, Noë 2004, Gallagher
2005, and Thompson 2007.
Insofar as the aggregates of sensation, volition, etc. fall under the general Abhidharma category of
dharma, they can be treated, at least under some scholastic interpretations (such as, for instance, of the
Sthaviravāda) as substances (dravya). As Ronkin (2005: 15) has convincingly argued – taking her lead
from Gombrich 1996 and Hamilton 1996 – the reductive analysis of human beings in terms of their
constitutive aggregates is meant to capture not what human beings are made of, but rather what human
experience is constituted as: specifically, as series of experiential events.
Detailed accounts of this twofold analysis of phenomena are found in Vasumitra’s Pañcavastukavibhāṣāśāstra [Wu shih p’i-p’o-sha Zun], T 28 (1555), p. 989b2, Vasubandhu’s Abhidharmakośabhāṣya,
II, 23–34, and Yaśomitra’s Sphuṭārthā Abhidharmakośavyākhyā. Cox (1995, ch. 4) offers the most
detailed account to date of the citta-viprayuktasaṃskāra. For a broader discussion of the process by
which mental factors that arise in conjunction with a given intentional object come to be associated
with the qualities of the respective object, see Waldron 2003: 57ff.
One may extend visual tropes to the domain of imagery, rational deliberation, and introspective awareness, but these are metaphorical rather than literal uses. Dignāga’s conception of mānasa-pratyakṣa
Christian Coseru
23
moving freely serves to contrast action from agency, and support the view that even
unreflective and habitual practices, if consciously undertaken, are constitutively agentic. As
the classical example of watching a dance performance while entertaining various thoughts
demonstrates, one may be solicited to respond in ways that are wholly unreflective. Being
consciously present to the situation at hand, however, means that these solicitations elicit
not merely a reflex but rather a subjective response. A brilliant dance performance does
not simply induce applauding; rather, the performance solicits appreciation, which serves
as an appropriate and deliberate response in such circumstances.6 Applause is not merely
a participant reactive attitude, but a learned subjective response to excellence (except,
perhaps, in instances where it reflects norms of audience participation).
It has been argued, most forcefully by Mark Siderits (2003, 2011), that Abhidharma
reductionism entails physicalism, the view that everything is or supervenes on the physical (where “physical” stands for the world as described by our best physics). Although
Dharmakīrti shares the empirical stance of Abhidharma, the naturalism that informs his
epistemological project is patently anti-physicalist. According to the Sautrāntika Abhidharma account of materiality that Dharmakīrti favors, entities reduce to their phenomenal
primitives: the particular (svalakṣaṇa) is a token of a type, not blue in general, but this
unique intensity of cerulean. Furthermore, the formal properties of material objects are
analyzed either in terms of how they are impacted by contact or as factors that oppose
resistance. These properties, however, do not extend to the atoms themselves, which according to the Abhidharma form the building blocks of materiality. As monadic units the
atoms are seen as devoid of any formal properties (rūpaṇa). It is only as atomic compounds
(saṃghātastha, saṃcita) that atoms are subject to the same properties of resistance and
destruction as composite material entities.7
The reductionist model of Abhidharma, like all philosophical attempts to carve reality
at its joints, works against the common conception that empirical awareness provides
access to an external, stable, and self-sustaining world: a world as is (captured by the
notion of svabhāva) rather than as it appears to an observer. But the human mind is not
(like) a clear mirror reflecting back the external world, as naïve realism would have it;
rather, its image is as projected by a mind not entirely free of its own propensities and
confabulations. What Abhidharma offers, then, is a metaphysics of experience, where the
irreducible elements of existence and/or experience (dharma) are not fixed substances but
activities, properties, or dynamic patterns of connectedness that are constitutive of the
world as perceived (lokasaṃjñā). As the Nikāyas clearly state (e.g., SN IV, 96), our sensory
organs (vision, hearing, etc.) operate in a world whose contours are disclosed in a dynamic
and mutually constituted setting of objects and meaning. In practical terms, that means
6
7
(lit. “mental perception”), thus, captures the distinct introspective awareness or attentiveness that
accompanies the perceptual occasion.
This example is meant to support an argument for minimal agency as an ineliminable feature of cognitive
awareness. The finer point is that ‘unreflective’ does not mean ‘unconscious’ or purely ‘behavioral.’
Skillful means, much like skillful coping, are treated here as modes of unreflective, but minimally
conscious and implicitly subjective, agency.
See, for example, AK I 13 and AKBh ad cit. (Pradhan 1975: 9): paramāṇurūpaṃ tarhi rūpaṃ na
prāpnoty arūpaṇāt / na vai paramāṇurūpam ekaṃ pṛthag bhūtam asti / saṃghātasthaṃ tu tad rūpyata
eva.
24
Consciousness and Causality: Dharmakīrti Against Physicalism
experience marks the boundary of what there is: the nexus of causes and conditions that
set the boundaries of lived experience are determined by the operations of our cognitive
architecture. Color, for instance, only exists for an organism that is sensitive to light.
How does this dynamic picture of what there is take on the characteristics of self and
other? And how do these emergent phenomena in turn create the conditions for grasping
and attachment? For the Buddhist, the answer does not lie primarily in the patterns of
conditioning that explain the aggregation of phenomena, but in certain defining characteristics that belong to the structure of experience itself. Not only are the senses conceived as
receptacles of experience (adhiṣṭhāna), they also serve as ground or support, joining the
external domain of sensory activity (bāhirāyatana) with the internal domain of perception
(ādhyātmikāyatana). We can make sense of Dignāga’s stance (at PS, I, 1 and PSV ad
cit.) that perception gives us the particular as such, without any conceptual mediation,
only insofar as the domain of sensory activity is reducible to its most basic physiological
function.
Now, does the central principle of Abhidharma reductionism apply to consciousness?
The principle states that things reduce to their component parts, which are ultimately real
only if they are further irreducible. If something can be reduced either by breaking it down
to more basic constituents or through conceptual analysis, then it is not ultimately real. Pots
are not ultimately real, nor are persons real in this ultimate sense.8 Consciousness too is but
a stream of momentary conscious events of different types (visual, auditory, introspective,
etc.). But reductionism about consciousness is problematic. Why? Because it cannot
explain its most basic features: its intentional, phenomenal, and self-reflexive character
(svasaṃvedana). The recognition that conscious awareness has these ineliminable structural
features creates a series of doctrinal problems for Buddhism in its post-Abhidharma stage
of development. For Mādhyamika thinkers like Nāgārjuna, notions such as ‘intrinsic’ and
‘ineliminable’ run counter to the cardinal Buddhist view that all phenomena are momentary,
impermanent, and interdependently arisen. Conceived largely as a response to Madhyamaka
dialectics, Yogācāra sets out to account for the nature of consciousness and cognition itself,
bracketing metaphysical assumptions about the kind of things that there are.
The relation between mind and world continues to be a subject of ongoing debate
between Buddhists and their opponents, and among rival Buddhist schools up to the
present day. The debates in Buddhist metaphysics of mind are not primarily exegetical
(the presence of a vast commentarial literature notwithstanding); rather, they often reflect
deep philosophical differences. When these differences are grounded in merely exegetical
claims, the text-critical method offers the best way to find solutions. However, if these
differences are also grounded in empirical and/or experiential claims, the analytic tools of
contemporary philosophy and advances in our empirical investigation of cognition ought
8
One important exception here are the Pudgalavādins, for whom persons, who are conceived in dependence upon the aggregates (although neither identical nor different from them), are nonetheless
real. In Vasubandhu’s summary of this position, for the Pudgalavādin “if a consciousness is aware
of a person in dependence upon a visible form known to exist by means of the eye, it is said that
a person in known to exist by means of the eye” (AKBh IX in Pradhan 1975: 463). Translation per
Duerlinger 2003: 77. Persons are ultimately real (for the Pudgalavādin) because, as Amber Carpenter has
convincingly argued, “perception-dharmas and consciousness-dharmas … are related to one another in
a … person-constituting way” (2015: 27).
Christian Coseru
25
to in principle help move this debate forward in more profitable directions. The mind-body
problem may be as intractable now as it was for Dharmakīrti in the seventh century, but
our conception of what counts as legitimate, reliable evidence, is less arbitrary, at least
with regard to the sort of things that can be said to exist both in a concrete and abstract
sense. Few philosophers today who are sympathetic to and influenced by Buddhist ideas
find traditional Buddhist beliefs about rebirth and cosmic bodhisattvas to be live options.
2. Physicalism and its discontents
The Pramāṇasiddhi section of the Pramāṇavārttika contains Dharmakīrti’s famous proof of
rebirth, better known for its ingenious attempt to answer, using mainly a priori arguments,
the Cārvāka’s challenge that consciousness originates, or otherwise has its causal basis,
in the body. The key verses (PV II vv. 34–72) and the extensive commentarial literature
thereon (from Devendrabuddhi, Prajñākaragupta, Ravigupta, and Manorathanandin, to the
Tibetan translations by Sa skya paṇḍita and Śākyaśrībhadra) have been explored at length
in Eli Franco’s superb monograph, Dharmakīrti on Compassion and Rebirth (1997). Its
key arguments have likewise been discussed in a series of recent, mainly philosophical,
engagements with this topic.9 The proof is occasioned by the claim, first advanced by
Dignāga in the Pramāṇasamuccaya, that the Buddha does not merely avail himself of
the right sources of knowledge (pramāṇa), but rather he in some fashion embodies them.
Dharmakīrti, as is well known, takes this claim one step further when he argues at PV II vv.
34–131 that the proof of the Buddha being a pramāṇabhūta lies in compassion cultivated
by practice over many lifetimes, and in the veracity of his teachings (upadeśa).10
Of course, the Buddhist epistemologist’s appeal to the Buddha as an enlightened knower
to justify the claim that perception and inference are trustworthy sources of knowledge –
because the Buddha declares them to be so – is unmistakably circular. Ernst Steinkellner
clearly explains this circularity:
9
10
See Vetter 1964, Hayes 1993, Taber 2003, and Arnold 2008.
There are two basic ways to make the case for the Buddha being a pramāṇabhūta: (i) demonstrate the
possibility of infinite compassion, and thus of previous lifetimes dedicated to perfecting such a goal, and
(ii) argue for the veracity and viability of the Four Noble Truths as proof that only someone motivated
by such compassionate aims could have gained this sort of knowledge. The second demonstration
has Dharmakīrti employ the so-called “no alternative” (agatyā) strategy: “Given this [teaching of the
Four Noble Truths], which leads to achieving human ends, which is reliable and worth practicing,
[we] accept that [this teaching] must be equally so with regard to the other domain [e.g., of scriptural
or imperceptible truths]. [Such teaching could] not [have been offered] with the intent of deceiving
[others], because it is not an obstacle [but rather an aid to knowledge]; and also because there is
no reason for the speaker to engage in aimless deception. Both instances, thus, demonstrate the
validity of reasoning on the basis of scriptural sources on account of there being no [other] way” (PV
I v. 217 and PVSV ad cit.: tasyāsya puruṣārthopayogino ’bhiyogārhasyāvisaṃvādād viṣayāntare ’pi
tathātvopagamaḥ, na vipralambhāya, anuparodhāt; niṣprayojanavitathābhidhānavaiphalyāc ca vaktuḥ.
tad etad agatyobhayathāpy anumānatvam āgamasyopavarṇitam). Tillemans (1993: 16ff.) thinks this
causal relationship between compassionate undertaking and the effectiveness of embodied epistemic
warrants only works for things that are empirically tractable (such as the Four Noble Truths). See
also Kataoka 2005: 256–59 for an illuminating discussion of Dharmakīrti’s second argument, and the
problematic issue of proving matters that fall outside the domain of empirical ascertainment.
26
Consciousness and Causality: Dharmakīrti Against Physicalism
(1) Our ordinary valid cognitions (pramāṇa) establish the authority of the Buddha’s teaching (buddha-vacana); (2) the validity of our cognitions (prāmāṇya)
is understood as their reliability (avisaṃvāditva); (3) reliability depends on
successful activity (puruṣārtha-siddhi); (4) all human goals are determined by
the “ultimate goal” (nirvāṇa); the “ultimate goal” is indicated in the Buddha’s
teaching (buddha-vacana) (Steinkellner 2003: 328).
So: perception and inference are taken to be instrumentally capable of demonstrating that the
Buddha is a trustworthy teacher. Because of his trustworthiness, his teachings are valid and
provide further proof that only perception and inference qualify as sources of knowledge
(again, to come full circle, because the Buddha has established that to be the case).11
Why claim that valid cognition establishes the authority of the Buddha’s teachings in the
first place? Why not simply be content with establishing knowledge on the best possible
foundation, whether or not the Buddha’s teachings are in agreement with it? If Buddhist
epistemology can lay claim to methodological universalism on account of its reliance on
perception and reason alone, reverence for the Buddha as a perfect embodiment of epistemic
excellence can seem redundant (except perhaps on political or religious grounds).12
Does appeal to the authority of the Buddha as a true embodiment of the sources
of knowledge, then, undermine the Buddhist epistemological stance? Not necessarily.
Indeed, as Franco notes, “Dharmakīrti argues here … that the Buddha used perception
and inference, not that they are valid because of him” (Franco 1999: 65). Precisely what it
means to ‘embody’ the sources of reliable cognition remains an open question, regardless
of whether Dharmakīrti’s argument is found to be circular or not. It is not enough to say that
the Buddha is a true embodiment of reliable cognition: one must also show in what ways.
Dharmakīrti’s own answer to this question invokes three distinct elements: the Buddha
embodies the sources of knowledge by means of (i) his compassion, (ii) his knowledge, and
(iii) the actions that bear testimony to this compassion and knowledge. But neither listing
the Buddha’s attributes nor the fact that he reasons and acts on the basis of compassion
and knowledge justify the veracity of our cognitions (and the sources thereof).
Whatever Dharmakīrti’s motives might have been in seeking to ground epistemic norms
in a proof of rebirth, the arguments against the Cārvāka claim that consciousness begins
and ends with the body offer interesting new ways to conceptualize the mind-body problem.
I will not dwell on the preamble to the proof, which takes compassion to act as a cause
in an effort to achieve the kind of knowledge Buddhas require for accomplishing their
mission. As John Taber (2003) has convincingly demonstrated, there is nothing particularly
original in articulating a conception of epistemic authority grounded on reliable testimony
(as provided by an āpta, a trustworthy person or, as in the case of Kumārila’s stance, of
the Vedas). One gets to limit the range of reliable sources of knowledge to perception and
11
12
Various formulations of this argument, first proposed by Nagatomi (1959) and Vetter (1964), are also
found in Franco 1997, 1999, and Dunne 2004.
The rhetorical implications of this sort of appeal to the Buddha’s embodiment of epistemic excellence are
well known. They concern the various models and proofs of omniscience, notwithstanding intramural
debates about what exactly constitutes the content of such omniscient (hence epistemically warranted)
states: the nature of things, their reality, or the knowledge of all things whatsoever. See McClintock
2010: chapters 4 & 5 for a detailed study of these proofs and their epistemic and rhetorical implications.
Christian Coseru
27
inference (accepted by most Indian schools of thought) and at the same time hold on to
tradition by granting the foundational figure (or text) privileged epistemic status.
Instead, I want to focus on the metaphysical considerations that ground the causal
account Dharmakīrti puts forward, and the specific conception of consciousness that thus
emerges. As noted above, the Buddhist holds that consciousness is but a stream of conscious
episodes of different types (visual, auditory, etc.). If conscious awareness is taken to be
made out of these discrete units, a dilemma arises: what accounts for the sense of continuity
of awareness and, more importantly, what could serve as the basis for the arising of each
instance of cognitive awareness from one moment to the next? The bundle theory stipulates
that every phenomenon is part of a complex causal web. Indeed, the Sanskrit notion of
skandha (lit., “heap”) captures rather well the aggregated nature of phenomena – something
fashioned by the collective combination of multiple causes and conditions (as Vasubandhu
glosses it in AKBh ad I, 7). The constitutive factors themselves exist only as part of a
causal continuum of interdependently arising phenomena. Of course, not all the constitutive
factors that ground causation contribute in equal measure: some are basic or necessary
and some are merely contingent. The Cārvāka claims that the body alone is the source of
cognition. But on the aggregate model of personal identity, the body is just one among
the five constitutive factors of agency. The principle that establishes effects as markedly
different from their cause or as not pre-exiting in the cause (asatkāryavāda), which Sāṃkhya
philosophers likewise confront, runs counter to empirical evidence. We observe that like
causes like: cows give birth to calves, and fermented milk yields yoghurt. Atypical cases,
such as the caterpillar’s metamorphosis into a butterfly, are just the exceptions that prove
the rule. Central to this model of causation is the so-called principle of “similar kind(s)”
(sajāti), which demands that phenomena arise not in an arbitrary manner, but thorough
homogeneous causal chains.
On this principle, then, cognitive awareness cannot arise from something non-conscious,
such as the physical body. As Dharmakīrti notes (PV II vv. 35–36a), there could be “unwarranted consequences” (atiprasaṅga) for presupposing otherwise, even as he does not spell
out what those consequences might be. One possible interpretation is that Dharmakīrti is
committed to a strict ontological difference between “cause” (kāraṇa) and “condition” or
“conditioning factor” (pratyaya): the former can only give rise to a specific type of effect,
while the latter can serve as a basis for the arising of multiple effects. The acorn can only
grow into an oak tree, but the same soil and climactic conditions may provide support for
various tree species.
On the reductionist Abhidharma model, all aggregate entities reduce to two kinds
of basic constituents: elemental atoms (dravya-paramāṇu) comprising the four primary
elements (mahābhūta), and the atomic totality (saṅghāta-paramāṇu), which includes the
secondary elements associated with each of the four sense spheres (with the exception of
sound). Although the position of the Sarvāstivāda – Vaibhāṣika differs somewhat from
that of the Sautrāntika (on which Dharmakīrti relies) in terms of how the lines are drawn
between primary and secondary existents, the sense spheres, as a domain of phenomenal
28
Consciousness and Causality: Dharmakīrti Against Physicalism
primitives, do belong in the Ābhidharmika’s ultimate ontology.13 It is worth noting that
this elemental domain is ascertained on the basis of different types of cognitive awareness
rather than as a mind-independent reality. For the Vaibhāṣika, sensible phenomena – say,
the experience of a pot’s color, weight and shape – are real, despite their reducibility to
more basic monadic elements of experience (e.g., phenomenal primitives). However, there
is an obvious tension between treating something both as a construct and as ultimately
real. In articulating the Sautrāntika position, Vasubandhu is keen to point out that shapes
and colors are ultimately real only insofar as they display a certain causal or pragmatic
efficacy, that is, only to the extent that they generate the appropriate cognitive event: in the
case of shape and color, a corresponding visual experience.14 In perceiving a pot, it is not
the pot itself that serves as the basis for the arising of the cognitive event but rather the
causal efficacy of material elements and phenomenal primitives.
Here too conditioning factors play an important role. Under certain conditions something solid may become liquid, like heat causing the melting of a block of ice into water.
Given the speculative nature of Abhidharma metaphysics, there should be no surprise in
finding disagreements about the specific ways in which properties attach to each aggregated
entity. For instance, while for the Vaibhāṣikas entities borrow their physical properties from
the elements themselves, Sautrāntikas take them to be present only as mere potentialities.
In a block of ice, the fire element is only potentially present, for without it, ice cannot
melt into water. Dharmakīrti works out this account of causal efficacy in terms of the strict
regularities that must obtain between elements in a causal series. These regularities act as
a kind of “restriction in causal potential” (śaktiniyama)15 – a notion that Dharmakīrti uses
to argue for the limited or restricted efficacy of causal elements. For instance, a lotus seed
cannot produce a cow and oil cannot be extracted from sand. The so-called essential nature
(or ‘nature-svabhāva’) of the causally efficient element in a causal chain suggests that
entities are not simply the product of a given causal chain or causal complex (hetusāmagrī ).
Rather, they are the product of specifically active elements within that chain and of the
conditions that make it possible for those active elements to manifest their potentiality.16
However, regardless of whether ice melts because an internal principle of preponderance
governs the transformation of physical substances, or because conditioning factors bolster
a particular causal chain of events, it simply cannot be the case that yoghurt could just as
easily come from clay as from milk and a gilt could give birth to a calf just as naturally as to
13
14
15
16
Atomism – the notion that matter reduces to some elemental constituents – finds expression for the
first time in Dharmaśrī’s Abhidharmahṛdaya (2nd c. C.E.), before receiving extensive treatment in the
Mahāvibhāṣā.
See, for example, AK k. 10–13 and AKBh ad cit., and discussion in Hattori 1988: 39–41 and Ronkin
2005: 56–59.
This notion appears in several places on the Pramāṇavārttika and its autocommentary (see, e.g., PV
I.43, I.195, and I.255 and PVSV ad cit.). As John Dunne notes, summarizing Dharmakīrti’s position,
the point of the restriction is to tie the causally efficient element to its specific causal antecedent: “it is
impossible for an apple seed to produce certain types of effects because it is impossible for it to arise
from certain kinds of causes” (2004: 162).
On the distinction between ‘nature-svabhāva’ of an entity and its location in a causal totality, see
Steinkellner 1971: 185f. and Dunne 2004: 163f. Dunne’s translation of the relevant passage from PV
I.7 and PVSV ad cit., slightly altered, reads: “The arising of an effect that is inferred by way of a causal
complex is characterized as a svabhāva of that causal complex, because the [the capacity for] the effect’s
production does not depend on anything else.”
Christian Coseru
29
a piglet. Given the widespread belief in pre-modern India that sentience can have multiple
origins (e.g., egg-born, sprout-born, womb-born, and moisture-born), Dharmakīrti does
in principle concede that the material elements could serve as a basis for the arising of
cognition. But empirical observation also yields restrictions: the principle of preponderance
may well apply to all kinds, but a cow is not just a collection of elements with a certain
predominant property like solidity, heat, or capacity to produce milk. Nor is it a conceptually
constructed entity like a forest, or a cart, that is analytically reducible to its constitutive
parts. There must be more than just the configuration of matter that accounts for the arising
of cognitive awareness (PV II vv. 37–38). The structural asymmetries within this aggregate
conception of personal identity are becoming obvious.
3. Causation and emergence
The canonical literature presents us with a standard formula for the dependently arising
phenomenon of consciousness:
Dependent on the eye and forms, visual-consciousness arises. The meeting of
the three is contact. With contact as condition there is feeling. What one feels,
that one perceives. What one perceives, that one thinks about. What one thinks
about, that one mentally proliferates. With what one has mentally proliferated
as the source, perception and notions resulting from mental proliferation beset
a man with respect to past, future, and present forms cognizable through the
eye.17
On this standard account, a specific type of consciousness accompanies each of the sense
modalities. In this specific case, what is occasioned is an instance of visual awareness.
No one constitutive factor in this nexus of interactions has causal priority: instead, the
association between perception and thinking results from the habitual tendency of the
mind towards conceptual proliferation. Note that while sense, object, and conscious apprehension come together as a consequence of past habituations and other conditioning
factors, the ensuing cognitive awareness is both sustained by and sustains these factors.
As stated, the principle of dependent arising would place consciousness alongside other
factors in the causal web in an interrelated, symmetric, and mutually supportive system
of relations. The sense would be as necessary for the arising of cognitive awareness as
this awareness would for the optimum functioning of the organism. But, as Dharmakīrti
points out, the class of internal mental states that comprise thoughts, memories, and affects
does not appear to depend on the senses. Introspective awareness (manovijñāna), which
Dharmakīrti (following Dignāga) categorizes as a distinct type of perception, specifically
“mental perception” (mānasa-pratyakṣa), does not depend on the sensory systems. Rather,
following the Yogācāra analysis of the afflicted mind (kliṣṭa-manas), Dharmakīrti takes
introspective awareness to be mired in the same confusion and ignorance that can also
cloud understanding and render sense perception ineffective. A deluded mind is incapable
of providing reliable testimony about matters of fact: the believer in ghosts is more likely
17
MN I, 111–112 in Ñāṇamoli/Bodhi 2001: 203.
30
Consciousness and Causality: Dharmakīrti Against Physicalism
to perceive the rustle in the bushes as a shadowy figure stalking them, rather than a gust of
wind.
The relational asymmetry between cognitive awareness and the other contributing
factors becomes obvious: for Dharmakīrti, introspective awareness cannot thus arise from
“the body together with all the senses” because its occurrence is observed even when one
or more of the senses are impaired (PV II v. 47). Dharmakīrti would welcome the wealth
of empirical evidence from clinical neuroscience about such phenomena as the “locked-in
syndrome” or the persistence of “minimal consciousness” in patients diagnosed as being
in a vegetative state.18 This sort of evidence, it seems, lends support to his thesis that
sentience, as a minimally conscious state, enjoys a certain degree of causal autonomy
from more specific higher-order modes of cognitive awareness.19 It also suggests that,
given the difficulty of diagnosing whether a patient is in a minimally conscious state rather
than a permanent vegetative state, the distinction between unconscious mental states and
states of consciousness with minimal cognitive and behavioral function is less clear than it
may seem. Rather than being unconscious, a cognitively and behaviorally non-responsive
individual could simply be minimally conscious. Most importantly, in the absence of a
better understanding of the tight correlation between mental and physical (e.g., brain) states,
such evidence sets the stage for developing a wider conception of causality than physicalism
allows.
Recall that Dharmakīrti does not reject the idea that the body can serve as a support
for cognition. Indeed, he acknowledges that in some circumstances the occurrence of a
sensation, say of pain, can simply be the result of a wound in the body. The pain has both
qualitative features or qualia (sharp, stingy) and intentional content, insofar as it discloses
the body as the locus of tissue damage. Furthermore, the co-occurrence of bodily processes
and specific mental states at best suggests that the body is a contributing factor in the arising
of cognition, not that it actually causes it. Presupposing otherwise would entail that there
is a closer causal connection between cognition and the body than even the physicalist is
willing to admit, one that ensures, for instance, that cognition could persist in the body
after death (PV II v. 51).
As Taber (2003: 492) notes in pursuing a similar line of inquiry, what we see here is a
clear example of Occam’s Razor: Dharmakīrti argues against taking cognitive awareness to
be a product of bodily functions because he thinks the mental domain is the natural place
18
19
For a descriptive account of the varieties of locked-in syndrome, see Bauer/Gerstenbrand/Rumpl
1979 and Laureys et al. 2005. A detailed review of the literature on minimal states of consciousness
experienced by coma patients, which also puts forward a model of consciousness that takes it to be an
emergent property of the collective functioning of widespread frontoparietal brain networks, is found in
Laureys/Schiff 2012.
Working out the implications of empirical research on borderline states of consciousness for a theory of
consciousness, Bayne, Hohwy and Owen (2016) point out the inadequacy of models that equate global
states of consciousness (e.g., alert wakefulness, dreaming, and such comatose conditions as vegetative
and minimally conscious states) with levels of consciousness. Unlike local states of consciousness,
typically associated with the contents of consciousness (e.g., perceptual states, thoughts, and desires),
global states are supposed to indicate that consciousness comes in degrees. But, as Bayne et al. conclude,
being conscious, much like being married or being a bachelor, does not come in degrees. Rather, being
conscious, at a minimum, is not merely a matter of occupying the first-personal stance, but of having
various cognitive capacities available for perceptual and behavioral tasks.
Christian Coseru
31
for cognitive awareness. The mental domain is sufficiently complex to support its own
operations. No need, therefore, to bring in an incongruous factor such as the body, which
obeys a different set of laws, to explain the arising and specific operations of cognitive
activity (PV II vv. 33–44). Nothing is closer to each instance of cognitive awareness than a
cognition immediately preceding it. Why not postulate that each state of cognitive awareness
serves as the antecedent cause for cognition? Hence Dharmakīrti’s dictum: “let only what
is observed as the cause always be considered the cause” (PV II v. 44cd). And what is
observed is the constant stream of mental states.
Furthermore, as the literature on meditative absorption testifies, while this stream of
cognitive awareness can be altered, it cannot be halted. In the Bhāvanākrama I, 212, for
instance, Kamalaśīla argues against those who think meditative cultivation essentially
amounts to casting aside all mental activity and achieving a state of unconscious concentration (asaṃjñīsamāpatti). What is achieved is a state of non-conceptual awareness, rather
than the cessation of all mental activity (manasikāra). Consciousness, it seems, persists so
long as the body is alive (even as the relation between life and mind remains somewhat
unclear). Indeed, concepts such as bhavaṅga-citta or “life-continuum mind” hint at an
intimate correlation between mind and life, despite the largely speculative nature of the
Abhidharma account in which it occurs.20
Dharmakīrti’s attempt to carve out a space for the autonomy of cognition from material causation, while retaining the efficient-causal model, showcases not only his logical
ingenuity but also his keen phenomenological sense. We may wonder, then, why he allows
his observations to be constrained by doctrinal considerations, rather than deferring to the
empirical evidence alone. The Cārvāka physicalist too is a keen observer, but – not saddled
with the sort of doctrinal commitments that press the Buddhist into a defense of rebirth –
paints a starker picture of the human condition. Just like fermented grain yields a liquid with
the capacity to intoxicate, so also consciousness must be regarded as nothing more than a
product of the type of material organization that is constitutive of biological organisms.
The Cārvāka’s response to the principle of similar kinds (sajāti) is a new conception of
causality: emergentism.21 Mental properties are ontologically novel emergent properties
that supervene on the physical.
20
21
The Pāli Abhidhamma typically glosses bhavaṅga as a mode or function of consciousness that captures its
receptive or transitional state, as when attention shifts from one object to another. The principal sources
(Buddhaghosa’s Visuddhimagga and Atthasālinī, Buddhadatta’s Abhidhammāvatāra, and Anuruddha
Abhidhammatthasaṅgaha) are quite explicit that, like all other aspects of consciousness, bhavaṅga too
is intentionally constituted, that is, it is consciousness of something. However, because it stands for
consciousness in its liminal state, its intentional content is not available for reflection. It is the kind of
consciousness that persists in the interval between more alert states of mind; hence, its association with
the continuum of life. Cf. Gethin 2005.
On the appropriation of ‘emergentism’ as a category for describing the Cārvāka account of consciousness
and cognition, see Ganeri 2011 and Coseru 2017.
32
Consciousness and Causality: Dharmakīrti Against Physicalism
4. The challenge of physicalism
As we noted above, Dharmakīrti’s statement about the relation between consciousness and
causal explanation points to the autonomy of cognition. His view is most clearly stated in
the following verse:
[Nor are the senses, or the body together with the senses, the cause of cognition,
for] even when every single one of the senses is impaired, the [corresponding]
cognitive awareness is not impaired. But when [the cognitive awareness] is
impaired, their (i.e., the senses’) impairment is observed.22
But cognitive awareness is obviously in some kind of dependency relation to the body, as
demanded by the causal principle of dependent arising. For instance, visual awareness can
only emerge in organisms that are sensitive to light. The Cārvāka does grant that cognitive
awareness can have novel properties not observed in the material substratum (the body)
that serves as its basis. But unlike the dualist picture the Buddhist paints, the Cārvāka
contends that as an awareness of a certain type (visual, auditory, etc.) consciousness
must be related to the body’s specific functionality in the respective cognitive domain.
Given that consciousness takes the form of an apprehension of objects (that is, given its
inherently intentional structure), and given that apprehension only occurs in specific modes
of cognizing such as perceiving, imagining, or remembering, consciousness can be present
neither when the sensory systems are not yet developed (e.g., in the embryonic stage) nor
when they are not responsive (e.g., in a state of coma). Is there a causal criterion for the
presence of consciousness? And, more importantly, can the Buddhist answer the challenge
of physicalism without appealing to the kind of evidence (e.g., “the remembrance of past
lives”) the Cārvāka would simply not accept?
I have discussed the Cārvāka’s objection to the autonomy of consciousness in detail
elsewhere (see Coseru 2017), specifically with reference to Śāntarakṣita’s Tattvasaṃgraha,
and Kamalaśīla’s commentary thereon. Here I will simply attempt to restate the arguments in
a formal description.23 The physicalist’s objection to any presumed continuity of awareness
is framed by some easily recognizable arguments, all of which have key Buddhist tenets as
their premise, but draw different conclusions. The first argument may be summarized as
follows:
P1: An individual is nothing but a bundle of aggregates.
P2: Aggregates, including consciousness, are reducible to their material substrata (viz., atoms).
C: ∴ Conscious awareness must be an emergent property of a certain type of
material aggregation (typical for biological organisms).
22
23
PV II. v. 39: pratyekam upaghāte ’pi nendriyānāṃ manomateḥ / upaghāto ’sti bhaṅge ’syās teṣāṃ
bhaṅgaś ca dṛśyate.
Śāntarakṣita’s summary of the Cārvāka position on consciousness and causation is found in TS vv.
1857–1870. For a detailed study of the relevant Cārvāka fragments that survive, see Bhattacharya 2009:
33–43 and Franco 1997: 253–256.
Christian Coseru
33
The Cārvāka is comfortable with the aggregate conception of personal identity, and welcomes its epistemic consequences (only irreducible elements are ultimately real). Since
consciousness is an aggregate phenomenon, essentially reducible to a stream of momentary
conscious events, it too must be the product of a specific kind of material organization,
perhaps the sort typical of organisms endowed with a nervous system. The Buddhist faces
two important problems: first, that of explanatory sufficiency: why shouldn’t material organization with its emergent properties and functionality suffice as an explanatory account for
the arising of consciousness? Second, that of causal relevance: what evidence is there that
consciousness, as the Buddhist claims, generates cognitive activity in ways that cannot be
accounted for by material causes and conditions? Empirical observation, it seems, yields
an altogether different picture, one where conscious states are tightly correlated with bodily
processes.
The second argument extends this critique, taking into account the intentional structure
of awareness, and pointing to its conditioning factors, specifically that cognition appears to
be tied to the development and maturation of the body.
P1: Consciousness is always consciousness of (i.e., it is intentional).
P2: Conscious apprehension occurs only in specific modes (perceiving, remembering, etc.).
C: ∴ Consciousness cannot be present if the cognitive systems are undeveloped
(embryonic stage) or unresponsive (comatose state).
Here the Cārvāka admonishes against taking adult experience to be normative for consciousness at all stages of biological development. For in utero, when the cognitive systems
are not yet formed, there is neither vision nor visual object. What sense would it make
then to talk about visual awareness without a visual system? Of course, the physicalist
is in no better position than the Buddhist to explain the arising of consciousness. But at
least, from the physicalist’s standpoint, recognizing that material causation must play a key
role in whatever properties or characteristics consciousness exhibits, is a step in the right
direction.
Finally, the third argument invokes the principle of positive and negative co-variance
(anvaya-vyatireka) to make the case that consciousness can only become manifest in one
cognitive chain:
P1: Different types of bodies (of human and non-human animals), and different
tokens of the same body, manifest different types of consciousness.
P2: Granted the principle of positive and negative co-variance.
C: ∴ Consciousness cannot apprehend that which is contrary to its mode of
realization (consciousness can only be associated with the one cognitive chain
of either human or non-human animals, that can serve as its basis).
If the arising of consciousness is grounded in the body, then it is specific to each body both
within and across species. That is, for the Cārvāka every concrete mental state arises from
a corresponding bodily process or function. The persistence of conscious awareness within
34
Consciousness and Causality: Dharmakīrti Against Physicalism
a given mental stream is only invariably concomitant with that stream as a specifically
embodied individual. At least in Śāntarakṣita’s reconstruction, the Cārvāka appears to hold
a version of the token identity theory of mental states: every concrete mental particular (e.g.,
a given sharp pain) can be identified with some concurrent physical (or neurophysiological)
state. This view is supported by two key principles that inform the Cārvāka’s philosophy
of mind: (i) the human being is just an aggregate of the four elements, the combination
thereof instantiating its mental properties; and (ii) mental properties thus instantiated
result from the specific ways in which the elements combine.24 But as Kamalaśīla notes in
his commentary,25 there is disagreement among the Cārvākas on how to interpret earlier
statements (attributed to Bṛhaspati) about the precise ways in which these instantiated
mental properties relate to the elements, either taken in isolation or combined.
Given these considerations, the Buddhist faces a dilemma: the aggregated conception
of personal identity seems to support the physicalist position that consciousness is an
emergent property of certain types of material organization. Consciousness cannot be both
part of this causal web and just an instance in a beginningless stream of conscious events.
Can the Buddhist answer the physicalist challenge while retaining a causal-explanatory
framework in accounting for the relation between cognition and the body? In a detailed
analysis of the explanatory role of causal generation, Kamalaśīla identifies an important
difference between the operations of causality in the physical domain and the limits models
of material causation face when extended to consciousness and cognition.
Whenever an effect is dependent on a collection of causes and conditions it
does not arise when even one of these conditions is absent, for it would not be
dependent upon them, if it did. It could be said, “All the atoms insofar as they
occur in [its] proximity are the cause of cognition.” In that case a difference
should be observable between the effect produced by a non-deficient cause
and that produced by a deficient cause, as the two are different. Otherwise,
a distinction in the [capacity of the] cause [to bring about different effects
depending on its fitness] would be futile. In effect, when a cause that has been
perfect in all its aspects becomes defective in some respect, it does not occasion
a difference in the mind and that which is mental (mano-mati), on account of
the fact that preceding auditory and other kinds of impression continue intact
[in the mental stream].26
We see here a clear acknowledgement that cognitive awareness depends upon the efficacy of
all underlying causal factors (perceptual, volitional, dispositional, etc.), and the recognition
that, in turn, these factors reduce to their causal totality (kāranasāmagrī ). On a strict
24
25
26
See Bhattacharyya 2002, and discussion in Ganeri 2011: 5.
See in particular, TSP ad TS 1885–1866 (in Shastri 1968: 450), and discussion in Coseru 2017.
TSP ad TS: v. 1886 (in Shastri 1968: 450): na hi sāmagrīpratibaddhaṃ kāryam anyatarābhāve bhavati;
tat pratibaddhasvabhāvatvahāniprasaṅgāt / atha yathā sanidhānam sarve ’pi caitanyasya hetavaḥ?
evam tarhi vikalāvikalāṅgadehajanitayor viśeṣena bhavitavyam kāranabhedāt, anyathā kāryasya bhedo
nirhetukaḥ syāt. na vā ’vikalāṅgasya sataḥ paścād vikalāṅgatāyām upajātāyām kaścin manomater
viśeṣo ’sti; śrutādisamskārasya tadānīm apy avikalasyaivānuvṛtteḥ.
Christian Coseru
35
account of causal generation, cognitive error would track closely deficient causation.27 But
that does not always happen. One might perceive a sparkling lake where there are only
naturally occurring conditions for an optical illusion. This perceptual illusion is not simply a
case of misapprehension, for the illusion persists even after it has been disambiguated (that
is, after one has come to apprehend the appearance of the lake as a mirage). What the error
argument targets is strict causal generation: the notion that each mental state is instantiated
by a suitably relevant combination of physical elements and processes. The persistence of
perceptual illusion even after disambiguation, and the possibilities of effective action such
disambiguation opens up (not chasing after a mirage), work against the strict causal model
of the Cārvāka physicalist, which reduces human agency to changes in the microphysical
structure of each individual.
When Dharmakīrti claims that a trustworthy cognition (avisaṃvāda) must also be
causally effective, he advances a different naturalistic account of cognition than the one put
forth by the Cārvāka, one that takes into account the intentional structure of awareness and
its phenomenal character: perception is not simply the apprehension of a unique particular
as such; rather it is the apprehension of a particular as perceived, which also discloses the
perceiver’s intentional stance. In the case of perceptual illusions such as mirages, it is not
only the object or content of the experience that gets disclosed, but also the perceiver’s
vantage point, who can ensure successful action through a shift in perspective.
5. Causality and the co-constitutive manifest
Has the Buddhist satisfactorily answered the challenge of physicalism? Before we attempt
an answer to this question, let us revisit once more Dharmakīrti’s contention that the senses
are rendered ineffective by an impaired consciousness. Regardless of whether we take him
to be arguing from a Sautrāntika or Yogācāra position,28 is it clear that even when he appears
to reject the notion that the intentional object is causally related to the experience of a
unique particular (as he does in PV III v. 320), Dharmakīrti is in fact pressing an important
phenomenological point: specifically, that considerations about the structure of awareness
must play a role in settling epistemological disputes. This point is necessary to support
his account of the efficacy of cognition. If one does not factor in the dual-aspect theory of
mental states in mapping out the relation between consciousness and causality, then one
cannot understand why causal explanation retains an element of ontological subjectivity.
The justification for taking reflexivity (svasaṃvedana, svasaṃvitti) as a condition for the
possibility of warranted cognition may indeed stem from Dharmakīrti’s commitment to
27
28
The assumption behind strict models of causal generation is that a suitably efficacious causal chain
generates each epistemically salient cognitive state: the state of quench is generated by water ingestion
or water metabolism. Cognitive error, as in the case of water mirages and such, results from a defective
cause: that is, the water in the mirage has the formal properties of real water (etc., reflectance) but lacks
the latter’s efficacy: illusory water does not quench thirst.
As Franco (1997: 87) notes, whether we attribute it to Dharmakīrti’s genius or the versatility of the
Sanskrit language, it is possible to read him as endorsing both the Yogācāra view of the luminosity
of the mind (prabhāsvaraṃ cittam) – which, consequently, means that one may have to take reflexive
awareness (svasaṃvedana) as the only warranted type of cognition – and the Sautrāntika notion that the
appearance of objects in cognition corresponds to eternal particulars.
36
Consciousness and Causality: Dharmakīrti Against Physicalism
the self-luminosity theory of consciousness. But what is important about this theory is
that it rules out the possibility of reducing cognition to subpersonal levels of explanation.
Walking is a complex motor skill, which, once learned, may appear involuntary. The same
can be said about language and basic perceptual function. But what ensures their epistemic
and pragmatic efficacy is the ongoing active presence of an implicit awareness that attends
to the object at hand.
The mind’s attentive capacity (manaskāra), then, makes a certain dimension of human
cognition not merely the effect of causal chains in the physical domain but a causal factor in
its own right in the domain of cognition. This self-reflexive aspect of cognition can remain
constant with respect to a given object of experience (say, a column of smoke), despite
it being prompted by a deficient cause (in this case, a dust column). On account of this
apparent variance, consciousness is neither entirely grounded in, nor explainable in terms
of, physical elements and processes. The Buddhist does not deny that cognitive states are
made manifest (abhivyajyate) when the body is present, only that their manifestation is to
be understood in strictly physical causal terms.
4. Conclusion or how to avoid the fallacy of ambiguity
Let me conclude by revisiting a question that was asked at the beginning of this paper:
are causal accounts of generation for material phenomena reconcilable with the seeming
irreducibility of consciousness? By reducing the analysis of consciousness and cognition
to transactions in the physical domain, the Cārvāka, much like contemporary physicalists,
is committing a category mistake: consciousness is a subjective phenomenon and thus its
manifestation cannot be accounted for in the impersonal language of causality for material
objects. Is it possible, then, that even Dharmakīrti, insofar as he seeks a causal explanation
for the epistemic reliability of certain cognitions, is guilty of the very charge he levels
against the physicalist? As I argued elsewhere (Coseru 2017), ontological reductionism
is not the same as epistemological reductionism. Indeed, there is a systematic ambiguity
between the ontological and the epistemic sense in using the word ‘empirical’ and its
cognates to capture causal relations.29 Sometimes ‘empirical’ stands for contingent states
of affairs, and sometimes for a method that can be used to establish something as factual.
Facts about one’s subjective experience, for instance, are not empirically accessible in
the way that facts about external objects (or their atomic constituents) are. The basis for
the epistemic subjective-objective distinction is an ontological distinction in modes of
existence.
In short, ontological subjectivity is no bar to epistemic objectivity. Consciousness,
unlike its contents, is implicitly manifest. Conscious awareness does not become manifest
by being reflected upon, as do its specific contents (which are only available when attended
to in perception or brought under a specific concept). That is to say: consciousness has an
observer-independent status. I do not become conscious by observing the occurrence of
my mental states. Rather, I become aware of the contents of my experience by virtue of
being conscious. Causality, on the other hand, is an observer-relative phenomenon: the
very notion of ‘event’ presupposes an observer. Events thus stand in a particular kind of
29
This distinction is examined at length by Searle (2015: 74ff.).
Christian Coseru
37
relation to their antecedents only to the extent that there is a conception of causality in
place. That causality should be an observer-relative phenomenon does not mean, however,
that it is arbitrary. Rather, its observer-relative status simply suggests that it contains an
element of ontological subjectivity.
Dharmakīrti (and his successors) may well admit that aggregated entities reduce to their
ontological primitives, which alone are real. But causally describable series of events are
not incompatible with treating such basic events as irreducibly mental. Buddhist conceptual
reductionism about consciousness, therefore, does not necessarily entail physicalism.
References and abbreviations
Primary sources
AK Abhidharmakośa: See AKBh.
AKBh Abhidharmakośabhāṣya: Abhidharmakośa and Bhāṣya of Āchārya Vasubandhu
with Sphuṭārtha Commentary of Āchārya Yaśomitra, ed. P. Pradhan. Vārāṇasī 1975.
Bhāvanākrama Bhāvanākrama: Āchāryakamalaśīlapraṇītaḥ Bhāvanākramaḥ, ed. Āchārya
Jamchen Namdol. Sarnath 1997.
MN Majjhimanikāya: Majjhima-Nikāya, vols. I–II, ed. Vilhelm Trenckner; vol. III, ed. R.
Chalmers. Oxford 1993–2004.
PS Pramāṇasamuccaya: Jinendrabuddhi’s Viśālāmalavatī Pramāṇasamuccayaṭīkā: Chapter I, ed. E. Steinkellner, H. Krasser, and H. Lasic. Vienna 2005.
PSV Pramāṇasamuccayavṛtti: See PS.
PV Pramāṇavārttika: Pramāṇavārttika, Ācāryamanorathanandivṛttiyutam (Dharmakīrtinibandhāvaliḥ), ed. Swami Dwarikadas Shastri. Vārāṇasī 1968.
PVSV Pramāṇavārttikasvavṛtti: The Pramāṇavārttikam of Āchārya Dharmakīrti with
the Commentaries “Svopajñavṛtti” of the Author and “Pramāṇavārttikavṛtti” of
Manorathanandin, ed. Ram Chandra Pandeya. Delhi 1989.
SN Saṃyuttanikāya: Saṃyutta Nikāya, ed. L. Feer. London 1975–2006.
TS Tattvasaṃgraha: Tattvasaṃgraha of Āchārya Shāntarakṣita with the Commentary
“Pañjikā” of Shrī Kamalaśīla, vols. I–II, ed. Swami Dwarikadas Shastri. Vārāṇasī
1968.
TSP Tattvasaṃgrahapañjikā: See TS.
Secondary sources
Arnold 2008 D. Arnold, Dharmakīrti’s Dualism: Critical Reflections on a Buddhist Proof
of Rebirth. Philosophy Compass 3.5 (2008) 1079–1096.
Bauer/Gerstenbrand/Rumpl 1979 G. Bauer, F. Gerstenbrand, and E. Rumpl, Varieties
of Locked-in Syndrome. Journal of Neurology 221 (1979) 77–91.
Bayne/Hohwy/Owen 2016 T. Bayne, J. Hohwy, and A. D. Owen, Are there Levels of
Consciousness? Trends in Cognitive Science 20.6 (2016): 405–413.
Bhattacharya 2002 R. Bhattacharya, Cārvāka’s Fragments: A New Collection. Journal
of Indian Philosophy 30.6 (2002) 597–640.
Bhattacharya 2009 R. Bhattacharya, Studies on the Cārvāka/Lokāyata. Firenze 2009.
38
Consciousness and Causality: Dharmakīrti Against Physicalism
Carpenter 2015 A. Carpenter, Persons Keeping Their Karma Together: The Reasons for
the Pugdalavāda in Early Buddhism. In: The Moon Points Back: Analytic Philosophy
and Asian Thought, ed. K. Tanaka, Y. Deguchi, J. L. Garfield, and G. Priest. New
York 2015, 1–44.
Clark 1997 A. Clark, Being There: Putting Brain, Body, and World Together Again. Cambridge, Mass. 1997.
Coseru 2017 C. Coseru, Consciousness and Causal Emergence: Śāntarakṣita Against
Physicalism. In: Oxford Handbook of Indian Philosophy, ed. J. Ganeri. Oxford 2017,
360–378. DOI: 10.1093/oxfordhb/9780199314621.013.24.
Cox 1995 C. Cox, Disputed Dharmas – Early Buddhist Theories on Existence – An Annotated Translation of the Section on Factors Dissociated from Thought from Saṅghabhadra’s Nyāyānusāra. Tokyo 1995.
Dreyfus 1979 H. Dreyfus, What Computers Still Can’t Do: A Critique of Artificial Intelligence. New York 1979.
Duerlinger 2003 J. Duerlinger, Indian Buddhist Theories of Persons: Vasubandhu’s “Refutation of the Theory of a Self.” London/New York 2003.
Dunne 2004 J. Dunne, Foundations of Dharmakīrti’s Philosophy. Sommerville, Mass.
2004.
Franco 1997 E. Franco, Dharmakīrti on Compassion and Rebirth. Vienna 1997.
Franco 1999 E. Franco, Two Circles or Parallel Lines? In: Dharmakīrti’s Thought and Its
Impact on Indian and Tibetan Philosophy: Proceedings of the Third International
Dharmakīrti Conference, Hiroshima, November 4–6, 1997, ed. Shoryu Katsura.
Vienna 1999, 63–72.
Gallagher 2005 S. Gallagher, How the Body Shapes the Mind. New York 2005.
Ganeri 2011 J. Ganeri, Emergentisms, Ancient and Modern. Mind 120.479 (2011)
671–703.
Gethin 2005 R. Gethin, Bhavaṅga and rebirth according to the Abhidhamma. In: Critical
Concepts in Religious Studies, ed. Paul Williams. New York 2005, 159–181.
Gombrich 1996 R. F. Gombrich, How Buddhism Began: The Conditioned Genesis of the
Early Teachings. London/Atlantic Highlands, N. J. 1996.
Hamilton 1996 S. Hamilton, Identity and Experience: The Constitution of the Human
Being According to Early Buddhism. London 1996.
Hattori 1988 M. Hattori, Realism and the Philosophy of Consciousness-Only. The Eastern
Buddhist 21.1 (1988) 23–60.
Hayes 1993 R. P. Hayes, Dharmakīrti on punarbhava. In: Studies in Original Buddhism
and Mahayana Buddhism, vol. 1, ed. Egaku Maeda. Kyoto 1993, 111–130.
Hurley 1998 S. L. Hurley, Consciousness in Action. Cambridge, Mass. 1998.
Hutchins 1995 E. Hutchins, Cognition in the Wild. Cambridge, Mass. 1995.
Kataoka 2005 K. Kataoka, Manu and the Buddha for Kumārila and Dharmakīrti. In:
Religion and Logic in Buddhist Philosophical Analysis: Proceedings of the Fourth
International Dharmakīrti Conference, Vienna, August 23–27, 2005, ed. Helmut
Krasser, Horst Lasic, Eli Franco, and Birgit Kellner. Vienna 2005, 255–270.
Laureys et al. 2005 S. Laureys, F. Pellas, P. Van Eeckhout, S. Ghorbel, C. Schnakers, F.
Perrin, J. Berré, M.-E. Faymonville, K.-H. Pantke, F. Damas, M. Lamy, G. Moonen,
Christian Coseru
39
and S. Goldman, The locked-in syndrome: what is it like to be conscious but paralyzed
and voiceless? Progress in Brain Research 150 (2005) 495–511.
Laureys/Schiff 2012 S. Laureys and N. D. Schiff, Coma and consciousness: Paradigms
(re)framed by neuroimaging. NeuroImage 61.2 (2012) 478–491.
McClintock 2010 S. McClintock, Omniscience and the Rhetoric of Reason: Śāntarakṣita
and Kamalaśīla on Rationality, Argumentation, and Religious Authority. Boston
2010.
Nagatomi 1959 M. Nagatomi, The Framework of the Pramāṇavārttika, Book I. Journal
of the American Oriental Society 79 (1959) 263–66.
Ñāṇamoli/Bodhi 2001 B. Ñāṇamoli and B. Bodhi, The Middle Length Discourses of the
Buddha: A Translation of the Majjhima Nikāya. Boston 2001.
Noë 2004 A. Noë, Action in Perception. Cambridge, Mass. 2004.
Pradhan 1975 See AKBh.
Ronkin 2005 N. Ronkin, Early Buddhist Metaphysics: The Making of a Philosophical
Tradition. London 2005.
Searle 2015 J. Searle, Seeing Things as They Are: A Theory of Perception. New York
2015.
Shastri 1968 See TS.
Siderits 2003 M. Siderits, Personal Identity and Buddhist Philosophy. Aldershot 2003.
Siderits 2011 M. Siderits, Buddhas as Zombies: A Buddhist Reduction of Subjectivity. In:
Self, No Self : Perspectives from Analytical, Phenomenological, and Indian Traditions,
ed. Mark Siderits, Evan Thompson, and Dan Zahavi. Oxford 2011, 308–332.
Steinkellner 1971 E. Steinkellner, Wirklichkeit und Begriff bei Dharmakīrti. Wiener Zeitschrift für die Kunde Südasiens 15 (1971) 179–211.
Steinkellner 2003 E. Steinkellner, Once More on Circles. Journal of Indian Philosophy
31 (2003) 323–341.
Taber 2003 J. Taber, Dharmakīrti Against Physicalism. Journal of Indian Philosophy 31
(2003) 479–502.
Thompson 2007 E. Thompson, Mind in Life: Biology, Phenomenology, and the Sciences
of Mind. Cambridge, Mass. 2007.
Tillemans 1993 T. J. F. Tillemans, Persons of Authority. Stuttgart 1993.
Varela/Thompson/Rosch 1991 F. J. Varela, E. Thompson, and E. Rosch, The Embodied
Mind: Cognitive Science and Human Experience. Cambridge, Mass. 1991.
Vetter 1964 T. Vetter, Erkenntsnisprobleme bei Dharmakīrti. Vienna 1964.
Waldron 2003 W. S. Waldron, The Buddhist Unconscious: The Ālaya-vijñāna in the
Context of Indian Buddhist Thought. London, 2003.
Action Theory and Scriptural Exegesis in Early
Advaita-Vedānta (2): Maṇḍana Miśra’s Excursus on the
Buddha’s Omniscience1
by
Hugo David
1. Introduction
1.1 The proof of the Buddha’s omniscience occupies, as is well-known, a prominent
place in Indian Buddhist philosophy, and also constitutes an essential part of Mahāyāna
Buddhism’s dogmatic construction, both as an essential component of its soteriological
ideal and as a foundation for the transmission of Buddhist teachings through Scripture.2 Yet,
surprisingly, we do not find a systematic consideration of this topic in the ‘pramāṇa’-school
until a comparatively late date. Although both Dignāga and Dharmakīrti deal in some
detail with the perception of yogins,3 their consideration of the ‘fourth’ type of perceptual
cognition is done independently of the question whether a yogin could become omniscient
by this means.4 One must wait until the 8th century and the works of Śāntarakṣita (725–788),
1
2
3
4
This article is the partial outcome of the project “Language and Action in Early Brahmanism” carried out
at the Austrian Academy of Sciences (Vienna) in 2013–2016 and funded by the Austrian Science Fund
(FWF-project no. P25287 – G15). Research for this paper also received support from the Royal Society
of Great Britain (Newton International Fellowship). I thank Vincent Eltschinger, Sara McClintock and
Vincenzo Vergiani for their very useful remarks on an earlier version of this paper, and Cristina Pecchia
for allowing me to use a draft of her unpublished article on omniscience according to Dharmakīrti.
On the early history of the concept of omniscience in Buddhism, see in particular Pandey 1972, Griffiths
1990 and Kawasaki 1992. For an overview in English, see McClintock 2010: 23–35. As she notes,
Buddhist attitudes toward omniscience might have been far more ambivalent than is the case in Jainism,
especially in the early period. On this point, see also Jackson 1991: 230–232.
For a synthesis on the question, see Eltschinger (2009), who discusses relevant passages of PS (1.6cd),
PVin 1 (27.7–28.8) and PV (3.281–286). Further references in Eltschinger 2009: 191 (n. 94).
By this I do not mean to say that Dharmakīrti, in particular, did not believe in the Buddha’s omniscience,
which is obviously not the case. His conception of yogipratyakṣa as a “vision of the [four] Nobles’
Truths” (āryasatyadarśana – see below § 3.5) also implies a form of ‘focused’ omniscience like the one
defined in PV 2.29–33. Nevertheless, omniscience did not constitute a major philosophical or religious
issue for him as it would for later Buddhist authors. As pointed out by Moriyama (2014: 17), “the
concept of omniscience does not occupy a special place in Dharmakīrti’s argument for establishing
the Buddha as pramāṇabhūta,” a point that sharply contrasts with later understandings of this attribute
of the Buddha. See also Franco 2011: 89 and Eltschinger (2005: 429–434), who explains this state of
affairs in terms of a conscious apologetic strategy: “Dharmakīrti aura élaboré une structure doctrinale
assez ouverte pour que les docteurs bouddhistes y lisent l’adhésion du maître à l’omniscience, mais
assez implicite pour que les pourfendeurs de ce dogme ne puissent lui en faire le reproche” (p. 434).
Equally significant is Dharmakīrti’s lack of interest for God’s omniscience in the section of the PV
refuting the existence of īśvara (PV 2.8–28). Specialists of Dharmakīrti still disagree as to what his
final position regarding total omniscience might have been, in particular in the case of the Buddha. See
Jackson 1991: 232–234, Eltschinger 2005: 434 and McClintock 2010: 135–138. The main passage
under discussion (PVin 2.55) is however unrelated to the definition of yogic perception formulated in
PVin 1 (see preceding note).
Birgit Kellner et al., eds., Reverberations of Dharmakīrti’s Philosophy: Proceedings of the Fifth International Dharmakīrti
Conference Heidelberg, August 26 to 30, 2014. Vienna 2020, pp. 41–76.
42
Maṇḍana Miśra’s Excursus on the Buddha’s Omniscience
Kamalaśīla (740–795) and Prajñākaragupta (750–810?)5 to find an articulate defence of
omniscience in the school claiming Dharmakīrti’s heritage, further developed by thinkers
like Śaṅkaranandana (800–980?), Jñānaśrīmitra (980–1040?) and Ratnakīrti (990–1050?).6
1.2 It is now generally admitted that the development of this new field of investigation
within the ‘epistemological’ school of Buddhism owes a lot to the critique propounded in
the 6th –8th centuries by Brahmanical thinkers, especially those belonging to the ritualistic school of Mīmāṃsā. From an early date, these thinkers regarded the possibility for a
person to grasp the totality of being perceptually as a serious threat to their conception
of the Veda as the unique means to know dharma, the ritual and ethical system defining
what we call ‘Brahmanism.’ As far as we know, the first Brahmanical thinker to present a
systematic attack against the belief in an omniscient being is Kumārila Bhaṭṭa (600–650?),7
both in the codanā-section (ad MīSū 1.1.2) of the Ślokavārttika (ŚlV) (vv. 110cd–155)
and in the corresponding fragments of the Bṛhaṭṭīkā (BṬ) quoted in Śāntarakṣita’s Tattvasaṃgraha (TS) (vv. 3127–3245).8 There, it is principally the idea of the Buddha as the
omniscient founder of a religion which is attacked, a fact that might explain the importance
accorded to Kumārila’s ideas in later Buddhist thought.9 Kumārila, however, was not the
only Mīmāṃsaka to have offered a critique of the Buddha’s omniscience by the beginning of
the 8th century. An important examination of this concept was also carried out by Maṇḍana
Miśra (660–720?) in the first, aporetical part – generally (though somewhat improperly)
known as ‘the pūrvapakṣa’ – of his treatise on action and injunction, the Vidhiviveka (ViV),
commented upon in the 10th century by Vācaspati Miśra in the Nyāyakaṇikā (NyK).10 In
5
6
7
8
9
10
On Śāntarakṣita and Kamalaśīla’s discussions of omniscience, see McClintock 2010. On Prajñākaragupta, see Moriyama 2014 and below § 4. Another early and potentially important document is the
(mostly unstudied) Sarvajñasiddhi of Śubhagupta (720–78), preserved only in Tibetan (see Bühnemann
1980: viii and Watanabe 1987).
For a preliminary edition and analysis of Śaṅkaranandana’s Sarvajñasiddhi (kārikās only), see
Eltschinger 2008. See also Eltschinger (2015: 323), who mentions, besides the Sarvajñasiddhi, a
shorter work called Sarvajñasiddhisaṃkṣepa, still unpublished. A study of fragments of Jñānaśrīmitra’s
otherwise lost Sarvajñasiddhi is found in Steinkellner 1977. For an in-depth study of Ratnakīrti’s treatise
of the same name, see Bühnemann 1980 and Goodman 1989. On the possible contribution of Jitāri
(9th –10th c.?) to this debate, see Bühnemann 1980: viii and Eltschinger 2008: 142.
Kumārila’s date is established in relation to that of Dharmakīrti, of whom he might have been an elder
contemporary. The date 600–650 proposed by Kataoka (20112 : 112) on the basis of Frauwallner’s
dating of Dharmakīrti (600–660) may have to be revised in case the great Buddhist logician should be
placed a few decades earlier, as proposed by Krasser (2012).
The numbering of verses in Śāntarakṣita’s Tattvasaṃgraha is that of Swami Dwarikadas Shastri’s edition
(Benares, 1968) (= TS), which slightly differs from that of Embar Krishnamacharya’s earlier edition
(Baroda, 1926, reprinted in 1988). For a detailed study of Kumārila’s critique of omniscience, see
Kataoka 2003a and Kataoka 2011. Equally central to this debate are vv. 26–33 of the pratyakṣa-section
of the ŚlV (ad Śābarabhāṣya 1.1.4), dealing with the perception of yogins (translation in Taber 2005:
54–56). Omniscience is again alluded to in ŚlV (saṃbandhākṣepaparihāra) 44ab and 114ab, but these
are simple reminiscences of the codanā-section, adding no new arguments.
Only on one occasion does Kumārila allude to omniscience as conceived by the Jains (ŚlV [codanā°]
141–142). No such allusion is found in the BṬ. The question of God’s omniscience is not touched upon
by Kumārila, and seems to have been introduced into Mīmāṃsā by Maṇḍana Miśra (see below § 1.3).
Other Mīmāṃsakas of the period might have been interested in the question of the Buddha’s omniscience,
but we do not have much evidence for this. The question whether Bhavya/Bhā(va)viveka, in the ninth
Hugo David
43
this work, the existence of an omniscient being is made the subject of a lengthy refutation
occupying more than a third of the whole pūrvapakṣa (ViV 15–25), which has not received
much attention so far though it constitutes one of our main sources for the history of this
debate in Mīmāṃsā before the time of Śāntarakṣita.11 My purpose in this essay is to give
the reader a first glimpse into this important text, concentrating on its treatment of the
Buddhist idea of omniscience.
1.3 Considering the ViV was written perhaps no more than a century after Kumārila’s
death, one is struck by how little Maṇḍana apparently owes to the old Master, whose works
he simply never quotes in that section. Kumārila’s almost exclusive preoccupation with
Buddhism is also not discernible in Maṇḍana’s work. For sure, the Brahmasiddhi (BS)
(presumably Maṇḍana’s last work)12 still mentions “the Buddha, wrongly believed to be
omniscient” (sarvajñābhimatabuddha) as the prototype of the false teacher of dharma.13
Vācaspati is also probably right in considering that the brief description of the Omniscient
in the prose introduction to ViVK 15 refers in priority to the Buddha:14 “an instructor of
11
12
13
14
chapter of his Madhyamakahṛdayakārikās (9.15–16), refers to Kumārila’s views or to those of an earlier
Mīmāṃsaka has been raised by Krasser (2012: 559–568), following a remark by Lindtner (2001: 3). I
fully agree with Krasser that “one can easily read Bhāviveka as refuting Kumārila” (p. 565), but the
passage in question is too brief to say much more. Krasser’s assumption (p. 567) that Bhavya targets an
early Mīmāṃsaka different from Kumārila while evoking his opponent’s belief in Jaimini’s omniscience
(Madhyamakahṛdayakārikā 9.163 and Tarkajvālā thereon; unavailable in Sanskrit, text and translation:
Krasser 2012: 566) requires additional caveats, for it holds only assuming Bhavya is perfectly accurate
in his critiques or always relies on a Mīmāṃsaka source, which is far from certain. A further unsolved
case is that of the two Mīmāṃsakas Sāmaṭa and Yajñaṭa (see McClintock 2010: 155–156, 225, 356–59).
These two enigmatic figures are known exclusively through their mention by Kamalaśīla, who ascribes
to them a series of opinions reported by Śāntarakṣita in TS 3246–60 (Japanese translation: Watanabe
1988). See TSP 1020.16–17: sāṃprataṃ sāmaṭayajñaṭayor matena punar api sarvajñadūṣaṇam āha;
“Now, he exposes yet another refutation of an omniscient [being], following the view of Sāmaṭa and
Yajñaṭa.” Some scholars suspected that these could be forged names, made up in reference to the Sāmaand Yajurveda (Kawasaki 1992: 255), but the fact remains that the opinions associated with them cannot
easily be traced to any known mīmāṃsaka text.
For a brief synopsis of the section, see Stern 1988: 28–45. The recent summary of the ViV by Potter
(2014: 289–295) unfortunately ignores that important section of the work, as did the small monograph
by Natarajan (1995) on which it is based. The only study of that part of the ViV-‘corpus’ so far is the
German translation of a fragment of Vācaspati Miśra’s NyK (ad ViV 15) dealing with yogic perception
by M. Pemwieser (1991). On this fragment, see also Steinkellner 1978. M. Biardeau’s unpublished
French translation of the whole pūrvapakṣa of the ViV and NyK, now kept at the archives of the Collège
de France in Paris, naturally includes a translation of ViV 15–25 and the corresponding NyK.
See David 2013: 281, n. 31.
BS 2.27cd–28 (84.9–10).
Although Vācaspati does not mention the Buddha by name, his characterisation of the Omniscient in
the NyK (445.5–446.6) has a definite Buddhist ring to it: for Vācaspati, the Omniscient is the “Blessed
Doctor” (bhagavān bhiṣaj – 445.5), “the Blessed One, whose all-pervasive compassion has become
his intimate goal, who has achieved the ultimate degree of detachment, untouched in the least by
[main] defilements such as desire or by minor defilements such as excitement or pride” (svārthībhūtaviśvavyāpikāruṇyo bhagavān vairāgyātiśayasaṃpanno mātrayāpi rāgādibhiḥ kleśair upakleśaiś ca
madamānādibhir aparāmṛṣṭaḥ [445.10–446.2]). As Stern (1988: 997) rightly points out, the description
of the NyK has an almost exact parallel in Vācaspati’s Nyāyavārttikatātparyaṭīkā ad Nyāyasūtra 2.1.68
(384.14–19, especially 384.16–19, which corresponds almost word for word to NyK 446.2–6), dealing
with the Naiyāyika definition of the āpta (see following note). There, we see the very same characteristics
44
Maṇḍana Miśra’s Excursus on the Buddha’s Omniscience
creatures (niyoktā bhūtānām), deserving to be obeyed, directly perceiving the means for
realizing the Supreme Good as prescribed [in the Scriptures] (sākṣātkṛtānuśravikaśreyaḥsādhana), who loves [to do] what is useful [to others] (hitakāma) and is omniscient
(sarvajña).”15 But in fact, only ViV 15 is directly concerned with Buddhism,16 while the
rest of the section discusses arguments from other schools17 and even contains the earliest
critique of divine omniscience in a Mīmāṃsā text, if not in Indian philosophy overall
(ViV 20–24).18 The tone of harsh religious polemics transpiring from Kumārila’s writings
15
16
17
18
applied to “the maker of products such as the body, the world, etc.” (tanubhuvanādilakṣaṇasya kāryasya
kartā), in other words to God, said to be “the knower of the real nature of all things” (samastavastutattvajña – 384.15) – i.e. omniscient –, “untouched by defilements and the store of the maturation of [past]
deeds” (kleśakarmavipākāśayāparāmṛṣṭa – 384.15) and “extremely compassionate” (paramakāruṇika
– 384.15). This striking similarity between the two descriptions might have motivated the assumption
by Stcherbatsky (1926) that Maṇḍana as well, in the ViV, is discussing the omniscience of ī śvara. I
find it unlikely, though, that Vācaspati should say in the NyK that God “achieved the ultimate degree
of detachment” (vairāgyātiśayasaṃpanna – NyK 445.10–446.1), a quality which, as far as I can see,
only fits a human being. This, together with the insertion immediately thereafter in the NyK of a long
defence of omniscience directly quoting Buddhist sources (NyK 447.2 – 458.7), makes me think that
Vācaspati, at least, is really speaking here of the Buddha.
ViV 15 (introduction): anuvidheyo niyoktā bhūtānāṃ sākṣātkṛtānuśravikaśreyaḥsādhanoa hitakāmaḥ
sarvajñaḥ (S 445.1–448.1 [≈ M 110.2–4/G 78.3–4]). a °anuśravika° S G: °anubhavika° M. This
description, in itself, has nothing specifically Buddhist. Of the four main characteristics of the omniscient
outlined in this passage (niyoktṛ[tva], sākṣātkṛta[…]sādhana[tva], hitakāma, sarvajña[tva]), the first
directly follows from the context of the ViV (see below § 2.1). The second and third may, of course, refer
to the first two epithets of the Buddha in the famous opening verse of Dignāga’s Pramāṇasamuccaya
(PS 1.1a): pramāṇabhūta (“authoritative”) and jagaddhitaiṣin (“seeking the benefit of the world”). But
Maṇḍana’s formulation of the second attribute – sākṣātkṛtānuśravikaśreyaḥsādhana (“having directly
perceived the means for realizing the Supreme Good as prescribed in the Scriptures”) – also recalls the
expression sākṣātkṛtadharma(n) (“having directly perceived the dharma(n)”) used to qualify Vedic ṛṣis
in the Nirukta (1.20), in Bhartṛhari’s Vākyapadīya (1.5 – SV 24.2) and in Bhavabhūti’s Uttararāmacarita
(see Ruegg 1994: 307–308). As is well-known, the quality of being sākṣātkṛtadharma(n) is also part
of Vātsyāyana/Pakṣilasvāmin’s definition of an āpta (“reliable speaker”) in the Bhāṣya ad Nyāyasūtra
1.1.7 and 2.1.68 (14.4 and 96.16). Pakṣilasvāmin’s commentary in sūtra 2.1.68 also mentions two
additional qualities of the āpta – bhūtadayā (“compassion for beings”) and yathābhūtārthacikhyāpayiṣā
(“desire to communicate about a real object”) (96.17) –, the first of which could very well be hinted
at by Maṇḍana while speaking of the Omniscient’s compassion toward others. Thus, although several
external clues plead for an identification of Maṇḍana’s omniscient being with the Buddha, one must
keep in mind that his description remains quite unspecific and corresponds to what most philosophical
traditions of his time would have expected of a reliable teacher, human or divine.
As Stern (1988: 28) rightly remarks, the section referred to by Vācaspati as bauddhasarvajñaparīkṣā
(“examination of the Omniscient [as conceived] by the Buddhists”) on two occasions in the NyK
(612.7–8 and 634.7, ad ViV 17) is certainly ViV 15.
Several of them are mentioned in Vācaspati’s commentary: “someone with a whiff of Prābhākara[doctrine]” (kaś cit prābhākaragandhī – NyK 570.10, ad ViV 16), Naiyāyikas (NyK 606.7–8, ad ViV 17;
679.9, ad ViV 21), “upholders of [the doctrine] of Svayaṃbhu [i.e. Patañjali’s Yoga]” (svāyaṃbhuvaḥ
– NyK 627.14, ad ViV 21). These identifications however testify to Vācaspati’s attempt to read in
the ViV a refutation of various ‘doctrines,’ while Maṇḍana’s progression is dialectic much more than
doxographic, so they should be taken with much caution.
See Moriyama (2014: 37), who also notes Maṇḍana’s influence on immediately later Buddhist thinkers
on that topic, in particular on Prajñākaragupta.
Hugo David
45
(especially the later ones)19 is also hardly discernible in Maṇḍana’s text. Thus we do not
find in the ViV anything comparable to Kumārila’s critical examination of non-Vedic
Scriptures (āgama) in the ‘appendix’ to Tantravārttika 1.3.4,20 a critique that also occupies
a substantial part of his discourse on omniscience in the ŚlV (codanā – 118–136) and
BṬ (= TS 3186–3213). In other words, it seems omniscience has become, in the hands of
Maṇḍana, less a matter for religious preachers and apologetes than for philosophers, the
latter more likely to find in his work tools to convince fellow dialecticians than the former
powerful incentives to attract faithful crowds and benevolent patrons.
1.4 The purpose of this study is to show that, in spite of all this, Maṇḍana plays a key role
in the early debate on the Buddha’s omniscience, and entertains an intense dialogue with
his two main predecessors in the field: Kumārila and Dharmakīrti. I will show, first, how
the beginning of ViV 15 can be read as a systematic rejection of Kumārila’s argumentative
strategy against omniscience, based on Dharmakīrti’s newly popularised logic (§ 2); second,
I will examine how Maṇḍana uses Kumārila’s main argument in terms of a relation between
perception and time to introduce a new type of epistemological consideration on the nature
and cause of the Omniscient’s cognition (§ 3). Finally, I will consider the possibility that
this evolution of the mīmāṃsaka discourse on omniscience influenced later stages of the
debate in Buddhism as well, by tracking possible echoes of Maṇḍana’s ideas in a slightly
later Buddhist work, Prajñākaragupta’s Vārttikālaṃkāra on Pramāṇavārttika 2.29 (§ 4).
2. Maṇḍana Miśra on non-apprehension and the Buddha’s speech: a
‘Dharmakīrtian’ response to Kumārila?
2.1 Every discussion of omniscience in Brahmanism must start from its prototype in
Kumārila’s ŚlV and BṬ. In these works, the great Mīmāṃsaka makes it clear that his
intention is not so much to prove that a human being cannot become omniscient (a possibility
he actually leaves open) but rather to maintain an absolute distinction between entities
that are accessible to the senses (aindriyaka) and others essentially beyond their reach
(nendriyagocaraḥ), like the relation between the elements of a Vedic sacrifice (dharma)
– actions, qualities and substances – and their expected result.21 What is most disturbing,
then, to Kumārila is that someone who would “see all things in a [single act of] perception”
(sarvapratyakṣadarśin), as he defines the Omniscient in the BṬ (= TS 3138c), would
also know dharma(s) perceptually: being a “knower of all things” (sarvajña), he would
also be a “knower of dharma” (dharmajña), and this would contradict the exclusivity of
19
20
21
See Kataoka (20112 : 11 and n. 8), who reports the view of Harikai (1985: 63) on a possible evolution
in Kumārila’s attitude towards Buddhism (less ‘logical,’ more ‘emotional’) between the ŚlV and the
TV. On this topic, see also Eltschinger 2014a: 66.
See La Vallée Poussin 1902, Eltschinger 2007: 38–46 and Eltschinger 2014a: 66–70.
See ŚlV (codanā°) 110cd–111, BṬ (= TS 3127), Kataoka 2003a: 42–43 and Kataoka 20112 : 320–324.
On Kumārila’s understanding of dharma, see ŚlV (codanā°) 13–14, translated in Kataoka 20112 :
206–209. As explained in these verses, actions, substances and qualities are not considered to be
dharma(s) in themselves, but only in so far as they are conducive to an expected result.
46
Maṇḍana Miśra’s Excursus on the Buddha’s Omniscience
the Veda on ethico-ritual matters.22 Maṇḍana shares the same preoccupation: for him,
the point is not the existence of an omniscient being in general, but of an “instructor
of creatures”23 promulgating a teaching on matters inaccessible to ordinary perception
in the form of injunctions (vidhi/codanā). The search for such a being is occasioned by
Maṇḍana’s reflection on Prabhākara’s concept of “commandment” (niyoga)24 which, in his
view, cannot be operative without supposing such an instructor:25
A commandment is [in itself] a mere instigation (pravartanāmātra), and [only]
that is the object of an injunctive suffix (liṅādi). What is grasped, then, through
[that] speech[-unit] is [only] that “I am prompted [to do this]” (pravartito
’ham); but the awareness that “I have to do [this]” (kartavyatāvagama)26
comes from the fact that someone who deserves to be obeyed (anuvidheya) is
the author of the commandment. [Only] when a commandment is given by
such a [person] do I understand that “I have to do this;” otherwise, I feel a mere
instigation, as it has been said [by Prabhākara]: “the commandment has the
obligation (kartavyatā) as its content, it does not affirm (āha) the obligation.”27
2.2 Given this essential agreement on the main point at stake, one would expect to find
at least an echo of Kumārila’s arguments in Maṇḍana’s text. Instead of that, the prose
development on ViVK 15 starts with the following statement, which seems at first entirely
foreign to Kumārila’s main argumentative strategy:28
22
23
24
25
26
27
28
Modern readers of Dharmakīrti’s statements on omniscience in PV 2.32–33 have rarely noticed how
close he actually stands to Kumārila on this point. Thus I would not necessarily interpret these verses of
the PV as a ‘response’ from Dharmakīrti to Kumārila (as suggested by Kataoka [20112 : 321]), but rather
as an essential agreement of both thinkers on the main point at stake: what matters is not the possibility
for a human being to know everything, but his capacity to access ethically/soteriologically relevant
matters by means of perception. As pointed out by Kataoka (20112 : 321), the distinction between
sarvajña and dharmajña is introduced by Kumārila only in the BṬ. Its use by Maṇḍana in the prose
introduction to ViVK 25 (S 733.4) might therefore point to his familiarity with Kumārila’s lost work.
ViV 15 (introduction) – S 445.1 (translated above § 1.3).
See ViV 12–14, David 2017 and David (forthcoming).
ViV 14: pravartanāmātraṃ niyogaḥ, sa ca liṅādyartha iti pravartito ’hama iti pratipattiḥ śabdāt. kartavyatāvagamas tu niyoktur anuvidheyatvāt, anuvidheyaniyoge mamedaṃ kartavyam ity adhyavasāyāt,
itaratra tu pravartanāmātrapratīteḥ. uktaṃ ca: kartavyatāviṣayo hib niyogaḥ, na niyogaḥ kartavyatām
āha (S 441.5–442.3 [≈ M 108.5–109.4/G 77.3–6]). a pravartito ’ham S; pravartito ’ham atra M G;
b
hi S: Ø M G.
A more literal translation would be “the awareness of an obligation.” As Vācaspati makes clear (NyK
442.9–10), the difference between a mere instigation (pravartanāmātra) and a proper obligation (kartavyatā) is that only the second can be the cause of an activity (pravṛttihetu) for a rational agent.
Bṛhatī 1.1.2 (38.8–9).
ViV 15: yady apy ātmapratyakṣanivṛttir viprakarṣavatām abhāvaṃ vyabhicarati, sarvapratyakṣanivṛttir
asiddhāsarvadṛśaḥ; sakalapramāṇanivṛttyā ca nārthābhāvasiddhiḥ, avyāpakanivṛttāva avyāpyanivṛtter
aniyamāt, avyāpakatvaṃ ca, ahetutvāt, tanmātrapratibandhābhāvācb ca, anyathā sarvasya sarvadarśitvaprasaṅgaḥ, aviśeṣāt c ; vacanādayaś ca yady apy avirodhād anivartakāḥ, tathāpi kāraṇanivṛttyā
kāryanivṛttipratīteḥ pramāṇānupapattyā tatkāryāyāḥ sarvārtheṣu saṃvido ’bhāvam anumimīmahe
(S 459.1–461.2 [≈ M 115.1–116.2/G 81.4–82.6]). a °nivṛttau S: °nivṛttyā M G: °nivṛttyā ca Mss (S);
b
°pratibandha° S: °anubandha° M G; c aviśeṣāt S: aviśeṣāc ca M G.
Hugo David
47
Even though (1’) the absence (nivṛtti)29 of one’s own perception does not prove
the non-existence (abhāva) of [objects] that are at a distance (viprakṛṣṭa),30
and [although] (1”) the absence of a perception for all [beings] is impossible
to establish unless one sees everything;31 (2) [even though] the absence of
all means of knowledge [regarding a certain object] does not prove that [this]
object does not exist – for (2A) the absence of a non-pervasive [property]
(avyāpaka) does not necessarily imply (ni-√yam) the absence of the [corresponding] non-pervaded [property] (avyāpya),32 and (2AI) [the existence of a
means of knowledge – pramāṇa] is [indeed] non-pervasive [with respect to the
existence of an object to be known – prameya], for it is not the cause (hetu) [of
the existence of the object to be known],33 and because there is [also] no essen29
30
31
32
33
Literally: the “cessation” or “non-activity,” as opposed to pravṛtti (“activity”). The term is used by
Dharmakīrti in a similar context in expressions like pramāṇatrayanivṛtti (“the absence of the three means
of knowledge” – PVSV 102.1), also found in Maṇḍana’s text. See Yaita (1985: 215): “the cessation
of the three means of knowledge;” Steinkellner (20131 : 81) “das Auffallen der drei Erkenntnismittel;”
Eltschinger/Krasser/Taber (2012: 9, n. 9): “the silence of the three means of valid cognition.” This last
translation is preferable, in my opinion, as the English term “cessation” implies previous activity, which
is not always the case of Dharmakīrti and Maṇḍana’s use of the word nivṛtti. See for instance Maṇḍana’s
(or rather, his opponent’s) definition of pleasure (sukha) as “the absence of pain” (duḥkhanivṛtti) in the
Brahmasiddhi (BS 1.1 [1.17]), which does not imply the previous existence of pain. I opt for a plainer
translation (“absence”) only to avoid confusion between a proper use of the word “silence” (in the case
of Scripture) and a metaphorical one (in the case of perception and inference).
According to Vācaspati (NyK 459.6–7), “distance” (viprakarṣa) is threefold: in space (deśa), time (kāla)
and nature (svabhāva). The same tripartition is found in chapter 2 of Dharmakīrti’s Pramāṇaviniścaya
(PVin 64.9–10) and in the Nyāyabindu (NB 2.27); the Pramāṇavārttika (PVSV 102.6–7, ad PV 1.199)
only mentions deśa and kāla.
Both published editions of the ViV suggest that asarvadṛśaḥ should be read with what follows, not with
what precedes, a solution also adopted by M. Biardeau in her unpublished translation: “(…) [quoi]que
l’absence de perception de tous ne soit pas établie, que l’on ne puisse établir la non-existence d’un
objet par l’absence de tous les moyens de perception chez quelqu’un qui n’est pas omniscient (…)”
(Ms. f. 123r). Although neither Stern’s edition nor Vācaspati’s commentary plead for either solution, I
find it easier to link the genitive with what precedes, since the position of ca would otherwise be odd.
The following argument (2) is directly borrowed from Dharmakīrti, as we shall see (§ 2.3), and stands
perfectly well on its own. On the other hand, it makes sense to argue that only an omniscient would be
able to establish a universal absence of perception concerning a given object.
In other words: the inference of the absence of a pervaded (vyāpya) property from the absence of the
pervading (vyāpaka) property is valid, as when we conclude to the absence of smoke (= vyāpya) on the
hill from the absence of fire (= vyāpaka): vyāpakābhāva → vyāpyābhāva (“a → b” = “valid inference
from a to b”). But this inference would not be valid if both properties were not in a relation of pervasion
(vyāpti), or if the pervasion was the other way round. Thus the inference of the absence of fire (=
vyāpaka) on the hill from the absence of smoke (= vyāpya) is not valid, for there are exceptions to this
conclusion. In the present case, the question raised is whether one can correctly conclude to the absence
in the world of an omniscient being – the object to be known (prameya) – from the absence of a means
to know (pramāṇa) such a being (pramāṇābhāva → prameyābhāva?). It follows from what precedes
that the inference is valid only if pramāṇa and prameya stand in a relation of pervasion (vyāpti), in
other words if we could correctly conclude to the existence of a means of knowledge from the existence
of the object of knowledge (prameya → pramāṇa?). Maṇḍana will now show that this is not the case,
by excluding the two only possible grounds for pervasion or invariable concomitance in Dharmakīrti’s
system: causality (tadutpatti) and identity (tādātmya) (argument 2AI).
See NyK 459.11–460.1: ahetutvād akāraṇatvāt.
48
Maṇḍana Miśra’s Excursus on the Buddha’s Omniscience
tial connection (tanmātrapratibandha) [between these two properties],34 for
otherwise35 everybody would see everything as there would be no difference
[between us and an omniscient being]36 –; and (3) even though speech, etc. do
not rule out [the existence of an omniscient] as there is no contradiction [in
his making use of speech, etc.]; even so, (4) since we understand the absence
of an effect (kāryanivṛtti) from the absence of [its] cause (kāraṇanivṛtti), from
the impossibility (anupapatti) of a means of knowledge [embracing the whole
realm of being] we infer the absence of its effect, namely an awareness of all
things (sarvārtheṣu saṃvid).
This dense passage is for the most part a discussion of inferential matters, and is indeed
a remarkable example of how logical considerations can successfully be applied to the
solution of a philosophical issue. Maṇḍana discusses here the use of the negative inferential
reason (hetu), “non-apprehension” (anupalabdhi), and its capacity to establish the nonexistence (abhāva) of a given object – in our case, an “awareness of all things” –, which is
the property to be proved (sādhya). His theoretical background is clearly Dharmakīrti’s
logical system, as the mention besides anupalabdhi of two possible grounds for ‘positive’
inference (causality and identity) suffices to prove. In substance, Maṇḍana proposes to
replace a series of illegitimate uses of anupalabdhi (conclusion to the non-existence of
an entity by mere lack of perception of it, etc.) (1–3) by another, legitimate one, known
to Buddhist logicians as kāraṇānupalabdhi (“non-apprehension of the cause”) (4).37 His
argument can be reconstituted as follows (the sign “←” indicates a logical relation: “justified
by”):
1. Absence of perception of an entity cannot establish its non-existence.
1’. Case of one’s own perception.
1”. Case of everyone’s perception.
34
35
36
37
The compound tanmātrapratibandha is equivalent to the expression tanmātrānubandha/°saṃbandha
used by Dharmakīrti on several occasions to speak of the relation underlying the use of an “essential
[inferential] reason” (svabhāvahetu). See for instance PVSV 6.26, 17.20, 18.1 and 18.21 (°anubandha),
PVSV 16.28 (°saṃbandha). See also NyK 460.3–5: hetudharmamātrānubandhī hi sādhyadharmas
tasya vyāpakaḥ, yathā vṛkṣatvaṃ śiṃśapātvasya; “For when the property to be proved (a) merely
depends on the property which is the [inferential] reason (b), it (= a) pervades the other (= b); for
instance, the quality of being a tree [pervades in this manner] the quality of being a śiṃśapā.” The
reading °anubandha° (instead of °pratibandha°), found in all Mss of the ViV and in some Mss of the
NyK (see Stern 1988: 1023–1024), is therefore equally plausible.
That is: if there was an essential connection between the existence of the object to be known (prameya)
and that of a means for knowing it (pramāṇa).
For a possible paraphrase of this difficult argument by Prajñākaragupta, see below § 4.5.
On kāraṇānupalabdhi, see NB 2.39: kāraṇānupalabdhir yathā nātra dhūmaḥ, vahnyabhāvād iti; “[Establishment of the non-existence of the effect through] non-apprehension of [its] cause is for instance:
‘Here, there is no smoke, for there is no fire’” (p. 135). In his commentary, Dharmottara remarks that
this particular use of anupalabdhi is restricted to cases where “the effect, even if it existed, would not be
perceived” (kāryaṃ sad apy adṛśyam bhavati – Nyāyabinduṭīkā 136.1), which is obviously the case of
an omniscient being. Although neither Maṇḍana nor Vācaspati uses exactly the term kāraṇānupalabdhi,
its equivalent kāraṇānupalambha is found in a pūrvapakṣa of Ratnakīrti’s Sarvajñasiddhi (SSiR 7.18),
which presents a reasoning identical in substance to Maṇḍana’s. The parallel between these two passages
would certainly require further exploration.
Hugo David
49
2. Absence of valid knowledge of an entity cannot establish its non-existence.
← 2A. A property p’s non-existence cannot justify positing another property q’s
non-existence unless p pervades (√vyāp) q.
2AI. Valid knowledge (pramāṇa) does not pervade its object (prameya).
← 2AIa. No relation of causality.
← 2AIb. No relation of identity.
3. Speech, etc. do not rule out omniscience in their possessor.
4. Omniscience can be negated, as an effect can be negated through the negation of its
cause.
Now this reasoning is clearly not Maṇḍana’s invention; in fact, the very structure of the
argument recalls Dharmakīrti’s ‘second’ consideration of anupalabdhi in the Pramāṇavārttika (PV) 1.198–204, especially PV 1.198–202 and the corresponding Svavṛtti (PVSV).38
What is comparatively new, however, is the application of these reflections to the particular
case of omniscience, which is not mentioned in this section of PVSV although Dharmakīrti
suggests other possible consequences of his theory for religious philosophy.39 The only text
I could find where Dharmakīrti applies a similar reasoning to omniscience (sarvajñatva) is
the Nyāyabindu (NB).40 I suspect this original and quite massive reinvestment of Dharmakīrti’s ideas must be interpreted in a polemical way. For the first victim of this exercise in
38
39
40
This passage has been translated twice, into English by Yaita (1985) and, more recently, into German
by Steinkellner (20131 and 20132 ). I am much indebted to the latter’s translation and rich annotation
of this text. Strictly speaking, Dharmakīrti’s discussion of anupalabdhi is much longer, finding its
conclusion only with PV 1.339 (thus practically with the end of the first chapter), including also his
lengthy digression on the authority of Scripture (āgama). See Eltschinger/Krasser/Taber 2012: 9, n. 9. I
am essentially concerned here with the initial part of this section.
Dharmakīrti remains quite vague in PV(SV) 1.198–204 about objects whose existence cannot be
disqualified by mere silence of Scripture, like “particular things (…) which are far away in time and
space” (deśakālavyavahitāḥ […] dravyaviśeṣāḥ – PVSV 102.6–7). He is more precise about objects
whose non-existence cannot be proved by the mere fact that we cannot infer them: “a mind free of passion”
(viraktaṃ cetas – PVSV 103.4), “a particular deity” (devatāviśeṣa – ibid.), “the capacity of intentions
[relative to] gifts and refraining from violent action to cause happiness” (dānahiṃsāviraticetanānām
abhyudayahetutā – PVSV 103.5; my translation of the compound in the genitive relies entirely on its
interpretation by Yaita [1985: 213] and Steinkellner [20131 : 84]). The closest approximation we find
in the Pramāṇavārttika to Maṇḍana’s reasoning is found in PV(SV) 1.311 (I thank V. Eltschinger for
drawing my attention to this important passage). In that portion of the SV, which forms a sort of ‘coda’
to his discussion on mantras, Dharmakīrti discusses possible objections against the assumption of an
“extraordinary person” (puruṣātiśaya) who could be the author of mantras on the basis of his “humanity”
(puṃstva), a property which must itself be inferred from his possessing an intellect (buddhi), senses
(indriya) and speech (vacana) (see PVSV 164.15). His main response is clearly similar to Maṇḍana’s
principal argument: na hy atīndriyeṣv ataddarśinaḥ pratikṣepaḥ saṃbhavati, satām apy eṣam ajñānāt;
“Those who do not see supersensible [objects] cannot confute (prati-√kṣip) them, for even if they exist,
they will not know them” (PVSV 164.17–18 [I do not translate hi]).
NB 3.69–71: yathāsarvajñaḥ kaś cid vivakṣitaḥ puruṣo rāgādimān veti sādhye vaktṛtvādiko dharmaḥ
saṃdigdhavipakṣavyāvṛttikaḥ, sarvajño vaktā nopalabhyata ity evaṃprakārasyānupalambhasyādṛśyātmaviṣayatvena saṃdehahetutvāt. tato ’sarvajñaviparyayād vaktṛtvāder vyāvṛttiḥ saṃdigdhā. vaktṛtvasarvajñatvayor virodhābhāvāc ca yaḥ sarvajñaḥ sa vaktā na bhavatīty adarśane ’pi na sidhyati,
saṃdehāt; “If what must be proved is, for instance, that a certain intended person is non-omniscient
(asarvajña), or is passionate, etc. (rāgādimant), one can doubt that a property like being a speaker
(vaktṛtva), etc. [establishing that conclusion] is absent from the negative instance [i.e. an omniscient
50
Maṇḍana Miśra’s Excursus on the Buddha’s Omniscience
‘applied logic’ is, no doubt, Kumārila, whose arguments against omniscience are – with
one exception to which I shall return later on (§ 3) – easily associated with one or the other
‘defective’ use of anupalabdhi. Thus, far from rejecting Dharmakīrti’s elaborations on
adṛśyānupalabdhi as ‘Buddhist,’ Maṇḍana appropriates them and adapts them in order
to free the classical Mīmāṃsaka argumentation against omniscience of some of its most
flagrant weaknesses.41 Let us now consider in more detail a few aspects of this strategy.
2.3 Kumārila’s refutation of omniscience, in the ŚlV and BṬ similarly, takes place in
three successive stages: after having established that omniscience is intrinsically impossible
(ŚlV42 112–115/TS 3157–3183), he shows that no evidence supports the assumption
(kalpanā) of an omniscient being in the past (ŚlV 117–136/TS 3184–3236) and finally
argues that, even if there had been such a being, he would have been unable to teach (ŚlV
137–140/TS 3237–3239).43 The second part of his demonstration, which is by far the
longest, starts with a very simple argument showing how our main means of knowledge
(pramāṇa) have no grip on an omniscient being:44
First, people like us do not see an omniscient being now; nor is it possible to
postulate that there was [such a being], as [one can] deny [it]. Nor [can one
postulate] an omniscient being on the basis of Scripture, for his [Scripture
would have the undesirable consequence of having] mutual reliance [with his
being an omniscient being]. If [Scripture] is composed by others, how is it
understood to be a means of valid cognition?
41
42
43
44
person]. For a non-apprehension like ‘we do not apprehend any omniscient speaker’ is the cause of
a [mere] doubt, for it is about [an object] whose nature cannot be seen (adṛśyātman). Therefore, the
absence of [the property of] being a speaker, etc. in [someone] other than a non-omniscient [i.e. in an
omniscient person] is subject to doubt. Moreover, since being a speaker and being omniscient are not
in contradiction (virodha), [the rule that] ‘whoever is omniscient does not speak’ is not established,
even when we do not see [anybody who is omniscient and speaks], for this is subject to doubt.” The
exemple of the property sarvajña(tva) is also used in NB 3.94–95, 3.125 and 3.130, with the same kind
of implications.
In a suggestive note of his recent study of PVSV (Steinkellner 20132 : 45–48, n. 49), E. Steinkellner
proposes to link the development of the theory of anupalabdhi in Dharmakīrti’s *Hetuprakaraṇa to the
debate on omniscience as known to us in particular through the works of Kumārila. See Steinkellner 2013:
“(…) ein wichtiger Anstoß für die kräftige Entwicklung der Lehre von der negativen Erkenntnis durch
Kumārila und stärker noch durch Dharmakīrti [ist] in der bei Kumārila sichtbar werdenden Polemik der
Mīmāṃsā gegen die Ansicht von der Existenz eines Allwissenden (sarvajña) zu identifizieren” (p. 46).
The ViV provides, in a way, a powerful confirmation of this insight, as do the passage of the Nyāyabindu
translated above (n. 40) and the statement from Kamalaśīla’s Nyāyabindupūrvapakṣasaṃkṣipta quoted
by Steinkellner (p. 47). I also fully agree that most of Kumārila’s arguments against omniscience do not
hold against Dharmakīrti’s elucidation of anupalabdhi (see below § 2.3–4). But one may also wonder
how far the question of omniscience was present to Dharmakīrti’s mind from the very beginning and,
if it was, why he never mentions it in the Pramāṇavārttika. It is not impossible that Maṇḍana’s text
reflects and elaborates on developments which are characteristic of Dharmakīrti’s later work.
All kārikā-numbers refer to the codanā-section (ad MīSū 1.1.2).
A useful synopsis of the section is found in Kataoka 20111 : xliv and Kataoka 20112 : 182–184. For a
systematic comparison of the ŚlV and BṬ, see Kataoka 2003a: 38 and Kataoka 20112 : 38.
ŚlV (codanā°) 117–118: sarvajño dṛśyate tāvan nedānīm asmadādibhiḥ / nirākaraṇavac chakyā na
cāsīd iti kalpanā // na cāgamena sarvajñas, tadīye ’nyonyasaṃśrayāt / narāntarapraṇītasya prāmāṇyaṃ
gamyate katham //. Translation: Kataoka 20112 : 332–336 (slightly modified).
Hugo David
51
Despite its simplicity and extreme popularity in later philosophical literature,45 the argument is not even mentioned in the ViV.46 A plausible explanation for this could lie in
Dharmakīrti’s newly introduced distinction of two kinds of non-apprehension (anupalabdhi): non-apprehension in general (anupalabdhimātra [PVSV 101.23]; anupalambhamātra [PVSV 103.3]) and non-apprehension of what, under normal circumstances, would
be fit for (perceptual) apprehension (upalabdhiyogya [PVSV 101.18]).47 As Dharmakīrti
explains it, the first type of non-apprehension can only produce doubt (saṃśaya) as to the
existence of the object, but cannot prove its non-existence (asattā);48 from our point of
view, an object we do not perceive (say, ghosts, or a particular deity) may as well exist or
not.49 The second type of non-apprehension, on the other hand, positively establishes its
non-existence,50 like when we do not see a pot in front of our eyes even in the clear light of
day. Following this important distinction, echoed in the passage of the ViV translated above
(§ 2.2), the fact that an omniscient being “is not seen” (na dṛśyate) has no value whatsoever
to prove that there is no such being, this regardless of whether we speak of the perception
of a single person (svapratyakṣa) or of everyone’s perception (sarvapratyakṣa).51 But the
point is not only about perception. Dharmakīrti further claims that even complete lack of
evidence about an object cannot persuade us of its non-existence (unless, of course, it fits
all conditions for present perceptual apprehension): “one cannot ascertain that [objects
that are at a distance] do not exist, even in the absence of [all] three means of knowledge
45
46
47
48
49
50
51
See the quite impressive list of quotations of these verses in Kataoka 20111 : 29–30. The fact that an
omniscient being “is not seen now” (na […] idānīṃ dṛśyate) is also, as we remember, the basic argument
of Bhavya’s pūrvapakṣa in the Mīmāṃsā-section of his Madhyamakahṛdayakārikā (15ab) (on Bhavya’s
relation with Kumārila, see above n. 10).
One could read an echo of Kumārila’s argument in ViV 19, where Maṇḍana underlines that an omniscient
being can neither be perceived, as he is “not within the reach of the senses” (indriyāṇāṃ na gocaraḥ
[ViVK 19b]), nor inferred, for the very same reason (ata eva [ViVK 19c]). Yet we should also pay
attention to the fact that Maṇḍana thereby only wants to prove our ignorance (ajñāna [S 686.2]) of
an omniscient being, while its non-existence or intrinsic impossibility (anupapatti) is considered
sufficiently proved by the preceding section (ViV 15–18). Kumārila, on the other hand, evaluates
which supposition (kalpanā) – that of the Veda’s authorlessness or that of an omniscient being – is
more dispendious (see ŚlV [codana°] 116). His reasoning is therefore against the probability for the
Omniscient’s existence, while Maṇḍana’s claim is only, on an epistemic level, about our knowledge of
that person.
On these two types of anupalabdhi, see Steinkellner (1967: 157–158), who proposes to distinguish
between non-apprehension in general (“Nicht-Beobachtung im Allgemeinen”) and non-apprehension
in particular (“Nicht-Beobachtung im Besonderen”) in the Pramāṇavārttika. Additional remarks on
this distinction are found in Steinkellner 20132 : 44.
See PVSV 101.19–20: na (…) asattāsādhanī (read °sādhanī instead of °sādhānī ); PVSV 103.10: ata
eva saṃśayo ’stu.
See PVSV 101.11: anupalabhyamāno na san nāsan, satām api svabhāvādiviprakarṣāt kadā cid anupalambhāt; PV 1.202a: sadasanniścayaphalā [anupalabdhiḥ]; PVSV 103.1–2: yasya kasya cit svabhāvo
nopalabhyate deśādiviprakarṣāt, na sa tadanupalambhamātreṇāsan nāma.
See PVSV 101.17–18: asattāyām (…) pramāṇam.
Although Dharmakīrti does not mention these two cases in the PVSV, the distinction is found in some
of his later works, as pointed out by Stern (1988: 1023). See for instance NB 2.27 ([…] deśakālasvabhāvaviprakṛṣṭeṣv artheṣv ātmapratyakṣanivṛtter abhāvaniścayābhāvāt) and Vādanyāya 10.12–14 (na
hy anumānādinivṛttir abhāvaṃ gamayati, vyabhicārāt, na sarvapratyakṣanivṛttiḥ, asiddheḥ, nātmapratyakṣāviśeṣanivṛttir api viprakṛṣṭeṣu).
52
Maṇḍana Miśra’s Excursus on the Buddha’s Omniscience
[i.e.: perception, inference, and teaching through Scripture] (pramāṇatrayanivṛttāv api).”52
Maṇḍana’s adoption of this argument without any change in the ViV can therefore be read
as a rejection of Kumārila’s whole strategy for denying the existence of an omniscient
being by mere lack of a pramāṇa capable of establishing it. To put it differently, Kumārila’s
proof against the probability of the existence of an omniscient being – quite sufficient for
the apologetic purpose of ‘weighing’ kalpanās – is systematically put aside by Maṇḍana,
whose only concern is now with the intrinsic impossibility of omniscience.53
2.4 The topic of non-apprehension is not the only one where Maṇḍana chooses to distance
himself from Kumārila by siding with Dharmakīrti. In v. 137 of the codanā-section of
the ŚlV, Kumārila famously points to a contradiction in the assumption of the Buddha’s
omniscience:54 had the Buddha really been omniscient, he would have been unable to teach;
teaching is a form of operation (vyāpāra) that naturally implies some kind of intention to
speak, thus a form of desire/passion (rāga), incompatible with the supposedly dispassionate (rāgarahita) state characterizing the Omniscient.55 As is well-known, Dharmakīrti is
familiar with Kumārila’s argument (or a similar one)56 and considers it a fundamentally
flawed use of the inference from the effect (kāryānumāna). Surely, some teachings are
52
53
54
55
56
PVSV 102.10: na ca te pramāṇatrayanivṛttāv api na santīti śakyante vyavasātum. The three means
of knowledge (pramāṇatraya) are enumerated in PV 1.199. This point, which is developed in the
whole SV on this verse, is reiterated at the very end of the first chapter of the PV. See PV 1.339cd and
PVSV 176.11–12: tenāsanniścayaphalānupalabdhir na sidhyati // tasmān na pramāṇatrayanivṛttāv api
viprakṛṣṭeṣv abhāvaniścayaḥ; “[PV:] Therefore, it is not established that non-apprehension results in the
ascertainment [of something] as non-existent. [SV:] Therefore, the non-existence of [things] beyond the
reach [of ordinary cognition] cannot be ascertained even if all three means of valid cognition [should]
fail to operate” (read asanniścaya° instead of asaṃniścaya°; translation: Eltschinger/Krasser/Taber
2012: 76–77; I modify “non-perception” into “non-apprehension”).
The same kind of refutation also forms the basis of immediately later Buddhist defences of omniscience, like that of Śāntarakṣita and Kamalaśīla. See McClintock 2010: 165–187. Śāntarakṣita’s use
of Dharmakīrti’s analysis of anupalabdhimātra as producive of a mere doubt (saṃśaya) is also clear.
See TS 3300–3301, translated in McClintock 2010: 186. I find it quite remarkable that Kumārila who,
according to the now (almost) consensual hypothesis of Frauwallner (1962), wrote the BṬ partly as a
response to Dharmakīrti’s *Hetuprakaraṇa, does not modify at all his strategy in what could be his
last great work. On the contrary, far from renouncing his proof of the Omniscient’s non-existence by
mere lack of evidence, Kumārila brings it to its perfection in the BṬ by adding to the examination of
perception (pratyakṣa), inference (anumāna) and speech (śabda) carried out in the ŚlV a consideration
of comparison (upamāna) and presumption (arthāpatti) as well. See BṬ (= TS 3214–3228).
As we have seen (§ 2.3), this constitutes the third part of Kumārila’s argument in that section.
ŚlV (codanā°) 137: rāgādirahite cāsmin nirvyāpāre vyavasthite / deśanānyapraṇītaiva syād ṛte pratyavekṣaṇāt //; “And when he is established as having no action because he lacks desire and so on, [his]
teaching could only have been composed by others without having [directly] observed [anything].”
Translation: Kataoka 20112 : 366–369.
See Kataoka 2003a: 55–63, Kataoka 20112 : 366–369 (nn. 425–426). Doubts about this identification
have been expressed by J. Taber (see Eltschinger/Taber/Krasser 2012: 119–120, n. 3) since Kumārila,
unlike Dharmakīrti, does not mention the Buddha’s speech (vacana) but only his operation (vyāpāra).
Though I agree that only a quote could settle the matter, I find the objection hardly convincing for, as
already pointed out by Steinkellner (20132 : 85), one fails to see to which ‘operation’ Kumārila could
possibly allude if not to the Buddha’s teaching (deśanā), which would otherwise have to be promulgated
by others (anyapraṇīta). Further arguments in favour of Kataoka’s identification on the basis of the
structure of PVSV have been voiced by Steinkellner (20132 : 84–87), who carefully concludes that
Hugo David
53
made out of desire for a benefit, as we observe among worldly teachers, but this need not be
the case for all teachings and does not rule out other reasons for teaching, less incompatible
with the state of omniscience, such as compassion (karuṇā): “Just as a passionate [person]
(rakta) speaks, so does the impassionate (virakta) one, too. Therefore it is not apprehended
from speech as such [whether one is passionate or dispassionate].”57 That this is precisely
the argument alluded to by Maṇḍana when he says in the ViV that “speech, etc. do not rule
out [the existence of an omniscient being], as there is no contradiction [in his making use
of speech, etc.].” (argument no. 3 in the above-quoted text) is proved, besides the explicit
mention of “speech, etc.” (vacanādi),58 by a further allusion to the SV on PV 1.12 in the
first book of the BS. In that passage, Maṇḍana distinguishes between two concepts of
desire – icchā (desire in general) and rāga (passion, which is an obstacle to liberation from
saṃsāra) –, and it is again Dharmakīrti’s definition of rāga in PVSV 9.5–6 that he calls for
support:59
Passion (rāga) is not mere desire (icchāmātra); they call “passion” that attachment to unreal qualities [of the object] brought about by nescience (avidyākṣiptam abhūtaguṇābhiniveśa).60 But the mind’s inclination (prasāda) towards
reality – i.e. its delight (abhiruci) [in reality] or desire (abhīcchā) [for it] –,
following [its] purification through the vision of reality, does not fall into the
category of “passion” (rāga), just as aversion produced by one’s vision of
that reality which is the worthlessness of transmigration does not fall into the
category of “hatred” (dveṣa).
57
58
59
60
Dharmakīrti must be attacking, if not Kumārila himself, at least some Mīmāṃsaka position concordant
with that of Kumārila.
PVSV 9.7–8 (ad PV 1.12): yathā rakto bravīti, tathā virakto ’pīti vacanamātrād apratipattiḥ. Translation:
Franco 2012: 231.
Although Vācaspati tells us nothing of the value of °ādi, the expression vacanādi may correspond to the
compound spandavacanādi (“movement, speech, etc.”) found at the beginning of the SV on PV 1.12
(PVSV 9.3), or else to the three properties of humanity enumerated in PV(SV) 1.311 (senses, mind and
speech). See above n. 39. The parallel passage in NB 3.71 only mentions vaktṛtva (see our translation
of this passage above, n. 40). Another possibility is that Maṇḍana alludes here to the contradiction
between contemplation (dhyāna) and teaching (upadeśana) underlined by Kumārila in the BṬ (= TS
3237–3239). It is unclear, in that case, how he intended to solve this apparent contradiction.
BS 1.1: na hīcchāmātraṃ rāgaḥ. avidyākṣiptam abhūtaguṇābhiniveśaṃ rāgam ācakṣate. tattvadarśanavaimalyāt tu cetasaḥ prasādo ’bhirucir abhīcchā na rāgapakṣe vyavasthāpyate, yathā saṃsārāsāratātattvadarśananiṣpanno nodvegas tato dveṣapakṣe (3.17–20). I slightly modify the translation of
this passage by Taber (2011: 443), who rightly points out the importance of this parallel for a correct
interpretation of Dharmakīrti’s concept of rāga in PV 1.12 and the corresponding SV.
Dharmakīrti’s definition of rāga in PVSV 9.5–6 is exactly similar if we except the mention, instead of
“nescience” (avidyā), of a list of objects of defilement typical of Buddhist thought, corresponding to the
four “aspects” of the first āryasatya (anityatā, duḥkha, anātmatā, śūnyatā – see Eltschinger 2014b):
nityasukhātmātmīyadarśanākṣiptaṃ sāsravadharmaviṣayaṃ cetaso ’bhiṣvaṅgaṃ rāgam āhuḥ; “They
call ‘passion’ the attachment of the mind, which has the defiled elements of existence as an object,
which is caused by seeing [erroneously, what is impermanent, suffering, not the self, and does not
belong to the self as] permanent, pleasure, the self and what belongs to the self.” Translation: Franco
2012: 231 (I modify “desire” into “passion,” to suit the convention followed here).
54
Maṇḍana Miśra’s Excursus on the Buddha’s Omniscience
2.5 As we can already see from our analysis of its initial portion, the discussion of
omniscience in ViV 15 reveals Maṇḍana’s deep familiarity with the works of Dharmakīrti
– especially PV 1 and the Svavṛtti, but also the Nyāyabindu and, as we shall see later on
(§ 3.5), the Pramāṇaviniścaya –, far deeper in any case than that of any Mīmāṃsaka before
(and perhaps even after) him. But it also testifies to the complexity of his engagement
with the famous Buddhist logician, which contrasts with Kumārila’s strategy of frontal
confrontation with Buddhism in general. It is noteworthy in this respect that the debate is
never on matters of pure logic, an impression that can only be confirmed by the remaining
part of ViV 15. The two kinds of non-apprehension, the negation of the effect through the
negation of its cause (see below, § 3), even Dharmakīrti’s conclusions about fallacious
uses of the inference from the effect: everything is accepted in block by Maṇḍana and
never questioned again in his work. His way to consider anew the topic of the Buddha’s
omniscience in the ViV is therefore to play, so to say, Dharmakīrti’s part, and to reconstruct
what could have been a ‘Dharmakīrtian’ answer to Kumārila from elements scattered
throughout Dharmakīrti’s works. But this is just a prelude to Maṇḍana’s real encounter
with Dharmakīrti, carried out once again with his own weapons. The field of this encounter,
however, would not be pure logic any more, but the epistemology of perception.
3. Time, perception and Scripture: on a mīmāṃsaka use of kāraṇānupalabdhi
3.1 If the bulk of Kumārila’s arguments against the Buddha’s omniscience, as we have
seen, finds no favour in the eyes of his most immediate successor in Mīmāṃsā, a small section of the ŚlV (codanā° – 112–115) and BṬ (= TS 3157–3183) remains to be investigated,
in which Kumārila argues for the incompatibility between omniscience and the very nature
of perception. That preoccupation, at least, is well in line with Maṇḍana’s philosophical
agenda and the problem is in fact at the heart of his enquiry in ViV 15. Once again, the
occasion for this reflection is given to him by a logical remark by Dharmakīrti. As we have
seen before (§ 2.3), Dharmakīrti insists in the PV that mere absence of perception (or even
of knowledge in general) of an object that is not fit for perceptual apprehension produces no
certitude (niścaya) of its non-existence, but only doubt (saṃśaya) regarding its existence.
However, there are other ways to produce such a proof of the non-existence of an object,
one of them being to find evidence for the absence of a cause (kāraṇa) of the object, as
explained in the following passage of the PV and SV:61
[PV:] But the fact that we do not know the efficient [cause] (kārakājñāna)
of [a given] effect proves that [such an effect] does not exist. [SV:] In case
the [property] to be proved is the non-existence (abhāva) of a [given] nature
(svabhāva), [we just] said62 that the [mere] absence of apprehension of that [na61
62
PV 1.201cd and PVSV 103.16–18: kārye tu kārakājñānam abhāvasyaiva sādhakam // svabhāvābhāve
sādhye tadanupalambha evāpramāṇam ucyate. kārakānupalambhas tu pramāṇam eva. na hy asti
saṃbhavo yad asati kāraṇe kāryaṃ syāt.
See PVSV 103.1–2: yasya kasya cit svabhāvo nopalabhyate deśādiviprakarṣān na sa tadanupalambhamātreṇāsan nāma, yathoktaṃ prāk. On the identification of the passage alluded to by Dharmakīrti
(PVSV 101.11 or 102.11–12?), see Steinkellner 20132 : 273 (n. 543).
Hugo David
55
ture] does not constitute a [valid] means of knowledge (apramāṇam). But the
non-apprehension of [its] efficient [cause] (kārakānupalambha)63 is a [valid]
means of knowledge (pramāṇam eva), for it is impossible (nāsti saṃbhavaḥ)
that an effect should take place without a cause.
Dharmakīrti’s reasoning is clearly alluded to in the passage quoted above (§ 2.2) by
Maṇḍana, who does not speak, however, of “non-apprehension of a cause” (kāraṇānupalambha/°anupalabdhi), but simply of the “absence of a cause” (kāraṇanivṛtti). And
indeed, a major issue of Dharmakīrti’s reasoning – which is not entirely clear from this
passage of the SV – is that “non-apprehension of the efficient [cause]” (kārakānupalambha)
cannot be mere non-apprehension (anupalabdhimātra), but has to be non-apprehension of
the second kind, where the object is fit for (perceptual) apprehension, lest the inference
becomes equally inconclusive.64 This, of course, is not the case of most objects placed at a
distance (viprakṛṣṭa) or supersensible (atīndriya) objects like particular deities, etc., whose
cause is very likely to be also beyond the reach of the senses. The case of omniscience is
somewhat peculiar, though, due to the presupposition – apparently shared by all participants in this debate – that (valid) knowledge of all things must be, to begin with, valid
knowledge over all. Just as “persons of exception” (puruṣātiśaya) remain persons all the
same, “exceptional visions” (darśanātiśaya) differ in degree (of precision, intensity, clarity,
etc.) from ordinary perceptions, but they obey the same principles as any other perceptual
cognition.65 Maṇḍana’s task is, then, to prove that the cause of perception is such that it
can never produce a knowledge of all things, and thereby to undermine the very possibility
(saṃbhava) of omniscience.66
3.2 Kumārila, who already had some thoughts on that topic, mostly insists on the mutual
delimitation of our senses, which disqualifies a cognition of all things at once: acute as it
63
64
65
66
The expression kārakānupalambha used in the PVSV is of course equivalent to the compound kāraṇānupalabdhi found in Dharmakīrti’s later works (see above n. 37).
This point is well made by Yaita (1985: 202, n. 65).
On puruṣātiśaya / sātiśayo naraḥ, see BṬ (= TS 3161/3159) and PV(SV) 1.311. The expression
darśanātiśaya is used, for instance, in Jayanta Bhaṭṭa’s Nyāyamañjarī (NM1 : 268.3) to define the
cognition of yogins. A similar use of atiśaya applied to cognition is found in the ŚlV (codanā° – 114)
and BṬ (= TS 3386).
Interestingly, this seems to be precisely the point where the problem of omniscience is taken up by
Śaṅkaranandana in his Sarvajñasiddhi, possibly written in the 9th century. See SSiŚ 2: jñātakāraṇasadbhāvā saṃbhaviny anyakāryavat / sarvārthaviṣayā saṃvit, sa hi saṃbhava ucyate; “A consciousness
encompassing all objects, like [any] other effect, is possible (saṃbhavin) [only] if the actual existence of its cause (kāraṇasadbhāva) is known; for this is what [we] call ‘possibility’ (saṃbhava).”
The corresponding prose portion, a preliminary edition of which is established by Eltschinger (2008:
140–141), is still too obscure (to me, at least) to provide any reliable information about the intellectual
background of this stanza. The proximity between Maṇḍana and Śaṅkaranandana’s way of speaking of
omniscience (sarvārtheṣu saṃvid [Maṇḍana]/sarvārthaviṣayā saṃvid [Śaṅkaranandana]) need not be
significant, but is nevertheless striking. The same kind of consideration is also found in Ratnakīrti’s
Sarvajñasiddhi: nanu kāraṇānupalambhād eva sarvajñatāpratiṣedhaḥ sidhyati; “[Objection:] but, the
negation of omniscience is established by the non-apprehension of its cause (kāraṇānupalambha)!”
(SSiR 7.19–20). See also Moriyama 2014: 64 (n. 29).
56
Maṇḍana Miśra’s Excursus on the Buddha’s Omniscience
may be, the eye – be it a Divine Eye –67 will never perceive sounds or smells; yet sounds or
smells are, no doubt, part of the totality of being.68 In order to grasp everything at once, the
senses of the Omniscient would therefore have to function simultaneously in an eminent
way, a possibility which contradicts the widely admitted principle that two cognitions
(say, the smell of a strawberry and the vision of its red colour) never take place exactly at
the same time.69 As before, but for reasons that are far less clear, this popular reasoning
did not find its way into the ViV.70 Maṇḍana prefers to concentrate on another aspect
of the mīmāṃsaka theory of perception, going back to Jaimini’s treatment of pratyakṣa
in MīSū 1.1.4, namely its relation to the present time (vidyamāna). Perception, so the
sūtra goes, cannot be a cause (nimitta) for our knowledge of dharma “because it grasps
[something] present” (vidyamānopalambhanatvāt). Applying this conclusion to the debate
67
68
69
70
As noted by Moriyama (2014: 60–61), the quotation of scriptural passages mentioning the Buddha’s
“Divine Eye” (divyacakṣus) by commentators on ŚlV (codanā°) 112–115 and by Kamalaśīla while
commenting on a verse of the BṬ (ad TS 3159–3160) in TSP 999.12–13 is certainly not done by chance.
It is indeed likely that Kumārila had this or a similar notion in mind while discussing this topic.
ŚlV (codanā°) 112–114: ekena tu pramāṇena sarvajño yena kalpyate / nūnaṃ sa cakṣuṣā sarvān
rasādīn pratipadyate // (…) yatrāpy atiśayo dṛṣṭaḥ sa svārthānatilaṅghanāt / dūrasūkṣmādidṛṣṭau syān
na rūpe śrotravṛttitā //; “However, if [you] postulate that he knows everything through a single means
of valid cognition, he surely grasps all tastes, etc. with the eye! Even when superiority of a particular
[pramāṇa] is seen, in so far as [a sense] does not go beyond [its] own object, that [superiority] can
[happen] in perceiving things that are remote, subtle, etc., [but] it is not the case that the ear should
grasp color.” Translation: Kataoka 20112 : 324–328 (modified). Cf. BṬ (TS 3157–3158), translated in
Kataoka 20112 : 328–329 (n. 368).
Although this last part of the argument is not voiced by Kumārila, it seems nevertheless a natural
consequence of his statements about the senses and their limited domain. It is found explicitly in
Kamalaśīla’s commentary on TS 3157. See TSP 998.7–9: na caitac chakyate vaktum mā bhūd ekena
jñānena yugapad aśeṣārthasya grahaṇam, anekena bhaviṣyatīti, yato yugapad anekavijñānāsaṃbhavāt;
“And you cannot say [the following:] ‘Maybe it is impossible to grasp all objects at the same time
(yugapad) in a single cognition (ekena jñānena), but this can happen in several [cognitions] (anekena),’
for it is impossible that several cognitions [should take place] at the same time.”
To the already long list of quotes of ŚlV (codanā°) 112–114 enumerated by Kataoka (20111 : 27–29), I
can only add the (somewhat unexpected) quotation of v. 112ab in Helārāja’s Prakīrṇaprakāśa (vol. 1, p.
54.17 – ad Vākyapadīya 3.1.46; I thank Vincenzo Vergiani for drawing my attention to this passage).
Helārāja’s response to Kumārila is quite unique in that the 10th -century Kashmiri grammarian directly
contests the Mīmāṃsaka’s claim that the domains of the senses are mutually impenetrable, and does so on
the basis of some hitherto unidentified Scripture (āgama): tac ca teṣāṃ śiṣṭānāṃ jñānaṃ sarvendriyam,
pratiniyamānapekṣaṇāt. sarvajñā hīndriyāntareṇāpīndriyāntaravyāpāraṃ kurvanti, tathā cāgamaḥ
nedānīm indriyair eva paśyanti, ghrāṇataḥ śabdaṃ śṛṇoti, pṛṣṭho rūpāṇi paśyati, apy aṅgulyagreṇa
sarvendriyārthān upalabhyate; “And this knowledge of the Learned Ones (śiṣṭa) is [produced] by all
the senses (sarvendriya), for omniscient [beings] accomplish the operation of a sense even by means
of another, as it is said in the [following] Scripture: ‘Now they do not see only by the senses. [In
that state,] one hears sounds by [the organ of] smell, sees forms [even] in [one’s] back. More than
that! One grasps all sensory objects even with the tip of a finger!’” (Prakīrṇaprakāśa 54.1–5). The
boldness of Helārāja’s statement appears by comparing it, for instance, to Jayanta Bhaṭṭa’s much milder
response to Kumārila: rasādigrāhīṇy api yoginām indriyāṇi cakṣurvad atiśayavanty eveti na rasādiṣu
cakṣurvyāpāraḥ parikalpyate; “The senses by which yogins grasp smells, etc. are also eminent, just like
[their] eyes, so there is no need to assume an operation of the eye towards smells, etc.” (NM1 270.1–2).
It is impossible to decide if Maṇḍana positively rejected Kumārila’s claim (and in that case, on which
basis), or simply considered it irrelevant to the present debate. I find it unlikely, in any case, that he
ignored it.
Hugo David
57
on omniscience, Maṇḍana makes it into a general statement as to the nature of sensory
perception:71
Perception, when brought about by the eye, etc. [does] not [apply] to all objects,
for the [eye, etc.] have a restricted domain: their domain is exclusively some
particular [object], which is present (vartamāna) and related (saṃbaddha)
[to the senses], and not all objects are like that. Now it is true that, since it is
possible to be aware of all sorts of knowable objects, a restriction (niyama) [of
the domain of perception] in terms of form (rūpa) is hardly defendable, and
so is also [its restriction] in terms of acuity, feebleness, etc. (paṭumandatādi).
[Moreover,] since we cognize (pari-√chid) objects at all sorts of distance and
in [all sorts of] measures, a restriction of relation in terms of distance (deśa)
[or] measure (parimāṇa) [is also not possible]. But [a restriction] in terms
of time (kāla) is defendable (nirūpyate), for in [the view that the eye, etc.]
operate by reaching [their object] (prāpyakāritve),72 they cannot reach it if [the
object] is not present; the same [is true] in [the view that the eye, etc.] operate
without reaching [their object] (aprāpyakāritve), since [in that case] one needs
a [special] capacity (sāmarthya) of the object [to be known], [and objects]
that have not come into being or have ceased to exist have no [such] capacity,
for they are inexpressible (anupākhyeya) [in terms of being and non-being].73
And therefore the eye, etc., should they have a special excellence (ati-√śī ),
may only make their own domain known in an eminent way (adhikam) in
terms of distance, measure [and] number, but not what is beyond their domain,
[namely] what has not come to existence, [and] what has ceased to exist.
Therefore it is said [in Mīmāṃsāsūtra 1.1.4]: “because it grasps something
present” (vidyamānopalambhanatvāt).
Both characteristics of the object of perception highlighted in this passage (saṃbandha/
vartamānatva) have their source in MīSū 1.1.4, where “contact with the senses of a person”
(saṃprayog[aḥ] puruṣasyendriyāṇām) is mentioned besides “being grasped at the present
time” (vidyamānopalambhanatva). Yet it is obviously the second characteristic that, above
all, captivates Maṇḍana’s attention. His four-fold suspension of restriction (niyama) in
71
72
73
ViV 15: na pratyakṣaṃ cakṣurādijanma tāvat sarvārtheṣu, teṣāṃ viṣayaniyamāt. kiṃ cid eva hi vartamānaṃ saṃbaddhaṃ ca tadviṣayaḥ, na ca sarve ’rthās tathāa . yady api cānekavidhaprameyasaṃvedanād
rūpato niyamo durnirūpaḥ, paṭumandatādibhedataś ca, nānādeśaparimāṇārthaparicchedād deśataḥb saṃbandhaniyamaḥ, parimāṇato niyamaś ca, kālatas tu nirūpyate, prāpyakāritve ’vartamānasya
prāptyabhāvāt, aprāpyakāritve ’py arthasāmarthyasavyapekṣatvāt, jñānotpattāv ajātanivṛttayor anupākhyeyatvād asāmarthyāt. tadāc cātiśayānād api cakṣurādayaḥ svaviṣayam eva deśataḥ parimāṇataḥ
saṃkhyāto vādhikaṃ bodhayeyuḥ, nāviṣayam ajātam ativṛttam. tad uktam vidyamānopalambhanatvād
iti (S 461.2–468.1 [≈ M 116.2–119.1/G 82.6–84.3]). a tathā S: Ø M G; b deśataḥ S: na deśataḥ M G;
c
tadā S: tathā M G; d atiśayānā(ḥ) S: atiśayānām M G. The text is trunked in M (118.4).
On this well-known divergence, see Chatterjee 1978: 138–141 and Bhatt 1989: 174–177. It opposes
thinkers (including Mīmāṃsakas, Naiyāyikas and Sāṃkhya-philosophers) who think that the senses
must “reach” (pra-√āp) their object in order to produce sensation and others (notably Buddhists) who
estimate that sensation can happen even while senses and object are at a distance.
Maṇḍana’s conception of past and future as “inexpressible [in terms of being and non-being]” (anupākhyeya) might be borrowed from Bhartṛhari’s Vākyapadīya. See David (forthcoming).
58
Maṇḍana Miśra’s Excursus on the Buddha’s Omniscience
terms of form, acuity, measure and distance (to which number can be added) also makes
the very idea of a relation with the senses practically useless. For what kind of ‘relation’
would there be between the senses and objects placed at an extreme distance, for instance,
if not their mere coexistence in one and the same moment? And what would be the point of
restricting the object of perception to what is ‘related’ to the senses if virtually everything
can be related to them? Thus it is possible to read in this text a form of thought experiment,
allowing us to discriminate between factual limitations of perception (in terms of form,
size, etc.) which, in principle, can be suspended, and a natural limitation of perception,
i.e. its relation to the present time, which no artificial extension of our faculties can allow
us to surpass. This insistance on the temporal dimension of perception might be a natural
consequence of Maṇḍana’s main thesis in the field of ontology, voiced in ViV 12, identifying
existence (sattā) with being present (vartamānatā): if only present things are perceptible,
it is perhaps because they only ‘exist’ in the true sense of the term.74 But this might also
be his one true link to Kumārila, whose core argument against the possibility of foresight
(obviously an essential component of omniscience) in ŚlV (codanā°) 115 is precisely the
natural limitation of sensory perception to the present moment.75
3.3 Maṇḍana’s strategy against omniscience thus appears, at this point, essentially as a
reduction of Kumārila’s arguments to a single one: perception, relying on the operation
of the senses, can only grasp things in the present. His use of this argument in ViV 15,
however, marks a radically different approach to Buddhist theories of perception. As we
have seen, all arguments of the ŚlV (codanā°) 112–114 and the corresponding verses of
the BṬ are based on the capacity of the senses to grasp all things in a single moment of
perception, and this may safely be extended to his remark in v. 115 as well. This presupposes
that perception can occur only through the senses, an assumption justified, in Kumārila’s
perspective, by his rejection of all kinds of supersensory perception in the chapter of the ŚlV
dealing with pratyakṣa (ad MīSū 1.1.4). Quite the opposite, Maṇḍana chooses to confront
Buddhist epistemologists on their own ground in order to show that even supersensory
perception as they conceive it is incompatible with omniscience. This ‘dialectical’ attitude,
so characteristic of Maṇḍana’s philosophical style, allows him to open an entirely new
field of philosophical enquiry into the nature of the Omniscient’s cognition, which was to
acquire some prominence in later stages of this debate.76
74
75
76
Maṇḍana’s equation between being and being present forms the topic of the third study in this series.
See David (forthcoming).
ŚlV (codanā°) 115ac: bhaviṣyati na dṛṣṭaṃ ca pratyakṣasya manāg api / sāmarthyam; “It is never
seen that perception has even a bit (manāg api) of capacity with regard to a thing in the future.”
Translation: Kataoka 20112 : 329. Unsurprisingly, Kumārila reads in MīSū 1.1.4, especially in the reason
vidyamānopalambhanatvāt, an implicit attack against the possibility of yogic perception. See ŚlV
(pratyakṣa°) 26–27ab (translated in Taber 2005: 54).
The earliest echo of this shift in the Mīmāṃsakas’ attitude towards omniscience in a Buddhist text
might be found in Kamalaśīla’s Pañjikā on TS 3156–3157 (= BṬ), where the learned Buddhist scholar
takes into consideration two hypotheses about the nature of the “complete knowledge of all things”
(sakalavastuparijñāna): sensory cognition (indriyajñāna) and mental cognition (manovijñāna). See
TSP 997.20–998.21. It is possible that Kamalaśīla’s comments on these stanzas should be read as
an effort to integrate Maṇḍana’s arguments into Śāntarakṣita’s mainly Kumārilan framework. His
examination of mental cognition (TSP 998.18–21), in particular, with his insistence on the mind’s
Hugo David
59
3.4 Three types of perceptual awareness are considered in ViV 15, corresponding to
Dharmakīrti’s four types77 with the exception of self-awareness (svasaṃvedana): perception
“born from the eye, etc.” (cakṣurādijanman), “mental” (mānasa) perception and perception
“born from meditation” (bhāvanāmaya), which is also how Dharmakīrti defines the cognition of yogins (yogināṃ jñānam).78 Among them, the greatest attention is not devoted to the
last kind of perception, as we would probably expect,79 but to mental cognition. For sure,
external senses are riveted to the present time, but the mind need not be; in fact we observe
that dreams, fantasies and other creations of the mind have no evident link to the world of
sensation, and also deal with past and future events.80 Could omniscience be a cognition
of that kind? We cannot be sure whether Maṇḍana had a particular Buddhist thinker or
school in mind while refuting that possibility, but I find it unlikely that his opponent should
be Dharmakīrti, whose conception of mental cognition (manovijñāna [NB 1.9]) explicitly
excludes independence of the mind from the senses. In fact it seems Maṇḍana chooses once
again not to refute Dharmakīrti’s ideas – at least, not at first –, but skilfully to take them
out of their original context to fit his own purpose. As is well-known, mental perception for
Dharmakīrti does not only cover internal mental phenomena such as awareness of pleasure
and pain, but also the (non-conceptual) moment of attention immediately following a
sensation, that of a patch of blue for instance, in which we become aware that there is
‘something’ in front of us without yet knowing that it is ‘blue.’ His main preoccupation in
PV 3.239–248 is precisely to show that the content (viṣaya) of that moment of perception is
different (anya) from what has been previously experienced (pūrvānubhūta) – so that it can
be considered valid knowledge (pramāṇa) –,81 but is also not entirely “unseen” (adṛṣṭa), so
that awareness of sound, for instance, cannot follow from a sensation of blue, or awareness
77
78
79
80
81
“dependence” (pāratantrya) on the senses, clearly reminds one of Maṇḍana (see below § 3.4). The case
of Prajñākaragupta will be dealt with further on (§ 4).
On these four types, see for instance NB 1.7–11.
See PV 3.281ab: prāg uktaṃ yogināṃ jñānaṃ teṣāṃ tad bhāvanāmayam (Translation: Eltschinger 2009:
192). On the ‘causal’ interpretation of bhāvanāmaya, see Eltschinger 2007: 85–86, n. 58.
This is, at least, the assumption made by Umbeka Bhaṭṭa (8th c.?), the oldest commentator on the ŚlV,
who begins his commentary on ŚlV (codanā°) 115 with the following objection: nanu heyopādeyagrāhakasya vijñānasyāsty ekaṃ kāraṇaṃ bhāvanā, kim ucyate kāraṇānupalabdhyā kāryābhāva iti?
bhāvanājanyapratyakṣaṃ dharmādharmagrāhakatvena tair iṣṭam!; “[Objection:] but, there is [indeed]
a cause for the cognition that grasps what is to be abandoned and what is to be appropriated, [namely]
meditation (bhāvanā)! Why do [you] say that ‘the effect does not exist since one does not grasp [its]
cause’? They [= Buddhists] maintain that perception born from meditation (bhāvanājanyaṃ pratyakṣam)
is what grasps merit and demerit!” (Tātparyaṭīkā 74.7–8; translation: Moriyama 2014: 64 [modified]).
The phrase kāraṇānupalabdhyā kāryābhāvaḥ recalls Maṇḍana’s formulation in ViV 15 (kāraṇanivṛttyā
kāryanivṛtti[ḥ] [S 460.2–461.1]), and it is possible that Umbeka, who is also the author of a commentary
on Maṇḍana’s Bhāvanāviveka, makes here an approximative quote of the ViV.
The possibility that dreams (svapna) manifest the mind’s capacity to grasp external objects independently
of the senses – a hypothesis Maṇḍana eventually rules out – is the topic of a separate investigation in
ViV 16 (S 583.2–590.3). This enquiry, carried out essentially with non-Buddhist arguments and only
loosely related to the question of omniscience, need not concern us here.
As is well-known, Dharmakīrti’s second characterization of “valid knowledge” (pramāṇa) in PV 2.5a
defines it as the “manifestation of an unknown object” (ajñātārthaprakāśa). On this definition, see
Katsura 1984, Krasser 2001: 185–190 and Kataoka 2003b. Further references in Krasser 2001: 184–185
(n. 45).
60
Maṇḍana Miśra’s Excursus on the Buddha’s Omniscience
of blue from no sensation at all, as in the case a blind man.82 Taking the best party of
these remarks, Maṇḍana insists, in his turn, on the dependence (pāratantrya) of the mind
(manas) on the senses when it comes to external objects (bahir):83
Even mental perception (mānasaṃ pratyakṣam) [cannot produce omniscience],84 for the mind (manas) has no autonomy (asvatantratvāt) [with
regard to what is] outside (bahir).85 If it had [such an] autonomy (svātantrya),
the undesired consequence would be that nobody would be blind, deaf, etc. To
explain: regarding perceptive awareness (pratyakṣa[ṃ] vedana[m]) of forms/
colours, etc. (rūpādi), the [mind] is dependent (paratantra) on [senses] like
the eye, [and] it is limited by their very limitation (niyama);86 otherwise, as
[we have just] said, the undesired consequence would be that nobody would
be blind, etc. If [you object] that [this undesired consequence, namely] that
nobody would be blind, etc. does not occur, for [mental perception] depends
on the [objective] correlate of [its] homogeneous and immediate cause [i.e. a
cognition] born from the [senses] (tajjasamanantarapratyayasahakāryapekṣaṇād),87 [our answer is that,] in all cases, the dependence [on the operation of
82
83
84
85
86
87
See PV 3.239–244: pūrvānubhūtagrahaṇe mānasasyāpramāṇatā / adṛṣṭagrahaṇe ’ndhāder api syād
arthadarśanam // (…) tasmād indriyavijñānānantarapratyayodbhavaḥ / mano ’nyam eva gṛhṇāti viṣayaṃ nāndhadṛk tataḥ // svārthānvayārthāpekṣaiva hetur indriyajā matiḥ / tato ’nyagrahaṇe ’py asya
niyatagrāhyatā matā //; “If mental [perception] grasps [an object] that has been experienced before
(pūrvānubhūta), then it is not a means of valid knowledge (apramāṇatā); if it grasps something [entirely] unseen (adṛṣṭa), then a vision of the object would occur also to a blind man, etc. (…) Therefore,
the mind [= mental cognition] (manas), born from the immediate cause (anantarapratyaya) that is
a sensory cognition (indriyavijñāna), grasps an entirely different object (anyam eva viṣayam) [with
respect to that sensation], so that [the undesired consequence that is] the vision [of the object] by a
blind man does not [occur]. [Still,] the sensory cognition (indriyajā matiḥ) that is [its] cause (hetu) is
entirely dependent (°apekṣaiva) on an object (artha) related to its own object [as its immediate cause]
(svārthānvaya°); so, even though it grasps something different, [we] consider that it grasps [only] a
delimited object (niyatagrāhyatā).” See also Vetter 1964: 40 and PVin 1.19 (19.1–7). A thorough
account of Dharmakīrti’s theory of mental cognition is given by Vācaspati in the NyK (471.2–473.8);
the passage has been translated into English by Stcherbatsky (19302 : 318–320).
ViV 15: mānasam api pratyakṣam, bahir manaso ’svatantratvāt, svātantrye ’ndhabādhirādyabhāvaprasaṅgāt. tathā hi: pratyakṣe rūpādivedane tac cakṣurādiparatantram tanniyamād eva niyatam,
anyathāndhādyabhāvaprasaṅgād ity uktam. tajjasamanantarapratyayasahakāryapekṣaṇād yadi nāndhādyabhāvaḥ sarvathā na pāratantryaṃa nivartate, tadviṣayajaviṣayatvāt b , anyathendriyāntarajasahakāriṇoc ’pi pravṛtteḥ sa evāndhādyabhāvaḥ (S 468.1–474.2 [≈ M 119.1/G 84.3–85.5]). a na
pāratantryam S: pāratantryaṃ na M G; b tadviṣayajaviṣayatvāt S: tadviṣayatvāt M G; c indriyāntaraja°
M G: indriyāntara° S. The order of the sentences differs widely between S and both printed editions
(M/G); I do not reproduce these variants here.
See NyK: mānasam api pratyakṣaṃ na sarvaṃ bodhayati (468.13).
Cf. ViVK 15d: paratantraṃ bahir manaḥ; “Regarding external [objects], the mind is dependent
(paratantra) [on the senses]” (S 458.3 [= M 114.3/G 81.3]).
As we have seen before, the essential limitation of the senses, in Maṇḍana’s view, is their incapacity to
grasp objects in the past or future.
Cf. PV 3.243ab (translated above, n. 82), PVin 1.19 (19.5–7) and NB 1.9. The latter’s definition
reads as follows: svaviṣayānantaraviṣayasahakāriṇendriyajñānena samanantarapratyayena janitaṃ tan
manovijñānam: “[M]ental perception is the product of a sense perception which forms its immediately
preceding homogeneous cause, and which cooperates with the immediately succeeding facsimile [i.e.
the second moment] of its proper object.” Translation: Kajiyama 1998: 45.
Hugo David
61
the senses] does not disappear, for the content [of mental perception] is born
of the content of a [particular sense].88 Were it not, since [mental perception]
would take place because of a [moment] that would cooperate with [the cognition] born from another sense (indriyāntarajasahakāriṇo ’pi),89 there would
indeed be no blind, etc.!
Despite the presentation of the arguments in a polemical form, there is probably little here
that Dharmakīrti would actually disagree with. I find it in general unlikely that a philosopher
arguing for the mind’s autonomy (svātantrya) in grasping external objects would really
draw any benefit from Dharmakīrti’s theory of mental awareness, and from his distinction
of two objective ‘moments’ corresponding to sensory and mental perception. Maṇḍana’s
detailed discussion of that theory in the ViV (S 474.2–542.1), leading him to reaffirm the
orthodox mīmāṃsaka view that “the mind never operates directly (sākṣāt) and independently
(svatantram) on an [object] that is not internal (āntara),”90 thus conscientiously fulfills the
epistemologist’s task, but has little bearing on the topic of omniscience proper.
3.5 A more profound divergence between the two philosophers comes out of Maṇḍana’s
brief discussion of perception “born from meditation” (bhāvanāmaya) at the end of ViV
15 (S 542.1–555.2). While Kumārila spoke in very general terms of “the perception of past
or future objects, or of one that is very small or obstructed, believed by some to belong to
yogins and liberated souls (muktātman),”91 Maṇḍana specifically addresses the Buddhist
epistemologists’ conception of yogic perception, especially their belief that it results from
“repeated practice” (abhyāsa) or habituation to cognitive contents reached by some other
means.92 As is well-known, Dharmakīrti thinks that the main cause of a yogin’s perception
is mental cultivation or “meditation” (bhāvanā), half way between ‘rumination’ of an idea in
view of its perfect assimilation and ‘imagination’ understood as the vivid representation of
something formerly conceived. In more Buddhist terms, meditation consists of the repeated
88
89
90
91
92
That is: the objective moment (kṣaṇa) which forms the content (viṣaya) of mental awareness of a patch
of blue (K2), for instance, is not identical with the preceding objective moment (K1), the content of
sensation. Yet both moments are not unrelated, since K1 is also the “homogeneous and immediate
cause” (samanantarapratyaya) of K2. In order for mental cognition to take place, then, we need two
things: a (sensory) cognition which is its samanantarapratyaya – or, in Vācaspati’s more oecumenic
terms, its “material cause” (upādāna – NyK 472.2) – and an objective correlate (sahakārin) which is the
immediate product of the objective moment (K1) grasped by that sensation. Thus, although sensation
and mental awareness have different contents, they are nevertheless indissolubly intertwined.
I exceptionally disagree with Stern’s choice to read °indriyāntarasahakāriṇo, and prefer the reading
°indriyāntarajasahakāriṇo transmitted in his Ms. B, in his own version of the NyK (474.5) and also
chosen by both published editions of the ViV (M 120.5/G 85.4). Vācaspati’s interpretation of the
compound indriyāntaraja° as referring to the cognition (indriyāntarajavijñāna°) which is the material
cause (or samanantarapratyaya) of mental awareness clearly supports that interpretation.
ViV 15: na kva cid a anāntare manaḥ sākṣāt svatantraṃ pravartate (S 541.3–4 [≈ M 140.6–141.1/G
100.3]). a na kva cid S: na kva cid api M G. Cf. ŚlV (pratyakṣa°) 160cd: pravṛttiḥ sukhaduḥkhādau
kevalasyaiva dṛśyate; “(…) a functioning of the mind by itself is observed in respect to pleasure, pain,
etc.” (text and translation: Taber 2005: 158 and 114).
ŚlV (pratyakṣa°) 26ac: atītānāgate ’py arthe sūkṣme vyavahite ’pi ca / pratyakṣam yoginām iṣṭaṃ kaiś
cin muktātmanām api // (text: Taber 2005: 152; translation: Taber 2005: 54 [adapted]).
On abhyāsa and its interpretation in Buddhist texts, see Eltschinger 2009: 184 (n. 57).
62
Maṇḍana Miśra’s Excursus on the Buddha’s Omniscience
presentation to the mind of the practitioner of the four Nobles’ Truths, culminating in their
direct apprehension (sākṣātkāra) or “vision” (darśana).93 To that activity Dharmakīrti
ascribes the power – well-attested in persons subject to hallucinations born from desire,
fear, madness, etc.94 – to produce an awareness with all external traits of perception:
clarity (sphuṭābhatā [PV 3.8b]/spaṣṭa[tva] [PV 3.281d]/spaṣṭābha[tā] [PV 3.284c], etc.),
non-conceptuality (kalpanāpo[h]a [PVinK 1.4a/PV 3.123]/akalpaka[tva] [PVinK 1.28d]/
akalpa[tva] [PV 3.285d], etc.). The difference between mere hallucination and the cognition
of a Buddhist Saint only lies, then, in the latter’s being “veridical” (saṃvādin [PV 3.286a])
or “non-erroneous” (abhrānta [PVinK 1.4b]), applying as it does to an object whose reality
(bhūta[tva] [PV 3.285]) has been ascertained through Scripture (āgama) and reasoning
(yukti). This last characteristic, which makes the cognition of yogins into valid knowledge
(pramāṇa) or perception (pratyakṣa) in the true sense of the term,95 is also the main topic
of Maṇḍana’s critique. For to claim, as Dharmakīrti does, that yogic perception is nonerroneous as it follows on hearing Buddhist Scriptures and pondering over their content
amounts to saying that meditation is essentially non-productive. As Maṇḍana puts in the
Brahmasiddhi: “[An injunction] concerning a cognition of the second type [= mental
cultivation]96 does not concern the comprehension of reality (tattvāvabodha), but only the
repeated practice (abhyāsa) [of that comprehension].”97 This ‘borrowed’ character of the
content of meditation, a warrant for its validity in Dharmakīrti’s view, is precisely what
leads Maṇḍana to disqualify it as mere second-hand knowledge:98
Even [perception] born from meditation (bhāvanāmaya) is about an object
[previously] heard about [in the Scripture] and/or known by inference (śrutānumitaviṣaya), for it is impossible to meditate at random; since it conforms
to a former cognition’s having a real object (bhūtārthatva) or the contrary
(viparyaya), it is dependent (paratantra) on Scripture and inference, and relies upon [another means of knowledge to ensure its validity] (sāpekṣatvāt).
Therefore it is not a means of valid knowledge (apramāṇam).
93
94
95
96
97
98
On the “vision of the [four] Nobles’ Truths” (āryasatyadarśana) as the specific goal of yogic perception
for Dharmakīrti, see PVin 1.28 (27.11) and Eltschinger 2014b: 250–251. As convincingly shown by
Eltschinger (2009: 199–200), that vision corresponds, in Dharmakīrti’s view, to the cognition of the
Buddhist practitioner after the “revolution of the basis” (āśrayaparivṛtti) has taken place.
See PV 3.282 (= PVin 1.29). Translations: Eltschinger 2009: 193, Franco 2011: 83.
See PVin 1.4ab (pratyakṣaṃ kalpanāpoḍham abhrāntam), NB 1.4 (tatra pratyakṣaṃ kalpanāpoḍham
abhrāntam). See also Steinkellner 1978: 126, Eltschinger 2009: 196–197 and Franco 2011: 86–88.
Maṇḍana alludes here to his own tripartition of Brahman-knowledge in the beginning of the Niyogakāṇḍa:
“knowledge born from speech” (śabdāt pratipattiḥ), knowledge “consisting of its continuous fixation”
(tatsaṃtānavatī [pratipattiḥ]) and knowledge “consisting of a direct apprehension” (sākṣātkārarūpā
[pratipattiḥ]). See BS 74.10–13. Meditation (bhāvanā) as understood by Dharmakīrti corresponds, of
course, to the second of these three stages.
BS 2.101–105ab (introduction): dvitīyapratipattiviṣayo hi na tattvāvabodhaviṣayaḥ, kiṃ tu tadabhyāsaviṣayaḥ (115.2–3 – I do not translate hi). Śaṅkhapāṇi’s commentary (Brahmasiddhivyākhyā 239.10–11)
makes the precision that the difference between both types of knowledge (pratipatti) is equivalent to
that between valid knowledge (pramāṇa) and memory (smṛti).
ViV 15: bhāvanāmayam apia śrutānumitaviṣayam akasmād bhāvanāyogād āgamānumānaparatantraṃ
pūrvajñānabhūtārthatvaviparyayānuvidhānāt sāpekṣatvād apramāṇam (S 542.1–555.2 [≈ M 147.1–3/G
104.1–3]). a bhāvanāmayam api S: bhāvanāmayam api vijñānam M G.
Hugo David
63
In these lines, which I propose to read as a direct response to Dharmakīrti’s opposite
statement in PVin 1.28,99 Maṇḍana inaugurates what would be the invariable position
of the Advaita tradition on the nature of meditation or “contemplation” (nididhyāsana)
for centuries to come. Meditation being only the repeated and progressively intensified
thought of an object, it cannot produce by itself any knowledge of that object. So, in order
to be mentally cultivated, the object must be reached by some other means, scriptural or
inferential. If meditation, then, has its use as a means of assimilation or ‘realization’ of
what has been grasped, it does not bring anything new and because of that it is “not a
means of valid knowledge” (apramāṇa).100 While Dharmakīrti interprets the transition
from conceptual knowledge originated from Scripture and reasoning to the immediacy and
non-conceptuality of the yogin’s insight as a passage from illusion to truth, Maṇḍana rather
insists on the identity of content of all three cognitions: no matter how we take it, it is always
the same reality that is “heard, thought and meditated upon,” as the Upaniṣad has it,101
and neither perception nor reasoning can grasp it without the help of Scripture. Beyond the
limited controversy about the yogins’s cognition and its capacity to account for omniscience,
we sense a deeper disagreement concerning the very function of Scripture and its place in
the path to liberation: from a mere preliminary (and to a certain point superfluous)102 stage
leading the adept to a more authentic and direct apprehension of reality in Dharmakīrti’s
view, the audition of Scripture has become for Maṇḍana the very centre of his Vedāntic
soteriology, the means of knowledge par excellence that other pramāṇas may of course
supplement, but never entirely replace.
99
100
101
102
See PVin 1.28: yoginām api śrutamayena jñānenārthān gṛhītvā yukticintāmayena vyavasthāpya bhāvayatāṃ tanniṣpattau yat spaṣṭāvabhāsi bhayādāv iva tad avikalpakam avitathaviṣayaṃ pramāṇaṃ
pratyakṣam; “Having first grasped objects through a cognition born from listening [to the treatises] and
[then] ascertained [them] through a [cognition] born of reflecting [upon them] by means of rational
enquiry, yogins cultivate [those objects]. The [cognition] which, at the completion of this [cultivation],
appears as vividly as in such cases as fear, etc. and [at the same time] is non-conceptual [and] has a true
object, is also a means of valid knowledge, [namely] perception” (27.9–11 – translation: Eltschinger
2009: 198 [modified]). The hypothesis of a direct response to Dharmakīrti is indirectly supported by
the paraphrase of PVin 1.28 in the corresponding portion of the NyK. Interestingly, Vācaspati does
not speak in general of the cognition of yogins but specifically of that of the Buddha (tathāgata), and
also relates Dharmakīrti’s remarks to the question of omniscience: śrutamayena vijñānena samastavastuviṣayaṃ nairātmyādi gṛhītvā yuktimayena ca bhūtatām asya vyavasthāpyāsakṛccetoniveśanarūpabhāvanāprakarṣaparyantajanma pratyakṣaṃ vijñānam anavayavenānātmādirūpaviśvālambanaṃ
karatalāravindaviṣayam ivātiviśadaṃ bhāvayiṣyati tathāgataḥ; “Having first grasped the absence of
a Self, etc., which concerns all beings, by means of a cognition born from listening [to the treatises]
and [then] established their reality by means of [a cognition] born from reasoning, the Tathāgata was
able to effectuate (bhāvayiṣyati) a perceptual cognition born of [His] intense meditation, consisting
of a repeated presentation to [His] mind [of the objects he reflected upon]. [That vision] had for its
objective correlate everything without exception possessing the property of being selfless, etc. and was
as entirely clear as [the vision] of a lotus on the palm of one’s hand” (S 544.6–545.4).
Recall that novelty or “manifestation of an unknown object” (ajñātārthaprakāśa) is one of the definitions
given by Dharmakīrti himself of “valid knowledge” in PV 2.5a. See above n. 81.
Bṛhadāraṇyakopaniṣad 2.4.5: ātmā vā are draṣṭavyaḥ śrotavyo mantavyo nididhyāsitavyaḥ; “Truly, it is
the Self that must be seen, heard, thought, meditated upon.” As is well-known, this passage is taken by
the later Vedāntic tradition to enunciate the three stages of the knowledge of Brahman, starting with its
“audition” (śravaṇa) in the Scripture, developing through “reflection” (manana) and “contemplation”
(nididhyāsana), and eventually leading to “direct perception” (sākṣātkāra).
See Steinkellner 1978: 127.
64
Maṇḍana Miśra’s Excursus on the Buddha’s Omniscience
3.6 With this last point it seems we have exhausted most of what Maṇḍana had to say on
the topic of the Buddha’s omniscience. It is now time to enquire whether his arguments
aroused any response from the Buddhist side in the following centuries, as was the case for
Kumārila. Our main field of investigation will be the work of an immediately later Buddhist
philosopher, Prajñākaragupta, on whom Maṇḍana’s influence – so is at least my contention
– is most easily discernible.
4. An early Buddhist paraphrase of ViV 15: Prajñākaragupta’s Pramāṇavārttikālaṃkāra (PVA) ad PV 2.29 (vv. 2.358–370)
4.1 Although Maṇḍana’s influence on later Buddhist thought is yet to be properly valued,
it is nevertheless certain that the ViV was read and extensively used by some at least
among later Buddhist logicians. Of the several texts one could invoke in support of this
claim,103 none is perhaps as revealing as Prajñākaragupta’s commentary on PV 2.29 (PVA
2.358–370).104 Prajñākaragupta is probably the first commentator on Dharmakīrti’s Pramāṇavārttika to regard the Buddha’s omniscience (sarvajñatva) as a central topic of the
Pramāṇasiddhi-chapter (= PV 2).105 His commentary on PV 2.29–33 is therefore, along
with chapter 26 of Śāntarakṣita’s Tattvasaṃgraha, among the oldest available testimonies
of that debate stemming from the Buddhist pramāṇa-tradition. Prajñākaragupta’s long
discussion of PV 2.29, where Dharmakīrti objects to the possibility of a knowledge of
objects beyond the senses (parokṣārthajñāna) in the absence of a means (sādhana) to
accomplish it, forms the pūrvapakṣa of that section, and is for the most part a web of
mīmāṃsaka arguments set against the possibility of omniscience. This section of the Vārttikālaṃkāra is thus of high significance for the early history of this debate in Buddhism and
Mīmāṃsā alike.106 In his recent study of that section, Sh. Moriyama (2014) rightly points
out Prajñākaragupta’s indebtedness to Kumārila’s works – both the ŚlV and the BṬ –107
103
104
105
106
107
Apart from the PVA, possible echoes of Maṇḍana’s arguments have been identified so far in the works
of Kamalaśīla (see above n. 76), Śaṅkaranandana (see n. 66) and Ratnakīrti (see nn. 37 and 66).
I am quoting here the recent edition of the passage by Moriyama (2014: 168–179) (= PVA), which
corresponds to PVAS 50.17–52.25. The numbering of kārikās is identical in both editions.
According to R. Jackson (1991: 235–236), Prajñākaragupta’s “conflating omniscience and authoritativeness” constitutes a decisive innovation of the Vārttikālaṃkāra with respect to earlier commentaries
on Dharmakīrti’s work. See also Franco (2011: 90, n. 44) and Moriyama (2014: 19–26), who reach the
same conclusion, the latter by an in-depth study of PVA ad PV 2.1–7. Interestingly, omniscience will be
regarded by some later Tibetan commentators as the fundamental topic, not only of those kārikās, but
of the whole second chapter. See Jackson (1991: 232 and 236), who mentions in particular the case of
rGyal tshab (15th c.). The latter’s indebtedness to Prajñākaragupta (which of course need not be direct)
is evident from the passage of his commentary on PV 2 translated in Jackson 1991: 241, which is little
more than a paraphrase of PVA ad PV 2.29.
Apart from a small portion of the text (PVA 168.9–10 and PVA 2.367–369), a rather bold adaptation
of an argument originally found in PV 1.335, all arguments of Prajñākaragupta’s pūrvapakṣin can be
traced back to earlier Mīmāṃsā works (see table below, § 4.6). Yamāri’s tentative identification of
Prajñākaragupta’s opponent as a materialist (tshu rol mdzes pa pa, Skt. *cārvāka?), on which Moriyama
(2014: 244, n. 5) already expressed serious doubts, can therefore be entirely discarded.
As he convincingly shows (pp. 58–59), the objection given in PVA 2.359 that an omniscient would
also experience the taste of impure things (aśuci), which is not found in the ŚlV, is certainly borrowed
from the BṬ (= TS 3144). Even if some parallels he draws between the PVA and the ŚlV might be
Hugo David
65
but also notes (pp. 59–60) that this cannot account for the pūrvapakṣin’s argumentation as
a whole, which has no clear equivalent in the works of the great Mīmāṃsaka. Adding to
Moriyama’s remarks, I shall argue that Prajñākaragupta’s model in building his pūrvapakṣa
is not only Kumārila, but also Maṇḍana, and that the first half of the text (PVA 2.358–363)
in particular can be read as a paraphrase of ViV 15.108 Incidentally, this identification of one
of Prajñākaragupta’s main opponents will help us, it is hoped, to solve certain difficulties in
the interpretation of that delicate passage, and also to highlight certain minor divergences
between the Buddhist scholar and his Brahmanical source.
4.2 Prajñākaragupta’s fundamental distinction, to begin with, between two possible interpretations of the word sādhana (“means”) in PV 2.29bc (tatsādhanasya ca / abhāvāt)109
– namely, as the (efficient) cause (hetu/kāraṇa) of the Buddha’s omniscience and as the
(informing) cause of our certitude (niścaye hetuḥ) of an omniscient being’s existence
–110 has generally been interpreted in reference to the ŚlV or BṬ.111 Yet nowhere does
Kumārila formulate such an opposition, which on the other hand closely corresponds to
Maṇḍana’s distinction, already found on the threshold of ViV 15, between the (efficient)
cause (hetu/kāraṇa) that should account (upa-√pad caus ) for omniscience112 and the cause
of our knowledge (jñāna) of an omniscient being.113 It is thus simpler to assume that
108
109
110
111
112
113
contested (especially in the case of PVA 2.358 and 2.359cd, as we shall see, but also in other cases like
PVA 2.362ab and 2.364, in which Prajñākaragupta might equally refer to the BṬ), the identification of
ŚlV (codanā°) 137 as the source of PVA 2.365, already proposed by Jayanta (see Moriyama 2014: 248,
n. 25), is in turn quite convincing. The parallel passage of the BṬ (= TS 3238–40) might indeed be
alluded to by Prajñākaragupta through his use of the expression vikalparahita (“devoid of conceptual
knowledge”), but the kārikā shows no evident formal similarity with that part of Kumārila’s late work,
while it is very close in wording to the verse of the ŚlV.
The possibility of Maṇḍana’s influence on Prajñākaragupta in this pūrvapakṣa is briefly considered by
Moriyama (2014: 63–65), who does not however engage in a systematic comparison of both texts.
The whole text of PV 2.29 runs as follows: prāmānyaṃ ca parokṣārthajñānaṃ tatsādhanasya ca (em.;
tat sādhanasya Ed) / abhāvān nāsty anuṣṭhānam iti ke cit pracakṣate //; “And the reliability [of any
religious authority] consists in [His/its] knowledge of objects beyond the senses, but because there is no
[possible] means to complete it, there is no [successful] practice in conformity with [its teaching]. Thus
claim certain [Mīmāṃsakas].” Translation: Moriyama 2014: 244.
See PVA ad PV 2.29: yas tāvad asarvajña eva sarvajño bhavati, tasya parokṣārthaparijñāne ko hetuḥ?
na khalv īdṛśaṃ kim api kāraṇam upalakṣitam, yato ’nuṣṭhānāt sarvavedanaṃ bhavati (…) nāpi
tanniścaye hetur asti; “First of all, if someone who is not omniscient becomes omniscient, what is the
cause (hetu) of his knowledge of objects that are beyond the senses? For sure, [you] cannot point out any
such cause (kāraṇa) from which, through practice, the knowledge of all [things] would arise (…); nor
is there any cause of [our] certitude (niścaye hetuḥ) that such a [being exists]” (168.5–9). Translation:
Moriyama 2014: 244 (modified).
See Inami 1996: 87, n. 5 and Moriyama 2014: 56.
See ViVK 15ab: hetvabhāve phalābhāvāt pramāṇe ’sati na pramā /; “No effect [takes place] without
a cause, [so] no valid knowledge [of all things takes place] without a [corresponding] means of
valid knowledge” (S 458.2 [= M 114.2/G 81.2]); ViV 15: (…) kāraṇanivṛttyā kāryanivṛttipratīteḥ
(S 460.2–461.1 – translated above, § 2.2).
See ViV 15: (…) na, tasyānupapatter ajñānāc ca; “No, for [an omniscient being] cannot be accounted
for, and because we would have no way to know [Him]” (S 445.1–458.1 [= M 114.1/G 81.1]); ViV
19: evaṃ tāvad anupapattiḥ, ajñānama api; “Thus [it has been shown], first of all, that [an omniscient
being] cannot be accounted for; now [we will see that] there is also no knowledge [of such a being]” (S
686.1–2 [≈ M 204.5/G 145.2]). a ajñānam S: jñānam M G.
66
Maṇḍana Miśra’s Excursus on the Buddha’s Omniscience
Maṇḍana’s distinction is the source of Prajñākaragupta’s twofold interpretation of the word
sādhana in Dharmakīrti’s verse.
4.3 The assumption of a debt to Maṇḍana further allows us better to understand the structure of Prajñākaragupta’s pūrvapakṣa, which already caused some difficulty to its Indian
commentators. Thus Yamāri (11th c.) tentatively identifies four parts in PVA 2.358–363,
corresponding to four possible causes (rgyu) of the Buddha’s omniscience: sensory cognition (dbang po’i shes pa; Skt. *indriyajñāna?), the senses and the object (?) (dbang po
dang don, Skt. indriyārtha?), mental cognition assisted by repeated practice (goms pa
dang bcas pa’i yid kyi shes pa, Skt. *abhyāsavanmanojñāna?) and inference (rjes su dpag
pa; Skt. *anumāna?).114 To this rather unlikely organisation of the pūrvapakṣin’s proof,
Moriyama (2014: 57–62) opposes his own twofold division, which sounds much more
promising: (1) refutation of omniscience as a form of sensory perception (v. 358–359), (2)
refutation of omniscience as a form of mental perception (v. 360–362). The problem is that
Kumārila, whom he considers to be Prajñākaragupta’s main model, never seems to consider
that omniscience could be something other than sensory perception, such as for instance
mental perception. So, either one considers that Prajñākaragupta himself introduces that
possibility115 or one has to admit that he draws from some other source, which is then very
likely to be Maṇḍana’s set of three (not two) hypotheses: omniscience as a form of sensory,
mental or yogic cognition (see § 3.4). The following table presents the various hypotheses
in presence regarding the nature of the Omniscient’s cognition:
ŚlV/BṬ
ViV 15
PVA 2.358–363
PVA
2.358–362
(Moriyama 2014)
[indriyapratyakṣa]
cakṣurādijanma
pratyakṣam (S 461.2–468.1)
mānasaṃ
pratyakṣam
(S 468.1–542.1)
bhāvanāmayaṃ pratyakṣam
(S 542.1–555.2)
indriyajñāna
(1.358–359 + 50.24)
manovijñāna (1.360)
indriyajñāna
(1.358–359 + 50.24)
manovijñāna
(1.360–362)
abhyāsāt [vijñānam]/
bhāvanā (1.361–363)
4.4 Arguments set against omniscience as a form of sensory or mental perception are
clearly similar in the ViV and PVA: the limitation of the domain (viṣaya) of sensory percep114
115
Tibetan text quoted in Moriyama (2014: 59, n. 12). Since the original Sanskrit of Yamāri’s Supariśuddhī
(on which see Steinkellner/Much 1995: xx) is still unpublished, it is not possible to check whether
oddities of this classification are due to Yamāri or to his Tibetan translator. In any case, Maṇḍana’s name
is not included in the list of authorities identified by M. Ono (2000: xxiv) in Yamāri’s commentary, so it
is quite possible that Prajñākaragupta’s commentator did not know the ViV. My very limited knowledge
of Tibetan did not allow me to consult Jayanta’s slightly earlier commentary.
This seems to be, in substance, the option chosen by Moriyama (2014: 59–62), who solves the difficulty by reading Prajñākaragupta’s argument as a reaction to Dharmakīrti’s newly elaborated theory
of yogic perception. I am not quite convinced by this explanation since Dharmakīrti, like Dignāga
before him, clearly distinguishes between mental and yogic perception, so there would be no point for
Prajñākaragupta to discuss mental perception specifically if that were really the point at stake.
Hugo David
67
tion to what is related/proximate (saṃbaddha [ViV]/saṃnihita [PVA]) to the senses,116 the
mind’s lack of autonomy (svatantra[tva]) from the senses in knowing external objects.117
A more delicate issue is whether there is any allusion to the perception of yogins in this
passage of the Vārttikālaṃkāra, as is my contention. The three kārikās at stake (PVA
2.361–363) read as follows:118
(361) The clarity (spaṣṭatā) of that [cognition acquired] through repeated practice (abhyāsa) cannot encompass all [objects]. If it (tasya) relies on Scripture,
[its] erroneousness (bhrāntatā) also (api) follows; (362) on the other hand
(tu), one does not reach all things (sarvaṃ vastu) as they are established by
an inference (anumānaprasiddha), so there is no meditation (bhāvanā) that
can encompass everything; how [then would one become] omniscient? (363)
Through repeated practice of the [sacred] treatises, etc. (śāstrādi) one can only
understand what is taught by them (śāstraprabhṛti); how [then] will anyone
(tasya) attain knowledge of the totality [of being] (sākalya)?119
Although these verses still pose considerable problems in the detail of their interpretation,
one can clearly recognize in them an elaboration on Maṇḍana’s main thesis regarding
meditation (see above, § 3.5): omniscience cannot result from meditation (bhāvanā –
362cd), for it only consists in the repeated practice (abhyāsa – 361a/363a) of what has
already been obtained by some other means (i.e. Scripture or inference). It is also possible
that v. 363ab should be read in reference to Maṇḍana’s idea of meditation as an essentially
non-productive activity, although this is far from certain.120 It seems in any case that
Prajñākaragupta substantially changes the nature of Maṇḍana’s argumentation by insisting,
above all, on the incapacity of the two ‘root-pramāṇas’ to apprehend all things,121 while
116
117
118
119
120
121
See PVA 168.11–12: (…) indriyajñānasya saṃnihitaviṣayasya darśanāt; “because one observes that a
sensory cognition has its object in [its] proximity.” Translation: Moriyama 2014: 245.
See PVA 2.360cd: svatantraṃ tu manojñānaṃ naiva kena cid iṣyate //; “On the other hand, a mental
cognition [that is] independent [of sense faculties] is never observed by anyone.” Translation: Moriyama
2014: 246. Cf. ViVK 15d: paratantraṃ bahir manaḥ // (S 458.3 – translated above, n. 85).
PVA 2.361–363: abhyāsāt spaṣṭatā tasya na sarvaviṣayā bhavet / āgamāśritatve tasya bhrāntatāpi
prasajyate // 361 // anumānaprasiddhaṃ tu vastu sarvaṃ na labhyate / tato na sarvaviṣayā bhāvanā
sarvavit katham // 362 // śāstrādyābhyāsataḥ śāstraprabhṛty evāvagacchatu / sākalyavedanaṃ tasya
kuta evāgamiṣyati // 363 // (p. 170).
My translation of the passage differs only punctually from that of Moriyama (2014: 246–247).
Even though this interpretation is clearly very tentative, such a solution would avoid the disturbing
redundancy of v. 363ab with respect to vv. 361–362. The hypothesis of an implicit reference to Kumārila’s
BṬ (= TS 3163), upheld by Moriyama (2014: 77), would be another way out of this difficulty, but I
must say I cannot easily read in Prajñākaragupta’s half-verse Kumārila’s objection that excellence or
superiority (atiśaya) in the knowledge of a treatise (śāstra) does not entail excellence in the knowledge
of another treatise (śāstrāntara): evaṃ śāstravicāreṣu dṛśyate ’tiśayo mahān / na tu śāstrāntarajñānaṃ
tanmātreṇaiva labhyate // “Thus we notice [in some people] a great superiority in the knowledge of
treatises, but this is not sufficient [to establish their] knowledge of other treatises.”
The pūrvapakṣin’s ground for refusing access to the totality of being to inference and Scripture is in
itself far from clear. It is almost certain, as rightly pointed out by Moriyama (2014: 59–60, n. 14), that he
discards inference in v. 362ab on the basis of Kumārila’s remark in ŚlV (codanā°) 115cd that inference
and similar pramāṇas cannot grasp objects in the future (bhaviṣyant). His argument against Scripture
in v. 361cd is in turn quite obscure, and I am not at all convinced by Yamāri’s recourse to the (typically
68
Maṇḍana Miśra’s Excursus on the Buddha’s Omniscience
Maṇḍana rather insisted on the “heteronomy” (pāratantrya) of meditation, preventing it
from becoming a pramāṇa in the full sense of the term. So, if the general structure of
this pūrvapakṣa seems to follow that of ViV 15, the detail of the argument is a blend of
Kumārila’s, Maṇḍana’s and – one may think – Prajñākaragupta’s own reflections.
4.5 Assuming, as I did, that most arguments in the first part of this pūrvapakṣa are
drawn from the ViV will also, it is hoped, allow us to solve problems in the detail of
the interpretation of that passage of the Vārttikālaṃkāra. Two verses are particularly
problematic, namely vv. 2.358 and 2.359cd. The first verse is interpreted by Moriyama
(2014: 57–58 and 245, n. 12) as an allusion to ŚlV (codanā°) 112–114 which is, as we
remember, a crucial group of stanzas dealing with the mutual limitation of the senses,
barring them the access to the totality of being. One has some difficulty, however, to read
this argument in Moriyama’s translation of PV 2.358:122
If an omniscient being arises despite the non-distinction regarding sense faculties and objects [between omniscient beings and ordinary people], everyone
would become omniscient because of the non-distinction regarding sense
faculties and objects. (Moriyama 2014: 245)
Although Moriyama does not provide much explanation for this, the logic behind his
translation seems to be the following: since the senses of the (putative) Omniscient – the
historical Buddha for instance – are not different from ours, they share the same limitations
(358ab); if we suspend this limitation (admitting, for instance, that the eye could grasp
sounds or smells), then there is no reason why everybody should not become omniscient
(358cd). The interpretation of the compound indriyārthāviśeṣa as the “non-distinction of
the senses and the object [in us and in an omniscient being]” looks quite forced though,
which makes me suspect that this translation somehow misses the point. The argument
becomes clearer, I think, if we relate Prajñākaragupta’s remark to Maṇḍana’s reasoning on
the relation of identity (tanmātrapratibandha/°anubandha) possibly underlying a relation
of invariable concomitance (vyāpti) between the means of valid knowledge (pramāṇa) and
the object to be known (prameya) (see above, § 2.2 [argument 2AIb] and nn. 31–32):123
should an object be identical with the means to know it (e.g., the senses), there would be no
122
123
Buddhist) argument of an “absence of connection [of speech] with external objects” (phyi rol gyi don
dang ’brel med pa – Tibetan text quoted by Moriyama [2014: 246, n. 18]), which I find very unlikely
to come from the mouth of a mīmāṃsaka opponent. Though many interpretations of this half-verse
are possible, I think it would make more sense for the pūrvapakṣin to underline, while speaking of
the “erroneousness” (bhrāntatā) of verbal cognition, its conceptual character. The opponent would
then reject Dharmakīrti’s claim that a conceptual (i.e. erroneous) cognition arising from the audition of
Buddhist Scriptures could lead by its mere repetition to a non-conceptual (i.e. non-erroneous) cognition.
The Sanskrit text of PVA 2.358 reads as follows: indriyārthāviśeṣe ’pi yadi sarvavidudbhavaḥ / sarvajña
eva sarvaḥ syād indriyārthāviśeṣataḥ // (p. 170).
Recall that such a relation would allow us, in the hypothesis considered by Maṇḍana, to infer the absence
of the object to be known (prameyābhāva) – i.e., an omniscient being – from the absence of a pramāṇa
establishing its existence, as when we infer for instance the absence of smoke (= vyāpya) from the
absence of fire (= vyāpaka).
Hugo David
69
difference between existing and being known, and everyone would become omniscient.124
On that basis, we can hopefully render Prajñākaragupta’s argument in a more faithful way:
Even if an omniscient being arose out of the absence of difference between a
sense and [its] object (indriyārthāviśeṣa), everybody would become omniscient
because of this absence of difference between a sense and [its] object!
This is not at all a central point in Maṇḍana’s argumentation, and I am struck by the amount
of knowledge of Mīmāṃsā Prajñākaragupta expects from his reader (unless, of course,
Maṇḍana himself is borrowing the argument from a Buddhist source). The same impression
results from another possible hint at ViV 15 in PVA 2.359cd. As we have seen above (n.
107), the first half of this verse (aśucyādirasāsvādasaṃgamaś cānivāritaḥ /) is almost
certainly a paraphrase of a verse of the BṬ (= TS 3144) arguing that an omniscient being,
who would experience all things, would also experience the taste of impure things, etc.
(aśucyādirasa).125 The second half of the verse (prāpyakārīndriyatve ca sarvavit katham
ucyate //) is read by Moriyama (2014: 245) as a continuation of this argument, and translated
as follows:
And [thus,] if [an omniscient being’s] sense faculties function after having
had a direct connection [with an object], how can he be [honorably] called an
omniscient being?
This translation is in itself impeccable, and it also makes perfect sense to say that the
perception of impure things is especially problematic if the senses operate while reaching
(prāpyakārin) their object. The presence of ca in pāda c is disturbing though,126 and
suggests another argument may be alluded to. As we saw (§ 3.2), the difference between
prāpyakārin and aprāpyakārin is also mobilised by Maṇḍana while dealing with sensory
perception to establish that neither explanation of perception (i.e. with and without a contact
between the senses and the object) can account for a knowledge of past and future objects
(S 465.1–466.2). I find it plausible that Prajñākaragupta reminds us of this argument, a
possibility that would also match our main hypothesis that he is following the chronological
order of ViV 15. If this proved correct, the allusion would be even more elliptic than in
the preceding case, and would presuppose a reader fully conversant with the detail of
Maṇḍana’s argumentation.
124
125
126
ViV 15: (…) tanmātrapratibandhābhāvāc ca, anyathā sarvasya sarvadarśitvaprasaṅgaḥ, aviśeṣāt
(S 459.4–460.1 – translated above, § 2.2). Supposing Prajñākaragupta is indeed alluding to that argument, it is possible that he interprets aviśeṣa in ViV 15 as well as the absence of difference between
the senses and the object (indriyārthāviśeṣa). This intepretation would differ from Vācaspati’s understanding of that term as referring to the absence of difference between us and the Omniscient. See NyK
460.7–8: aviśeṣād asmadādīnāṃ bhavadabhimatena saha sarvavidā. My translation of the passage
(above, § 2.2) follows this last interpretation.
PVA 2.359ab: aśucyādirasāsvādasaṃgamaś cānivāritaḥ /; “And [for an omniscient being who perceives
everything through the sense faculties], the connection with the experience of tasting an impure [thing],
etc. is unavoidable.” Translation: Moriyama 2014: 245.
As rightly pointed out by Moriyama (2014: 246).
70
Maṇḍana Miśra’s Excursus on the Buddha’s Omniscience
4.6 All this suggests that the ViV was not only known to Buddhist scholars, but that
its contents were also fairly widespread in learned Buddhist circles by the end of the 8th
century. The following table, which also integrates evidence found in previous scholarship,
summarizes my hypotheses concerning the sources of this pūrvapakṣa:
ŚlV/BṬ/PV/ViV
PVA
ViV 15 (S 445.1–458.1)
2.358–370
(prose
168.5–9
2.358–370
(prose
168.11–12
2.358
2.359ab
2.359cd
2.360
2.361–363
2.362ab
2.364
2.365
2.366
2.358–370
(prose
168.9–10/2.367–369
2.370/PVAS 114.26
ViVK 15a/ViV 15 (S 461.2–462.1)
ViV 15 (S 459.4–460.1)
BṬ (= TS 3144)
Idem/ViV 15 (S 465.1–466.2)
ViVK 15d/ViV 15 (S 468.1–470.1)
ViV 15 (S 542.1–555.2)
ŚlV (cod°) 115cd/BṬ (= TS 3173cd)
ŚlV (cod°) 134/BṬ (= TS 3191)
ŚlV (cod°) 137/BṬ (= TS 3238–3240)
≈ TS 3249/ViV 18 (S 675.1–676.1)
PV(SV) 1.335
ViV 18 (S 676.1–3)
introd.)
introd.)
introd.)
5. Conclusion
Having reached the term of this enquiry, Maṇḍana Miśra appears to us, without contest,
as the other great voice of Mīmāṃsā in the early debate on the Buddha’s omniscience.
Less massive, less uncompromising, less influential also than Kumārila’s, his critique is
nevertheless more complex, and philosophically more ambitious. It may also have served a
slightly different purpose. For sure, Maṇḍana’s final view essentially coincides with that
of his predecessor: no Omniscient can legitimately claim to instruct people about their
religious duties, their origin or destiny, or about the path leading them to beatitude. Yet
this reaffirmation of the basic Mīmāṃsā position on religious authority does not imply,
in the case of Maṇḍana, a complete hostility to the ideal of omniscience, as shown by the
following passage of the Brahmasiddhi:127
127
BS 2.106cd–107: nanu prapañcaśūnyasyādvaitasya brahmarūpasya jñeyābhāvād īśitavyābhāvāc ca
vijñānam aiśvaryaṃ cānupapannam, tatra sarvajñaḥ, sarveśvara ity api śrutī samādheya eva. – naitat
sāram, yato neśitavyakṛtam īśvaratvam, jñeyakṛtaṃ vā jñātṛtvam, kiṃ tu siddhena jñānarūpeṇa siddhayā ceśanaśaktyā jñeyam avāpnoti, īśitavyaṃ ca viniyuṅkte praśāsti ca, prakāśadāhavat. siddhena
hi prakāśarūpeṇa prakāśyaṃ prakāśayati vivasvān, na tu prakāśyādhīnaṃ tasya prakāśarūpam, dāhyādhīnā vāgner dāhaśaktiḥ. tathā ca taccaitanyenaiva kṛtsnasya prapañcasyāvabhāsanāt tasya bhāsā
sarvam idaṃ vibhāti, nānyo ’to ’sti draṣṭetyādiśruteḥ sarvajñatvam (127.5–13).
Hugo David
71
[Objection:] but, for that non-dual [entity] having the nature of Brahman,
for whom there is no proliferating [universe], there is nothing to be known
(jñeya) and nothing to be ruled over (īśitavya), so knowledge (vijñāna) and
sovereignty (aiśvarya) are unaccountable [in its case]. If it is so, scriptural
passages [mentioning Brahman as] “omniscient” (sarvajña) and “lord of all”
(sarveśvara) must be trusted blindly! – This is not true, because sovereignty is
not brought about by those who are ruled over, nor is it the known [object]
that makes one into a knower (jñātṛ). Quite the contrary! [Only] when the
form of knowledge is established, or the ability to reign, may one attain a
knowable [object] or assign tasks and govern those to be ruled over. [It all
happens] as in the case of light (prakāśa) and burning (dāha): [only] when
the sun’s luminosity is established may it shed light on [objects] to be illumined; its luminosity is not due to [there being] something to illumine, no
more than a fire’s capacity to burn is due to [there being] something to burn.
Thus, since the whole proliferating [universe] manifests itself only thanks to
His consciousness, as stated in scriptural passages like “All this shines only
through His splendour,”128 “There is no other seer than Him [= the ātman],”129
[Brahman] is [said to be] omniscient (sarvajña).
In this text, quite unique in Maṇḍana’s work, we see the lineaments of an alternative
concept of omniscience, virtually escaping the objections raised in the ViV. Omniscience
for Maṇḍana is not a matter of apprehending past and future, or perceiving the extremely
large or extremely small; indeed it is not at all about knowing objects. Omniscience is
understood here, in a negative way, as the absence of obstruction of the natural property
of awareness (jñāna) pertaining to Brahman, inversely proportional to the presence of a
multitude of knowable objects (jñeya).130 As such it would be vain to ask for its cause, and
the means to ascertain it is, of course, none but the eternal Veda. To put it differently, we
can read in ViV 15–25 an attempt to release omniscience from its ties with the problem
of dharma, which are as tight in the case of the Buddha as they are in the case of Īśvara
for those who uphold Him. In that sense, his critique certainly contributes to Kumārila’s
apologetic enterprise of (re)affirming the Veda as the one source of all religious authority.
But at the same time it also paves the way for a reevaluation of omniscience as part of a
Vedāntic soteriology, and thereby for its integration into the conceptual architecture of
Uttara-Mīmāṃsā.
References and abbreviations
Bhatt 1989 G.P. Bhatt, The Basic Ways of Knowing. An In-depth Study of Kumārila’s
Contribution to Indian Epistemology. Delhi 2 1989.
128
129
130
Kaṭhopaniṣad 5.15.
Bṛhadāraṇyakopaniṣad 3.7.23.
See also Ānandapūrṇa’s comment: jāḍyapāratantryābhāvopalakṣitaṃ rūpaṃ jñānādi, na tu saviṣayam;
“Knowledge and [sovereignty] are natures marked by the absence of insensibility and dependence, they
do not imply [the existence of] an object” (Bhāvaśuddhi 446.11).
72
Maṇḍana Miśra’s Excursus on the Buddha’s Omniscience
Bhāvaśuddhi Bhāvaśuddhi of Ānandapūrṇamuni and Abhiprāyaprakāśikā of Citsukhamuni. Two Commentaries on Brahmasiddhi, ed. N.S. Anantakrishna Sastri. Madras
1963.
Brahmasiddhivyākhyā See BS.
Bṛhadāraṇyakopaniṣad Bṛhad-Āraṇyaka-Upaniṣad (traduite et annotée par Émile Senart). Paris 2 1967.
Bṛhatī Bṛhatī of Prabhākara Miśra (on the Mīmāṃsāsūtrabhāṣya of Śabarasvāmin) with
Ṛjuvimalāpañcikā of Śālikanātha. Vol. 1.1, ed. S.K. Rāmanātha Śāstrī. Madras 1934.
BS Brahmasiddhi by Ācārya Maṇḍanamiśra with [the] Commentary by Śaṅkhapāṇi, ed.
S. Kuppuswami Sastri. Madras 1937.
BṬ Bṛhaṭṭīkā (Kumārila). See TS.
Bühnemann 1980 G. Bühnemann, Der Allwissende Buddha. Ein Beweis und seine Probleme. Ratnakīrtis Sarvajñasiddhi (übersetzt und kommentiert). Vienna 1980.
Chatterjee 1978 S. Chatterjee, The Nyāya Theory of Knowledge. A Critical Study of some
Problems of Logic and Metaphysics. Calcutta 4 1978.
David 2013 H. David, Action Theory and Scriptural Exegesis in Early Advaita-Vedānta
(1). Maṇḍana Miśra on upadeśa and iṣṭasādhanatā. In: Scriptural Authority, Reason,
and Action. Proceedings of a Panel at the 14th World Sanskrit Conference, Kyōto,
September 1st –5th , 2009, ed. V. Eltschinger and H. Krasser. Vienna 2013.
David 2017 H. David, Une ontologie du commandement ? Réflexions sur l’idée d’existence, l’impératif et l’objet du Veda. ThéoRèmes 11 (2017). https://doi.org/10.4000/
theoremes.1216
David (forthcoming) H. David, Action Theory and Scriptural Exegesis in Early AdvaitaVedānta (3). Maṇḍana Miśra on Ontology, Time and Commandment. In: A road less
travelled, ed. V. Eltschinger, B. Kellner, and E. Mills.
Eltschinger 2005 V. Eltschinger, Recherches sur la philosophie religieuse de Dharmakīrti.
I. Le Bouddha comme Śāstṛ et comme Sugata. Asiatische Studien/Études Asiatiques
59 (2005) 395–442.
Eltschinger 2007 V. Eltschinger, Penser l’autorité des Écritures. La polémique de Dharmakīrti contre la notion brahmanique orthodoxe d’un Veda sans auteur. Autour de
Pramāṇavārttika I.213–268 et Svavṛtti. Vienna 2007.
Eltschinger 2008 V. Eltschinger, Śaṅkaranandana’s Sarvajñasiddhi. A Preliminary Report.
In: Manuscripta Buddhica I. Sanskrit Texts from Giuseppe Tucci’s Collection Part I,
ed. F. Sferra. Rome 2008, 115–155.
Eltschinger 2009 V. Eltschinger, On the Career and the Cognition of Yogins. In: Yogic
Perception, Meditation and Altered States of Consciousness, ed. E. Franco and D.
Eigner. Vienna 2009, 169–213.
Eltschinger 2014a V. Eltschinger, Buddhist Epistemology as Apologetics. Studies on the
History, Self-understanding and Dogmatic Foundations of Late Indian Buddhist
Philosophy. Vienna 2014.
Eltschinger 2014b V. Eltschinger, The Four Nobles’ Truths and their 16 Aspects: On
the Dogmatic and Soteriological Presuppositions of the Buddhist Epistemologists’
Views on Niścaya. Journal of Indian Philosophy 42 (2014) 249–273.
Hugo David
73
Eltschinger 2015 V. Eltschinger, Latest News from a Kashmirian “Second Dharmakīrti.”
In: Cultural Flows across the Western Himalaya, ed. P. Mc Allister, C. ScherrerSchaub, and H. Krasser. Vienna 2015, 303–363.
Eltschinger/Krasser/Taber 2012 V. Eltschinger, H. Krasser, and J. Taber, Can the Veda
speak? Dharmakīrti against Mīmāṃsā exegetics and Vedic authority. An annotated
translation of PVSV 164,24–176,16. Vienna 2012.
Franco 2011 E. Franco, Perception of yogis – Some epistemological and metaphysical
considerations. In: Religion and Logic in Buddhist Philosophical Analysis. Proceedings of the Fourth International Dharmakīrti Conference (Vienna, August 23–27,
2005), ed. H. Krasser, H. Lasic, E. Franco, and B. Kellner. Vienna 2011, 81–98.
Franco 2012 E. Franco, Once Again on the Desires of the Buddha. In: Devadattīyam.
Johannes Bronkhorst Felicitation Volume, ed. F. Voegeli, V. Eltschinger, D. Feller,
M-P. Candotti, B. Diaconescu, and M. Kulkarni. Bern 2012, 229–245.
Frauwallner 1962 E. Frauwallner, Kumārila’s Bṛhaṭṭīkā. Wiener Zeitschrift für die Kunde
Südasiens 6 (1962) 78–90.
Frauwallner 1968 E. Frauwallner, Materialien zur ältesten Erkenntnislehre der Karmamīmāṃsā. Vienna 1968.
Goodman 1989 S.B. Goodman, A Buddhist Proof for Omniscience: The “Sarvajñasiddhi”
of Ratnakīrti. Unpublished PhD Thesis, Temple University 1989.
Griffiths 1990 P. Griffiths, Omniscience in the Mahāyānasūtrālaṃkāra and its Commentaries. Indo-Iranian Journal 33 (1990) 85–120.
Harikai 1985 K. Harikai, Kumārila no Shakosun Kan (Kumārila’s views on Śākyamuni
Buddha). Journal of the Nippon Buddhist Research Association 50 (1985) 49–65 (in
Japanese).
Inami 1996 M. Inami, Puramanavarttika Puramanasiddi sho no kenkyu (5) (*A study
of PV II [5]). Shimane Kenritsu Kokusai Tankidaigaku Kiyo 3 (1996) 77–100 (in
Japanese).
Jackson 1991 R. Jackson, Dharmakīrti’s Attitude Toward Omniscience. In: Pt. Dalsukh
Bhai Malvania Felicitation Volume I, ed. M.A. Dhaky and S. Jain. Varanasi 1991,
230–246.
Kajiyama 1998 Y. Kajiyama, An Introduction to Buddhist Philosophy: an annotated
translation of the Tarkabhāṣā of Mokṣākaragupta. Vienna 2 1998.
Kataoka 2003a K. Kataoka, Kumārila’s Critique of Omniscience. Indo Shisōshi Kenkyū
15 (2003) 35–69.
Kataoka 2003b K. Kataoka, The Mīmāṃsā Definition of pramāṇa as Source of New
Information. Journal of Indian Philosophy 31 (2003) 89–103.
Kataoka 20111 /20112 K. Kataoka, Kumārila on Truth, Omniscience, and Killing. Part
1: A Critical Edition of Mīmāṃsāślokavārttika ad 1.1.2 (codanāsūtra); Part 2: An
Annotated Translation of Mīmāṃsāślokavārttika ad 1.1.2 (codanāsūtra). 2 vols.
Vienna 2011.
Katsura 1984 Sh. Katsura, Dharmakīrti’s Theory of Truth. Journal of Indian Philosophy
12 (1984) 215–235.
Kawasaki 1992 Sh. Kawasaki, Issaichi-shiso no kenkyu (*A Study of the Omniscient Being
[sarvajña] in Buddhism). Tokyo 1992 (in Japanese).
74
Maṇḍana Miśra’s Excursus on the Buddha’s Omniscience
Krasser 2001 H. Krasser, On Dharmakīrti’s Understanding of pramāṇabhūta and His Definition of pramāṇa. Wiener Zeitschrift für die Kunde Südasiens 45 (2001) 173–199.
Krasser 2012 H. Krasser, Bhāviveka, Dharmakīrti and Kumārila. In: Devadattīyam. Johannes Bronkhorst Felicitation Volume, ed. F. Voegeli, V. Eltschinger, D. Feller, M-P.
Candotti, B. Diaconescu, and M. Kulkarni. Bern 2012, 535–594.
La Vallée Poussin 1902 L. de La Vallée Poussin, On the Authority of Buddhist Āgamas.
Journal of the Royal Asiatic Society 34.2 (1902) 363–376.
Lindtner 2001 Ch. Lindtner, Bhavya on Mīmāṃsā. Mīmāṃsātattvanirṇayāvatāraḥ.
Madras (Chennai) 2001.
Madhyamakahṛdayakārikā See Lindtner 2001.
McClintock 2010 S.L. McClintock, Omniscience and the Rhetoric of Reason. Śāntarakṣita and Kamalaśīla on Rationality, Argumentation and Religious Authority. Boston
2010.
MīSū Mīmāṃsāsūtra. See Frauwallner 1968.
Miyasaka 1972 Y. Miyasaka, Pramāṇavārttika-kārikā (Sanskrit and Tibetan). Acta Indologica 2 (1971/72) 1–206.
Moriyama 2014 Sh. Moriyama, Omniscience and Religious Authority. A Study on Prajñākaragupta’s Pramāṇavārttikālaṅkārabhāṣya ad Pramāṇavārttika 2.8–10 and 29–33.
Berlin/Zürich 2014.
Much 1991 M.T. Much, Dharmakīrtis Vādanyāya (Teil 1: Sanskrit-Text). Vienna 1991.
Natarajan 1995 K. Natarajan, The Vidhi Viveka of Maṇḍana Miśra: understanding Vedic
Injunctions. Delhi 1995.
NB Nyāyabindu (Dharmakīrti). In: Dharmottarapradīpa (being a sub-commentary on
Dharmottara’s Nyāyabinduṭīkā, a commentary on Dharmakīrti’s Nyāyabindu), ed.
Dalsukhbhai Malvania. Patna 2 1971.
NM1 /NM2 Nyāyamañjarī of Jayanta Bhaṭṭa, with Ṭippaṇī – Nyāyasaurabha by the editor,
ed. K.S. Varadācārya. 2 vols. Mysore 1969 and 1983.
Nyāyabinduṭīkā See NB.
Nyāyavārttikatātparyaṭīkā Nyāyavārttikatātparyaṭīkā of Vācaspatimiśra, ed. Anantalal
Thakur. New Delhi 1996.
Nyāyasūtra/Nyāyabhāṣya Gautamīyanyāyadarśana with Bhāṣya of Vātsyāyana, ed. Anantalal Thakur. New Delhi 1997.
NyK Nyāyakaṇikā. See ViV/ViVK (S).
Ono 2000 M. Ono, Prajñākaraguptas Erklärung der Definition gültiger Erkenntnis
(Pramāṇavārttikālaṃkāra zu Pramāṇavārttika II 1–7) (Materialien zur Definition gültiger Erkenntnis in der Tradition Dharmakīrtis 3). Teil 1: Sanskrit-Text und
Materialien. Vienna 2000.
Pandey 1972 L. Pandey, The Buddhist Conception of Omniscience. Unpublished PhD
Thesis, McMaster University, Hamilton (Ontario) 1972.
Pemwieser 1991 M. Pemwieser, Materialien zur Theorie der yogischen Erkenntnis im
Buddhismus. Unpublished M.A. Thesis, University of Vienna 1991.
Potter 2014 K.H. Potter, Philosophy of Pūrva-Mīmāṃsā. Delhi 2014.
Prakīrṇaprakāśa Vākyapadīya of Bhartṛhari with the Commentary of Helārāja. Kāṇḍa
III. Vol. 1, ed. K.A. Subrahmania Iyer. Poona 1963.
Hugo David
75
PS Dignāga’s Pramāṇasamuccaya, Chapter 1. A hypothetical reconstruction of the Sanskrit text with the help of the two Tibetan translations on the basis of the hitherto
known Sanskrit fragments and the linguistic materials gained from Jinendrabuddhi’s
Ṭīkā, ed. E. Steinkellner. Vienna 2005. https://www.oeaw.ac.at/fileadmin/Institute/I
KGA/PDF/forschung/buddhismuskunde/dignaga_PS_1.pdf, last visited 10-12-2019.
PV 1/PVSV The Pramānavārttikam of Dharmakīrti. The first Chapter with the Autocommentary, ed. R. Gnoli. Rome 1960.
PV 2–4 See Miyasaka 1972.
PVA See Moriyama 2014.
PVAS Pramāṇavārtikabhā[ṣ]yam or Vārtikālaṅkāraḥ of Prajñākaragupta (being a commentary on Dharmakīrti’s Pramāṇavārtikam), ed. R. Sāṅkrityāyana. Patna 1953.
PVin/PVinK 1–2 Dharmakīrti’s Pramāṇaviniścaya. Chapters 1 and 2, ed. E. Steinkellner.
Beijing/Vienna 2007.
Ruegg 1994 D. Seyfort Ruegg, Pramāṇabhūta, *pramāṇa(bhūta)-puruṣa, pratyakṣadharman and sākṣātkṛtadharman as Epithets of the ṛṣi, ācārya and tathāgata in Grammatical, Epistemological and Madhyamaka Texts. Bulletin of the School of Oriental
and African Studies, University of London 57.2 (1994) 303–320.
ŚlV Mīmāṃsāślokavārtikam. śrīmatkumārilabhaṭṭapādaviracitam (…)śrīmatpārthasārathimiśrapraṇītayā nyāyaratnākarākhyayā vyākhyayānugatam, ed. Mānavallyupāhvatailaṅgarāmaśāstrī. Benares (Kāśī) 1898.
ŚlV (codanā°) See Kataoka 20111.
SSiR Sarvajñasiddhi (Ratnakīrti). In: Ratnakīrtinibandhāvaliḥ (Buddhist Nyāya Works of
Ratnakīrti), ed. Anantalal Thakur. Patna 2 1975, 1–31.
SSiŚ Sarvajñasiddhi (Śaṅkaranandana). See Eltschinger 2008.
Stcherbatsky 1926 Th. Stcherbatsky, Über die Nyāyakaṇikā des Vācaspatimiśra und die
indische Lehre vom kategorischen Imperativ. In: Beiträge zur Literaturwissenschaft
und Geistesgeschichte Indiens. Festgabe Hermann Jacobi zum 75. Geburtstag, ed.
W. Kirfel. Bonn 1926, 369–380.
Stcherbatsky 19301 /19302 Th. Stcherbatsky, Buddhist Logic. 2 vols. Leningrad 1930.
Steinkellner 1967 E. Steinkellner, Dharmakīrti’s Hetubinduḥ. Teil II: Übersetzung und
Anmerkungen. Vienna 1967.
Steinkellner 1977 E. Steinkellner, Jñānaśrīmitra’s Sarvajñasiddhiḥ. In: Prajñāpāramitā
and related systems: Studies in honor of Edward Conze, ed. L. Lancaster and L.
Gómez. Berkeley 1977, 383–393.
Steinkellner 1978 E. Steinkellner, Yogische Erkenntnis als Problem im Buddhismus. In:
Transzendenzerfahrung, Vollzugshorizont des Heils, ed. G. Oberhammer. Vienna
1978, 121–134.
Steinkellner 20131 /20132 E. Steinkellner, Dharmakīrtis frühe Logik. Annotierte Übersetzung der logischen Teile von Pramāṇavārttika 1 mit der Vṛtti (Introduction, Übersetzung, Analyse). 2 vols. Tokyo 2013.
Steinkellner/Much 1995 E. Steinkellner and M.T. Much, Texte der erkenntnistheoretischen Schule des Buddhismus. Göttingen 1995.
Stern 1988 E. Stern, Vidhivivekaḥ of Maṇḍana Miśraḥ, with commentary Nyāyakaṇikā
of Vācaspatimiśraḥ and supercommentaries Juṣadhvaṅkaraṇī and Svaditaṅkaraṇī
76
Maṇḍana Miśra’s Excursus on the Buddha’s Omniscience
of Parameśvaraḥ. Critical and annotated edition: the pūrvapakṣaḥ (Sanskrit text).
Unpublished PhD Thesis, University of Pennsylvania 1988.
SV svavṛtti
Taber 2005 J.A. Taber, A Hindu Critique of Buddhist Epistemology. Kumārila on Perception. The “Determination of Perception” chapter of Kumārila Bhaṭṭa’s Ślokavārttika.
Translation and Commentary. London/New York 2005.
Taber 2011 J. Taber, Did Dharmakīrti think the Buddha had desires? In: Religion and
Logic in Buddhist Philosophical Analysis. Proceedings of the Fourth International
Dharmakīrti Conference (Vienna, August 23–27, 2005), ed. H. Krasser, H. Lasic, E.
Franco, and B. Kellner. Vienna 2011, 437–448.
Tantravārttika Śrīmajjaiminipraṇīt[aṃ] mīmāṃsādarśan[am]. Vol. 2 (ad MīSū 1.2–2.1),
ed. Gaṇeśaśāstrī Ambādāsa Jośī. Poona 3 1981.
Tātparyaṭīkā Ślokavārttikavyākhyā Tātparyaṭīkā of Umveka Bhaṭṭa, ed. S.K. Rāmanātha
Śastrī. Madras 2 1971.
TS Tattvasaṅgraha of Ācārya Śāntarakṣita, with the commentary Pañjikā of Śrī Kamalaśīla,
ed. Dwarikadas Shastri. 2 vols. Benares (Varanasi) 1968.
TSP See TS.
Vādanyāya See Much 1991.
Vākyapadīya (SV) svavṛtti Vākyapadīya of Bhartṛhari with the Commentaries Vṛtti and
Paddhati of Vṛṣabhadeva. Kāṇḍa I, ed. K.A. Subrahmania Iyer. Poona 1966.
Vetter 1964 T. Vetter, Erkenntnisprobleme bei Dharmakīrti. Vienna 1964.
ViV/ViVK M = Vidhivivekaḥ śrīmadācāryamaṇḍanamiśraviracitaḥ pūjyapādaśrīmadvācaspatimiśranirmitayā nyāyakaṇikākhyayā vyākhyayā samalaṃkṛtaḥ, ed. Mānavallyupāhvatailaṅgarāmaśāstrī. Benares 1907 (reprint from “The Pandit”). G = Vidhivivekaḥ of Śrī Maṇḍana Miśra, with the Commentary Nyāyakaṇikā of Vācaspati
Miśra, ed. Mahaprabhu Lal Goswami. Benares 1978. S = Stern 1988.
Watanabe 1987 Sh. Watanabe, Śubhagupta’s Sarvajñasiddhikārikā. Journal of the Naritasan Institute for Buddhist Studies 10 (1987) 55–74 (in Japanese).
Watanabe 1988 Sh. Watanabe, Tattvasaṃgraha XXVI, kk. 3247–3261 and kk. 3622–3646.
Journal of the Naritasan Institute for Buddhist Studies 11 (1988) 501–533 (in
Japanese).
Yaita 1985 H. Yaita, On anupalabdhi, annotated translation of Dharmakīrti’s Pramāṇavārttikasvavṛtti (I). Taishō Daigaku Daigakuin Kenkyū Ronshū 9 (1985) 216–199.
On the Determination of Causation by Dharmakīrti
by
Eli Franco
One of the benefits of reading new translations is not only to learn new things, but also
to find out that things one took for granted and thought to be generally accepted were not
at all so. I had such an experience last year when reading the awe-inspiring translation of
the logical portions of the PVSV by Ernst Steinkellner.1 To my surprise, his interpretation
of the relationship between cause and effect, and especially their putative resemblance
in the famous discussion in PVSV on v. 34, were not quite what I took to be the case
when I wrote on the same passage some twenty five years ago. I am, therefore, very
grateful to Steinkellner to have inspired me to take a fresh look at an old problem. My (not
insubstantial) disagreement with Steinkellner on the topic of this paper detracts neither
from my admiration for the great scholar who has been a singular force in our discipline
for the past half a century nor from my gratitude for his friendship and support ever since I
took his seminar on Vādanyāya at the University of Vienna in 1981.
On the alleged similarity between cause and effect
The passage in Steinkellner’s book that took me by surprise concerns the determination of
the relationship between cause and effect, where Steinkellner argues at some length that
cause and effect must be of the same kind rather than similar. I do not know to what extent
this opinion is prevalent. He refers specifically to John Taber and Toshikazu Watanabe,2
who propose different notion of similarity. However in Watanabe’s case, I fail to see that
he assumes the similarity or identity to be between causes and effects. Taber informs me in
personal communication that his suggestion of similarity was only tentative. It is based
on the fact that in the Pramāṇasiddhi chapter Dharmakīrti argues that cognition can only
arise from cognition, senses only from senses, breath only from breath and so on. This
is certainly true, but the question that I want to raise here is whether such similarity is
essential or accidental to causal relationship.
Summarising his interpretation that cause and effect must belong to the same kind,
Steinkellner concludes (2013 II: 213):
Anders gesagt: “Rauch” ist eine Art der Gattung “Feuer.” Das heißt, er ist
etwas, das “feuerartig” ist. Er ist eben so nicht “ähnlich” oder “gleich” einer
Ursache, sondern kann, weil er seine wesentlichen Beschaffenheiten dieser
Ursache verdankt, als eine Art der Gattung der Ursache beurteilt werden.
1
2
Steinkellner 2013.
See Taber 2003: 490 and 492f., and Watanabe 2004: 58–60 referred to in Steinkellner 2013 II: 211f., n.
366.
Birgit Kellner et al., eds., Reverberations of Dharmakīrti’s Philosophy: Proceedings of the Fifth International Dharmakīrti
Conference Heidelberg, August 26 to 30, 2014. Vienna 2020, pp. 77–90.
78
On the Determination of Causation by Dharmakīrti
Put differently, “smoke” is a species of the genus “fire.” That is to say, it is
something “fiery.” It is thus not “similar” or “equal” to [its] cause; rather,
because it owes its special properties to its cause, it can be judged to be as a
species of the genus of the cause.
My first contention does not only concern the understanding of the similarity – whether it
means appertaining to the same kind, the same species or the same genus – but is more
radical: Cause and effect are not, or do not have to be, of the same species or the same
kind. They don’t even have to be similar. Furthermore, I claim that this question is not
at all addressed in the passage in question (PVSV on v. 34). Rather, what Dharmakīrti
maintains in this passage is that causes of the same kind produce effects of the same kind.
In other words, the similarity, or the appertainment to same kind, is between causes among
themselves and effects among themselves, not between causes and effects.3 To substantiate
this claim I suggest to take a close look at the passage in question (PVSV 22.10–19 on
34cd):
anyahetukatvān nāhetukatvam iti cet / na / tatrāpi tulyatvāt / tadabhāve ’py agnau bhavatīti / kathaṃ vā tato ’nyato vātajjananasvabhāvād bhavet / svayam
atatsvabhāvasyājananāt / tasyāhetutā syāt / na vai sa eva bhavati tādṛśasya
bhāvāt / anyādṛśād bhavan kathaṃ tādṛśaḥ syāt / tādṛśād dhi bhavan tādṛśaḥ
syāt / anyādṛśād api tādṛśo bhāve tacchaktiniyamābhāvān na hetubhedo bhedaka ity akāraṇaṃ viśvasya vaiśvarūpyaṃ syāt / sarvaṃ vā sarvasmāj jāyet
(read: jāyeta) / tasmāt kāraṇabhedābhedābhyāṃ kāryabhedābhedau / tan na
dhūmo ’rthād dṛṣṭākāravijātīyād bhavaty ahetukatvaprasaṅgāt /
Steinkellner’s translation4 (2013 I: 55):
(Einwand:) “Weil er eine andere Ursache haben (kann), ist er [auch dann,
wenn er ohne Feuer vorhanden ist] nicht ohne Ursache.” (Antwort:) (Das
ist) nicht (der Fall), denn auch im Falle dieser (anderen Ursache) [wäre die
Ursachelosigkeit] die gleiche, denn, auch wenn diese (andere Ursache) fehlt,
ist (der Rauch) beim (Vorhandensein von) Feuer (dennoch) vorhanden. Oder
weshalb könnte er aus diesem oder einem anderen entstehen, wenn (beide)
nicht das WesenN haben ihn hervorzubringen? Weil das, was selbst nicht dieses
WesenN (ihn hervorzubringen) hat, ihn nicht hervorbringt, wäre dieser (Rauch
nichts als) ursachelos.
(Einwand:) (Aus der anderen Ursache) entsteht ja nicht gerade dieser (gewöhnlich vom Feuer hervorgebrachte Rauch), denn es entsteht ein derartiger (tādṛśa). (Antwort:) Wieso ist er ein derartiger (tādṛśa), wenn er aus einem
nicht Derartigen entsteht? Weil er aus Derartigem entsteht, wäre dieser
3
4
Though the formulation is a bit ambiguous, I believe that this is what Mookerjee and Nagasaki also
mean in their translation (Mookerjee and Nagasaki 1964: 84): “There is no exception to the rule that
similars produce similars and dissimilars produce dissimilars.” Dunne (2004: 335–336) also does not
seem to share Steinkellner’s interpretation. Gillon’s translation (2009: 202) leaves the possibility open.
The emphasis indicates where our interpretations differ.
Eli Franco
79
(Rauch) nämlich ein derartiger. Wenn auch aus nicht Derartigem ein
derartiger (Rauch) entstünde, (würde) eine Verschiedenheit der Ursache
keine Verschiedenheit der Wirkung veranlassen, weil die Kräfte dieser (derartigen und andersartigen Ursachen) nicht [auf die jeweiligen Wirkungen]
eingeschränkt wären. Somit wäre die Vielseitigkeit des Universums ohne
Ursache oder alles würde aus allem entstehen. Daher ergeben sich die Verschiedenheit und Nichtverschiedenheit der Wirkungen aus der Verschiedenheit
und Nichtverschiedenheit der Ursachen. Infolgedessen entsteht der Rauch nicht
aus einer Sache, die von der [als Ursache] gesehenen Erscheinungsform [nämlich des Feuers] verschiedenartig ist, weil (sonst seine) Ursachelosigkeit folgen
würde.
My tentative translation:5
[Objection:] Because [smoke] has [also] another cause [than fire],6 it is not
without a cause [when it arises without fire].
[Reply:] No, because it’s the same in this case too.7 [In this case, one would
consider that smoke] arises when fire is present, even when that [other cause]
is absent.
Or [given different causes of smoke, which have different natures, if the one,
e.g., fire, has the nature of producing smoke, the other, having a different
nature, would not have the nature of producing smoke. Thus] how could
[smoke] arise either from that [fire that has the nature of producing smoke] or
from something else, which does not have the nature of producing it? What
itself does not have this nature [of producing smoke] does not produce [it].
Consequently [since that the other thing does not produce smoke, if smoke
would arise without fire] it would have no cause.
[Objection:] It is not the case that exactly that [smoke] arises [from the other
thing] because something of the same kind [as smoke] arises.
5
6
7
I thank Karin Preisendanz for making this translation more precise.
For instance, a termite mound or an anthill.
The argument is not entirely clear to me. How does it differ from the second argument which begins
with kathaṃ vā? Perhaps one can understand the first argument epistemologically: If smoke arises also
from other cause than fire, then just as one would not be able to infer smoke from fire, one would equally
not be able to infer smoke from the other cause, for one knows that even when the other cause is absent,
smoke could arise from fire. Thus, what Dharmakīrti would be claiming in the first argument (up to
kathaṃ vā) is not that smoke would actually be without a cause, but that its cause would not be inferable.
The second argument is clearer. Dharmakīrti defines (or in fact reduces) fire to that whose nature is
to produce smoke. If smoke arises from non-fire, it would arise from something whose nature is not
to produce smoke and thus without a cause. One may mention here that Dharmakīrti recognizes that
the determination of causation is not always possible. In a case of a so-called general effect, or effect
common to more than one causal complex (kāryasāmānya), the determination of the cause may not be
possible, for instance in PVSV on v. 12, the fact that one speaks may be based on desire or on compassion
(see also Franco 2012). Similarly, activity after rest, a special configuration (saṃsthānaviśeṣa) and so
on (see PV II 10f.) may prove a conscious agent, but not that this agent is an eternal God and not a
human being. How are such statements compatible with what is stated here?
80
On the Determination of Causation by Dharmakīrti
[Reply:] Inasmuch as it arises from something of a different kind [than fire],
how could it be of the same kind [as smoke]? Indeed, inasmuch as it arises
from something of the same kind [as fire], it is/must be8 of the same kind (i.e.,
not the same kind as the cause, but of the same kind as the other effects of
fire, namely, smoke; viz., effects of the same kind must arise from causes of
the same kind). If [an effect] of the same kind [as smoke] would also arise
from [a cause] of a different kind [than fire], a difference in cause would not
distinguish [effects] because there would be no restriction to the capacity of its
[cause to produce all kinds of effects]. Thus, the diversity of the world would
be without a cause, or everything would arise from everything. Therefore, the
difference and identity of effects are due to difference and identity of causes.
Therefore, smoke does not arise from something which is of a different kind
than that thing whose form has been observed [before whenever smoke arose]
because [in this case smoke] would be without a cause.
The upshot of Dharmakīrti’s argument in this passage is that if causes of different kinds
can produce the same effect, we will not be able to infer the cause from the effect. To make
such inference, the question whether the effect is similar to the cause is quite irrelevant.
The effect does not have to be similar or of the same kind as the cause in order to enable
such an inference. The fact that Dharmakīrti does not assume that the effect is similar to the
cause in all cases is more than clear in the example of the lotus and the cow dung, which
appears a bit later in this context. Dharmakīrti refers to an apparently wide-spread belief
that some lotuses do not arise from a lotus seed but from cow dung. It will be difficult to
maintain that a lotus and the cow dung are similar or of the same kind, all the more so
when Dharmakīrti himself expressly says that they belong do different kinds (vijātīya).9 If
one were to maintain that even cow dung and lotus belong to the same kind, the notion of
appertaining to same kind would become arbitrary, tautological, and circular when used
for the determination of causation. For if appertaining to same kind means, as Steinkellner
argues, the fact that the properties of the effect are due to the cause, then to be of the same
kind as something simply means to be produced by that thing. In other words, if we then
argue that a cause must be of the same kind as the effect, we would actually argue that the
cause must be a cause.
Modes of causation
My second and perhaps more important contention is that what Dharmakīrti states at the
beginning of PV I (and in the parallel passages in the Pramāṇaviniścaya, Hetubindu and
Vādanyāya) is not meant as complete doctrine of determination of causality. I do not wish
8
9
See the emendations suggested by Steinkellner (2013 I: 55, nn. 30–31).
See PVSV 22: 22–23: yad api kiṃcid vijātīyād bhavad dṛṣṭaṃ gomayādeḥ śalūkādi. Karṇakagomin
(PVSVṬ 106: 13–14) adds the example of śara grass (?) arising from a cow horn and a scorpion
arising from cow dung: ādiśabdād gauśṛṅgāc charo gomayād vṛścikaḥ. Note that the discussion here
concerns everyday practice. In Abhidharma causality one could indeed maintain similarity between
cause and effect and Dharmakīrti certainly accepts the notion of samanantarapratyaya. This is however
not relevant to the present context, which deals with inferences in everyday practice.
Eli Franco
81
to deny the importance of these passages, or even their paradigmatic role, but they do not
tell us the whole story. Curiously, studies of Dharmakīrti’s notion of causation10 have not
raised the question as to how Dharmakīrti puts his own theory into practice. I suggest,
therefore, to broaden the scope of the discussion and open up some new perspectives. It
is important to note that Dharmakīrti’s practice in determining causation is considerably
more complex and varied than his statement in PVSV on v. 34 and in the parallel passages
in his later works would suggest.
Determining permanent causes
According to the usual interpretation, Dharmakīrti’s suggestion for the determination of
causation (as based on PVSV on v. 34) consists in identifying a cause in a given situation
by isolating it from the environment. Given that all other conditions remain the same, and
upon the introduction of a certain new element the effect arises, while when this element
is removed, the effect does not arise, one can determine that the one is the cause of the
other.11
Thus, according to this statement, one would not be able to determine causal relation, if
an entity is permanent or constantly present, for one would be unable to observe whether the
effect is absent when the cause is absent. Yet Dharmakīrti admits that the earth, of course
along with other factors, is a cause of the sprout. For all practical purposes, the earth is
eternal and always present. How could it be determined as a cause of sprouts? Dharmakīrti’s
reply indicates that presence and absence are indeed not necessary to determine causation
in all cases. In the case of earth, it suffices to observe a transformation that brings about
changes in the result. For instance, by perfecting the earth with manure, ploughing it and
so forth, one observes changes in the quality of the sprout (PV II 25). These changes allow
one to determine that the earth is part of the causal complex of the sprout. (So when it
comes to God, the reason why he – at least for Dharmakīrti and his opponents it’s a he
– cannot be determined as a cause is, in the final analysis, not because he is eternal and
all-pervasive, but because he is changeless.)
Determining a “permanent” material cause
Similarly, one would hardly expect the process of introducing and removing the cause in the
case of a material cause (upādāna). In this case, just like the case of the earth and the sprout,
it’s the transformation in the material cause which allows its identification. This is stated
several times and in various forms, for instance in PV II 60–61: Without transformation in
the material cause (upādāna), there is no transformation in the effect, just as a plate does
not change without transformation in the clay. One would not seriously expect the enquirer
to remove the clay in order to observe whether the plate continues to exist.
10
11
The most important studies are mentioned in Steinkellner 2013 II: 185–186.
PVSV 22.3–5: yeṣām upalambhe tallaklṣaṇam anupalabdhaṃ yad upalabhyate, tatraikābhāve ’pi
nopalabhyate, tat tasya kāryam. See the translation below.
82
On the Determination of Causation by Dharmakīrti
Determining a “permanent” non-material cause
Dharmakīrti’s notion of causation is often apparent when he denies causal connection in
specific cases and we must pay attention to such cases as well. For instance, cognition and
body are always present together, at least throughout one’s life. How can one deny that
the body is the cause of cognition or that they are causes of each other? In this case too
Dharmakīrti relies on the same principle: a transformation of a cause must bring about
a transformation in the effect. Thus, if we observe a transformation of the one without a
transformation in the other, we can exclude the possibility that the two are causally related.
Therefore, when one observes a change in the cognition without a change in the body, one
can conclude that the cognition does not produce the body and vice versa. This does not
mean that the body is not at all a cause of cognition, but that it is not its so-called material
cause.
Temporality
Cause and effect are also connected by a temporal aspect: The effect must last as long as
its cause. For instance, if the body were the cause of cognition, the cognition would last as
long as the body, and thus there would be no dead body (PV II 51). Further, temporality
allows one to distinguish between material/main cause and auxiliary cause (upakāraka).
Fire may change the color of a pot, but the pot and its new color, unlike smoke, continues to
exist when the fire ceases (PV II 50). The auxiliary cause is responsible for some specific
aspect of the result, not for the result as such.
The influence of the body on cognition is explained by Dharmakīrti in this way, that is,
as an auxiliary cause. For instance, the transformation of the body due to poison causes a
mental transformation in the form of pain. Dharmakīrti claims that in this case the body is
only the object of cognition and the pain results from perceiving the body, not directly from
the body (PV II 48). He does not explain though why perceiving someone else’s body does
not cause pain in the same manner; one may assume that he would anchor the distinction
in the false apprehension of the body as belonging to oneself.
Gradation
Another important aspect to which cause and effect have to conform is the gradual arising.
If the cause does not change, the effect cannot arise gradually (or after a time). Thus, lack in
gradation allows one to exclude causal relation between breath and body: “How can breath
be gradual without its cause be gradual?” (PV II 107) Similarly, if the body is constant
throughout one’s life, cognitions cannot arise from it one after the other. They would have to
arise all at once. The gradual arising cannot be due to co-producers unless they bring about
changes in the cause (e.g., the body, PV II 43). I am not sure how Dharmakīrti would justify
this statement in some cases. For instance, light, senses and object, which are co-producers
of a cognition, do not bring about changes in the previous cognition.
Eli Franco
83
Locating or locative cause (special case of auxiliary cause)
Dharmakīrti’s causal theory uses not only upakāraka, but also ādhāra/āśraya, which may
be translated as locating cause. Dharmakīrti considers this cause from two aspects. If things
are momentary, the ādhāra causes them to be located in the same place. For instance, the
plate causes the berries, which would tend to fall on the floor and disperse in all directions,
to be produced in the same place. Similarly, the jug keeps the water from spilling (PV I
144, see also PV II 67–68, 74). If things are not momentary, the ādhāra is the cause that
prevents their movement. In this manner, Dharmakīrti explains the role of the body in
mental phenomena such as amnesia (smṛtibhraṃśa, PV II 76).
Reversibility of process
Reversibility of process also allows one to determine causation. For instance, the Cārvākas
argue that changes in the humours can account for the changing capacity of the body to
produce cognitions, i.e., when the humours are in strong disequilibrium which causes death,
the body is no longer capable of producing cognitions, just like a wick becomes incapable
to produce a flame, but Dharmakīrti retorts that when the humours of a dead body regain
their equilibrium, just as they do when fever is cured, life would arise again (PV II 54–55).
If the Cārvāka argues that the process is not reversible, like the transformation of wood
into charcoal, this is not correct because one applies medicine to reverse the transformation
of humours.
Increase and decrease
An important aspect for the determination of something as a cause is whether its increase
and decrease brings about the corresponding increase and decrease in the result. For
instance, lamp and light. However, one observes increase in the properties of cognition
such as wisdom, compassion, etc., without increase or decrease in the body. Therefore, the
body cannot be the support/cause of cognition (PV II 73). It is impossible that the effect
would be destroyed when the cause increases, e.g., when pitta increases, fever does not go
away (PV II 151). (The anomaly of water and homeopathic medicine would have presented
a challenge in this case.)
The Cārvāka claims that when the humours are balanced, the production of sperm
increases, which causes the increase of desire. This would be a valid argument, but Dharmakīrti retorts that a sick person may have strong desires and a healthy person none. Further,
one may have strong desires even without sperm. Increase in desire arises from increase in
pleasure, even when there is no increase in the humours (PV II 151). Further, one observes
sick persons with unbalanced humours and strong desires, and also that one ejaculates
blood when sperm is exhausted (PV II 153). (I’m afraid I cannot vouch for this observation.)
Another case where increase and decrease are used to reject causal relation is this:
Even if the material elements were the cause of consciousness, they cannot be the cause
of desire. The causal mode of the elements is not characterized by increase and decrease
because one cannot say that one living being is more alive or less alive than another, but
some living being have strong desires and some weak. Therefore the cause of desire must
be characterized by increase and decrease (PV II 167–169).
84
On the Determination of Causation by Dharmakīrti
Generality and specificity
This is an argument that I have only seen once. The modal correspondence between cause
and effect presupposed by Dharmakīrti is so strong that if the cause is general, the effect
cannot be specific. Sperm does not explain the nature of desire which is directed towards a
specific woman. If the Cārvāka claims that the beauty of the woman also plays a role, this
is not correct because one desires also ugly women. The argument looks like an aside, but
if we take it seriously, Dharmakīrti claims that if desire is specific towards a single women,
so should also be the sperm (PV II 154).
The numerical aspect
If the cause, or causal complex, is one, the effect is one, if the causes are many, the effects
are also many: If each atom is capable of producing a cognition, there would be as many
cognitions as there are atoms in the body. Similarly, if breath is a product of the atoms of
the body singly, there would be as many breaths as atoms (PV II 103–104).
Restricting, hindering and regulating aspects of causality
It is theoretically possible that although each atom of the body is capable of producing a
cognition, and thus there would be as many cognitions as atoms, the breath in the body
restricts the capacity of atoms to produce only one cognition at the time (PV II 103–104).
Dharmakīrti is not explicit about how this could work, but one can assume perhaps that
breath, being a part of the causal complex with one atom, cannot be a part of a causal
complex with another atom. The possibility is of course rejected by Dharmakīrti. There is
no restriction that one breath produces one cognition because one observes that several
cognitions arise during one long breath.
Mental properties such as compassion grow out of their own seed (svabīja); consequently since saṃsāra has no beginning, everyone should have become a Buddha by now.
This would indeed be the case if the same did not apply to the opposite negative properties
as well. Just as compassion arises from its own seed, so do hatred or aversion (dveṣa),
and so on. Thus, compassion and aversion obstruct each other’s development. It is for this
reason that great effort is needed to suppress aversion, etc., with their antidotes (pratipakṣa)
so that compassion can flow unhindered and reach its utmost degree, as in the case of the
Buddha (PV II 131).
Limited and unlimited causal processes
Interesting is Dharmakīrti’s distinction between limited and unlimited causal process. The
distinction depends on whether or not the causal process has a stable or an unstable result.
Certain results continue by themselves, by their own essence (svarasena pravartate), e.g.,
the change of color in burned wood or the increase in mental properties such as compassion.
Certain are limited because their causes are limited like jumping and its causes force and
effort. Others like boiling water have unstable support. Some are reversible, like heating
gold (PV II 124–126).
Eli Franco
85
Presence and absence
Finally, we should not forget our starting point. As long as the cause remains, the result
does not cease to exist (PV III 133cd). This is indeed the principle discussed in PVSV 34
and applied to smoke and fire.
Intermediary conclusion
To conclude this section, we see all kinds of correspondences between cause and effect.
For lack of a better word, I would like to call them modal correspondences. Existence
and inexistence or presence and absence are just one of them; others are transformation,
gradation, increase and decrease, reversibility and non-reversibility, generality and specificity, temporal aspects, numerical aspects and so on. I do not pretend to be exhaustive here.
However, what we do not see is similarity or appertaining to the same kind as an argument
or a consideration in the determination of causal relationship or in its denial.12 Nowhere is
it said, for instance, that the body cannot be the cause of cognition because it is of different
kind than the cognition.
If we would like to generalise the underlying principle behind the different modalities,
we could say that a change in the cause must bring about a change in the effect. PV II 111
puts it explicitly: if A is the cause of B, B changes when A changes. Or more literally:
“What does not change because of the change of something else is not the result of that
thing” (na hi tat tasya kāryaṃ yad yasya bhedān na bhidyate).
The problem of induction
I come now to my third contention, which concerns the problem of induction. Steinkellner
suggested two new interpretations of Dharmakīrti’s words, which he sees as complementary
rather than contradictory (2013 II: 210). The core of the discussion is Dharmakīrti’s famous
and enigmatic statement in PVSV 22.3–5 (German transl. in Steinkellner 2013 I: 54): yeṣām
upalambhe tallakṣaṇam anupalabdhaṃ yad upalabhyate, tatraikābhāve ’pi nopalabhyate,
tat tasya kāryam “When a perceptible thing unperceived [before] is perceived when several
[other things] are perceived, and is not perceived when even one among these [things] is
absent, it is the effect of that [one of these things]/of these [several things].”13
No matter whether one understands tat to refer to tatra or to eka, the straightforward
reading of this statement does not single out one thing such as fire as the cause of smoke,
and this in contradistinction to the verse on which this statement comments: kāryaṃ dhūmo
hutabhujaḥ. “Smoke is the result of fire” (as well as to later formulations where the plural is
changed to singular). Rather, it is clear that Dharmakīrti focuses here on a causal complex.
Perhaps for this reason, Steinkellner suggests that fire in the verse should be taken for the
12
13
Again, this is not to deny Dharmakīrti’s acceptance of samanantrapratyaya and Abhidharma causality.
The discussion above deals with inference on the level of everyday practice.
Steinkellner’s translation opts for the first alternative, as do Mookerjee and Nagasaki (1964: 82); Gillon
(2009: 201) opts for the second: “their effect.” Grammatically eka is subordinate and should not be
referred to by tat, but we know that Dharmakīrti does not respect this rule (e.g., Nyāyabindu 1.1). For
the time being, I would like to leave this issue open.
86
On the Determination of Causation by Dharmakīrti
entire causal complex, namely, fire (i.e., “heat atoms”), wood, air and so on (2013 II: 189,
199). He further argues that each of the causes referred to by yeṣām, tatra, etc., is not a
single thing, but a causal complex. In other words, Dharmakīrti’s statement means that
several causal complexes are perceived and then any one of them may be absent and not be
perceived. I am not sure what we would gain by this hypothesis; I am also not sure that we
don’t have a terminological problem here. Usually according to Dharmakīrti one causal
complex, being complete and unhindered, is alone capable to produce an effect (see also
“the numerical aspect” above). If several complexes cooperate in producing something,
this means, according to Dharmakīrti, that they form one larger and encompassing causal
complex. Be that as it may, Steinkellner’s interpretation goes clearly against the straightforward understanding of the verse and is contradicted by the commentaries. Nevertheless,
it could be accepted if it would have some advantage, for instance, in making Dharmakīrti’s
statements clearer or more persuasive. However, the contrary is the case. If accepted, it
would make the determination of the causal relation practically impossible. How could one
ascertain that even one of the causal complexes is absent? Steinkellner himself says (2013
II: 189) that every causal complex is “in principle quantitatively infinite” (“grundsätzlich
quantitativ unendlich”). Thus, one would be able to ascertain at most that a causal complex
is incomplete; not that it is absent.14 It is of course true that in the final analysis causes for
Dharmakīrti are not individual things, but complexes, but going down to this atomic level
of analysis when considering inferences of everyday practice such as from smoke to fire
is counter-productive, for it would mean that for practical purposes in everyday practice
causation could never be determined.
Whatever the case may be, Steinkellner considers this to be the first step of a proof,
which has an inductive character (2013 II: 201). The second step, which consists in a
prasaṅga (34cd+Vṛtti), is supposed to bestow the necessity and general validity (2013 II:
201: “Notwendigkeit und allgemeiner Gültigkeit verleihen” with references to Dunne 2004:
174f. and Lasic 2003: 186–191) upon the cognition gained from the first step. This may
be true, but one should note the price, and a very high price it is. Basically the prasaṅga
argument makes non-fire into fire. For fire is now not what looks like fire, heats like fire
and burns like fire, but whatever produces smoke. If it is an anthill, then an anthill is fire,15
if it arises from rubbing together two pieces of wood (PVin II 85.9) then the invisible heat
atoms are fire, and if it is water then water is fire. Thus, the understanding and definition of
fire becomes arbitrary, and Dharmakīrti more or less admits it. Furthermore, Dharmakīrti
14
15
Consequently under this interpretation absence (abhāva) of causal complex and its incompleteness
(vaikalya) become, at least in practice, conflated.
See PV I, v. 36. Dharmakīrti probably refers to a popular belief that anthills or termite mounds contain
fire and emit smoke. As termite mounds are humid, one can imagine that vapors, that look like smoke,
rise out of them when they are heated by the sun. Dharmakīrti may also be alluding to the fire ritual,
where an anthill or termite mound symbolizes the head of the sacrifice. On the role of anthills or
termite mounds in the “establishing of fire” (agnyādheya), see Krick 1982: 139ff., esp. 141–142 and
König 1984: 170ff.; the belief that fire, as well as the god of fire (Agni), and the sun reside in termite
mounds appears already in Vedic literature, cf. König 1984: 171. The Majjhimanikāya 23 (Valmīkisutta)
mentions an anthill that “smokes by night and blazes up by day” (Horner 2004 I: 183, repeated with
explanation p. 185; I owe this reference to Antonio Rigopoulos): ayaṃ vammiko rattiṃ dhūpāyati divā
pajjalati. However, this may refer to an unusual appearance and normally it would be the other way
round.
Eli Franco
87
admits that certain effects may arise from different causes; would he admit that the lotus seed
and the cow dung have the same nature inasmuch as they produce lotuses? His suggestion
that the nature of lotuses produced by seeds is different from the nature of lotuses produced
by cow dung (PVSV 23.23–24) sounds like an axiomatic assumption rather than one based
on observation. And would scorpions and lotuses have the same nature inasmuch as they
are both produced by cow dungs or would one have to distinguish between different types
of cow dung? Put differently, Dharmakīrti’s prasaṅga makes the inference from smoke to
fire certain, but arbitrary and tautological. If we call fire whatever smoke arises from, then
the inference of fire from smoke is only an inference that smoke has a cause. We are not
actually inferring fire from smoke, but only that smoke has a cause, which we call fire.
Steinkellner considers Dharmakīrti’s proof to be a stroke of genius (2013 II: 204:
“… nichts weniger als für genial”) and that Dharmakīrti may have solved or ‘avoided’ the
problem of induction “at least for his own purposes.” It goes without saying that Dharmakīrti
did neither avoid nor solve the problem of induction. It would be naïve to expect him to
solve what is clearly an insoluble problem. Moreover, the basis of Dharmakīrti’s proof is
not particularly original. What he actually does is to revive an old Abhidharma idea which
appears in the AKBh. Furthermore, the way Vasubandhu mentions the determination of
causation as a matter of course indicates that he too is not its original author, but relies on
a well-known Abhidharma definition:
AKBh 461:8–9: tatredam anumānam sati kāraṇe kāraṇāntrasyābhāve kāryasyābhāvo dṛṣṭo bhāve ca punar bhāvas, tadyathāṅkurasya.
There is an inference in relation to these [senses]: When cause(s are) present
and another cause is absent, the effect is observed not to arise, and on the other
hand when [that other cause] is present, [the effect] arises, for instance [the
seed] for the sprout.16
However, this does not mean that Dharmakīrti is simply repeating Vasubandhu. Certainly
the philosophical problems he faced in the seventh century were different from those of
Vasubandhu in the fourth. Rather, we have to appreciate Dharmakīrti’s advance upon
the doctrine of his predecessors (especially Kumārila and Īśvarasena), who, like many
philosophers in the Western tradition up to the 21st century, seem to have regarded the
inductive process as merely or basically cumulative. While sporadic accumulation of facts
is certainly used in everyday practice to form general judgements, Dharmakīrti’s (and
in that respect also Vasubandhu’s) method depicts not only what we actually often do in
everyday practice, but resembles the one used (of course with much more elaboration
and refinement) in scientific determination of causality, for instance, by pharmaceutical
companies to determine the causes of pathologies, the effect(s) of particular substances,
16
See also AKV 1190.22–24 thereon: sati kāraṇe kṣetrodakādike, kāraṇāntarasya bījalakṣaṇasyābhāve
kāryasyāṅkurasyābhāvo dṛṣṭaḥ. bhāve ca tasya bījasya punarbhavo (read punar bhāvo; it does not
make sense to say that the sprout arises again) aṅkurasya dṛṣṭaḥ. “When causes such as the field, water
and so on are present [and] another cause such as the seed is absent, the absence of the effect, namely
the sprout, is observed, and on the other hand when this seed is present the presence of the sprout is
observed.”
88
On the Determination of Causation by Dharmakīrti
and so on.17 Dharmakīrti’s advance over the older Abhidharmic formulation can be seen
not only in his reformulation in PVSV on v. 34 and the repeated insistence that mere
non-observation is an unreliable basis for inference, but also in the supplements of this
procedure by multiple other aspects such as increase and decrease, gradation, and so on, as
indicated above. Although rudimentary and unsystematic in its formulation, the various
aspects of causation that Dharmakīrti uses in practice contain in a nutshell the procedure we
still employ today: experiment, strength of association, consistency, specificity, temporality,
gradation and coherence. Interestingly, we should also note what it does not contain:
plausibility and probability. To repeat, Dharmakīrti neither circumvented the problem
of induction, nor did he solve it. Yet, in his perception that inductive knowledge is not
merely an accumulation of observations, he comes as close to dealing with the problem of
induction as we are today.
What is it all about?
Finally, my fourth and last contention: What is it all about? In the second part of his study,
Steinkellner suggests that Dharmakīrti’s statements (in PVSV 22.2–4) can be understood
as having an entirely different purpose (Zielrichtung, 2013 II: 205). They are not at all
aiming at explaining how to determine a causal connection, but what should be understood
under the words “cause” and “effect.” In other words, what Dharmakīrti is talking about
are the conditions for the usage of the words. Steinkellner calls this “the linguistic turn”
(2013 II: 210 “linguistische Wende”) and justifies this move by the statement the objects
one investigates are not real entities, but only conceptual constructions of something as
“cause” and as “effect.” Under certain conditions one can conceptualize something as a
“cause” or as an “effect.”
To be sure, the term “linguistic turn” has more meanings than one. The article about
Relativism in the SEP (Stanford Encyclopedia of Philosophy) considers the linguistic
turn to be characterized by “questions about properties and concepts being replaced by
questions about words and linguistic usage.” For example, some philosophers spoke of the
role of language or, more generally, “systems of symbols in structuring our experience,
thought, or even reality itself.” In a lengthy introduction to a volume entitled The Linguistic
Turn (1967), Rorty characterized the linguistic turn as the opinion that “a ‘philosophical
problem’ was a product of the unconscious adoption of a set of assumptions built into the
vocabulary in which the problem was stated – assumptions which were to be questioned
before the problem itself was taken seriously.” (SEP s.v. Richard Rorty, see Ramberg 2009).
In this sense, Dharmakīrti should not be considered to introduce a linguistic turn into the
17
The example of pathologies was suggested to me by John Taber in a personal conversation and I
would like to take this opportunity to express my gratitude to him for many stimulating conversations.
However, Taber has his own opinion on the problem of induction (or indeed the lack thereof) in Indian
philosophy and is not to be held responsible for anything suggested here. For the determination of
causation in medical sciences see for instance the often quoted Hill 1965. Among the criteria specified,
Hill mentions the strength of the association, its consistency, specificity, temporality, biological gradient
(which corresponds to “gradation” in Dharmakīrti’s terminology), plausibility (as far as I can see, not
used, at least not explicitly, by Dharmakīrti or anyone else in the Indian tradition), coherence, experiment
and analogy.
Eli Franco
89
problem of causation. At least I cannot see him distinguishing concepts from words and
linguistic usage or that the assumption built into a vocabulary would have to be articulated
and questioned before the problem could be solved. I also fail to see that Dharmakīrti
would fit into the linguistic turn of the Vienna Circle, which considers representation as
the proper subject matter of philosophy (Vienna Circle, SEP).
The key to our question lies in Dharmakīrti’s purpose. What does he aim to achieve
with his new theory of reason in PVSV? As far as I can see, he neither aims at “saving
the phenomena,” nor was he concerned with establishing any scientific theory (in sense of
natural sciences), nor was he a pure logician concerned with the logical problems per se, nor
did he aim at establishing the validity of everyday practice, which involves inferences from
smokes to fires. I think that what really troubled him were inferences on doctrinal matters.
Such inferences appear in the beginning of the PVSV, and in this sense we cannot say that
he keeps his cards close to the chest. Already in v. 11 he addresses the Mīmāṃsā inference:
The Buddha had desires because he had a body, like a common man. And immediately
after that (v. 12) we have an extensive discussion of a somewhat similar inference: The
Buddha was not free from desires (vītarāga) because he spoke, like a common man. Another
inference, perhaps put forward by a Naiyāyika, appears in v. 18: The living body is not
without a soul because otherwise it would not have breath and so forth. Such inferences
were unacceptable to the Buddhists, and yet they were valid in the sense of complying
with the trairūpya theory. I suggest that it is probably in response to such inferences that
Dharmakīrti developed his theory of three kinds of reasons, and the primary aim of his
theory was to show why these and similar inferences were not valid.
Dharmakīrti’s dealing of the vyavahāra inference from smoke to fire may be seen
therefore just a by-product for his doctrinal concerns. For in classical India, no theory
of inference would get off the ground without dealing with this paradigmatic inference.
However, it is clear that his interest lies elsewhere. As Lasic (2003: 190) has already
noticed, “[h]e [Dharmakīrti] does not seem to have aimed [in PVSV 22.2–4, etc.] at a
detailed discussion of the procedure of establishing the causal dependence in an individual
case.” On the other hand, he was very much interested in the pragmatic situation of what
should count as a valid proof in the inter-religious debate. His interest in causation was
thus not directed at causation in natural phenomena, but in causal connection between
body and consciousness, between desires, compassion and speech, between body and soul.
And in dealing with these topics, he was as much interested in the possibility of denying a
causal relation (notably between body and cognition) as in establishing one.
References and abbreviations
AKBh Abhidharmakośabhāṣya of Vasubandhu, ed. P. Pradhan. Patna 1975.
AKV Abhidharmakośavyākhyā of Yaśomitra, ed. D. Sastri. Patna 2 1977.
Dunne 2004 John Dunne, Foundations of Dharmakīrti’s Philosophy. Boston 2004.
Franco 2012 Eli Franco, Once Again on the Desires of the Buddha. In: Devadattīyam
– Johannes Bronkhorst Felicitation Volume, ed. François Voegeli et al. Bern 2012,
229–245.
90
On the Determination of Causation by Dharmakīrti
Gillon 2009 Brendan Gillon, The Role of Knowledge of Causation in Dharmakīrti’s Theory of Inference. In: Buddhist Philosophy. Essential Readings, ed. Williams Edglass
et al. Oxford/New York 2009, 197–204.
Hill 1965 Austin Bradford Hill, The Environment and Disease: Association or Causation?
Proceedings of the Royal Society of Medicine 58 (1965) 295–300. http://www.ncbi
.nlm.nih.gov/pmc/articles/PMC1898525/, last visited 15-05-2016.
Horner 2004 I.B. Horner, Middle Length Sayings. Delhi 2004.
König 1984 Ditte König, Das Tor zur Unterwelt. Mythologie und Kult des Termitenhügels
in der schriftlichen und mündlichen Tradition Indiens. Wiesbaden 1984.
Krick 1982 Hertha Krick, Das Ritual der Feuerbegründung. Wien 1982.
Lasic 2003 Horst Lasic, On the Utilisation of Causality as a Basis for Inference. Dharmakīrti’s Statements and Their Interpretations. Journal of Indian Philosophy 31 (2003)
185–197.
Mookerjee and Nagasaki 1964 Satkari Mookerjee and Hujun Nagasaki, The Pramāṇavārttikam of Dharmakīrti. Patna 1964.
PV Pramāṇavārttika of Dharmakīrti, ed. Y. Miyasaka. Acta Indologica 2 (1971/72) 1–206.
PVin Dharmakīrti’s Pramāṇaviniścaya, Chapters 1 and 2, ed. E. Steinkellner. Beijing/
Vienna 2007.
PVSV The Pramāṇavārttikam of Dharmakīrti, ed. R. Gnoli. Roma 1960.
PVSVṬ Pramāṇavārttikasvavṛttiṭīkā of Karṇakagomin, ed. Rahula Sāṅkṛityāyana. Ilāhābād
1943.
Ramberg 2009 Bjørn Ramberg, Richard Rorty. In: The Stanford Encyclopedia of Philosophy, ed. Edward N. Zalta. Spring 2009 Edition. http://plato.stanford.edu/archives/
spr2009/entries/rorty/, last visited 15-05-2016.
Steinkellner 2013 Ernst Steinkellner, Dharmakīrtis frühe Logik: Annotierte Übersetzung
der logischen Teile von Pramāṇavārttika 1 mit der Vṛtti. I: Introduction, Übersetzung,
Analyse. II: Introduction, Anmerkungen, Anhänge etc. Tokyo 2013.
Taber 2003 John Taber, Dharmakīrti Against Physicalism. Journal of Indian Philosophy
31 (2003) 479–502.
Watanabe 2004 Toshikazu Watanabe, On the Problem of the Determination of Causality:
bādhakapramāṇa and bheda. Hikaku Ronrigaku Kenkyū 2 (2004) 57–61.
Reinterpretation of the Compound svabhāva-pratibandha in
Dharmakīrti’s Logical Theory
by
Yoichi Fukuda
Introduction
Many scholars agree that svabhāva-pratibandha forms the actual basis of Dharmakīrti’s
logical theory. However, there are different opinions about what svabhāva-pratibandha
means.1
In his article “Svabhāvapratibandha Again,” Steinkellner wrote that “since the word
pratibandha has only a formal meaning, the word svabhāva is responsible for connecting
the reality needed” (1984: 458). He does not distinguish the meaning of sambandha from
that of pratibandha, understanding both as having only the formal meaning of “connection.”
Before examining the usage of the word svabhāva-pratibandha in Dharmakīrti’s texts,
I would like to consult how pratibandha/pratibaddha is presented in the Mahāvyutpatti
and Abhidharmakośa, two authoritative texts on Buddhist terminology. As the Tibetan
equivalents for pratibandha, the Mahāvyutpatti lists phyir ’jil ba’am bgegs byed pa’am bar
chad byed pa, which mean hindrance or obstruction (Mahāvyutpatti, 6483). The same is
the case in the Abhidharmakośa, where pratibandha is translated into Tibetan as gegs byed
pa / bgegs su gyur pa and into Chinese as 障, 障礙, 遮 (Index to the Abhidharmakośa, p.
248).
On the other hand, in Mahāvyutpatti 6481, rag lus pa’am ’brel ba’am bgegs su gyur pa
are given as the Tibetan equivalents of pratibandha, while in the Abhidharmakośa (Index,
p. 248) in four cases “X-pratibaddha” is translated into Tibetan as “X la rag lus/las pa,”
and in one case as “X dang ’brel ba.” In five cases, it is translated into Chinese as 繋属,
属, 随. These translations in the Abhidharmakośa all mean “dependence upon.” Thus, in
these traditional texts, pratibandha is never used to mean connection, but is rather chiefly
used to mean dependence.
In one-third of the cases where pratibandha is used in Dharmakīrti’s Pramāṇavārttikasvavṛtti (hereinafter referred to as PVSV), the word means “obstruction,” translated into
Tibetan as gegs/gags byed (pa). However, in two-thirds of the cases pratibandha is translated
as ’brel ba or rag las/lus pa. I have collected fifty-one passages that use the latter meaning
for pratibandha/pratibaddha (List C). There is no distinction between pratibandha and
pratibaddha in these Tibetan translations. The only difference between them is the syntactic
1
Most articles on svabhāva-pratibandha have been published in Japanese. A historical survey of these
discussions was written by Kei Kataoka (2012). The idea I present here dates back thirty years. At that
time I wrote two short papers on this theme in Japanese (“An Inquiry into the Structure of Dharmakīrti’s
Logic,” 1987; “On the meaning of svabhāvapratibandha in Dharmakīrti’s Logic,” 1989). In 2012, I
wrote two additional longer papers (“On the Interpretation of the Compound svabhāvapratibandha,”
2012a; “On the Meaning of pratibandha in Pramāṇavārttikasvavṛtti,” 2012b), in which I rearranged
the former presentations and supplemented them with more citations to demonstrate my idea.
Birgit Kellner et al., eds., Reverberations of Dharmakīrti’s Philosophy: Proceedings of the Fifth International Dharmakīrti
Conference Heidelberg, August 26 to 30, 2014. Vienna 2020, pp. 91–109.
92
Reinterpretation of the Compound svabhāva-pratibandha
difference in Sanskrit.2 In twenty-seven of these passages, pratibandha/pratibaddha is explicitly related to expressions of the logical nexus, for example, gamya, gamaka, gamayate,
gamyate, avyabhicāra, avisaṃvāda, vyāpti, anvaya, vyatireka, ekanivṛttyānyanivṛtti, and so
on.3 Another eighteen passages mention pratibandha/pratibaddha without svabhāva. From
these usages I could conclude that pratibandha has a more important role than svabhāva
as the basis of the logical nexus.
Of these fifty-one examples of pratibandha/pratibaddha in the PVSV, I believe that none
conflicts with the meaning of dependence, which is the traditional meaning of pratibaddha
as seen above. Later I would like to return to this point and add another piece of indirect
evidence.
Analysis of the compound svabhāva-pratibandha
To understand the meaning of the compound svabhāva-pratibandha, I will discuss the
assumptions regarding the syntactical value of svabhāva in this compound. Until now,
there have been three interpretations of this compound, namely as instrumental tatpuruṣa,
genitive tatpuruṣa, and locative tatpuruṣa. The first interpretation is the most common
and the one used by Dharmottara in paraphrasing the compound in the Nyāyabinduṭīkā:
svabhāvena pratibandhaḥ svabhāvapratibandhaḥ (NBṬ 110,1). However, I could not find
any textual evidence in Dharmakīrti confirming that pratibandha is used with words in the
instrumental case. Dharmakīrti mentions pratibandha with one word in the genitive case
and another in the locative case; the former indicates a possessor of pratibandha and the
latter indicates an object upon which the possessor of pratibandha is dependent. A similar
situation can be seen in the example of pratibaddha, which is used with a word in the
locative case indicating the object of dependence and a notional subject in the appositional
case. In order to explain the usage of related words used with pratibandha, I would like to
consult the Nyāyabindu (NB) instead of the PVSV for reasons of simplicity.
NB 2.19: svabhāvapratibandhe hi saty artho(X) arthaṃ(Y) gamayet /4
NB 2.20: tad(Y)-apratibaddhasya(X) tad(Y)-avyabhicāraniyamābhāvāt /5
NB 2.21: sa ca pratibandhaḥ sādhye arthe(Y) liṅgasya(X) /
NB 2.22: [liṅgasya(X)] vastutas tād(Y)-ātmyāt tad(Y)-utpatteś ca /
NB 2.23: a-tad(Y)-svabhāvasya(X) a-tad(Y)-utpatteś(X) ca tatra(Y) apratibaddha-svabhāvatvāt /
In these statements, X (= hetu or liṅga) is the object that infers another thing, and Y (=
sādhya) is the object to be inferred.
According to NB 2.21, X (liṅga) in the genitive case is a notional subject of pratibandha,
which is the repetition of svabhāva-pratibandha in NB 2.19. In other words, X (liṅga)
2
3
4
5
Hereinafter I will mention these terms as pratibandha/pratibaddha without distinction.
In the actual texts of Dharmakīrti, these words are often part of negative expressions.
Cf. PVSV 2,19–20: svabhāvapratibandhe hi saty artho arthaṃ na vyabhicarati /.
Cf. PVSV 8,12–13: apratibaddhasvabhāvasya avinābhāvaniyamābhāvāt /; PVSV 12,24–25: apratibaddhasya tadabhāve sarvatra abhāvāsiddheḥ saṃśayād avyatireko vyabhicāraḥ śeṣavataḥ /.
Yoichi Fukuda
93
is a possessor of pratibandha. sādhye arthe (= Y) of the locative case in NB 2.21 and
its substitute pronoun tatra in NB 2.23 represent the object, which is connected to or
dependent upon X (liṅga). In other words, Y (sādhya) is an object upon which X (liṅga)
is dependent. According to NB 2.23, X (liṅga), which is implicitly assumed to be in
the same case (genitive) as tatsvabhāvasya or tadutpatteś, is a notional subject of the
bahuvrīhi compound pratibaddha-svabhāva. The same usage of pratibaddha-svabhāva
is found in PVSV: apratibaddha-svabhāvasyāvinābhāvaniyamābhāvāt (8,12f), or na ca
tadapratibaddhasvabhāvo bhāvo anyaṃ gamayati (107,25). In both cases, X (liṅga or hetu)
is supplemented in the appositional case.
One could naturally interpret this compound as meaning that something (X = liṅga)
has its svabhāva, which is dependent upon another thing (Y = sādhya); in other words,
this would mean that the svabhāva of X (liṅga) is dependent upon Y (sādhya). From
these usages of pratibaddha-svabhāva, one could conclude that the first component of the
compound svabhāva-pratibandha is the svabhāva of X (liṅga) and that the compound is a
genitive tatpuruṣa.
However, and inconveniently, one exceptional passage exists that possibly suggests
that svabhāva expresses the dependent object.6 In this case svabhāva-pratibandha must
be interpreted as a locative tatpuruṣa. These examples show that Dharmakīrti was not
particularly rigorous in his interpretation of the compound, but that he generally considered
svabhāva to be a liṅga and the compound thus a genitive tatpuruṣa.
The meaning of pratibandha
Next, I would like to return to the investigation into the meaning of the dependence of X
(liṅga) upon Y (sādhya). Because Dharmakīrti does not explain the meaning of pratibandha
explicitly, we must try to understand it from its context. The word is very closely related to
ekanivṛttyānyanivṛtti (PVSV 10,23; PV I.24), which is an improved version of avinābhāva,
itself a traditional expression for a logical nexus. While avinābhāva means the inevitable
absence of one thing (X = hetu) in the sphere of the absence of another thing (Y = sādhya),
ekanivṛttiyānyanivṛtti means that the disappearance of one thing (X) is caused by the
disappearance of another thing (Y).7 This causality of disappearance is expressed by the
instrumental case or the causative verb, while the inclusive relation of the absence of
two things is expressed by the locative case. Dharmakīrti introduces this causality of
disappearance as a condition of an inevitable logical nexus, and insists that the causality of
disappearance necessarily requires the pratibandha relation, meaning the dependence of
the existence of one thing (X = hetu) upon another thing (Y = sādhya). If X is dependent
in this way upon Y, the disappearance of Y will inevitably cause the disappearance of X.
6
7
PVSV 17,1–3: svaṃ ca svabhāvaṃ parityajya kathaṃ bhāvo bhavet / svabhāvasyaiva bhāvatvād iti tasya
(= liṅgasya) svabhāvapratibandhād avyabhicāraḥ /. In this case, svabhāva is consistently the object to
be conferred and bhāva is the object conferring it. Śākyabuddhi commented on the second sentence:
tasyātmabhūtasya sādhaasya śiṃśapādeḥ svabhāvapratibandhād eva svabhāve sādhyābhimate vṛkṣādau
… pratibandhād evāvyabhicāraḥ / (KG 75,3ff = PVṬ 39b7f).
The Tibetan equivalent of avinābhāva is med na mi ’byung ba, while on the other hand the equivalent
of ekanivṛttyānyanivṛtti is gcig log pas gzhan ldog pa. As will be shown below, in Tibetan logic the
expression corresponding to the latter is de ldog pa’i stobs kyis khyod ldog pa.
94
Reinterpretation of the Compound svabhāva-pratibandha
Such dependence is, in itself, a connection between real objects and therefore to express
this meaning Dharmakīrti does not need to mention svabhāva.
tasmād ekanivṛttyā ’nyanivṛttim icchatā tayoḥ kaścit svabhāvapratibandho
’py eṣṭavyaḥ / anyathā ’gamako hetuḥ syāt / (PVSV 10,23–25)
This passage shows that the logical reason (= hetu) is based on the relation of ekanivṛttyānyanivṛtti and, in turn, this relation is based on svabhāva-pratibandha. Conversely, if
svabhāva-pratibandha exists between two objects, the disappearance of one object causes
the disappearance of the other and thus one can infer the existence of the former object
from the existence of the latter. One can find this used in a similar way several times in the
PVSV, as detailed below.
PVSV 16,28f: tanmātrasambandhaḥ svabhāvo (= sādhya) bhāvam (= liṅga)
eva nivartayet (k.23a–c) /.
PVSV 17,5: kāraṇaṃ nivartamānaṃ kāryaṃ nivartayati /.
PVSV 17,7f: ubhayathā (= tādātmyena tadutpattyā vā)8 svabhāvapratibandhād eva nivṛttiḥ / anyathā ekanivṛttyānyavinivṛttiḥ kathaṃ bhavet (k.24ab) /
.
PVSV 19,25: tāv (= kāraṇa and vyāpaka) eva hi nivartamānau svapratibaddhaṃ (= kārya and vyāpya) nivartayata iti /.
PVSV 20,1f: apratibandhe hi katham ekasya nivṛttir anyasya nivṛttiṃ sādhayet /.
This nivartaka power does not represent a kind of logical nexus, but rather the power to
affect another object existentially. Based on these expressions, I would like to suggest that
pratibandha, or existential dependence,9 is confirmed by means of the realization of the
causality between the disappearances of two objects. If svabhāva-pratibandha does not exist,
then it is not possible to confirm the causality between the two disappearances. Conversely,
if one can find causality between the two disappearances, then svabhāva-pratibandha must
exist.
This is the very idea evoked by Dar ma rin chen (1364–1432) when he said
byas pa mi rtag pa la ’brel ba’i don ldog gi phyogs gcig de gnyis tha dad
mngon sum gyis grub cing / mi rtag pa log pa’i stobs kyis byas pa ldog pa’i
tshul ni mi rtag pa log pa’i gzhi la byas pa ’gog pa’i gnod pa can gyi rtags la
brten nas bsgrub pa yin …10
8
9
10
According to Śākyabuddhi, KG 75,23.
The translation “existential dependence” coincides with the translation by F. Th. Stcherbatsky (1930).
He translated the first occurrence of svabhāva-pratibandha in NB II.19 (p. 69) as “Because one thing can
convey the (existence of) another one when it is existentially dependent (on the latter)” (my emphasis,
Y. F.).
rGyal tshab rje Dar ma rin chen, tshad ma rnam ’grel gyi tshig le’ur byas pa’i rnam bshad thar lam
phyin ci ma log par gsal bar byed pa, Zhol ed., cha, 22b4–5. That the defining character of ’brel
pa is closely related to the causality of two disappearances was first pointed out in the presentation
“rGyal tshab’s Understanding of Svabhāvapratibandha” by Choi Kyeon-jin at the Conference of the
International Association of Tibetan Studies, 2013, Ulaanbaatar.
Yoichi Fukuda
95
don ldog is an equivalent of mtshan nyid, that is, the “defining character.” Therefore in
this passage Dar ma rin chen is postulating that there are two conditions that define the
character of the dependence of the produced object upon an impermanent object: (1) the
two objects must be different, and (2) the disappearance of the former object is caused by
force of the disappearance of the latter object.
It is obvious that the second condition is based on ekanivṛttyānyanivṛtti. don ldog
literally means the distinguishing character in the object, and therefore this ekanivṛttyānyanivṛtti is a distinguishing character of pratibandha. This is indirect evidence of the close
relationship between pratibandha and ekanivṛttyānyanivṛtti.
Problems in the Tibetan translation of the PVSV and the commentaries
on Dharmakīrti
As shown above, there is little that remains speculative about the meaning of pratibandha
in the texts of Dharmakīrti. Now I would like to point out some interesting matters that can
be drawn from the Tibetan translation of the PVSV and its commentary.
I have already mentioned that in the Abhidharmakośa and Mahāvyutpatti, pratibaddha
is in most cases translated as rag las pa. In passages no. 1 to no. 25 of fifty-one citations
in List C, below with the exception of no. 5, pratibandha/pratibaddha is translated as
’bral ba/pa and in the rest of the passages, with the exception of nos. 27 and 34, it is
translated as rag las/lus pa. The same situation is found in the Tibetan translation of the
commentary on the PVSV by Śākyabuddhi (PVṬ). In half of the cases translated into
rag las/lus pa, pratibandha/pratibaddha is used in the context of the basis for the logical
nexus. For example, svabhāva-pratibandha in passages nos. 44 and 46 is translated as rang
bzhin la/gyi rag lus pa, not as rang bzhin (dang) ’brel ba, which is the translation in the
first half of the cases of svabhāva-pratibandha/pratibaddha-svabhāva. In passage no. 40,
tadapratibaddhasvabhāva is properly translated as de la rag lus pa med pa’i rang bzhin
can gyi, with the meaning of the bahuvrīhi compound. As mentioned above, there is no
distinction made between svabhāva-pratibandha and pratibaddha-svabhāva in the Tibetan
translation. This transition of the Tibetan translation of the PVSV and PVṬ might mean that
the translator(s)11 of these two texts realized Dharmakīrti’s intentions behind this term in the
process of their translation work, and changed their translation of pratibandha/pratibaddha
halfway through.
But other Tibetan scholars who did not consult the Sanskrit texts must not have noticed
that rag las/lus pa was the translation of pratibandha. Amazingly, they properly recorded
the defining character of ’brel pa as I have mentioned above, even in such a restricted
or incomplete situation. I cannot but concede that they had a very profound ability to
understand Dharmakīrti’s intentions.
This change in the Tibetan translation of pratibandha/pratibaddha may have been
influenced by an annotation of Śākyabuddhi, which paraphrases pratibaddha as āyatta and
11
The colophon of the PVSV does not mention a translator, but that of the PVṬ mentions lo ts’a ba rma
dge ba’i blo gros. Perhaps he also translated the PVSV during the same period.
96
Reinterpretation of the Compound svabhāva-pratibandha
pratibandha as āyattatva on several occasions (passages nos. 1,12 5, 24, 32, 35, 38, 39, 41,
42, 46).
In the midst of paraphrasing pratibandha in NB II.21: sa ca pratibandhaḥ sādhye
’rthe liṅgasya, Dharmottara comments that liṅgam parāyattatvāt pratibaddham / sādhyas
tv artho ’parāyattatvāt pratibandhaviṣayo na tu pratibaddha ity arthaḥ / (NBṬ 112,1f);
“liṅga is pratibaddha because it is dependent upon another [object], while on the other hand
sādhya artha – because it is not dependent upon the other [namely the liṅga] – is an object
of pratibandha, [but] it is not pratibaddha.” Dharmottara clearly interprets pratibaddha as
āyatta and pratibandha as āyattatva. He also puts liṅga and pratibaddha in the appositional
case and explains that sādhya is pratibandhaviṣaya. Moreover, Durvekamiśra annotates
pratibandha three times with pratibandhaḥ pratibaddhatvam āyattatvam (NBṬ 96,21;
110,20; 115,15f). These explanations correspond with what I have explicated above in this
paper.
Conclusion
To summarize my arguments in this paper:
1. pratibandha is more significant than svabhāva in the compound svabhāva-pratibandha.
2. Generally svabhāva-pratibandha is interpreted as pratibaddha-svabhāvatva, which
is an abstract noun of a bahuvrīhi compound. Therefore, svabhāva stands for the liṅga’s
svabhāva and svabhāva-pratibandha should be understood as a genitive tatpuruṣa compound.
3. pratibandha/pratibaddha is closely related to the causality of the dual disappearance
of two objects (ekanivṛttyānyanivṛtti).
4. Therefore, pratibandha/pratibaddha express the existential dependence of one object
upon another, of which the former is liṅga or hetu and the latter is sādhya.
References and abbreviations
AKBh Vasubandhu, Abhidharmakośabhāṣya: Abhidharmakośabhāṣyam of Vasubandhu,
ed. P. Pradhan. Patna 1967.
AKBhT Vasubandhu, Abhidharmakośabhāṣya: chos mngon pa’i mdzod kyi bshad pa.
Derge ed. Tohoku No. 4090. mNgon pa’i bstan bcos, gu 27b6–ngu 109a8.
Choi 2013 Choi Kyeon-jin, rGyal tshab’s Understanding of Svabhāvapratibandha. Paper
presented at the Conference of the International Association of Tibetan Studies,
2013, Ulaanbaatar.
Dar ma rin chen rGyal tshab rje Dar ma rin chen, tshad ma rnam ’grel gyi tshig le’ur
byas pa’i rnam bshad thar lam phyin ci ma log par gsal bar byed pa. Zhol ed.
12
At this first mention of svabhāva-pratibandha, Śākyabuddhi annotates that rang bzhin bsgrub par bya
ba’i dngos po dang ’brel pa ni de la rag las pa nyid de de yod na’o. But Karṇakagomin comments
differently, that svabhāvena pratibandhaḥ sādhanaṃ kṛtaiti samāsa / svabhāvena pratibaddhatvaṃ /
pratibaddhasvabhāvatvam iti yāvat / (KG 13,18f), which is the almost same paraphrase as that in
Dharmottara’s Nyāyabinduṭīkā (NBṬ 110,1f).
Yoichi Fukuda
97
Fukuda 1987 Y. Fukuda, An Inquiry into the Structure of Dharmakīrti’s Logic. Journal
of Indian and Buddhist Studies 33-1 (1987) 347–345.
Fukuda 1989 Y. Fukuda, On the meaning of svabhāvapratibandha in Dharmakīrti’s Logic.
Journal of Indian and Buddhist Studies 35-2 (1989) 888–885.
Fukuda 2012a Y. Fukuda, On the Interpretation of the Compound svabhāvapratibandha.
Indian Logic 4 (2012) 205–222.
Fukuda 2012b Y. Fukuda, On the Meaning of pratibandha in Pramāṇavārttikasvavṛtti.
Indian Logic 5 (2012) 73–107.
Index to the Abhidharmakośabhāṣya Index to the Abhidharmakośabhāṣya. Pt. 1: SanskritTibetan-Chinese, ed. Akira Hirakawa et al. Tokyo 1973.
Kataoka 2012 K. Kataoka, A Sketch of Svabhāvapratibandha Studies. Indian Logic 5
(2012) 163–204.
KG Karṇakagomin, Pramāṇavārttikasvavṛttiṭīkā: Ācārya-Dharmakīrteḥ Pramāṇavārttikaṃ (svārthānumānaparicchedaḥ) svopajñavṛttyā Karṇakagomiviracitayā taṭṭīkayā
ca sahitam, ed. Rāhula Sāṃkṛtyāyana. Allahabad 1943.
Mahāvyutpatti A New Critical Edition of the Mahāvyutpatti: Sanskrit-Tibetan-Mongolian
Dictionary of Buddhist Terminology, ed. Yoichi Fukuda and Yumiko Ishihama. Tokyo
1989.
NB Dharmakīrti, Nyāyabindu. See NBṬ.
NBṬ Dharmottara, Nyāyabinduṭīkā: Paṇḍita Durveka Miśra’s Dharmottarapradīpa, ed.
D. Malvania. Patna 2 1971.
PV Dharmakīrti, Pramāṇavārttika. See PVSV.
PVSV Dharmakīrti, Pramāṇavārttikasvavṛtti: The Pramāṇavārttikam of Dharmakīrti, the
First Chapter with the Autocommentary, ed. R. Gnoli. Roma 1960.
PVṬ Śākyabuddhi, Pramāṇavārttikaṭīkā: tshad ma rnam ’grel gyi ’grel bshad, tr. rma dge
ba’i bo gros. Derge ed. Tohoku No. 4220. Tshad ma je 1–328a7, nye 1–282a7.
Stcherbatsky 1930 F. Th. Stcherbatsky, Buddhist Logic: In Two Volumes. Vol. II Containing a Translation of the Short Treatise of Logic by Dharmakīrti, and of Its
Commentary by Dharmottara, with Notes Appendices and Indices. Leningrad 1930.
Steinkellner 1984 E. Steinkellner, Svabhāvapratibandha Again. Acta Indologica 6 (1984)
457–476.
SVTib Dharmakīrti, Pramāṇavārttikavṛtti: tshad ma rnam ’grel gyi ’grel pa. Derge ed.
Tohoku No. 4216. Tshad ma ce 261b1–365a7.
98
Reinterpretation of the Compound svabhāva-pratibandha
List A: Abhidharmakośabhāṣya
1.
AKBh
AKBhT
2.
AKBh
AKBhT
3.
AKBh
AKBhT
4.
AKBh
AKBhT
5.
AKBh
AKBhT
icchāmātrapratibaddho hi teṣāṃ sarvaguṇasampatsaṃmukhībhāvaḥ / (Ch.2,
70.23)
de dag gi yon tan ’byor ba mngon sum du ’gyur ba thams cad ni bzhed pa tsam
la rag lus pa yin pas … (Ku, 76a7)
一切圓德,隨樂而起故。
yasya yatpratibaddha utpādaḥ sa tasyānantaram utpadyate / (Ch.2, 99.13)
gang zhig gang la rag las te skye ba de ni de’i ’jug thogs su skye ste / (Ku 100a5)
若此法生,繫屬彼法,要彼無間,此乃得生。如芽等生,要藉種等。
tatra pañcavidhamanaskārānantaram āryamārgasammukhībhāvo ’nyatropapattipratilambhikebhyaḥ / prayogapratibaddhatvāt / (Ch.2, 109.1f)
de la skyes nas thob pa dag ma gtogs pa yid la byed pa rnam pa lnga’i mjug
thogs su ’phags pa’i lam mngon du ’gyur te / sbyor ba la rag lus pa’i phyir
ro // (Ku 107b5f)
此中五種,作意無間,聖道現前。除生所得聖道。繫屬加行心故。
yadā cāsyāśrayo vipariṇantum ārabhate tadāvaśyam asya tadāśrayapratibaddhaṃ cittaṃ saṃmukhībhūya paścāt pracyavet nānyathā / (Ch.3, 156.6f)
gang gi tshe de’i lus yongs su ’jug par rtsom pa de’i tshe yang de las gdon mi za
bar de’i lus dang ’brel pa’i sems mngon du gyur nas phyis ’chi ’pho bar ’gyur
gyi gzhan du ni ma yin no // (Ku, 143a1f)
若所依身將欲變壞。必定還起屬所依心。然後命終。更無餘理。
icchāmātrapratibaddhaḥ sarvaguṇasaṃpatsaṃmukhībhāvaḥ / (Ch.7, 421.3)
de dag gi yon tan ’byor pa thams cad mngon du ’gyur ba ni bzhed pa tsam la
rag lus so // (Khu, 122b5)
諸佛功德・・・隨欲能引現前。
List B: Mahāvyutpatti
6483
6481
pratibandhaḥ / phyir ’jil ba’am bgegs byed pa’am bar chad byed pa /
pratibaddhaḥ / rag lus pa’am ’brel ba’am bgegs su gyur pa /
Yoichi Fukuda
99
List C: PVSV, commentaries of Śākyabuddhi and Karṇakagomin
1.
PVSV
SVTib
KG
PVṬ
2.
3.
PVSV
SVTib
KG
PVSV
SVTib
KG
PVṬ
4.
PVSV
SVTib
KG
PVṬ
5.
PVSV
SVTib
KG
PVṬ
6.
PVSV
SVTib
KG
svabhāvapratibandhe hi saty artho ’rthaṃ na vyabhicarati / (2,19)
rang bzhin dang ’brel pa yod pa don gyis don ’khrul pa med de / (262a6f)
svabhāvena pratibandhaḥ sādhanaṃ kṛteti samāsaḥ / svabhāvena pratibaddhatvaṃ / pratibaddha-svabhāvatvam iti yāvat / (23,16f)
(≠) rang bzhin bsgrub par bya ba’i dngos po dang ’brel pa ni de la rag las pa
nyid de de yod na’o // (10a7)
kāryasyāpi svabhāvapratibandhaḥ / (3,3f)
’bras bu yang rang bzhin dang ’brel pa yin te / (262b2)
– (24,25); PVṬ: – (11a2)
tatrānubhayasyāpratibandhāt tadabhāve anyena na bhavitavyam iti kuta etat /
(5,19f)
de la gnyis ka ma yin pa ni ’brel pa med pa’i phyir te / de med na gzhan med
par ’gyur ro // zhes bya ba der lta ga la ’gyur // (263b7)
anubhayasyākāryakāraṇātmakasya niṣedhyena saha pratibandhābhāvāt tadabhāve ’pratibaddhasyābhāve / (38,17f)
gnyi ga ma yin pa ste rgyu yang ma yin la ’bras bu yang ma yin pa’i bdag nyid
ni dgag par bya ba dang lhan cig ’brel pa med pa’i phyir de med na ste ’brel
pa med pa med na … (16b2)
apratibaddhasvabhāvasyāvinābhāvaniyamābhāvāt / (8,12f)
rang bzhin ma ’brel pa ni med na mi ’byung bar nges pa med pa’i phyir ro //
(265a6)
tādātmyatadutpattibhyāṃ liṅginy apratibaddhasvabhāvasyāvinābhāvaniyamābhāvāt / (49,12)
de’i bdag nyid dang de las byung ba dag gi rtags can dang / rang bzhin ma
’brel pa ni med na ma ’byung bar nges pa med pa’i phyir ro // (22b2f)
nāntarīyakam eva kāryaṃ kāraṇam anumāpayati / tatpratibandhāt / (10,5f)
med na mi ’byung ba’i ’bras bu kho nas rgyu rjes su dpog par byed do // de la
rag las pa yin pa’i phyir ro // (266a6)
tatpratibandhāt / tatra kāraṇe āyattatvāt / (55,4f)
med na mi ’byung ba’i ’bras bu kho nas zhes bya ba ni rgyu med na mi ’byung
ba kho nas so // de la rag las pa yin pa’i phyir zhes bya ba ni rgyu de la rag
las pa yin pa’i phyir ro // (26b2f)
ekanivṛttyā ’nyanivṛttim icchatā tayoḥ kaścit svabhāvapratibandho ’py eṣṭavyaḥ / anyathā ’gamako hetuḥ syāt / (10,23f)
gcig log pas gzhan ldog par ’dod pas ni de gnyis rang bzhin ’brel pa ’ga’ zhig
kyang ’dod par bya dgos so // de lta ma yin na gtan tshigs go bar byed pa ma
yin par ’gyur ro // (266b6)
tayoḥ sādhyasādhanayoḥ kaścit svabhāvena pratibandhas tādātmyatadutpattilakṣaṇo [’]py eṣṭavyaḥ / (57,7f)
100
Reinterpretation of the Compound svabhāva-pratibandha
PVṬ
7.
8.
9.
PVSV
SVTib
KG
PVSV
SVTib
KG
PVSV
SVTib
KG
PVṬ
10.
PVSV
SVTib
KG
PVṬ
11.
PVSV
SVTib
KG
PVṬ
12.
PVSV
SVTib
KG
(;) bsgrub par bya ba dang sgrub pa de gnyis de’i bdag nyid dang / de las
byung ba’i mtshan nyid kyis ’brel pa ’ga’ zhig kyang ’dod par bya dgos te /
(28a2f)
na hy asati pratibandhe ’nvayavyatirekaniścayo ’sti / (10,28)
’brel pa med par ni rjes su ’gro ba dang ldog pa nges pa med de / (266b7)
– (59,27); PVṬ: – (28a6)
anyathā hy asati pratibandhe ’darśanamātreṇa vyatireke / (11,5)
de lta ma yin te / ’brel pa med par ma mthong ba tsam gyis ldog na / (267a2)
– (58,23); PVṬ: – (28b6)
anupalambhāt tu kvacid abhāvasiddhāv apy apratibaddhasya tadabhāve sarvatrābhāvāsiddheḥ / (12,23f)
mi dmigs pa’i sgo nas ni ’ga’ zhig la med pa grub tu zin kyang ’brel pa med pa
ni de med kyang thams cad la med pa mi grub pas / (267b4f)
apratibaddhasya hetoḥ sādhye / tadabhāve sādhyābhāve sarvatra vipakṣe ’bhāvāsiddheḥ… (62,9f)
gtan tshigs bsgrub par bya ba dang ’brel pa med pa ni sgrub par byed pa de
med kyang mi mthun pa’i phyogs su gyur ba’i dngos po thams cad la med par
mi ’grub pa’i gtan tshigs kyis na the tshom za ba yin no // (31a5)
tannivṛttau kvacin nivṛttāv api prāṇādīnām apratibandhāt / sarvatra nivṛttyasiddher agamakatvam / (13,10f)
’ga’ zhig la de log ste / srog la sogs pa log kyang ’brel pa med pa’i phyir / thams
cad las ldog pa mi ’grub pas rtogs par byed pa ma yin pa nyid do // (268a2)
nivṛttāv api prāṇādīnām apratibandhād ātmanā saha sambandhābhāvāt /
(63,23f)
srog la sogs pa log kyang ’brel pa med pa’i phyir / bdag dang lhan cig ’brel
pa med pa’i phyir … (32a4)
yadi kathaṃcid vipakṣe ’darśanamātreṇa apratibaddhasyāpi tadavyabhicāraḥ /
(15,11)
gal te ji zhig ltar mi mthun pa’i phyogs la ma mthong ba tsam gyis ’brel pa med
pa yang de ’khrul pa med do zhe na / (269a1)
yadi vipakṣe hetor adarśanamātreṇāpratibaddhasya svasādhye tadavyabhicāraḥ sādhyāvyabhicāra iṣyate / (69,2f)
gal te mi mthun pa’i phyogs la gtan tshigs ma mthong ba tsam gyis bsgrub par
bya ba dang ’brel pa med pa’i yang bsgrub par bya ba der ’khrul pa med par
’dod na ni … (35a6f)
svaṃ ca svabhāvaṃ parityajya kathaṃ bhāvo bhavet / svabhāvasyaiva bhāvatvād iti tasya svabhāva-pratibandhād avyabhicāraḥ / (17,1f)
rang gi ngo bo yongs su bor nas kyang ji ltar yod par ’gyur te / rang gi ngo bo
kho na ngo bo yin pa’i phyir ro // de’i phyir de ni rang bzhin ’brel pa’i phyir
’khrul pa med do // (270a3)
tasyātmabhūtasya sādhanasya śiṃśapādeḥ svabhāvapratibandhād eva svabhāve sādhyābhimate vṛkṣādau yathoktena prakāreṇa pratibandhād evāvyabhicāraḥ / (75,3f)
Yoichi Fukuda
PVṬ
13.
PVSV
SVTib
KG
PVṬ
14.
PVSV
SVTib
KG
PVṬ
15.
PVSV
SVTib
16.
17.
KG
PVSV
SVTib
KG
PVSV
SVTib
KG
PVṬ
18.
19.
PVSV
SVTib
KG
PVSV
101
sgrub pa shing sha pa la sogs pa bdag nyid du gyur pa de ni rang bzhin ’brel
pa kho na’i phyir te / rang bzhin bsgrub par bya bar ’dod pa shing la sogs pa
dang ji skad bshad pa’i rnam pas ’brel pa kho nas ’khrul pa med do // (39b7f)
ubhayathā svabhāvapratibandhād eva nivṛttiḥ / (17,7)
gnyis ga rang bzhin ’brel pa kho nas ldog go // (270a4f)
ubhayatheti tādātmyena tadutpattyā vā yaḥ svabhāvapratibandhas tasmād eva /
sādhyanivṛttyā hetor nivṛttiḥ / anyatheti yadi pratibandho neṣyate / ekasyāpratibandhakasya sādhyasya nivṛttyānyanivṛttiḥ / apratibaddhasya sādhanadharmasya nivṛtiḥ katham bhavet / (75,11f)
(;) gnyi gar zhes bya ba ni de’i bdag nyid dang / de las byung ba’i mtshan nyid
kyis rang bzhin ’brel pa kho nas bsgrub par bya ba log pas gtan tshigs ldog
go // de lta min na gcig log pas zhes bya ba ni bsgrub par bya ba ma ’brel pa
log pas / ji ltar gzhan mi ldog par ’gyur sgrub pa’i chos ma ’brel pa ldog par
’gyur mi ’gyur ba kho na ste / (40a3f)
svabhāvapratibandhād eva hetuḥ sādhyaṃ gamayati / sa ca tadbhāvalakṣaṇas
tadutpattilakṣaṇo vā / (17,12f)
rang bzhin ’brel pa kho nas gtan tshigs kyis bsgrub par bya ba go bar byed
de // de’i ngo bo’i mtshan nyid dam // de las byung ba’i mtshan nyid yin no //
(270a6)
svabhāvapratibandhād eva sādhyābhimate vastuni pratibaddhatvād eva hetuḥ
svasādhyaṅ gamayati / (76,1f)
rang bzhin ’brel pa kho nas bsgrub par bya bar ’dod pa’i dngos po dang ’brel
pa kho nas gtan tshigs kyis bsgrub par bya ba go bar byed do // (40a6)
yadā punar dṛṣṭāntena nāgnidhūmayoḥ kāryakāraṇabhāvaḥ pradarśyate / tadā
yatra dhūmas tratrāgnir ity eva na syāt / pratibandhābhāvāt / (19,16f)
gang gi tshe dpes me dang du ba dag rgyu dang ’bras bu’i dngos por mi ston
pa de’i tshe ni ’brel pa med pa’i phyir / gang na du ba yod pa de na me yod do
zhes bya ba nyid du yang mi ’gyur na / (271a7)
– (85,5f); PVṬ: – (44a2f)
dṛṣṭāntenāyam eva yathoktasvabhāvapratibandhaḥ pradarśyate / (19,21f)
dpes ji skad bshad pa’i rang bzhin ’brel pa ’di kho na rab tu ston pa / (271b1f)
– (85,8f); PVṬ: – (44a4)
tāv eva hi nivartamānau svapratibaddhaṃ nivartayata iti / (19,25)
de dag kho na ldog pa na rang gi ’brel pa zlog par byed pas … (271b2)
tāv eva hi kāraṇavyāpakau nivarttamānau svapratibaddhaṃ kāryaṃ vyāpyaṃ
ca svabhāvaṃ nivarttayata … (85,14f)
rgyu dang khyab par byed pa de dag kho na ldog pa na / rang dang ’brel pa
’bras bu dang rang bzhin khyab par bya ba zlog par byed pas / (44a5f)
apratibandhe hi katham ekasya nivṛttir anyasya nivṛttiṃ sādhayet / (20,1f)
ma ’brel na ni ji ltar gcig log pas gzhan ldog pa sgrub par byed / (271b3)
– (85,18f); PVṬ: – (44a7)
yadi nāmaite śabdāḥ puruṣaiḥ kvacit praṇinīṣitā api na śakyante praṇetuṃ
vastupratibandhād dhūmādivat / (35,20f)
102
Reinterpretation of the Compound svabhāva-pratibandha
SVTib
KG
PVṬ
20.
PVSV
SVTib
KG
PVṬ
21.
PVSV
SVTib
KG
PVṬ
gal te sgra ’di dag skyes bu rnams kyis ’ga’ zhig la bya bar ’dod du zin kyang
du ba la sogs pa bzhin du dngos po dang ’brel pa’i phyir bya bar mi nus na …
(280b4f)
vastupratibandhāt / vastvāyattatvāt / dhūmādivat / na hy agnipratibaddho
dhūmo vahnipratyāyanasamarthas tadvaiparītyena jalapratyāyane niyoktuṃ
pāryate / tadā vastupratibaddhatve śabdānām ayam upālambhaḥ syād asati
vyatireke kathaṃ ṣaṣṭhyādaya iti // (157,14f)
(;) dngos po dang ’brel pa’i phyir dngos po tha dad pa la rag las pa’i phyir
bya bar mi nus na / sgra rnams dngos po dang ’brel pa de’i tshe / tha dad pa
med na ji ltar drug pa la sogs par ’gyur zhes bya ba’i klan ka ’di ’gyur ba zhig
na / (80a1f)
sarvaś cāyaṃ svalakṣaṇānām eva darśanāhitavāsanākṛto viplava iti tatpratibaddhajanmanāṃ vikalpānām atatpratibhāsitve ’pi vastuny avisaṃvādo …
(43,2f)
’di thams cad rang gi mtshan nyid dag kho na mthong bas gzhag pa’i bag chags
kyis byas pa’i bslad pa yin pa las de dang ’brel pa las skye ba’i rnam par rtog
pa rnams ni de snang ba nyid ma yin yang dngos po la mi bslu ste / (285a2f)
sarvaś cāyam ityādi / sarvo viplava iti sambandhaḥ / viplavo bhrāntiḥ / ayam
iti sāmānyādirūpaḥ / svalakṣaṇānām eva yad darśanan tenāhitā yā vāsanā tatkṛtaḥ / paramparayā sarvavikalpānām vastudarśanadvārāyātatvāt / tathā hi
nityādivikalpā api vastudarśanenaivotpannāḥ sadṛśāparāparotpattidarśanāyātatvāt / tatra tulye sarvavikalpānām vastudarśanadvārāyātatve / tatpratibaddhajanmanām anityādivikalpānām atatpratibhāsitve [’]pi svalakṣaṇāpratibhāsitve [’]pi vastuny avisamvādaḥ /(183,4f)
’di thams cad ces bya ba la sogs pa smos te / ’di zhes bya ba ni rtag pa la
sogs pa’i rnam par rtog pa’o // thams cad bslad pa yin no zhes bya bar sbyar
te / bslad pa’i rnam par rtog pa’o // rang gi mtshan nyid dag kho na mthong
ba gang yin pa des gzhag pa’i bag chags des byas pa yin te / rnam par rtog pa
thams cad ni dngos po mthong ba’i sgo nas brgyud de ’ongs pa’i phyir ro // ’di
ltar rtag pa la sogs pa’i rnam par rtog pa yang dngos po mthong ba kho na las
byung ba yin te / ’dra ba gzhan ’byung ba la sogs pas ’khrul ba ’ba’ zhig tu zad
do // de la rnam par rtog pa thams cad dngos po mthong ba’i sgo nas ’ongs pa
nyid du ’dra ba las de dang ’brel pa las skye ba mi rtag pa la sogs pa’i rnam
par rtog pa rnams ni de snang ba nyid ma yin yang / rang gi mtshan nyid snang
ba nyid ma yin yang mi rtag pa nyid la sogs pa’i ngo bo de dngos po la yod pa’i
phyir dngos po la mi slu ste / (95b4f)
vastusaṃvādas tu vastūtpattyā tatpratibandhe sati bhavati / (49,2)
dngos po ’thob pa ni dngos po las skye bas de dang ’brel pa yin na ’gyur gyi
… (288b6f)
tasmin sādhye pratibandhe sati (/) pratibandha eva kutaḥ / vastūtpattyā sādhyavastūtpattyā hetubhūtayā anyathety asati pratibandhe / (204,8f)
(≠) brgyud de dngos po las skye ba’i rgyur gyur pas dngos po dngos po de
dang ’brel te / de lta yin na dgnos po ’thob par ’gyur gyi zhes bya bar sbyar
ro / (110a1)
Yoichi Fukuda
22.
PVSV
SVTib
23.
KG
PVṬ
PVSV
SVTib
KG
PVṬ
24.
PVSV
SVTib
KG
PVṬ
25.
PVSV
SVTib
KG
PVṬ
26.
PVSV
SVTib
KG
PVṬ
103
vikalpānām arthapratibandhaniyamābhāvāt / na hi vikalpā yathārtham eva
jāyante / (51,8f)
rnam par rtog pa rnams don dang ’brel par nges pa med pa’i phyir te / rnam
par rtog pa rnams ni don ji lta ba bzhin skye ba med pa nyid do // (290a4f)
arthapratibaddho ’rthākārānuvidhānenaotpattiḥ / (212,10)
don dang ’brel pa ni don gyi rnam pa’i rjes su byed par skye ba’o // (116a2)
upakāre ’pi tatraiva tatpratibandha iti kim anyas tatkaraṇāt tadupakārī /
(53,24f)
phan ’dogs na yang de kho na la de ’brel pas de byed pa’i phyir de la phan par
byed pa gzhan ci dgos / (291b1)
upakāre vā sāmānyakṛte upakārasyābhyupagamyamāne / tatraiva sāmānye
tasyopakārasya pratibandha iti kim anyo vyaktibhedas tasyopakārasya karaṇāt
tasya sāmānyasyopakārī … (218,15f)
phan ’dogs na yang zhes bya ba ni spyis byas pa’i phan pa khas len na spyi de
kho na la phan pa de ’brel pas de byed pa’i phyir spyi de la phan par byed pa
gsal ba’i khyad par gzhan ci dgos te … (120a1f)
tadutpattidharmā bhāvaḥ svabhāvapratibandhād apekṣate nāma / (53,26f)
de las skye ba’i chos can ni rang bzhin ’brel pa’i phyir ltos pa zhes bya’o //
(291b2)
tasmād apekṣaṇīyād utpattiḥ sā dharmaḥ svabhāvo yasya sa tadutpattidharmmabhāvaḥ / svabhāvasya pratibandhād āyattatvād apekṣate nāma upakāriṇaṃ …
(218,20f)
gang la chos de’i ngo bo nyid yod pa de ni de las skyes ba’i chos can gyi dngos
po’o // rang bzhin ’brel cing rag las pa’i phyir phan par byed pa la ltos pa zhes
bya ste / (120a3f)
yaḥ kaścit kasyacit kvacit pratibandhaḥ sa sarvo janyatāyām evāntarbhavati /
(54,1f)
gang cung zad ’ga’ zhig la lar ’brel pa de thams cad ni bskyed par bya ba’i
khongs su gtogs so // (291b3)
yaḥ kaścid bhāvaḥ pratibandhaḥ kasyacid vastunaḥ kvacid āśraye sa sarvo
janyatāyāṃ kāryatāyām evodbhavati / (219,3f)
gang cung zad dngos po ’ga’ zhig rten la lar ’brel pa dngos po las gyur pa
de thams cad ni bskyed par bya ba ste / ’bras bu kho na’i khongs su gtogs
so // (120a6f)
katham anyonyasya sāmyam / tatsaṃbandhād iti cet / na / apratibaddhasya
saṃbandhāyogāt / (55,18f)
ji ltar gzhan dang gzhan du mtshungs pa yin / gal te de dag dang ’brel pa’i
phyir ro // zhe na ma yin te / rag ma las pa la ’brel par mi rung ba’i phyir
dang / (292b1f)
apratibaddhasya vyaktāv anāyattasya tābhir anupakṛtasyety arthaḥ / (224,21f)
rag ma las pa la gsal ba la rag ma las pa la ste / de dag gis phan ma btags pa
la zhes bya ba’i tha tshig go // (125a1f)
104
27.
Reinterpretation of the Compound svabhāva-pratibandha
PVSV
SVTib
KG
PVṬ
28.
PVSV
SVTib
KG
PVṬ
29.
PVSV
SVTib
KG
PVṬ
30.
PVSV
SVTib
KG
PVṬ
31.
PVSV
SVTib
KG
tathābhūtārthadarśanadvāreṇāyaṃ nānaikadharmabhedābhedapratibhāsaviplavānusārī vyavahāra iti tasya tatpratibandhe sati tadavyabhicāraḥ / (65,3f)
tha snyad ’di ni yang dag pa’i don de bzhin du mthong ba’i sgo nas chos du ma
dang / gcig dang / tha dad pa dang / tha dad pa ma yin par snang ba ’khrul
pa’i rjes su ’brang ba’i phyir / de dang der ’brel par ’gyur ba la de (DN: las
du) ma ’khrul pa yin no // (297a5f)
tasya vyavahārasya tatpratibandhe tasmiṃs tathābhūte svalakṣaṇe pāramparyeṇotpattipratibandhe sati tadavyabhicāraḥ / vastvavyabhicāraḥ / (260,12f)
de ste tha snyad der ’brel par gyur pa de lta bur gyur pa’i rang gi mtshan nyid
der brgyud pas ’byung ba’i ’brel pa yod na de ma ’khrul te / dngos po la ma
’khrul pa yin no // (153b3f)
samavāyasaṃyogāv ekārthasamavāyādayo ’pi vastusaṃbandhāḥ kāryakāraṇabhāvān na vyatiricyante / parasparam anyato vānupakāriṇām apratibandhāt /
apratibaddhasya ca asambandhāt (70,3f)
phrod pa ’du ba dang ldan pa dang / don gcig la ’du ba zhes bya ba la sogs
pa dngos po’i ’brel pa dag kyang rgyu dang ’bras bu’i dngos po las tha dad
pa ma yin no // gcig la gcig gam gzhan dag gis phan ’dogs pa med pa ni rag
lus pa med pa’i phyir dang / rag lus pa med pa yang ’brel pa med pa’i phyir
ro // (300a6f)
parasparam anyonyam upakāriṇām anyato vā ’śrayābhimatād anupakāriṇām
apratibandhād anāyattatvāt / apratibadhannasya [=apratibandhasya] cāsambandhāt kāraṇāt … (278,1f)
gcig la gcig te / phan tshun nam rten du mngon par ’dod pa gzhan gyis phan
btags pa med pa’o // phan btags pa med pa ni rag lus pa med pa’i phyir te / ltos
pa med pa’i rgyu’i phyir … (167a5)
tatrāpi kāryakāraṇabhāvakṛta eva pratibandhaḥ / (70,9f)
de la yang rgyu dang ’bras bu’i dngos por byas pa kho na rag lus pa yin no //
(300b1)
tatrāpy ekārthasamavāyini kāryakāraṇabhāvakṛta evāśrayeṇa saha yaḥ kāryakāraṇabhāvas tatkṛta eva yaddvāreṇāropita eva sambandhaḥ / (278,15f)
(;) don gcig dang ’phrod pa ’du ba de la yang rgyu dang ’bras bu’i dngos por
byas pa kho na ste / rten dang bcas pa’i rgyu dang ’bras bu’i dngos po gang yin
pa des byas shing des sbyar ba kho na rag lus pa’i ’brel pa yin no // (167b4f)
tām eva sa āśrayaḥ karoti / sā cāpratibaddhā sāmānya iti kim sāmānyasyāśrayeṇa / (72,2f)
de nyid de’i rten byed pa yin la / de yang spyi la rag ma lus pa’i phyir / spyi’i
rten gyis ci zhig byed / (301b2)
sety arthāntarabhūtā sthitiḥ / na hi tasyāḥ sāmānye pratibandhakāraṇaṃ kiṃcid
asti … (285,18)
de zhes bya ba ni gnas pa don gzhan du ’gyur pa’o // de la spyi la rag lus pa’i
rgyu cung zhig kyang yod pa ma yin no // (172a1f)
pratibaddhe vā kaḥ pratibandha iti vācyam / (72,3f)
rag las pa kho na la rag las pa zhes ji ltar brjod par bya / (301b2)
abhyupagamyata eva sthiteḥ sāmānye pratibandha iti ced āha / (285,10)
Yoichi Fukuda
PVṬ
32.
PVSV
SVTib
KG
PVṬ
33.
PVSV
SVTib
KG
PVṬ
34.
PVSV
SVTib
KG
PVṬ
35.
PVSV
SVTib
KG
PVṬ
36.
PVSV
SVTib
KG
105
gal te gnas pa ni spyi la rag lus par khas blangs pa kno na yin no zhe na /
(172a2)
apekṣeti hi tatpratibandhaḥ sa cānādheyaviśeṣasya sāmānyasyāyukta iti /
(72,7f)
ltos pa zhes bya ba ni de la rag lus pa yin na de yang spyi khyad par med pa la
mi rigs pa dang … (301b3f)
asmin vastuny asyāpekṣeti yeyam apekṣā sā tasminn apekṣye pratibandhas
tadāyattatā / (286,2f)
dngos po ’di la ’di ltos pa zhes bya ba la ltos pa gang yin pa de ni ltos par bya
ba ’di dang ’brel cing de la rag lus pa nyid de // (172b3)
nāpi vyatirekiṇas tasya kvacid anāśrayād anyasyāpi vyaṅgyavyañjakabhāvādeḥ
saṃbandhasya kenacid anupakāryasyāpratibandhenābhāvāt / (75,27f)
tha dad pa las kyang ma yin te / de ’ga’ zhig la yang brten pa med pa’i phyir
ro // ’ga’ zhig gis phan btags pa med par rag lus pa med pas na bsal bar bya
ba dang / bsal bar byed pa’i ngo bo la sogs pa las ’brel pa gzhan yang med pa’i
phyir (303b6f)
kiṅ kāraṇaṃ / nityatvāt kenacid vyaktibhedenānupakāryasya sāmānyasyāpratibandhena / na hy apratibandhasya kaścit sambandho ’stīty uktaṃ / (297,8f)
ci’i phyir zhe na / rtag pa nyid kyis gsal ba’i khyad par ’ga’ zhig gis phan btags
pa med pa’i spyi rag lus pa med pa’i phyir ro // rag lus pa med pa’i ’brel pa
’ga’ zhig kyang yod pa ma yin no // (180b3f)
kvacid avisaṃvādo ’syā vastuni kāryakāraṇabhāvapratibandhān … (76,20f)
’ga’ zhig la ’di’i mi slu ba ni dngos po la rgyu dang ’bras bu’i ngo bo’i (D adds
’bras bu’i) ’brel pa las yin gyi / (304a6)
kvacid vastuny asya buddheḥ sakāśād avisamvādo yasmāt kāryakāraṇasambaddhād … (300,4f)
dgnos po ’ga’ zhig la ’di la blo las mi bslu ba gang yin pa de ni / rgyu dang
’bras bu’i ’brel pa las yin te / (182b6)
tad dhi kiṃcid upalīyeta na vā yasya yatra kiṃcit pratibaddham apratibaddhaṃ
vā / (99,27f)
de ni cung zhig ’ga’ zhig la ’brel pa ma yin te / (D: no //) gang gang la cung
zhig rag las pa ma yin no // (317a6f)
tad dhi vastu / kiṃcid upalīyet āśrayet / yasya yatra kiṃcid utpādādikaṃ pratibaddham āyattaṃ / na copalīyeta yasya yatrāpratibaddhaṃ / (366,2f)
dngos po de ni rten du gyur pa ’ga’ zhig la ’brel ba ste brten par ’gyur ba’am /
gang gang la skyed par byed pa po la sogs pa la cung zhig rag lus pa ste ltos
pa’am / ’brel pa med pa ma yin zhing gang gang la rag lus pa ma yin no //
(226b1f)
seyaṃ sattā apratibandhinī cet / niyamavatī na syāt / (99,28f)
gal te yod pa nyid de ’di ni ’ga’ zhig la rag lus pa can ma yin no zhe na / nges
pa can du mi ’gyur ro // (317a7)
seyaṃ sattā kvacid apratibandhinī cet / dravyakālāpekṣayā na niyamavatī syāt /
(366,4)
106
Reinterpretation of the Compound svabhāva-pratibandha
PVṬ
37.
PVSV
SVTib
KG
PVṬ
38.
PVSV
SVTib
KG
PVṬ
39.
PVSV
SVTib
KG
PVṬ
40.
PVSV
SVTib
KG
PVṬ
41.
PVSV
SVTib
KG
gal te yod pa nyid de ’di ni ’ga’ zhig la rag lus pa can ma yin no zhe na / rdzas
dang yul dang dus la ltos par nges pa can du mi ’gyur te / (226b2)
yathāsvaṃ vyāpini sādhye tayor eva pratibandhāt liṅgayor liṅgini / (101,1f)
bdag nyid ji lta ba bzhin du khyab (D: khyad) par byed pa bsgrub par bya ba’i
rtags can la rtags de dag rag lus pa’i phyir … (318a1f)
kāryasya kāraṇaṃ vyāpakaṃ sādhyaṃ (/) svabhāvasyāpi svabhāvo vyāpakaḥ
sādhyas tasmin sādhye liṅgini tayor eva kāryasvabhāvayor liṅgayoḥ pratibandhāt / (370,5f)
’bras bu’i rgyu khyab par byed pa bsgrub par bya ba yin la / rang bzhin gyi
yang rang bzhin khyab par byed pa bsgrub par bya ba yin te / bsgrub par bya
ba rtags can de la ’bras bu dang / rang bzhin gyi rtags de dag de nyid rag lus
pa’i phyir ro // (228a2f)
sadasatpakṣabhedena śabdārthānapavādibhiḥ // vastv eva cintyate hy atra pratibaddhaḥ phalaodayaḥ // (106,27f)
sgra don bsnyon pa med rnams kyis // yod med phyogs kyi bye brag gis // dngos
po nyid ni dpyad bya ste // ’di la ’bras ’byung rag las phyir // (321b1f)
yasmād atra vastuni pratibaddhaḥ phalodayaḥ / (387,11)
ci’i phyir zhe na / ’di ltar dngos po ’di la ’bras bu rag las phyir sgra’i don la ni
ma yin no // (240b7f)
tad ayaṃ pravartamānaḥ sarvadā sadasaccintāyām avadhīritavikalpapratibhāso vastv evādhiṣṭhānīkaroti / yatrāyaṃ puruṣārthaḥ pratibaddho … (107,3f)
de bas na ’di ’jug pa na thams cad du yod pa dang med pa dpyod pa na rnam
par rtog pa’i snang ba la ltos pa med par gang la skyes bu’i don ’di rag lus pa’i
dngos po nyid la dmigs par byed de … (321b3f)
vastv evādhiṣṭhānīkaroti viṣayīkaroti yatra vastuny ayaṃ puruṣārthaḥ pratibaddhaḥ / (388,4f)
dngos po gang la skyes bu’i don ’di rag lus pa’i dngos po nyid la dmigs par
byed cing yul du byed de / (241a5)
na ca tadapratibaddhasvabhāvo bhāvo ’nyaṃ gamayati / (107,25)
de la rag lus pa med pa’i rang bzhin can gyi dngos po ni gzhan go bar byed
pa ma yin no // (322a3)
tasmin vastuny apratibaddhaḥ svabhāvo yasya śabdalakṣaṇasya / anyaṃ yatrāsau na pratibaddhaḥ / (390,7f)
(;) dngos po de la rag lus pa med pa’i rang bzhin gyi sgra’i bdag nyid can
gyi dngos po gang la yod pa zhes bya bar tshig rnam par sbyar ro // gang la
rag lus pa med pa ni gzhan de’i go bar byed pa ma yin no zhes bya bar sbyar
ro // (242b2f)
paryavasthānajanmapratibaddhaduḥkavivekāt / (111,5)
kun nas dkris pa dang skye ba la rag lus pa’i sdug bsngal dang bral ba’i phyir
dang / (324a3)
tayor yat pratibaddhan duḥkhan tasya vivekāt / rāgādyutpattikāle yad duḥkhaṃ
kāyacittaparidāhalakṣaṇan tat paryavasthānapratibaddhaṃ jātijarāvyādhyādiduḥkhan tu janmapratibaddhaṃ / (400,18f)
Yoichi Fukuda
PVṬ
42.
PVSV
SVTib
KG
PVṬ
43.
PVSV
SVTib
KG
PVṬ
44.
PVSV
SVTib
KG
PVṬ
45.
PVSV
SVTib
KG
PVṬ
46.
PVSV
107
de dag la rag lus pa’i sdug bsngal gang yin pa de dang bral ba’i phyir ro // ’dod
chags la sogs pa’i skye ba’i tshe lus dang sems yongs su gdung ba’i mtshan nyid
can gyi sdug bsnagl gang yin pa de ni kun nas dkris pa la rag lus pa’o // skye
ba dang rga ba dang na ba’i sdug bsngal ni skye ba la rag lus pa’o // (251b3f)
tataḥ katham idānīṃ tatpratiniyamasaṃsādhyaṃ tadanvayaṃ sādhayeyuḥ / na
hy apratibaddhas tatsādhana iti / (114,28f)
de bas na de la so sor nges par bsgrub par bya ba de’i rjes su ’gro ba ji ltar
sgrub par byed par ’gyur te / rag las pa med pa de’i (D adds ni) sgrub par byed
pa ma yin no // (326a7)
bāhye ’rthe ’pratibandhena niyamābhāvāt / (417,5)
(;) don gang la rag las pa med pa de ni sgrub par byed pa’i rigs pa ma yin
no // (260b2)
parāśrayo hi sambandho ’pratibandhe tayoḥ sambandhitā ’yogāt / (115,14)
gzhan la (D adds / ) rten pa can ’brel pa yin na ni rag las pa med pa la de dag
gi ’brel pa nyid mi rung ba’i phyir ro // (326b4)
parāśraya iti parassambandhī āśrayo [’]syeti kṛtvā / sambandhini sambandhasyāpratibandhe sati tayoḥ sambandhinos sambandhitāyā ’yogāt / na hy apratibaddhena kenacit kaścit tadvān bhavati gaur ivāśvena / (418,15f)
(;) gzhan la brten pa can ’brel pa yin te / ’brel pa ’di la gzhan la brten pa
yod pa’i phyir ro // ni’i sgras (D adds ni) nges par gzung ba’o // ci’i phyir zhe
na / ’brel pa can la ’brel pa rag las pa med pa yin na ’brel pa can de dag gi
’brel pa nyid mi rung ba’i phyir ro // rta ba lang dang ’dra bar ltos pa med pas
don ’ga’ zhig de dang ldan par mi ’gyur ro // (261b3f)
svabhāvāpratibandho ’nyattvam iti cet / (147,3f)
gal te rang bzhin la rag lus pa med pa ni gzhan nyid yin no zhe na / (346b5f)
nanv atatsvabhāvatve ’pi parasparaṃ svabhāvāpratibandho [’]nyattvam iti
cet / sa ca pratibandhaḥ pudgalasya skandheṣv asti tato tatsvabhāvatve [’]pi
nānyattvaṃ skandhebhyaḥ pudgalasyeti / (527,16f)
gal te rang bzhin la rag lus pa med pa ni gzhan nyid yin na gang zag gi rag
lus pa de yang phung po dag la yod pa de bas na de’i rang bzhin ma yin kyang
phung po dag las gang zag gzhan nyid ma yin no zhe na / (326a6f)
ko ’yaṃ pratibandho nāma yena sa ca na syāt / nānyasvabhāvaś ca / (147,4f)
rag lus pa zhes bya ba ’di gang yin / gang gis der yang mi ’gyur ba dang / rang
bzhin gzhan du’ang mi ’gyur / (346b6)
ko [’]yaṃ pratibandho nāma pudgalasya skandheṣu yena pratibandhena / sa
ca na syād iti skandha-svabhāvaś ca pudgalo na syāt / nānyasvabhāvaś ca
skandhebhyaḥ / (527,19f)
phung po dag la gang zag gi rag lus pa zhes bya ba ’di’i gang yin / rag lus pa’i
rgyu gang gis der yang mi ’gyur ba zhes bya ba gang zag phung po’i rang bzhin
du mi ’gyur ba dang gzhan gyi rang bzhin can du mi ’gyur ba ste / (326a7f)
na ca tajjanmalakṣaṇāt svabhāvapratibandhād anyaḥ pratibandho nāma / anāyattasya vyabhicārā-virodhāt / (147,10f)
108
Reinterpretation of the Compound svabhāva-pratibandha
SVTib
KG
PVṬ
47.
PVSV
SVTib
KG
PVṬ
48.
PVSV
SVTib
KG
PVṬ
49.
PVSV
SVTib
KG
PVṬ
50.
PVSV
de’i skye ba’i mtshan nyid kyi rang bzhin gyi rag las pa las rag lus pa gzhan
yang yod pa ma yin te / rag lus pa med pa’i ’khrul pa la ’gal ba med pa’i phyir
ro // (347a1f)
na cānyaḥ pratibandhaḥ pudgalasya skandheṣu / yasmān na hi janmalakṣaṇāj
janmasvabhāvāt svabhāvapratibandhād anyaḥ pratibandho nāma / kiṃ kāraṇam (/) anāyattasya tadutpattyā tatrāpratibaddhasya / tena saha yo vyabhicāras
tasyāvirodhāt / (528,8f)
(;) gal te skye ba’i mtshan nyid can gyi rag lus pas phung po dag las gang zag
gzhan nyid du yod pa ni ma yin mod kyi ’on kyang rag lus pa gzhan nyid du
yod do zhe na / de’i zhes bya ba la sogs pa smos te / skye ba’i mtshan nyid can
te / skye ba’i dbang po can gyi rang bzhin gyi rag las pa las rag lus pa zhes
bya ba gzhan yod pa ma yin no // ci’i phyir zhe na / de las byung ba’i rag lus
pa med pa de ni de la rag lus pa med pa de dang lhan cig ’khrul pa gang yin
pa de la ’gal ba med pa’i phyir ro // gang zhig gang la rag lus pa med pa de ni
der ’khrul pa’i phyir zhes bya ba’i tha tshig go // (326b5f)
jñānakṛtaḥ pratibandha iti cet / syād etat / yatpratipattināntarīyakaṃ yajjñānaṃ tadgatau niyamena tatpratibhāsanāt tad atadrūpam apy avācyam iti /
(147,12f)
gal te shes pas byas pa’i rag lus pa yin no zhe na / de ltar ni ’gyur na gang
rtogs pa med na mi ’byung ba’i shes pa gang yin pa des rtogs pa la nges par de
snang ba’i phyir de de’i ngo bo ma yin du zin kyang / brjod par bya ba ma yin
no zhe na / (347a2f)
pudgalasya skandheṣu jñānakṛtaḥ pratibandha iti cet / … / yasya rūpādeḥ pratipattir yatpratipattis tayā nāntarīyakam avinābhāvi yajjñānaṃ yasya pudgalasya
jñānaṃ / tadgatāv iti rūpādigatau niyamena tasya pudgalasya pratibhāsanāt /
jñānakṛtaḥ pratibandhas … / tena jñānakṛtāt pratibandhāt / (528,13f)
(≠) phung po dag las gang zag gi shes pas byas pa’i rag lus pa yin no zhe na / …
(327a4)
nanu saivāsati pratibandhe na yuktety ucyate / (147,26)
de nyid rag lus pa (DN inserts med pa) med par rigs pa ma yin no zhes brjod
pa ma yin nam / (347a6)
nanu saiva pratītes tannāntarīyakatā / rūpādiṣu pudgalasyāsati pratibandhe
na yuktety ucyate / (530,4f)
de med na mi ’byung ba’i shes pa de nyid ni gzugs la sogs pa gang zag rag lus
pa med par rigs pa ma yin no zhes brjod pa ma yin nam / (328a4)
nākāryakāraṇayoḥ kaścit pratibandha iti coktam / (148,1)
rgyu dang ’bras bu ma yin pa dag la ni rag lus pa cung zad kyang ma yin no
zhes bya ba yang bshad zin to // (347a6f)
akāryakāraṇayor na kaścid vāstavaḥ pratibandha ity asakṛd uktaṃ … (530,7)
rgyu dang ’bras bur gyur pa ma yin pa dag la ni dngos su ’brel pa cung zad
kyang yod pa ma yin no zhes mang du bshad zin to // (328a5)
vivakṣayā śabdo ’rthe niyamyate / na svabhāvataḥ / tasya kvacid apratibandhena sarvatra tulyatvāt / yatrāpi pratibandhas tadabhidhānaniyamābhāvāt /
(172,19f)
Yoichi Fukuda
SVTib
KG
PVṬ
51.
PVSV
SVTib
KG
PVṬ
109
brjod par ’dod pas sgra ni don dag la nges par bya ba yin gyi ngo bo nyid kyis
ni ma yin te / de ni ’ga’ zhig la yang rag lus pa med par thams cad la mtshungs
pa nyid kyi phyir ro // gang la rag las pa yin na yang de rjod par byed pa’i nges
pa med pa’i phyir ro // (363a4f)
tasya śabdasya kvacid vastuny apratibandhena sambandharahitatvena kāraṇena sarvatrārthatulyatvāt / yatrāpi śabdasya pratibandhaḥ sthānakaraṇeṣu
tataḥ śabdānām utpatter abhivyakter vā / (606,3f)
de ni dngos po ’ga’ zhig la rag lus pa med pa ste / ’brel pa med par don thams
cad la mtshungs pa nyid kyi phyir ro // gang la sgra’i rag las pa gnas dang byed
pa dag gi yin te / de las (DN: la) sgra rnams skye ba’i phyir ram rtag par smra
ba’i gzhung gis mngon par gsal ba’i phyir ro // (nye 61a4f)
uktam atrāpratibandhād aniyama iti // (172,27)
’dir rag lus pa med pa’i phyir / nges pa yod pa ma yin no zhes bshad zin to //
(363a6f)
uktam atrottaraṃ / kvacid vastuny apratibandhād aniyata iti / (606,17)
dir lan ni dngos po ’ga’ zhig la yang rag lus pa med pa’i phyir nges pa yod pa
ma yin no zhes bshad zin to // (nye 61b3)
Jain Conceptions of Non-Cognition: A Dialogue with
Dharmakīrti on Inferential Evidence
by
Marie-Hélène Gorisse
1. Introduction
1.1. Philosophical background
This paper aims at presenting and comparing the way non-cognition can serve as the
basis of inference according to Buddhist and Jain philosophers, as well as to indicate the
philosophical relevance of the main divergences between the two conceptions. I will focus
on the view of the Buddhist Dharmakīrti, as it is found in his Pramāṇavārttikasvavṛtti
and Nyāyabindu, as well as on the view of the Jains Akalaṅka in his Laghīyastraya and
Māṇikyanandi in his Parīkṣāmukham.
First of all, the teachings of Akalaṅka (720–780) represent an important step in the
development of Jain philosophy, especially in the constitution of a paradigm clearly distinct
from the Buddhist one. More precisely, Akalaṅka has founded a systematic Jain theory of
knowledge in answer to Dharmakīrti’s attacks against the Jain theory of non-one-sidedness.
Following the style of his Buddhist opponent, Akalaṅka expresses his theses in a very
concise way. For example, in his Laghīyastraya (henceforth LT), the Three Short [Treatises],
instead of presenting all the types of inference he grants, he presents only the discrepancies
between the ones Dharmakīrti accepts and the one he himself accepts. This concise style
explains our recourse to his commentators. Firstly, the Jain Māṇikyanandi (9th c.) has
organized Akalaṅka’s mature philosophy in the Parīkṣāmukham (PM), the Introduction
to Philosophical Investigation. What is more, this work has itself been commented on by
the Jain Prabhācandra (980–1065) in his Prameyakamalamārtaṇḍa (PKM), the Sun that
Grows the Lotus of Knowable, as well as by Vādi Devasūri (12th c.) in his Pramāṇanayatattvālokālaṃkāra (PNT), the Commentary on the Explanation of the Nature of Knowledge
and Viewpoints. These three works constitute a lineage of commentaries and share the
same conception of inference. I will refer to them as “the tradition of Akalaṅka.” Since PM
is the first work in this tradition, I will mainly refer to this text, and will quote from PKM
and PNT respectively only when considering matters which are absent from earlier works.
Dharmakīrti’s texts are implicitly referred to in the sections on inference in LT and in
PM, and explicitly in PKM and PNT. There, the most frequently quoted work of Dharmakīrti
is the Pramāṇavārttikasvavṛtti (PVsV), the Auto-commentary on the Essay on Knowledge.
In this paper, following Prabhācandra and Vādi Devasūri, I will mainly refer to the PVsV.
But I will also make use of Dharmakīrti’s Nyāyabindu (NB), the Drop of Logic, because
one finds there the most extensive list of correct types of inference he grants.
To begin with, it is useful to sketch the aforementioned framework of theories of
inference, so as to agree on concepts as well as on the means to express them. First of all,
inference is the cognitive process by which an epistemic agent acquires new knowledge
Birgit Kellner et al., eds., Reverberations of Dharmakīrti’s Philosophy: Proceedings of the Fifth International Dharmakīrti
Conference Heidelberg, August 26 to 30, 2014. Vienna 2020, pp. 111–128.
112
Jain Conceptions of Non-Cognition
using reasoning. It is specifically used in situations in which cognitive processes like
perception fail. Entities being dealt with in metaphysical inquiries are typical examples
of such situations. Inferential reasoning consists more precisely of the transmission of
certainty from the established knowledge of the fact that a property, referred to as “the
evidence-property” (hetu), is ascribed to a given object, to the new knowledge of the fact
that another property, referred to as “the target-property” (sādhya), is ascribed to the same
object.1 For example, from the previous knowledge that there is smoke on the hill, a person
can know that there is fire on the hill.2 As Shah (1967: 248) noticed, the Sanskrit expression
for inference, anumāna, refers to this transmission, since it means “the knowledge that
follows [another knowledge].” And this transmission of certainty is based on the vyāpti
between the target-property and the evidence-property. Vyāpti means “pervasion” and is
traditionally translated by the technical expression “invariable concomitance” in order to
refer to the situation in which whenever the evidence-property is present, the target-property
is present too. Because such a situation is usually granted by the fact that the target-property
is pervading the evidence-property, as we find, for example, between the property of being
a tree and the property of being a Sissoo tree.
What is more, stating such an inferential reasoning is the core mechanism of philosophical debates, in which the aim is to convince the interlocutor that a given piece of knowledge,
even if not directly agreed on, is a valid one. In consequence, since every participant in
such a debate is to seek convincing inclusions, the determination of the extension of the
domains of predicates is one of the core issues in those debates.
1.2. Types of inferential evidence in Buddhism and in Jainism
In this conceptual framework, philosophers were aiming towards a theory of the proper
relationship between the target-property and the evidence-property. That is to say that
they considered it insufficient that the target-property is always present when the evidenceproperty is present, and they wanted to be able to distinguish between arbitrary and necessary universal relationships. An important step towards such an achievement was Dignāga’s
introduction of the particle eva, which functions similarly to the operator of focus “only.”3
One of Dharmakīrti’s subsequent breakthroughs in this dynamic was to seek the precise
reasons why a target-property is always present when its evidence-property is present,
and as a consequence to accept as good evidence only the properties that are ‘essentially’
connected to the target-property. The requirement that inferential reasoning relies only on
1
2
3
In the following, I will abbreviate sentences as “the property of ‘being endowed with fire’ is ascribed to
the subject ‘here’” by saying simply “there is fire here.”
In my presentation of inferential reasoning, I make explicit the epistemic conditions by writing “the
knowledge that there is smoke on the hill,” and not “there is smoke on the hill.” But I defend the position
that although these epistemic conditions are present, it is important to keep them implicit in a formal
representation. I do so because in contemporary logic, expressing these epistemic conditions within the
object language is usually a technique used in order to have a level of description in which it is possible
to deal only with the pure relation between propositions. But this is important to keep in mind that logic
in India is concerned with the relation between an epistemic subject and a proposition.
Dignāga introduced this particle in his attempt to combine the theory of the triple characteristic of the
evidence-property (trairūpya) with the theory of the necessity of the absence of the target-property
when the evidence-property is absent (avinābhāva).
Marie-Hélène Gorisse
113
necessary relationships led Dharmakīrti to consider only two types of essential connections
as ensuring correct inferences: the connection between an effect and it cause, and the
connection between two natural properties. In doing so, Dharmakīrti provides “the ontic
foundation for valid reasoning” (Katsura 1992: 224). More concretely, only three types
of inferential evidence are granted: (i) natural properties (svabhāva); (ii) effects (kārya);
and (iii) non-cognitions (anupalabdhi). Indeed, with such a conception, it is not accidental
that whenever there is a Sissoo tree, there is also a tree. This is due to the very nature of
the Sissoo, for which being a tree is a natural property. And this is also not accidental that
whenever there is smoke, there is also fire. This is due to the very nature of smoke, which
is the effect of fire. What is more, cases of non-cognition are consequences of this state of
affairs also, since it is not accidental that whenever there is no tree, there is also no Sissoo
and that whenever there is no fire, there is no smoke.
Two remarks are important for what follows. First of all, we should keep in mind the
difference between the first two types of evidence, which are used to infer a presence; and
the last one, which is used to infer an absence. The second remark is that when Dharmakīrti
speaks about “natural property” he first intends the relation between, e.g., the property
of being a cow and the property of being an animal. In other words, he intends a relation
between two predicates that do not have the same extension. If we consider that these
predicates denote natural kinds, then “pervaded properties” are species, and “pervasive
properties” are genera.4 What is more, only pervaded properties are good evidence to
infer the presence of their respective pervasive properties, and not the other way around,
since knowing that there is a Sissoo is sufficient to know that there is a tree, but knowing
that there is a tree is not sufficient to know that there is a Sissoo, for there might be an
oak. Contrary to this, only pervasive properties are good evidence to infer the absence of
their respective pervaded properties, and not the other way around, since knowing that
there is not a tree is sufficient to know that there is no Sissoo, but knowing that there is
no Sissoo is not sufficient to know that there is no tree, for there might be an oak. What is
more, when Dharmakīrti speaks about “natural property” he also intends to speak about
the relation between, e.g., the property of being perishable (anityatva) and the property of
being a product (kṛtakatva). In this case, we have a relation in which the two predicates are
co-extensive. Here, no restriction needs to be done in order to draw correct inferences. Both
conceptions are in the same category “natural property,” because in both cases it concerns
the description of the nature of a thing, and because in both cases there is a numerical
identity between what is characterized by the pervaded property and what is characterised
by the pervasive property.
Jain philosophers of Akalaṅka’s lineage have a different conception of evidence. More
precisely, they consider that there are especially six situations in which the presence of an
invariable concomitance is unquestionable, namely when the evidence-property is (i) a
property pervaded (vyāpya) by the target-property; (ii) an effect (kārya) of it; (iii) a cause
(kāraṇa) of it; (iv) a predecessor (pūrvacara) of it; (v) a successor (uttaracara) of it; or
(vi) a co-existent (sahacara) with it. My paper deals with the causes and consequences
4
We are used to conceiving the species ‘cow’ as the set of all cows. But in Vaiśeṣika, the universal
‘cowness’ is a characteristic possessed by all cows. This is how genus and species should be considered
here also.
114
Jain Conceptions of Non-Cognition
of one discrepancy between Akalaṅka’s tradition and that of Dharmakīrti,5 namely the
fact that “non-cognition” is not part of this list. In fact, instead of being considered as
good inferential evidence, as it is by Dharmakīrti, “non-cognition” is considered by Jain
philosophers as part of the general form an inference might have.
1.3. Non-cognition as a cognitive or as a linguistic process
In order to understand why “non-cognition” is not listed by Jain philosophers as good
evidence, but as part of the general form an inference might have, let us consider the
motivations of its introduction for both Dharmakīrti and Akalaṅka. First, when introducing
non-cognition as a type of evidence, Dharmakīrti’s program is quite specific: he intends to
prove the possibility of knowing absences from inference. Such a possibility is essential
for Buddhist soteriology, which relies on the awareness of the absence of a persistent
soul. This, in turn, enables him not to commit himself to the existence of a third kind of
means of knowledge in addition to perception and inference, since everything – absences
included – can be known from one of these two means.6 Therefore, he is only interested in
non-cognition as good evidence to infer an absence. Contrary to this, Jain philosophers
additionally investigate the situations in which it is possible, from non-cognition, to infer
a presence. Such considerations on the possibility to infer presences from absences are
not new. They can already be found in the Vaiśeṣikasūtra 3.1.9 and, approximately at the
same period, in Jain canonical literature in the Ṭhāṇaṃgasutta (Sthānāṅgasūtra, Possibilities) 4.3.336. Besides, Dharmakīrti criticizes these conceptions in his PVsV in the
chapter on inference for oneself, svārthānumānapariccheda 20ff. If we compare Akalaṅka,
Māṇikyanandi, Prabhācandra and Vādi Devasūri, we can see an evolution of their focus.
First, Akalaṅka in his LT is clearly responding to Dharmakīrti. He does refer to noncognition also as a means to know absences and diverges from his Buddhist opponent
mainly on the following:
LT.15. People [think they] know the non-existence of the mind of others
because of its invisibility. [But this is incorrect,] because the modifications
of the forms of this [mind] would be impossible otherwise [than with the
postulation of the existence of their minds].7
The relevance of this quote is intelligible only if we recall that Dharmakīrti claimed that it is
only the non-cognition of something that is usually perceptible and that meets the conditions
for cognition that can be used as the basis of an inference.8 In LT 15, Akalaṅka is criticizing
this on the ground that it is wrong to think that non-cognition can be used to infer the
absence only of perceptible entities, since it can be used to infer the absence of imperceptible
entities as well. For example, from the absence of the recognised characteristics of human
cognitive abilities at the moment of death, it is possible to infer the absence of the mind
5
6
7
8
For a more thorough survey of Jain theories of inference in Akalaṅka’s tradition, see Gorisse 2017.
For more on this topic, see Katsura 1992: 228.
adṛśya-paracitta-āder abhāvaṃ laukikā viduḥ / tad-ākārā-vikāra-āder anyathā anupapattitaḥ // Edited
by Shastri (1939: 6). Also translated in German by Balcerowicz (2005: 175).
For more on this topic, see Kellner 2003.
Marie-Hélène Gorisse
115
that causes them (see Shah 1967: 276). We won’t go further in this discussion, as it is
sufficient for our purpose to understand that for both authors, the concern is primarily an
epistemological one.
Contrary to this, non-cognition is by the time of Māṇikyanandin primarily conceived
as a negative premise in the stated form of an inference. We have mentioned the fact that
inference is not only a cognitive process by which one can acquire knowledge, it is also the
rational means used in debates in order to be assured that from a true input, the output is
always true as well. In this context of convincing, the inferential process has to be stated, and
when this is done so it is commonly followed by a regulated argumentation aiming to defend
or refute it. The stated form of an inference is what Indian philosophers call “inference
for others” (parārthānumāna), in opposition to “inference for oneself” (svārthānumāna).
And whereas Dharmakīrti has introduced non-cognition already in the section presenting
the inference for oneself, Māṇikyanandi and after him Prabhācandra and Vādi Devasūri
introduce non-cognition only in the section presenting the inference for others, that is to
say, in the section that deals with inference firstly as a linguistic device to transfer values
with certainty. And this switch from an epistemological to a formal concern becomes fully
explicit with Vādi Devasūri, who introduces cognition and non-cognition in terms of basis
for, respectively, affirmation and negation:
PNT.3.55. Cognition and non-cognition are the basis for the establishment of
affirmation and negation.9
This focus on the linguistic form of inference is what explains why in this precise section of
the treatises of Māṇikyanandi, Prabhācandra and Vādi Devasūri, nothing is said concerning
epistemological problems related to the knowledge of absences. In other words, the Jain
focus is on non-cognition as a negation, that is to say as a linguistic device usable to reverse
the truth value of a sentence. This work on the relationship between negative (respectively
affirmative) premises and negative (respectively affirmative) conclusions led them to single
out four forms an inference might have, namely:
(i) Cognition of [evidence] compatible [with the target-property] in the case of affirmation [of the thesis] (aviruddha-upalabdhir vidhau);
(ii) Cognition of [evidence] incompatible [with the target-property] in the case of
negation [of the thesis] (viruddha-upalabdhiḥ pratiṣedhe);
(iii) Non-cognition of [evidence] compatible [with the target-property] in the case of
negation [of the thesis] (aviruddha-anupalabdhiḥ pratiṣedhe);
(iv) Non-cognition of [evidence] incompatible [with the target-property] in the case of
affirmation [of the thesis] (viruddha-anupalabdhir vidhau).
Since the goal of this paper is to compare what can be inferred from non-cognition
within this classification of forms of inference with what can be inferred from non-cognition
in Dharmakīrti’s classification of types of evidence, it is important to have his classification
in mind before we explain the Jain one.
9
upalabdhir vidhi-niṣedhayoḥ siddhi-nibandhanam anupalabdhiś ca / Bhattacharya (1967: 226).
116
Jain Conceptions of Non-Cognition
1.4. Types of non-cognition in Dharmakīrti
First of all, according to Dharmakīrti, there are several means to infer that a property
does not pertain to a given object. On one hand, this can be achieved from the previous
knowledge that another property does not pertain to this object as well. For example, if there
is no tree in a given place, then there is also no Sissoo in this place. This first type of noncognition is “non-cognition” (anupalabdhi) properly speaking. Here, knowledge of absence
is gained by means of absence of knowledge. On the other hand, to infer that a property
does not pertain to a given object can also be achieved from the previous knowledge that
a property incompatible with the one that one seeks to know does pertain to the object
under discussion. For example, it is sufficient to know that there is a fire in the room, in
order to know that the room is not cold. This is so because there cannot be heat and cold at
the same place at the same time. For this second type of non-cognition, even though the
process does involve a cognition (of the fire) properly speaking, Dharmakīrti speaks also
about “non-cognition,” because the cognition of the presence of a property incompatible
with another one is equivalent to the knowledge of the impossibility for at least one of the
conditions of existence of this other property to be the case. And this, in turn, amounts to
the non-cognition of this precise condition.
Now, next to this distinction between non-cognition properly speaking and cognition
of a property incompatible with the one that one seeks to know, Dharmakīrti also grants
that it is possible to know absence of different types of thing. For example, one may know
the absence of something, the absence of its effect, or the absence of its cause. From these
distinctions, eleven types of non-cognition are recognised in NB.10 For the sake of the
coming comparison, I will not follow the order of Dharmakīrti’s presentation in NB, but I
will present the eleven types of non-cognition following the Jain thematic classification:
Cognition of an [evidence] incompatible [with the target-property]
Type 1 (viruddha-x-upalabdhi)
(ix) Cognition of a [property] pervaded by a [property] incompatible [with the targetproperty] (viruddha-vyāpta-upalabdhi)
(x) Cognition of the effect of [something] incompatible [with the target-property]
(viruddha-kārya-upalabdhi)
Cognition of an [evidence] incompatible [with the target-property]
Type 2 (x-viruddha-upalabdhi)
(v) Cognition of a [property] incompatible with [the target-property] itself (svabhāvaviruddha-upalabdhi)
(vi) Cognition of a [property] incompatible with a [property] pervading [the targetproperty] (vyāpaka-viruddha-upalabdhi)
(vii) Cognition of a [property] incompatible with the effect [of the target-property]
(kārya-viruddha-upalabdhi)
(viii) Cognition of a [property] incompatible with the cause [of the target-property]
(kāraṇa-viruddha-upalabdhi)
10
Again, I have chosen the presentation of types of non-cognition as it is in NB, because it is the most
comprehensive list, so it furnishes more material to work with in the line of a comparison. But other
presentations are found in the different texts of Dharmakīrti. Especially, three types of evidence involving
non-cognition are granted in HB; and four types in PVsV.
Marie-Hélène Gorisse
117
Non-cognition [of an evidence for the target-property] (anupalabdhi)
(i) Non-cognition of [the target-property] itself (svabhāva-anupalabdhi)
(ii) Non-cognition of a [property] pervading [the target-property] (vyāpaka-anupalabdhi)
(iii) Non-cognition of the effect [of the target-property] (kārya-anupalabdhi)
(iv) Non-cognition of the cause [of the target-property] (kāraṇa-anupalabdhi)
Complex cognition of an [evidence] incompatible [with the target-property]
(xi) Cognition of the effect of a [property] incompatible with the cause [of the targetproperty] (kāraṇa-viruddha-kārya-upalabdhi)
We will examine these types of negative evidence one by one in the course of the presentation
of the classification offered by Māṇikyanandi.
2. The four forms of inference according to Māṇikyanandi
In what follows, I will focus on Māṇikyanandi, because Akalaṅka does not offer such an
extensive list of the forms of inference and of the different types of evidence valid in each
form. In addition, the conceptions of Prabhācandra and Vādi Devasūri on this topics are
substantively the same as those of Māṇikyanandi.
2.1. Cognition of [evidence] compatible [with the target-property] in the case of affirmation [of the thesis] (aviruddha-upalabdhir vidhau)
The first pattern is the plain affirmative one. This is the form I have been using by default
until now. It is used to express inferences of the presence of a property due to knowledge of
appropriate evidence for it. As we have seen, Jain philosophers grant six types of evidence
as being appropriate in this context. In Māṇikyanandi’s words:
PM.3.59. In the case of affirmation [of the thesis], there are six kinds of
cognition of [evidence] compatible [with the target-property], namely [the
evidence can be a property] pervaded (vyāpya) [by it], an effect [of it], a cause
[of it], a predecessor [of it], a successor [of it] or [a property] co-existent [with
it].11
Māṇikyanandi gives the example “sound is subject to change, because it is a product.”12 In
the Indian paradigm of philosophy, this is an example with co-extensive predicates. But
it is important to recall that with predicates of different extensions, only the one which is
pervaded constitutes correct evidence in this first form. To compare with Dharmakīrti’s
theory, this is in this plain affirmative form that he acknowledges evidence-properties that
are natural properties (svabhāva) or effects of the target-property. In PVsV.1.1.6, as in
NB.16–17, the examples are, respectively, “this is a tree, because this is a Sissoo” and “there
11
12
aviruddha-upalabdhir vidhau ṣoḍhā vyāpya-kārya-kāraṇa-pūrva-uttara-sahacara-bhedāt / Ghoshal
(1940: 122).
pariṇāmī śabdaḥ kṛtakatvāt / Ghoshal (1940: 127).
118
Jain Conceptions of Non-Cognition
is fire there, because there is smoke.”13 In conclusion, cognition of evidence compatible
with the target-property in the case of affirmation of the thesis is twofold in Dharmakīrti’s
theory and sixfold in the Jain one.
I will not discuss predecessor, successor or co-existent in this paper,14 which focuses on
the divergences between the two frameworks only when non-cognition is involved, because
Dharmakīrti never accepts predecessor, successor or co-existent as correct evidence, no
matter whether a non-cognition is involved or not or whether the conclusion of the inference
is being stated in a positive or in a negative form. Contrary to this, there are interesting
divergences to reflect upon when cause is being considered, because the efficiency of causal
evidence is not the same for affirmative and negative forms. Therefore, we will have a special
focus on it. First of all, in the affirmative form, Dharmakīrti and Māṇikyanandi agree on
the fact that the presence of something can be inferred from the knowledge of the presence
of its effect, as in “there is intelligence in this living being, because [it shows activities]
like speech.”15 And the divergence appears when the Jain tradition considers as well that
the presence of something can be inferred from the knowledge of the presence of its cause,
like in “there is shade here, because there is an umbrella.”16 In consequence, the causal
relationship is a symmetric one as far as the Jain conception is concerned, whereas for
Dharmakīrti, only the effect, and not the cause, can serve as evidence in a correct inference.
The fact that, e.g., the presence of a seed is not sufficient evidence for the future presence
of a plant, is due to the fact that it is impossible to be sure that no impediment is blocking
the potency of the given cause to produce its effect.17 If combinatory considerations alone
are being developed, the same divergence, namely the fact that the Jain tradition under
consideration accepts both cause and effect, and that Dharmakīrti accepts only the effect,
as correct evidence, should be seen in the second form of inference as well. Let us have a
look at this second form to see if this is what happens.
2.2. Cognition of [evidence] incompatible [with the target-property] in the case of
negation [of the thesis] (viruddha-upalabdhiḥ pratiṣedhe)
The second form is used to express inference of the absence of a property from knowledge
of evidence incompatible with it.
13
14
15
16
17
ta ete kārya-svabhāva-anupalabdhi-lakṣaṇās trayo hetavaḥ / yathā ’gnir atra dhūmāt / vṛkṣo ’yaṃ
śiṃśapātvāt / Gnoli (1960: 2); svabhāvaḥ […] yathā vṛkṣo ’yaṃ śiṃśapātvād iti / kāryaṃ yathā vahnir
atra dhūmād iti / Desai (1991: 52).
This has been done in Gorisse 2015.
asty atra dehini buddhir vyāhāra-ādeḥ / Translated by Ghoshal (1940: 128).
asty atra chāyā chatrāt / Ibid.
This discussion exceeds the aim of this paper. Let me just indicate that Māṇikyanandi, as well as
Naiyāyika philosophers, will counter-attack Dharmakīrti by offering a more finely grained definition
of a ‘cause’ as being what already consists of the totality of conditions needed for the emergence of
the effect. In other words, as what already ensures that the pre-requisite that nothing is blocking its
potency is fulfilled. Since this conception implies that the effect is already present when the cause is
being investigated, both conceptions actually agree.
Marie-Hélène Gorisse
119
2.2.1. What is an incompatible (viruddha) evidence?
In all Dharmakīrti’s examples, something is incompatible with the target-property if it is
its precise contrary. For example, “constant” (dhruva) is incompatible with “inconstant”
(adhruva), or “cold” (śīta-sparśa) with “fire/heat” (agni). As far as pervasion and causality
are concerned, Māṇikyanandi also uses contraries to illustrate incompatibility. But when it
comes to succession, a property recognised as being incompatible with the target-property
is not necessarily its contrary. For example, the present rising of the star Revatī is recognised
as being incompatible with the rising of the star Rohiṇī in a muhūrta,18 because Revatī is
the group of stars in the constellation of Pisces which is the last group of stars to rise in the
sky. And in this way, it is impossible for Rohiṇī to rise after it:
PM.3.75. Rohiṇī won’t rise in a muhūrta, because Revatī has just risen.19
As a consequence, at least for Māṇikyanandi and his commentators, what is incompatible with the target-property is “anything whose presence prevents the presence of the
target-property.” This being understood, the types of incompatible evidence which ensure
inferential knowledge are exactly the same types of evidence as in the previous form. In
Māṇikyanandi’s words:
PM.3.71. In the case of negation [of the thesis], the types of cognition of
evidence incompatible [with the target-property] are the same [as the types
of cognition of evidence compatible with the target-property in the case of
affirmation of the thesis].20
To state it in an explicit way, a property pervaded by something incompatible with the
target-property, an effect, a cause, a predecessor or a successor of something incompatible
with the target-property, as well as a property co-existent with something incompatible
with the target-property, all these are correct evidence of the absence of the target-property.
In PM.3.72, the example “there is no feeling of cold here, because there is warmth”21 is
found. In this example, warmth is a species of heat, which is incompatible with cold.
This form of inference with negative conclusions is a means to know absences. As
such, it should interest Dharmakīrti. And indeed, he recognizes as correct evidence two
types which belong to this form, namely the cognition of a property pervaded by something incompatible with the target-property and the cognition of an effect of something
incompatible with the target-property.22 The example presented in NB.2.36 is not the same
as that found in PM.3.72: the disappearance of a being – even of one that has come into
18
19
20
21
22
A muhūrta is a unit of measurement in classical Indian astrology, representing approximately 48
minutes.
na-udeṣyati muhūrta-ante śakaṭaṃ revaty-udayāt / Ghoshal (1940: 130). Śakaṭaṃ (the Chariot) is
another name for Rohiṇī (the Rising one). Its Arabic name is Aldebaran (the Follower), because its rise
follows the one of the famous Pleiades. Concerning Revatī, it means “the Prosperous.”
viruddha-tad-upalabdhiḥ pratiṣedhe tathā / Ghoshal (1940: 129).
na asty atra śīta-sparśa auṣṇyāt / Ghoshal (1940: 129).
See paragraph 1.4 of this paper.
120
Jain Conceptions of Non-Cognition
being – is not inevitable, because it requires a further cause.23 In this example, “requiring a
further cause” is pervaded by “being evitable;” and the properties of “being evitable” and
of “being inevitable” are incompatible ones. As for the cognition of an effect of something
incompatible with the target-property, the example in NB.2.35 is the same as that found in
PM.3.73, namely “there is no feeling of cold, because there is smoke here.”24 Here, “being
endowed with smoke” is an effect of “being endowed with fire/heat;” and the properties of
“being endowed with heat” and of “being endowed with cold” are incompatible ones.
To sum up, Māṇikyanandi accepts the same six types of evidence for the first and
for the second form of an inference. In the same way, my presentation of Dharmakīrti’s
theory indicates that he accepts pervaded (vyāpta)25 properties and effects of a given targetproperty as sufficient evidence to infer its absence, as he did in the affirmative form. This
means that here again, the Buddhist and the Jain tradition disagree on the status of the
cause, since Māṇikyanandi recognizes the cause as correct evidence, as in PM.3.74 “there
is no happiness in this creature, because it has grief,”26 whereas for Dharmakīrti, only the
effect, and not the cause, can serve as incompatible evidence to infer an absence:
PV.1.5. The cognition of the causal conditions of what is incompatible with
that [property which is to be established] is erratic when used when there
is no incompatibility between the causal conditions [of the property to be
disestablished and the property that is incompatible with it].
PVsV.1.5.1. For example, [a bad inference based on this kind of erratic sign
is]: because there is firewood here, there is no feeling of cold.27
In other words, it is not sufficient to know the presence of the cause of what is incompatible
with a given target-property in order infer the absence of the latter, because something
might block the potency of the cause. On the contrary, knowing the presence of what is
incompatible with the cause of a given target-property is sufficient in order to infer the
absence of the latter, since it cannot be present without its cause.28 First, Māṇikyanandi is
avoiding this problem, since in PM.3.74 he provides with an example of the cause of what is
incompatible with a given target-property, which turns out to be also what is incompatible
with the cause of this target-property. Second, the discrepancies between the cause of what
is incompatible with a given target-property and what is incompatible with the cause of
this target-property calls for another type of non-cognition evidence, as we will see in the
next paragraph.
23
24
25
26
27
28
viruddha-vyāpta-upalabdhir yathā na dhruva-bhāvī bhūtasya api bhāvasya vināśo hetv-antaraapekṣaṇād iti / Desai (1991: 71).
viruddha-kārya-upalabdhir yathā na atra śīta-sparśo dhūmād iti / Desai (1991: 70).
The difference between natural property (svabhāva) and pervaded (vyāpta) will be considered in the
section on non-cognition.
na asmin śarīriṇi sukham asti hṛdaya-śalyāt / Ghoshal (1940: 130).
tad-viruddha-nimittasya yā-upalabdhiḥ prayujyate / nimittayor viruddhtva-abhāve sā vyabhicāriṇī //
yathā na śīta-sparśo ’tra kāṣṭhād iti / Edition in Gnoli 1960: 6; English translation by Gillon and Hayes
(1991 : 10).
In the next paragraph, examples of these types of inference will be considered and schemata will be
provided.
Marie-Hélène Gorisse
121
By the way, these complex inferential forms are not needed in this framework, since they
can be dealt with thanks to transitivity. Indeed, when we know how to get new knowledge
from previous knowledge of the effect, and that we know how to get new knowledge from
previous knowledge of a property incompatible with the target-property, then we know how
to get new knowledge from previous knowledge of what is incompatible with the effect of
this target-property. And Māṇikyanandi explicitly accepts transitivity when he writes:
PM.3.90. The pieces of evidence which arise one after the other should be
included here (in this list).29
2.2.2. Dharmakīrti’s extra type of incompatibility
If we go through the eleven types of non-cognition listed by Dharmakīrti in his NB, then
we notice that he deals with a second type of incompatibility, in which what is at stake
is not the effect, etc. of what is incompatible with the target-property, but rather what is
incompatible with the effect, etc. of it. Strangely enough, Jain philosophers do not even
mention this second type of incompatibility. In this type of form of inference, not two, but
four sub-types depending on the type of evidence are granted by Dharmakīrti, namely what
is incompatible with the target-property itself, what is incompatible with a pervader of
it, what is incompatible with an effect of it and what is incompatible with a cause of it.
Let us only consider the third situation of this list, namely what is incompatible with an
effect of the target-property. This situation is interesting, because in NB.2.37, Dharmakīrti
uses the example “there are no causes of cold whose potency is unimpeded here, because
there is fire,”30 and this example is easily comparable to the one “there is no feeling of cold,
because there is smoke here” he used in NB.2.35 for the effect of what is incompatible
with the target-property. More precisely, let us draw a schemata representing the cause
and effect of heat in the first line, and the cause and effect of cold in the second line. The
elements in bold in this schemata are the ones being considered in NB.2.35:
Causes of fire → heat/fire → smoke
Causes of cold → cold → goose flesh
Whereas NB.2.37 is concerned with the following elements:
Causes of fire → heat/fire → smoke
Causes of cold → cold → goose flesh
By the way, it is interesting to notice that in the second example, there is the restriction
“whose potency is unimpeded.” This means that in a normal case, the presence of a fire in
a room is a sufficient sign for the absence of cold in this room. Another example using this
restriction, is found in NB.2.32 on the occasion of a situation displaying non-cognition as
evidence, namely: “there is no fire (lit., ‘causes of smoke’) whose potency has not been
29
30
paraṃ-parayā saṃbhavat sādhanam atra eva antarbhāvanīyam / Ghoshal (1940: 135).
na iha apratibaddha sāmarthyāni śīta-kāraṇāni santi vahner iti / Desai (1991: 74).
122
Jain Conceptions of Non-Cognition
obstructed, because there is no smoke.”31 This example illustrates the fact that in a normal
case, that is to say in a case in which no external parameter is blocking the spreading of the
smoke, its absence is a sufficient sign for the absence of the fire.
As we explained, Dharmakīrti accepts four sub-types of such incompatible evidence.
The extra two are due, first to the fact that he distinguishes between natural property
(svabhāva) and pervaded property (vyāpta); and second, to the fact that he here accepts
not only the effect, but also the cause, as correct evidence. These two differences are
important ones. Since they are also present in the non-cognition type of evidence; and
since Jain philosophers do not develop the second type of incompatibility, but develop the
non-cognition one, we will investigate their importance when considering non-cognition.
2.3. Non-cognition of [evidence] compatible [with the target-property] in the case of
negation [of the thesis] (aviruddha-anupalabdhiḥ pratiṣedhe)
The third form advocated by Māṇikyanandi is the one used to express inferences of the
absence of a property from the non-cognition of appropriate pieces of evidence. This
is non-cognition properly speaking. According to Māṇikyanandi, there are six types of
evidence that ensure inferential knowledge in this form:
PM.3.78. In the case of negation [of the thesis], there are seven kinds of noncognition of [evidence] compatible [with the target-property], namely [the
evidence can be the target-property] itself, a [property] pervading (vyāpaka)
[it], an effect [of it], a cause [of it], a predecessor [of it], a successor [of it] or
a [property] co-existent [with is].32
Two main differences with the precedent forms are to be noticed. First, only a property
pervading the target-property can function as correct evidence, whereas only a pervaded
property could in the previous forms. The reason for this has been fully developed in 1.2.
The second difference from the previous forms is that “the target-property itself” is
added to the list of correct evidence. Indeed, in reasonings aiming at inferring an absence
from a non-cognition, knowledge is gained through the inference from “I do not know
the presence of the target-property” to “I know that the target-property is absent.” But in
the affirmative, nothing would have been gained through the inference from “I know the
target-property” to “I know the target-property.” As for incompatibility, the situation is more
complex and we need to come back to Dharmakīrti in order to understand what happens.
In Dharmakīrti’s presentation, whereas he is speaking of natural property (svabhāva) in
general for the first type of evidence, he distinguishes between svabhāva and vyāpaka in the
situations involving non-cognition and cognition of incompatible properties. I have therefore
translated svabhāva-viruddha-upalabdhi by “cognition of [a property] incompatible with
[the target-property] itself,” and vyāpaka-viruddha-upalabdhi by “cognition of [a property]
incompatible with [a property] pervading [the target-property].” Now, we have seen that
the two inferential forms involving incompatibility which are granted by Dharmakīrti are,
31
32
na iha pratibaddha-sāmarthyāni dhūma-kāraṇāni santi dhūma-abhāvād iti / Desai (1991: 67).
aviruddha-anupalabdhiḥ pratiṣedhe saptadhā svabhāva-vyāpaka-kārya-kāraṇa-pūrva-uttara-sahacaraanupalambha-bhedād iti / In Ghoshal 1940: 131.
Marie-Hélène Gorisse
123
on one hand, cognition of a [property] pervaded by, etc. a [property] incompatible [with
the target-property] and, on the other hand, cognition of a [property] incompatible with a
[property] pervading, etc. [the target-property]. And when it comes to “[with the targetproperty] itself,” the difference between the two types of incompatibilities vanishes. This
is probably the reason why Dharmakīrti, in NB.2.34 “there is no feeling of cold, because
there is fire/heat,”33 has treated this type of evidence only one time, as an incompatible of
the own nature of the target-property (and not as the own nature of something incompatible
with it). This, in turn, might be what explains that “incompatible with the target-property
itself” is not being considered as correct evidence in the second form by Jain philosophers
either, since they do not consider at all the second type of incompatibility presented by
Dharmakīrti.
Now that we have seen the two differences between this third form and the two previous
ones in the Jain framework, let us consider Dharmakīrti’s conception on the matter. When
it comes to the knowledge of absence of the target-property from the non-cognition of
sufficient evidence for it, Dharmakīrti grants more types of properties that can function
as sufficient inferential evidence than in the two previous forms. Indeed, non-cognition of
the target-property itself, non-cognition of a property pervading it, non-cognition of an
effect of it and non-cognition of a cause of it are all correct evidence to infer its absence.
First, we have already said one word on the non-cognition of a property pervading the
target-property (vyāpaka-anupalabdhi), for which NB.2.33 gives the same example as
PM.3.80, namely “there is no Sissoo here, because there is no tree.”34 Second, we have also
presented non-cognition of the target-property itself. In NB.2.31 the following example is
found “[It is known that] there is no smoke here, because there is non-cognition of that
which meets the conditions for cognition,”35 which is equivalent to the example in PM.3.79
“there is no pot here, because it is not known.”36 Third, concerning causality, something
unexpected happens. More precisely, we are used to reading Dharmakīrti’s writings on the
fact that an effect can be used as good evidence for the presence of its cause. We are less
used to seeing him write that a cause may also be one. And yet, in NB.2.39 and PM.3.82,
the inference of the absence of the target-property from the non-cognition of its cause
(kāraṇa-anupalabdhi) is presented with the following correct example “there is no smoke,
because there is no fire.”37 Second, the inference of the absence of the target-property from
the non-cognition of its effect (kārya-anupalabdhi) is presented in NB.2.32 and PM.3.81
with the following correct example “there are no causes of smoke whose potentials are
unimpeded here, because there is no smoke.”38 If we conceive negations as inverting the
values, how then are we to give an account of the fact that Dharmakīrti accepts the non33
34
35
36
37
38
svabhāva-viruddha-upalabdhir yathā na atra śīta-sparśo vahner iti / In Desai 1991: 70.
na atra śiṃśapā vṛkṣa-abhāvāt iti / In Desai 1991: 69. na asty atra śiṃśapā vṛkṣa-anupalabdheḥ / In
Ghoshal 1940: 132.
na atra dhūma upalabdhi-lakṣaṇa-prāptasya anupalabdhir iti / In Desai 1991: 67.
na asty atra bhūtale ghato ’nupalabdheḥ / In Ghoshal 1940: 132.
na atra vahny agny-abhāvāt iti / In Desai 1991: 75. na asty atra dhūmo ’n-agneḥ / In Ghoshal 1940:
132.
na iha apratibhaddha-sāmarthyāni dhūma-kāraṇāni santi dhūma-abhāvāt iti / In Desai 1991: 67. na
asty atra apratibhaddha-sāmarthyo ’gnir dhūma-anupalabdheḥ / In Ghoshal 1940: 132.
124
Jain Conceptions of Non-Cognition
cognition of the effect of a target-property as good evidence for its absence? Indeed, if only
combinatory purposes are intended, then:
A → B is equivalent to ¬ B → ¬ A39
Therefore:
K (cause x) → K x40 is equivalent to ¬ K x → ¬ K (cause x)
Which, by definition of a cause and of an effect, is equivalent to:
¬ K (effect x) → ¬ K x
It seems therefore suspicious to accept that the non-cognition of the effect of a targetproperty entails the knowledge of its absence, without accepting that the knowledge of
the presence of the cause of a target-property entails the knowledge of its presence, as we
have seen in 2.1. But if we have a closer look at Dharmakīrti’s refusal of the fact that cause
constitutes correct evidence in the affirmative form, we see that the point of disagreement
is elsewhere. Indeed, in his PVsV, Dharmakīrti accepts the following:
PVsV.1.7.1. For in that case, it is only the possibility of the effect’s arising
from the complete cause that is inferred, because there is an inference of the
aptitude of the collected [causes] to produce an effect. And the aptitude is
dependent on nothing more than the totality [of causes], so it is only a virtual
natural property (svabhāva-bhūta) that is inferred.41
In other words, Dharmakīrti accepts that one is legitimised to infer the potentiality of the
effect from the presence of its cause. Because when we deal with future events, we deal with
potential phenomena, not actual ones, since “the beautifully coloured apple that showed
promise of tasting sweet may turn out to have a bitter taste.”42 Therefore, Dharmakīrti
rescued cause as good piece of evidence given appropriate restrictions. What he is saving
in doing so is our ability to make predictions. Indeed, if cause could never be used as
an evidence in order to infer its future effects, no prediction could be made by means of
inference. And since inference and perception are the only two ways to acquire knowledge
according to Buddhist philosophers, and since perception can be of no use in relation with
future events, it would not have been possible for us to make predictions at all. And this, in
turn, would have had bad consequences, especially for Buddhist soteriology. In conclusion,
what Dharmakīrti does not accept is not that the causality relation is not symmetric, but
that it is possible to speak about a future event as if it was an actual one. And since in
the form exemplified by “there is no smoke, because there is no fire,” no knowledge of a
39
40
41
42
“¬ B → ¬ A” is to be read “non B entails non A.”
“K cause x” is to be read “knowledge of the presence of the cause of x.”
tatra hi kevalaṃ samagrāt kāraṇāt kārya-utpatti-saṃbhavo ’numīyate samagrāṇāṃ kārya-utpādanayogyatā-anumānāt / yogyatā ca sāmagrī-mātra-anubandhinī iti svabhāva-bhūta eva anumīyate /
Edition in Gnoli 1960: 6, translation by Gillon and Hayes (1991: 11).
Gillon and Hayes 1991: 69.
Marie-Hélène Gorisse
125
future event is concerned, there is no reason why it should not be accepted as a correct
form of inference. In conclusion, also epistemological considerations and not only formal
ones are here guiding Dharmakīrti’s choices. By the way, a possible explanation for the
fact that Jain philosophers do not mention the second form involving incompatibility is
that they are more concerned with formal considerations and that this is a form which, as
far as the transmission of truth values is concerned, is redundant with the one involving
non-cognition.
2.4. Non-cognition of [evidence] incompatible [with the target-property] in the case
of affirmation [of the thesis] (viruddha-anupalabdhir vidhau)
And finally, the last form advocated by Māṇikyanandi is the one used to infer the presence
of a target-property from the non-cognition of evidence incompatible with it. In such a
challenging form, inference can rely on only three types of evidence, namely the targetproperty itself, its cause and its effect. In Māṇikyanandi’s words:
PM.3.86. In the case of affirmation [of the thesis], there are three kinds of
non-cognition of [evidence] incompatible [with the target-property], namely
the non-cognition of [evidence] incompatible with [the target-property] itself,
an effect [of it], or a cause [of it].43
This last form involves both non-cognition and incompatibility. It is considered that such
a combination of two negations entails an affirmation. In this form, one cannot draw as
many types of correct inferences as in the other forms, because there exist situations in
which “non-non-A” does not equal “A.” This is especially a problem for Jain philosophers,
who considered the present rising of the star Revatī as being incompatible with the present
rising of the star Rohiṇī. Let us take an intuitive example to see the problem: if Tuesday
can be considered as non-Monday, then non-non-Monday may admittedly be Monday, but
it might also be Wednesday, or any day of the week provided it is not Tuesday. Actually,
the only way for “non-non-A” to be the equivalent of “A” is if “non-A” refers to the whole
list of things that are non-A, and not only to one item of this list. To escape this problem,
Māṇikyanandi deals only with predicates that divide the domain into two parts when he
investigate this last form. More precisely, the example presented in PM.3.89 to illustrate the
inference of the presence of a target-property from the absence of something incompatible
with it is the following one “all things possess several aspects, because something having
only one aspect is never found.”44 By the way, this example is an important one, because
this inference is meant to establish the Jain doctrine of non-one-sidedness (anekāntavāda).
Second, the example presented in PM.3.87 to illustrate the inference of the presence of
a target-property from the absence of the effect of something incompatible with it is the
following one “as for example, some disease exists in this animal, because the actions of a
healthy body are not found.”45 And third, the example presented in PM.3.88 to illustrate the
43
44
45
viruddha-anupalabdhiḥ vidhau tredhā viruddha-kārya-kāraṇa-svabhāva-anupalabdhi-bhedāt /
Ghoshal (1940: 133).
anekānta-ātmakaṃ vastv-ekānta-svarūpa-anupalabdheḥ / Ghoshal (1940: 134).
yathā asmin prāṇini vyādhi-viśeṣa asti nirāmaya-ceṣṭa-anupalabdheḥ / Ibid.
126
Jain Conceptions of Non-Cognition
inference of the presence of a target-property from the absence of the cause of something
incompatible with it is the following one “there is pain in this individual, because he has no
connection with his dear ones.”46 In these three examples, the incompatible properties are,
respectively, unique/non-unique (ekāntasvarūpa/anekāntasvarūpa), healthy/non-healthy
(āmaya/nirāmaya) and happiness/unhappiness (sukham/duḥkham). That is to say, couples
of contraries whose contrary nature is reflected already in the grammatical formation of
the words. In this way, there is no third option. Someone is either happy or unhappy. In
consequence, in this framework not-unhappy is happy, and nothing else.
As for Dharmakīrti, we have seen that his aim in PVsV is to give an account on the
means to know absences by means of inference. From this, it is only natural that this form
leading to an affirmative conclusion did not receive his attention in this passage.
3. Conclusion
After having compared the way non-cognition is involved in the inferential process in
the Buddhist and in the Jain traditions, I would like to summarize their main points of
divergence, as well as to address the consequences of these divergences.
First, Dharmakīrti considers non-cognition as a type of evidence employable to infer
the absence of a normally perceptible entity, and Akalaṅka extends this conception to
non-perceptible entities as well. In distinction to this, Māṇikyanandi and his commentators
consider non-cognition mainly as a negative premise in the stated form of an inference,
without being specifically concerned with the possibility to know that something is absent
thanks to inference. In this new dynamic, non-cognition is introduced only in the section on
inference for others, whereas Dharmakīrti introduced it already in the section on inference
for oneself. In other words, the concern on non-cognition in Jainism lies more in the fact
that its linguistic counterpart is a negation, that is to say as a linguistic device usable to
reverse the truth value of a sentence.
The second line of divergence concerns the forms of inference involving non-cognition
on which each tradition focuses. Only Dharmakīrti distinguishes between the cognition
of a property incompatible with the effect of the target-property and the cognition of the
effect of a property incompatible with the target-property. It is interesting to notice that the
first type of incompatibility is strictly equivalent to non-cognition as far as the transmission
of truth values is concerned, which might be one explanation of the Jain disinterest for it.
Conversely, only Māṇikyanandi is interested in the possibility of inferring the presence of
the target-property from the non-cognition of a property incompatible with it since, again,
Dharmakīrti here focuses on the possibility of inferring an absence.
The third important line of divergence between the two frameworks concerns the types
of evidence that are active in inferences involving non-cognition. Whereas 11 forms of
inference involving non-cognition are admitted in Dharmakīrti’s Nyāyabindu, 16 forms are
admitted in Māṇikyanandi’s Parīkṣāmukham. First, this is due to the fact that Dharmakīrti
rejects succession and coexistence as inferential evidence, therefore he also rejects noncognition of succession and non-cognition of coexistence. Second, when it comes to the
cognition of evidence incompatible with the target-property, which is considered by both
46
asti atra dehini duḥkham iṣṭa-saṃyoga-abhāvāt / Ibid.
Marie-Hélène Gorisse
127
traditions as a type of non-cognition, Māṇikyanandi accepts to infer the absence of a given
target-property, both from an effect incompatible with it and from a cause incompatible
with it. As expected, Dharmakīrti accepts to infer the absence of a given target-property
only from an effect incompatible with it. What is less expected is that he considers both
the non-cognition of the cause and the non-cognition of the effect of a target-property as
correct evidence for its absence. My hypothesis is that this is due to the fact that when the
absence of a target-property is known thanks to the non-cognition of its effect or of its cause,
no discourse on future event is involved; and that this was Dharmakīrti’s main objection
concerning cause as correct inferential evidence. By the way, this is also probably one good
explanation for the fact that the material implication used by contemporary logicians is not
a good candidate to express in a formal way the relationship between the evidence-property
and the target-property. Indeed, material implication has several properties, amongst which
is symmetry, and this epistemological concern that no discourse should be made on future
events prevents such a symmetry. Actually, in an attempt towards a formal representation of
these theories, as many logical connectors as there are types of evidence would be needed.
I would like to stress the fact that these considerations are really at the junction between
logic (recognition of certain patterns, and rules describing them, such as transitivity, types
and functions of negations), epistemology (what a person can know) and argumentation
(how to convince a given interlocutor).
In conclusion, this presentation is conceived as a first step for a more thorough analysis
of the divergences between Jain and Buddhist conceptions of inference after Dharmakīrti.
What would be especially useful, in relation to the conceptions of non-cognition in this
line of analysis, would be to see if later Buddhist philosophers adopt a classification
into different forms of inference; whether they drop the second type of incompatibility
(viruddha-upalabdhi); whether they gain an interest in the non-cognition of properties
incompatible with one another; or whether they state a rule of transitivity.
References and abbreviations
Balcerowicz 2005 P. Balcerowicz, Akalaṅka und die buddhistische Tradition: von der
Nichtwahrnehmung von Unsichtbarem (adṛśyānupalabdhi) zur Allwissenheit. Wiener Zeitschrift für die Kunde Südasiens 49 (2005) 151–226.
Bhattacharya 1967 See PNT.
Desai 1991 See NB.
Ghoshal 1940 See PM.
Gillon and Hayes 1991 B. Gillon and R. Hayes, Introduction to Dharmakīrti’s theory of
inference as presented in Pramāṇavārttikasvopajñavṛtti 1–10. Journal of Indian
Philosophy 19 (1991) 1–73.
Gnoli 1960 See PV and PVsV.
Gorisse 2015 M.H. Gorisse, The taste of the mango: a Jaina-Buddhist controversy on
evidence. International Journal of Jaina Studies 11/3 (2015) 1–19.
Gorisse 2017 M.H. Gorisse, Logic in the tradition of Prabhācandra. In: Oxford Handbook
of Indian philosophy, ed. J. Ganeri. Oxford 2017, 486–506.
128
Jain Conceptions of Non-Cognition
Katsura 1992 S. Katsura, Dignāga and Dharmakīrti on adarśanamātram and anupalabdhi.
Asiatische Studien/Études asiatiques 47 (1992) 222–231.
Kellner 2003 B. Kellner, Integrating negative knowledge into pramāṇa theory: the development of the dṛśyânupalabdhi in Dharmakīrti’s earlier works. Journal of Indian
Philosophy 31 (2003) 121–159.
Jain 1939 See LT.
LT Akalaṅka’s Laghīyastrayam: Akalankagranthatrayam, ed. M.K. Shastri. Ahmedabad
1939.
NB Dharmakīrti’s Nyāyabindu, ed. N.R. Desai. Ahmedabad 1991.
PKM Prabhācandra’s Prameyakamalamārtaṇḍa, ed. M.K. Shastri. Delhi 1912/1990.
PM Māṇikyanandin’s Parīkṣāmukham, ed. and tr. S.C. Ghoshal. Lucknow 1940.
PNT Vādi Devasūri’s Pramāṇanayatattvālokālaṃkāra, ed. and tr. H.S. Bhattacharya.
Bombay 1967.
PV and PVsV Dharmakīrti’s Pramāṇavārttika and Pramāṇavārttikasvavṛtti. Chapter 1,
ed. R. Gnoli. Roma 1960.
Shah 1967 N.J. Shah, Akalaṅka’s criticism of Dharmakīrti philosophy. A study. Ahmedabad 1967.
Shastri 1912 See PKM.
Universals, Demons’ Pots, and Demons’ Permanent Pots:
Phya pa Chos kyi seng ge on Unestablished Subjects in
Arguments by Consequence
by
Pascale Hugon
Introduction
According to Dharmakīrti, prasaṅga arguments – i.e., arguments that rely on a logical reason
that is merely accepted by the opponent to draw an unacceptable conclusion – are legitimate
means of proof provided that they are indicative of a correct, i.e., triply characterized,
logical reason on which the reverse form of the prasaṅga (the prasaṅgaviparyaya) is
based.1 The relevant passage of the Pramāṇaviniścaya (hereafter: PVin) has been subject to
conflicting interpretations by Indian commentators who notably disagree about the nature
of the example that Dharmakīrti gives in this context – namely, whether it is meant to
illustrate a prasaṅga argument or its reverse form – and about the status to ascribe to
the subject in this example – i.e., whether the subject “universal” is a real universal (as
upheld by the intended addressee of the argument) or a non-entity (as accepted by the
Buddhist proponent).2 The second point is crucial for the acceptance of the prasaṅga as a
probative argument insofar as the validity of the reverse form would be threatened by the
non-existence of the subject. Indeed, the non-existence of the subject would prevent the
fulfillment of a required characteristic of the logical reason, being a property of the subject
(pakṣadharmatva) – a fallacy known as that of the “unestablished basis” (āśrayāsiddha).
Whereas this fallacy allows the Buddhists to discard proofs by opponents upholding,
for instance, the reality of space (ākāśa), primordial nature (pradhāna) or universals
(sāmānya), it also threatens their own arguments aimed at the refutation of such pseudoentities. The logical reason in the Buddhist proof would indeed also fail to satisfy the
requirement of pakṣadharmatva.3 The status of the subject poses a similar problem in the
1
2
3
The work on this paper has been generously supported by the Austrian Science Fund (FWF) in the
context of the project P23422-G15 “Early bKa’ gdams pa scholasticism.” I wish to thank my colleague
Toshikazu Watanabe for stimulating discussions on the topic, and Katharine Apostle for improving my
English.
Dharmakīrti’s short mention of such arguments in PV 4.12 (on which see Tillemans 2000: 21–24)
is supplemented by a longer discussion in the PVin (PVin 3 4,4–6,12; the references to the Tibetan
translation of this discussion in the sDe dge, Peking and sNar Thang bstan ’gyur are: D188a5–189a5;
P286a5–287a5; N300a6–301b1).
For an in-depth study see Iwata 1993. The main points appear in a summarized form in English in Iwata
1997a.
In NB 3.65 dharmyasiddha (“non-establishment of the property-possessor”) is listed as one type of
fallacy of the logical reason. An additional issue with such proofs is that the Buddhist proponent could
be accused of annulling his own thesis when negating the subject.
Birgit Kellner et al., eds., Reverberations of Dharmakīrti’s Philosophy: Proceedings of the Fifth International Dharmakīrti
Conference Heidelberg, August 26 to 30, 2014. Vienna 2020, pp. 129–153.
130
Universals, Demons’ Pots, and Demons’ Permanent Pots
Madhyamaka proof establishing the emptiness of all things,4 and also impacts arguments
for the establishment of the negative entailment of logical reasons, typically in the proof
of momentariness. Various answers to the problem of the unestablished basis have been
developed in the Buddhist tradition.5 The question of the subject in the context of prasaṅga
arguments was taken up in particular by Dharmottara and Prajñākaragupta.6
The present paper deals with the views on this topic of the Tibetan logician Phya pa Chos
kyi seng ge (1109–1169).7 Phya pa, as can be expected, was influenced by the discussions
of Indian thinkers but he nevertheless came up with his own creative method to deal with
this issue. The examination of his views also sheds light on later Tibetan developments that
obviously heavily borrowed from his works. In what follows I examine Phya pa’s commentary on the prasaṅga passage of Dharmakīrti’s Pramāṇaviniścaya (’Od zer 149a5–150b3)
to establish his take on the question of the nature of Dharmakīrti’s example and on the
status of the subject. I then situate his interpretation within the framework of his overall
theory of arguments by consequence (thal ’gyur) – of which we find two parallel versions
in his commentary on the PVin (’Od zer 145a2–149a5) (preceding his commentary on the
prasaṅga passage) and in his epistemological summa (Mun sel 83a1–95b5) – and relate
this issue to discussions relevant to the problem of the “unestablished basis” (āśrayāsiddha)
in the broader context of inference.
Before turning to Phya pa’s view, it will be useful to provide some references to the
solutions developed by Indian thinkers to the problem of the unestablished basis.
Some methods developed by Indian thinkers to solve problems caused
by unestablished subjects
I refer the reader to Watanabe’s paper in this volume for a detailed discussion of the solutions
developed by Dignāga and Dharmakīrti and their link to the context of prasaṅga, as well as
for further references to primary and secondary literature. I list below (adopting Watanabe’s
terminology) some solutions that will be mentioned in my discussion of Phya pa’s position:
The ‘Method of Conceptual Subject:’ the actual or intended subject is the conceptual
representation of the entity in question. This method is adopted by Dharmakīrti in the
passage of PVin 3 dealing with the refutation of the Sāṃkhya’s primordial nature (to
4
5
6
7
On Madhyamaka proofs see, for instance, Tillemans 1982 and 1984 and Iwata 1999. Kobayashi 1987
(in Japanese) exposes mainly Kamalaśīla’s view in the Madhyamakāloka (the relevant section of the
Madhyamakāloka is translated into English in Keira 2004).
For a panorama see Iwata 1999 and Watanabe in this volume. For the views of specific authors see also
Tillemans 1999, chap. 8, Tillemans and Lopez 1998, and the articles mentioned in n. 4.
In addition to Iwata 1993 and 1997a, see Watanabe’s paper in this volume, which discusses the various
solutions to the problem of āśrayāsiddha developed by Dignāga and Dharmakīrti and their connection
with prasaṅga arguments. I summarize some of them below.
Since the surfacing of eighteen of his works, the views of this thinker have been the subject of an
increasing number of publications, so that a formal introduction may no longer be needed. For a
recapitulation of the information available on his life and works the reader may refer to my “Compiled
information on the life and works of Phya pa Chos kyi seng ge and bibliographical ressources,” which
is available online at https://www.oeaw.ac.at/ikga/forschung/tibetologie/materialien/materials-for-thestudy-of-phya-pa-chos-kyi-seng-ge-1109-1169/.
Pascale Hugon
131
which I will refer as the “pradhāna passage”).8 The core idea can be traced to Dignāga’s
discussion on the refutation of pradhāna in the Nyāyamukha, where he describes the
subject as “imagined” (kalpita). Dharmakīrti makes clear in what way one can prove the
non-existence of pradhāna without negating the subject in a way that would make the proof
liable to a fallacy: what is being proven is not that the subject qua concept is inexistent, but
that this concept does not have a real substratum.9 The concept itself is well established for
both debaters and not liable to refutation.10 It is not actually the concept that is targeted
by the discussion but the substratum (the entity) which, unlike the concept, is apt to fulfill
some human goal.
The ‘Method of Paraphrase:’ the reasoning is reformulated in such a way that the
actual subject is acceptable to the proponent. This idea, found in PV 4.136–148 in the
commentary on the word svadharmiṇi in Dignāga’s definition of the thesis in PS 3.2,
is linked to the distinction between nominal subject (kevaladharmin) and actual subject
(svadharmin) when discussing the fallacy of the non-established basis as a fallacy of the
thesis.11 I will refer to this context as the “svadharmin passage.”
The ‘Method of Simple Negation:’ the attribution of negative properties does not
require the subject to exist, so no fault occurs as long as the properties attributed to the
non-existent subject are simple negations (prasajyapratiṣedha). This method was adopted
by Dharmottara in the context of prasaṅga and by Devendrabuddhi and Śākyabuddhi when
discussing the svadharmin passage. Prajñākaragupta criticized this position, arguing that
it does not resolve the fallacy of the unestablished subject because an inexistent subject
cannot be qualified by any property whatsoever.
Iwata (1999) discerns two other methods found in the Madhyamaka works of Śāntarakṣita and Kamalaśīla along with the Method of the Conceptual Subject and the Method
of Simple Negation; they consist in taking the subject to be existent (i) as a convention
(vyavahāra) and (ii) as an appearance. In Śāntarakṣita’s Madhyamakālaṅkāra(/vṛtti) and in
Kamalaśīla’s Madhyamakālaṅkārapañjikā one finds the idea that the verbal designations
of the elements of the inference are based on a subject that appears in the cognition of both
disputants – for instance for the subject “sound,” the appearance in the auditive cognition
upon hearing a sound.12 In these texts the authors restrict the notion of “appearance” to the
context of perception and do not deal with non-existent subjects such as pradhāna. The
issue is mainly to avoid that the subject becomes unestablished when it is proven to be
essenceless.
8
9
10
11
12
PVin 3 67,4–70,6; Tib. P306a6, D208b3. This passage is parallel to PVSV 105,15–107,14. In both texts
it is part of the anupalabdhi section. Verses PVin 3.53–54 = PV 1.205–206 are, however, frequently
cited in discussions pertaining to the apoha theory.
Cf. PVin 3.54 = PV 1.206 and the auto-commentary. The property to be ascertained is thus not bhāva or
abhāva but bhāvopādānatva/bhāvānupādānatva (cf. PVin 3 68,6–7: kim ayam pradhānaśabdapratibhāsy
artho bhāvopādāno na veti).
Cf. PVin 3 68,7–8: tasya bhāvānupādānatve sādhye sa ca pratyātmavedyatvād apratikṣepārho ’rtho
dharmī and 69,5: na tu punar atrāyam eva śabdavikalpapratibhāsy artho ’pahnūyate.
See Tillemans 2000: 194–198, Tillemans 1999: chap. 8, and Tillemans and Lopez 1998.
See Iwata 1999: 165.
132
Universals, Demons’ Pots, and Demons’ Permanent Pots
Kamalaśīla addresses the issue of the inexistent subject in his Madhyamakāloka. In this
work he mainly invokes the Method of Simple Negation to deal with such subjects.13 But he
also draws a parallel between the previous idea of positing as the subject a commonly shared
appearance and the refutation of pradhāna in which one posits as a subject “something
existing in the mind” (blo la yod pa nyid).14 Actually, the idea that the subject can be a
mental appearance is well represented in the pradhāna passage of the PVin. Notably, in his
prose commentary to PVin 3.53–54 Dharmakīrti reformulates the notion of the concept qua
“object of words” (śabdārtha) in terms of “the object that appears to conceptual thought,”
an appearance that emerges upon hearing a given word.15 This method is thus akin to
the Method of Conceptual Subject but emphasizes the mental appearance of the concept
rather than the nature of conceptual construction. I will refer to it in terms of ‘Method
of Conceptual Appearance.’ As mentioned above, Dharmakīrti clearly indicated that the
subject in this form is established for both disputants and not liable to refutation. Among
Madhyamaka authors, Jñānagarbha similarly points out that appearances are not negated
insofar as they are established to be experienced.16
1. Phya pa on the prasaṅga passage of the Pramāṇaviniścaya
1.1. Phya pa’s interpretation of the example in the prasaṅga passage of the Pramāṇaviniścaya
In the prasaṅga passage of the PVin, Dharmakīrti introduces with the words “for example”
(yathā) an argument against Naiyāyika and Vaiśeṣika opponents who accept the reality of
universals. This argument is complex enough that one cannot avoid, for its very translation,
relying on a commentarial interpretation.17 My tentative translation follows here the understanding of the role of various portions of the Sanskrit sentence by Dharmottara, an
interpretation which is also adopted by Phya pa.18
13
14
15
16
17
18
See Iwata 1999: 167–168. The relevant section of the Madhyamakāloka is translated in Keira 2004:
120–152.
The latter view is presented in the form of an objection by an Alīkākāravādin, but is accepted by
Kamalaśīla. See the translation and notes in Keira 2004: 142–145 (objection) and 146–148 (reply).
See PVin 3 67,12–68,1 (ad 3.53): anādikālavāsanāprabhavavikalpapratibhāsinam arthaṃ (Tib. rtog pa
la snang ba’i don); PVin 3 69,5: śabdavikalpapratibhāsy artho (Tib. sgra’i rnam par rtog pa la snang
ba’i don); PVin 3 68,6–7: pradhānaśabdapratibhāsy artho (Tib. gtso bo’i sgra las snang ba’i don).
See, for instance, the passage of the Satyadvayavibhaṅgakārikā and Satyadvayavibhaṅgavṛtti translated
in Keira 2004: 35–37.
A degree of interpretation is already present in the Tibetan translation by rNgog Blo ldan shes rab and
Parahitabhadra, which was the reference for Tibetan interpreters who relied on the Tibetan version of
the PVin. In this regard one can note that the Tibetan version known to Phya pa, which one can partially
reconstruct from his direct quotes of the text, occasionally differs from the version preserved in the
canonical collections. The absence of corresponding direct quotes by rNgog Blo ldan shes rab leaves
open the question of the original translation and whether it was Phya pa or the editors of the canon (or
maybe both) who adopted a slightly modified version of rNgog Blo ldan shes rab’s translation.
Translations of the prasaṅga passage of the PVin can be found in Tani 1987 (in English) and Iwata
1997b (in German). See also Watanabe’s translation and his discussion in this volume. For the details
of Dharmottara’s interpretation see Iwata 1993: 50 and Iwata 1997a: 429–430. I will present the details
of Phya pa’s interpretation in a forthcoming paper.
Pascale Hugon
133
a→
A single thing that occurs in many←a , b→ because it is devoid of another
essential property which is not characterized [as] mixed (or: which is not
separate, due to being mixed) with a single instance determined in location,
time and status,←b c→ cannot be connected with something else that differs in
location, etc.←c
(PVin 3 4,4–7: b→ deśakālāvasthāviśeṣaniyataikadravyasaṃsargāvyavacchinnasvabhāvāntaravirahād ←b a→ anekavṛtter ekasya←a c→ na deśādiviśeṣavatānyena yogaḥ←c )
Phya pa notably follows the broad lines of Dharmottara’s interpretation on the following
points:
1. This example is taken to illustrate the reverse form of a prasaṅga (i.e., a prasaṅgaviparyaya);19
2. The portion marked as “a→ …←a ” represents the subject, that marked as “b→ …←b ”
represents the logical reason of the prasaṅgaviparyaya, and that marked as “c→ …←c ”
the property to be proven of the prasaṅgaviparyaya;
3. The type of the logical reason of the prasaṅgaviparyaya consists in the non-apprehension of the pervader (vyāpakānupalabdhi);
4. The type of the logical reason of the prasaṅga is an essential property (svabhāva).
When commenting on this passage of the PVin, Dharmottara and Phya pa also similarly
reformulate the subject, logical reason and property to be proven (respectively, the derived
conclusion), as indicated in the schematic representation provided below.20 According to
them, the above passage should be understood as the prasaṅgaviparyaya:
A universal does not occur in many, because it is not multiple.
19
20
Iwata (1993: 50) notes that this interpretation is shared by Jñānaśrībhadra, Bu ston and rGyal tshab Dar
ma rin chen. This interpretation qualifies as counter-intuitive insofar as the topic of discussion in this
passage of the PVin is prasaṅga. One would thus expect the example introduced by the expression “for
example” to illustrate a prasaṅga, not a prasaṅgaviparyaya. The option that Dharmakīrti exemplifies
a prasaṅga is considered by Prajñākaragupta and, reportedly, by Vinītadeva and Śāntabhadra. The
ascription to these two authors of alternative interpretations (which are discussed by Dharmottara)
of Dharmakīrti’s example as a prasaṅga is made by Bu ston (Iwata 1993: 51) and found also in an
anonymous interlinear note in the manuscript of Zhang Thang sag pa’s dBu ma tshig gsal gyi ti ka
(Yoshimizu and Nemoto 2013: 58). Phya pa ascribes the first alternative interpretation to Vinītadeva,
the second to Prajñākaragupta.
Dharmottara’s reformulation of the prasaṅgaviparyaya is as schematized in Iwata 1993: 42. The
reformulation of its logical reason as na cānekam is found in PVinṬ-Skt 7a5. In PVinṬ-Skt 6b3
Dharmottara speaks of the logical reason and the property to be proven in terms of “the absence
of the pervader ‘multiplicity’” (vyāpakanānātvābhāva) and the “absence of the pervaded property
‘occurrence in many’” (vyāpyasyānekavṛttitvasyābhāva). For Dharmottara’s formulation of the prasaṅga,
see PVinṬ-Skt 6a1: sāmānyāder anekavṛttitva iṣṭe ’nekatvam aniṣṭam āsaṃjyate and PVinṬ-Skt 7a5:
yad anekadeśādivṛtti tad anekaṃ tathā ca sāmānyam ity anekaṃ syān. For Phya pa’s reformulation see
’Od zer 149a6–7: spyi gcig gsal ba du ma dang ’brel par kun brtags pas du mar thal zhes bsgrub pa
thal bar bsgrub pa’o // des rang rgyud gang ’phen zhe na / du mas stong pas du ma dang ’brel pas
stong zhes pa khyab byed mi dmigs pa ’phen te /
134
Universals, Demons’ Pots, and Demons’ Permanent Pots
which is induced by the prasaṅga (not stated by Dharmakīrti):
Because (according to you) a universal occurs in many, it follows that it is
multiple.
prasaṅga
Subject
Reason
universal
occurrence
many
multiplicity
in
Dharmottara
Phya pa
sāmānyādi
anekavṛttitva
spyi cig
gsal ba du ma dang ’brel pa
Derived
anekatva
property
Dharmakīrti’s example (= prasaṅgaviparyaya)
Subject
universal
sāmānya
na…aneka = anekatvābhāReason
non-multiplicity
vamātra, nānātvābhāva
Property to non-occurrence
*na anekavṛttitva, anekabe proven
in many
vṛttitvasyābhāva
du ma
spyi cig
du mas stong pa
du ma dang ’brel pas stong, gsal
ba du ma dang ma ’brel ba
1.2 Phya pa on the subject “universal” in Dharmakīrti’s example
A key issue in Dharmakīrti’s example is the nature of the subject “universal.” Indeed,
the Buddhist proponent does not accept the reality of universals and the non-existence
of the subject would threaten the validity of the logical reason in the reverse form: the
fault of “non-establishment of the basis” (āśrayāsiddha) could be invoked. The case under
consideration is typical of the drawbacks of the fallacy of the unestablished basis, which
Buddhist scholars have attempted to palliate through various methods.
In the prasaṅga passage of the PVin the question of the subject arises indirectly in an
objection pointing out that the logical reason of the prasaṅga is unestablished (asiddha),
i.e., it does not satisfy the first characteristic of a correct reason – being a property of
the subject (pakṣadharmatva). This, is, according to Dharmottara, because the Buddhist
proponent does not accept that “a universal occurs in many” insofar as he does not accept
that universals exist in the first place.21 Although here the non-establishment of the logical
reason is a consequence of the non-existence of the subject, the argument appears to be
hinting at a more general issue: in any proper prasaṅga, whether the subject is accepted to be
existent by both debaters or not, the qualification of the subject by the logical reason is only
accepted by the opponent and corresponds to the opposite of what the proponent wishes to
establish.22 Accordingly, in my understanding, Dharmakīrti’s answer does not touch on the
21
22
See PVin 3 5,1–6,1: nanu tathāpy asiddhir hetoḥ… and PVinṬ-Skt 8a6: bauddho hi sāmānyābhāvavāde
sthitaḥ prasaṅgasādhanam āha / tadāśrayāsiddher asiddham anekavṛttitvaṃ nāma hetuḥ /
For instance, in the proof of the presence of fire on a hill where smoke is perceived via the prasaṅga “if
there were (as you claim) no fire on the hill, there would be no smoke,” the status of the ‘hill’ is not
Pascale Hugon
135
way to deal with an inexistent subject, but first of all invokes the indeterminacy that should
prevail in an inquiry pertaining to a given subject. In the given argument, the “universal”
should not be taken either as “the Buddhists’ universal” or “the Naiyāyika/Vaiśeṣika’s
universal.” Further, Dharmakīrti points to the hypothetical nature of the pakṣadharmatva
in the prasaṅga – it is not something that is required to be established by a valid cognition
as it is in a direct proof.
The discussion in this context revolves solely around the prasaṅga.23 The question of
the subject in the reverse form is not addressed by Dharmakīrti, thus leaving the interpreters
free to opt for their preferred scenario on this point. Dharmottara applied the Method of
Simple Negation in this context (Iwata 1997a: 430). While Phya pa follows the major lines of
Dharmottara’s interpretation of the prasaṅga passage and, like Dharmottara, identifies the
logical reason and the property to be proven in the prasaṅgaviparyaya as simple negations
(using the formulation “void of…”), he does not invoke this as a solution to the potential
problem of the unestablished subject. His solution is not influenced by Prajñākaragupta
either.24
Phya pa’s solution, when commenting on the prasaṅga passage, is to appeal conjointly
to the Method of Conceptual Subject and the Method of Conceptual Appearance, both of
which find support in the pradhāna passage of the PVin.
The Method of Conceptual Appearance stands out notably in his commentary on the
passage of the PVin referred to above discussing a potential non-establishment of the
logical reason:
When one investigates the nature of things, at the time of an initial examination
of philosophical tenets, one has not [yet] accepted the Buddhist philosophical
tenets and one does not accept the non-Buddhist tenets. Therefore one does
not posit as a basis a universal that is either an entity or a non-entity. The
mental appearance is the basis. Therefore, even though [the universal] is not
23
24
controversial, but the “absence of fire on the hill” is not established for the proponent, who precisely
aims at establishing the opposite.
Obviously it was a concern to ascribe legitimacy to an argument that had the same form as an inferencefor-others but whose logical reason did not satisfy the requirements of the former.
While it is doubtful that Phya pa would have been well acquainted with Prajñākaragupta’s commentary
on the PV on a first hand basis, he might have known about his position via rNgog Blo ldan shes rab’s
commentarial work on the Pramāṇavārttika cum Pramāṇavārttikālaṅkāra or rNgog Blo ldan shes
rab’s discussion of prasaṅga in his larger PVin commentary (both works are currently not available
to us). rNgog Blo ldan shes rab’s explanation of the difficult points of the PVin (dKa’ gnas) does
not offer a word commentary on the prasaṅga passage, but it explicitly enjoins the reader to follow
Dharmottara’s interpretation and reject all others (dKa’ gnas 393). It is thus certain that rNgog Blo ldan
shes rab’s position did not match that of Prajñākaragupta. In his 1997a paper (433–435) Iwata evokes
the possibility that Prajñākaragupta influenced the Tibetan classification of arguments by way of a
consequence, in which a distinction is made between arguments in which the subject is non-established
– i.e., those that are “non-probative” – and arguments in which it is established, i.e., that qualify as
probative. This suggestion, however, must be rejected. The classification in question, whether that made
by gTsang nag pa (considered in Iwata’s paper) or Phya pa (see below § 2.1), does not distinguish
between two ways of dealing with the subject in a single given argument in the way Prajñākaragupta
does, but rather distinguishes those arguments in which the subject is established by a valid cognition
(for instance the subject “hill” when refuting the absence of fire) from those in which it is not (gTsang
nag pa does not give an example – see § 2.3 for Phya pa’s example).
136
Universals, Demons’ Pots, and Demons’ Permanent Pots
established as an entity for the Buddhist and is not established as a non-entity
for the non-Buddhist, this mere mental appearance is established for both.25
As for the Method of Conceptual Subject, it is visible in Phya pa’s reference to the subject as
the “object of a word” (sgra don, Skt. śabdārtha), i.e., the concept,26 on the basis of which
the characteristic of the logical reason being a property of the subject (pakṣadharmatva) is
established:
Our [i.e., the Buddhist’s] position is correct: That the pakṣadharmatva of
[the logical reason] ‘void of multiplicity’ for the object of the word [as it is]
accepted is established by reflexive awareness [is explained by Dharmakīrti]
in the passage “But the singular thing…” (PVin 6,6–7: ekasya tu…).27
Phya pa’s commentary also refers the subject “the singular thing” as an object of the
word for which the property of “appearing as singular” can be established by reflexive
awareness.28
Using this joint method Phya pa is able to guarantee the legitimacy of Dharmakīrti’s
example in spite of the problematic status of the subject “universal:” whether universals
exist in reality or not, debaters who use the term sāmānya have a conceptual representation,
the appearance of which can be established by reflexive awareness and cannot be refuted.
The Method of Conceptual Appearance and/or Conceptual Subject is successful when
dealing with cases where the author wants to avoid the fallacy of the unestablished basis. But
such a method appears to have far-reaching consequences. Indeed, if all verbal expressions
generate conceptual representations and just any conceptual representation may qualify as
being “established” insofar as its appearance can be established by reflexive awareness, there
should be no unestablished bases, and hence no occasion for the fallacy of āśrayāsiddha to
arise.29 However, as I discuss in what follows, Phya pa preserves this fallacy in the context
of inference-for-oneself, and the requirement that the subject must be established also
plays a role in the identification of probative consequences (i.e., instances of arguments by
consequence whose features correspond to Dharmakīrti’s understanding of a legitimate
prasaṅga).
25
26
27
28
29
’Od zer 150b2–3: shes bya’i gnas lugs dpyad nas (read: na) grub mtha’ gzod tshol ba’i dus su sangs
rgyas pa’i grub mtha khas blangs pa myed la / mu stegs pa’i grub mtha khas myi len pas spyi dngos po
dang dngos med gzhir ma bkod de blo snang gzhi’ yin pas dngos por sangs rgyas pa la ma grub la
dngos myed du mu stegs pa la ma grub kyang blo snang tsam de gnyi’ ga la grub po zhes pa’o //
One may note that Phya pa does not use the term “concept” (don spyi) in this context. Cf. n. 65.
’Od zer 150a9: rang gi lugs la ’thad pa ni khas blangs pa’i sgra don la du mas stong pa’i phyogs chos
rang rig pa’i shugs la grub pa […] gcig la ni zhes pa […].
’Od zer 149a7: chos can ni du ma la yod pa’i gcig po ste gcig du snang pa’i chos sgra don la yod
par rang rig pa’i shugs la grub pa’o / The establishment of the properties of “unicity” and “void of
multiplicity” by reflexive awareness evoked in these passages stands out in a more detailed way in the
portion of Phya pa’s excursus on arguments by consequence dealt with in § 3.
This is not to say that other fallacies would not arise, allowing the Buddhist to criticize his opponent’s
proof.
Pascale Hugon
137
2. Unestablished subjects in Phya pa’s theory of argumentation by consequence
2.1 Phya pa’s classification of arguments by consequence
Phya pa’s theory of arguments by consequence includes a much broader range of arguments
than the type of prasaṅga taken into account by Dharmakīrti as a legitimate argument.
According to Phya pa, arguments by consequence include any argument of the form
“Because S is P, it follows that it is Q.”30 A distinction is then made between genuine (rnal
ma) arguments by consequence and fallacious ones (ltar snang). Fallacious consequences
are to be understood as “non-pertinent.” They are arguments to which the opponent is able
to retort that he does not accept the premises – he does not accept that S is P and/or that
everything that is P is Q, in which case he is not bound to accept the conclusion “therefore
S is Q” – or is able to retort that he accepts the conclusion (S is Q), which the proponent
intended to be ‘absurd’ or at least unacceptable for the opponent. If the opponent is unable
to retort, the consequence is qualified as “genuine.” Genuine consequences are then divided
into probative and non-probative consequences. Only the first type, in which Dharmakīrti’s
prasaṅga is to be included, amounts to an inference-for-others that proves something. The
logical reason of its reverse form satisfies the triple characteristic (trairūpya) – put shortly,
the qualification of the subject (pakṣadharmatva) and the pervasion (vyāpti) are ascertained
by a valid cognition.31
2.2 The definition of a probative consequence
Phya pa gives as a definition of a probative consequence:
The indication of a logical reason pertaining to a subject in the context such that
the pervasion is determined by a valid cognition and the explicit conclusion is
eliminated by a valid cognition.32
This definition brings two requirements to the fore. For a consequence of the form “Because
S is P, it follows that it is Q” to be probative,
(a) the pervasion of the consequence (P is pervaded by Q) must be established by a valid
cognition;
(b) the derived conclusion (S is Q) must be eliminated by a valid cognition.33
30
31
32
33
For an overview of Phya pa’s theory of argumentation by consequence see Hugon 2013. As I make clear
in this paper (675–676), the form “Because S is P, it follows that it is Q” is a reference to a consequence
statement, which should not be confused with the actual consequence statement, that typically should
have the form “all that is P is Q, S is P.”
This classification of consequences and the various issues linked with it are analyzed in details in Hugon
2016.
Mun sel 90a9–b1, ’Od zer 146b6: skabs su bab pa’i chos can la dngos kyi dam bca’ (Mun sel sgrub bya,
to be emended to dam bca’) la tshad mas bsal pa dang / khyab pa tshad mas nges pa’i rtags ston pa’o //
Note that this definition does not include any explicit requirement pertaining to the qualification of the
subject by the logical reason in the consequence (i.e., to the premise “S is P”). Phya pa disagrees in this
regard with some of his predecessors. See below n. 44. But according to Phya pa’s gloss of the terms of
138
Universals, Demons’ Pots, and Demons’ Permanent Pots
These requirements guarantee that the reverse form of the consequence, which has the form
“Since S is notQ, it is notP,” is a correct proof: (a) guarantees that its pervasion (notQ is
pervaded by notP) is established, (b) guarantees that its logical reason qualifies the subject
(S is notQ).
There is no explicit requirement pertaining to the establishment of the subject in this
definition nor is there any mention of this issue in Phya pa’s gloss of the individual terms.
Only in a later section of the Mun sel does one finds the mention that the expression used
in this definition for the subject, i.e., “property-possessor” (chos can), implies that the basis
is established by a valid cognition.34 This requirement is otherwise highlighted in Phya
pa’s classification of arguments by consequence.35
Why is the establishment of the subject required? From the point of view of the reverse
form, it is required in order to avoid the fallacy of the “unestablished subject,” which
would affect the characteristic of pakṣadharmatva of the “root logical reason.” From the
point of view of the consequence, the non-establishment of the subject does not affect its
qualification by the logical reason, because the premise “S is P” merely needs to be accepted
by the opponent.36 But it can affect the negation of the conclusion of the consequence, so
that the requirement (b) present in the definition of a probative consequence will not be
fulfilled.
2.3 Example of a consequence failing to be probative due to an unestablished subject
In his classification of consequences Phya pa adduces the following example to illustrate
the category of arguments by consequence that fail to be probative because of the nonestablishment of the subject:
(1) “Because an ultimate entity is produced, it follows that it is impermanent.”
The reverse form of this consequence would be:
(1’) “Since an ultimate entity is not impermanent, it is not produced”
According to Phya pa, the pakṣadharmatva of (1’) fails to be established. Indeed, one
cannot establish that “an ultimate entity is not impermanent” due to the non-establishment
of the subject “ultimate entity.” This is in line with his take on unestablished subjects when
discussing inference-for-oneself. There Phya pa lists seven cases where the “qualification
34
35
36
his definition, this feature must be understood from the expression “subject in the context” (skabs su
bab pa’i chos can), which refers to a subject (S) such that the opponent entertains the erroneous belief
that S is P.
Mun sel 91a7–8: chos can zhes pas gzhi la lta bu tshad mas grub par bstan “With the word ‘propertypossessor’ one indicates that a basis such as ‘a mountain pass’ is established by a valid cognition.”
For the details see Hugon 2016.
Phya pa seldom uses the term phyogs chos (the Tibetan rendering of pakṣadharmatva) for this feature
in the argument by consequence. He just speaks of “the nature of the reason being established” or “the
reason being established.” See Hugon 2013: 679.
Pascale Hugon
139
of the subject” (phyogs chos) is not fulfilled for the logical reason.37 The first three are due
to the subject (or basis):
i. Unfounded basis (gzhi gtan med pa), for instance, ‘ultimate sound’ or a ‘self consisting in a pudgala;’38
ii. Basis not ascertained by valid cognition (gzhi tshad mas ma nges pa), for instance,
‘a demon’s pot’ or ‘a nymph’s song;’
iii. Basis with regard to which there is no desire to know (gzhi la shes ’dod med pa), for
instance, ‘sound’ when proving audibility.
The first two are relevant to the fallacy of āśrayāsiddha (even though Phya pa does not
use the corresponding term gzhi ma grub here). Phya pa does not explain in detail the
distinction between (i) and (ii), but a passage in the discussion pertaining to demons’ pots
in the context of arguments by consequence suggests that instances of the second category
are things that may exist, but whose existence cannot be known to (human)39 cognizers.40
The subject of the consequence (1) under consideration (“ultimate entity”) would fit
into the first category: “unfounded basis” by analogy with the example ‘ultimate sound.’
In the section on inference, the subject is posited as something that the proponent wants
to prove to be impermanent via the logical reason ‘produced.’41 In the reverse form of
the consequence (1’), however, both the property to be proven and the logical reason are
simple negations. This does not prevent the non-establishment of pakṣadharmatva: Phya
pa rejects the possibility that an unestablished subject may be qualified by any property
whatsoever, even a simple negation. We can note here a major difference to Dharmottara
and other proponents of the Method of Simple Negation.
If the non-establishment of the subject “ultimate entity” affects the establishment of
pakṣadharmatva, it does not prevent the elimination of the conclusion of the consequence
by a valid cognition. According to Phya pa, it is possible to refute that “ultimate entity
is impermanent.” This is achieved by way of negating the conjunction of ultimate entity
and impermanent (mi rtag pa don dam pa’i dngos po dang tshogs pa khegs). To do so,
one considers the fact that what is produced is pervaded by impermanence, which is
itself pervaded by emptiness. There can thus be no connection between impermanence
and something outside the range of what is empty, such as an ultimate entity; hence
37
38
39
40
41
See Mun sel 46a3–5, ’Od zer 84b6–8.
In Mun sel 28.1a6, the “unfounded definitional basis” (mtshan gzhi gtan med pa) is similarly responsible
for the fault of “impossible definiens” (mi srid pa’i mtshan nyid). It is exemplified by the definitional
basis “permanent pot” to which someone applies the definiens “causally active” to define it as real.
Possibly demons can see their own pots and nymphs can hear their own songs.
Cf. ’Od zer 147a4: rtag pa’i sha za’i bum pa dang spyi gcig po’i dngos po chos can du byed na gzhi de
dag myed pa dang yod kyang shin du lkog du gyur pas de la ’jug pa’i tshad mas dam bca’ la bsal pa
myed pa bden pa zhig na /
This is the case for the other examples of (i) and (ii) also. The formal applications mentioned here are
(i) “Ultimate sound is impermanent because it is produced” (don dam pa’i sgra byas pa’i phyir mi rtag),
or “A self consisting in a pudgala is impermanent because it is produced” (gang zag gi bdag gzhir byas
te byas pas myi rtag pa bsgrub pa lta bu’o); (ii) “A demon’s pot is impermanent because it is produced”
(sha za’i bum pa byas pa’i phyir mi rtag), or “a nymph’s song is impermanent because it is produced”
(dri za’i dbyangs chos can du byas te byas pas myi rtag pa bsgrub pa lta bu’o).
140
Universals, Demons’ Pots, and Demons’ Permanent Pots
one can negate the association of these two.42 But this does not amount to establishing
pakṣadharmatva. Phya pa holds (contra to some of his predecessors) that pakṣadharmatva
must consist in the attribution of a property (here consisting in a simple negation) to a basis.
In dealing with this example, he thus acknowledges a difference between pakṣadharmatva
and the elimination of the conclusion.43 The distinction amounts to the one we may draw
in terms of “internal negation” and “external negation” when considering the form of the
elimination of the conclusion and that of the pakṣadharmatva:
Consequence: Because S is P, it follows
that it is Q
Reverse form: Since S is notQ, it is notP
Elimination of the conclusion = it is not
the case that S is Q
Pakṣadharmatva = S is notQ
The distinction mostly goes unnoticed in the formulation of the examples in Tibetan:
apart from the context where Phya pa speaks of “negation of the conjunction,” the formulation of the elimination of the conclusion in examples is not usually distinguished from
that of the pakṣadharmatva. And the distinction is not actually crucial since apart from
cases involving a non-existent subject, the two features are either both ascertained or both
not ascertained by a valid cognition. For instance, a hill is either a thing-with-smoke or a
thing-without-smoke. By eliminating, for instance by a perception, the conclusion “there
is no smoke on the hill” that derives from the acceptance that there is no fire, one also
warrants the establishment that “there is smoke on the hill.”
2.4 Comparison of four examples of consequences with potentially unestablished subjects
It would obviously be problematic if Dharmakīrti’s example were to fall into the same
category as case (1) due to the nature of the subject “universal.” We have already seen in
§ 1.2 that the Method of Conceptual Appearance would enable Phya pa to ‘save’ this case.
In what follows, I will consider two further examples of consequences with problematic
subjects in an attempt to establish whether there is a criterion that directs the application or
non-application of this method.
Altogether, there are four examples of consequences in Phya pa’s excursus that are
relevant to the question of the subject. The first is case (1) considered above; the fourth
corresponds to Dharmakīrti’s example. I list below their constitutive elements and those
of their putative reverse form. Note that all four examples are genuine consequences –
42
43
Mun sel 85b4–5: dang po ni don dam pa’i dngos po byas pa yin no zhes smra ba la don dam pa’i dngos
po byas pa’i phyir mi rtag par thal lo zhes brjod pa na byas pa la mi rtag pas khyab pa tshad mas nges
la mi rtag pa la stong pa nyid kyis khyab pa nges pas mi rtag pa stong pa nyid kho na la yod par nges
pas don dam pa’i dngos po dang tshogs pa khegs pas dam bca’ la tshad mas bsal yang gzhi don dam
pa’i dngos po nyid med pas de la mi rtag pas stong pas byas pas stong zhes pa ’phen mi nus pa yin no //
Curiously, the question of the subject is not pointed out in Mun sel in the discussion on “correspondence”
(gnad cig) preceding the classification of arguments by consequence, where Phya pa establishes that the
negation of the conclusion of the consequence and the pakṣadharmatva of the reverse form are either
both established or both not established. In ’Od zer it appears in an interlinear note that, to my opinion,
represents a later addition to the text. See Hugon 2016 for the details.
141
Pascale Hugon
arguments to which the opponent is unable to retort. It must therefore be granted that the
opponent accepts the qualification of the subject by the logical reason in the consequence
and does not agree with the derived property that comes to be ascribed to the subject. In
all four cases the pervasion must be taken to be established by a valid cognition for both
debaters.
1
2
3
4
Consequence
Subject
ultimate entity
demon’s pot
singular universal
Reason
existent
demon’s permanent
pot
existent
impermanent
impermanent
produced
impermanent
Derived
property
Reverse form
Subject
ultimate entity
Reason
Property to
be proven
linked with multiple
instances
multiple
demon’s permanent singular universal
pot
void of imper- void of im- void of imperma- void of multiplicity
manence
permanence nence
void of being void of exis- void of existence
void of link with
produced
tence
multiple instances
demon’s pot
At first sight these four cases look very similar:
1. They are all consequences that a Buddhist could put forward to refute an opponent
and/or negate in a proactive way the attribution of a property.
2. In all four cases the subject appears to be non-established. In particular, in Phya
pa’s discussion of pakṣadharmatva in the inference-for-oneself, ‘demon’s pot,’ the
subject of (2), is mentioned as an instance of a basis that is not established by a valid
cognition, and ‘ultimate entity,’ the subject of (1), is akin to the example illustrating
an unfounded basis. Note that the distinction between these two types of unestablished
basis does not play a role in the section on arguments by consequence: Phya pa does
not differentiate in this context between bases that are termed “not determined by
valid cognition” (tshad mas nges pa med pa), “unfounded” (gtan myed pa), or simply
“not established” (gzhi ma grub).
3. The logical reason and the derived property of the consequence are positive properties,
whereas the logical reason and the property to be proven of the reverse form have
the form of a simple negation. As we have seen in the analysis of case (1), this is not
a feature that justifies the application of the property to an unestablished subject.
4. Hence in all four cases one can expect that pakṣadharmatva is not established, so
that the reverse form is not correct.
But Phya pa treats these four cases very differently: (3) and (4) are held to be probative
consequences, while (1) and (2) are held to be non-probative due to a failure to establish
142
Universals, Demons’ Pots, and Demons’ Permanent Pots
pakṣadharmatva in the reverse form. Further, (1) and (2) differ as to the possibility of
eliminating the conclusion of the consequence.
The difference between the four cases in terms of elimination of the conclusion and
establishment of the pakṣadharmatva of the reverse form can be summarized as follows:
Elimination of the conclusion
1
2
3
4
Rejection of the conjunction of impermanence and ultimate entity
Refutation that the demon’s pot is
impermanent
Refutation that the demon’s permanent pot is impermanent
Refutation that the singular universal
is multiple
Establishment of the pakṣadharmatva of the reverse form
✔
✘
✔
✔
Establishment that an ultimate entity is
void of impermanence
Establishment that a demon’s pot is void
of impermanence
Establishment that a demon’s permanent
pot is void of impermanence
Establishment that a singular universal
is void of multiplicity
✘
✘
✔
✔
It is understandable that Phya pa would want (4) to be probative since it is the case
discussed in the prasaṅga passage of the PVin. But what distinguishes it from (1) and (2)?
Can the method adopted for (4) in the commentary on the prasaṅga passage not also be
applied to (1) and (2)? And further, what distinguishes (2) from (3)?
(2) “Because the demon’s pot is (according to you) existent, it follows that
it is impermanent.”
The second example (2) appears in two contexts in the Mun sel (there is no parallel for the
second context in the ’Od zer). In the first Phya pa uses this example to refute the definition
of probative consequence given by other scholars in terms of “qualification of the subject
merely accepted by the opponent and pervasion established by a valid cognition,”44 in the
second he argues that his own definition of probative consequence (which he inherits from
rNgog Blo ldan shes rab) does not allow for the inclusion of this case.45
The main argument in both discussions revolves around the fact that the conclusion
of the consequence “the demon’s pot is impermanent” cannot be eliminated by a valid
cognition.46 As we have seen in case (1), the non-establishment of the subject does not
necessarily prevent the elimination of the conclusion – there may be a way to negate the
conjunction of the subject and the derived property. But no similar solution is offered here.
The failure to satisfy the feature of elimination of the conclusion disqualifies this
example from being probative according to Phya pa’s definition, which includes it as a
44
45
46
Mun sel 90a7–9; ’Od zer 146b4–6. Their definition, he argues, would lead them to include this case
among “probative consequences,” whereas it cannot be so. The discussion has a precedent in rNgog
Blo ldan shes rab’s dKa’ gnas 385.
Mun sel 91b3.
The formulation in Mun sel is sha za’i bum pa mi rtag pa la tshad ma’i gnod pa med pa; in ’Od zer on
the other hand, the formulation matches rather that of the pakṣadharmatva of the reverse form: myi rtag
pas stong par tshad mas ma nges pa.
Pascale Hugon
143
necessary criterion. For other scholars who do not include it in the definition, this failure may
threaten the consequence to become fallacious. Indeed, unless the opponent is convinced
of the permanence of demons’ pots,47 he is likely to retort that he accepts the derived
conclusion that the demon’s pot is impermanent. The consequence would hence not even
be genuine.
As a final blow, Phya pa points out that the logical reason of the reverse form would
be unestablished (rtags ma grub). Although it is not specified explicitly what makes the
reason unestablished, the fallacy of the non-established basis is a likely candidate as it was
in (1).
(3) “Because the demon’s permanent pot is existent, it follows that it is
impermanent.”
(4) “Because a singular universal is linked with multiple instances, it follows that it is multiple.”
Demons’ permanent pots were probably not a significant topic of debate in philosophical
circles but universals certainly were. Nevertheless, case (3) is the key to understanding the
difference between cases where the nature of the subject is problematic from those where it
is not. On the one hand, its form and constitutive elements are considered by Phya pa to be
perfectly parallel to the fourth application – the results obtained from the analysis of this
example thus apply to the example from Dharmakīrti’s prasaṅga passage.48 On the other
hand, it only differs from (2) in one aspect, namely the subject is “demon’s permanent pot”
rather than “demon’s pot” – a detail that turns out to be of major importance.
The argument by consequence about the demon’s permanent pot is introduced as a
potential counter-example to Phya pa’s definition of a probative consequence. Phya pa’s
definition, as we have seen, includes the requirement of “elimination of the conclusion by
a valid cognition,” which is a necessary (but not sufficient) condition for the establishment
of the pakṣadharmatva in the reverse form. Phya pa invoked this criterion to exclude case
(2) from the range of probative consequences. A (hypothetical) opponent now argues that
cases (3) and (4) should be excluded for the same reason: the subject being unestablished,
there can be no elimination of the conclusion by a valid cognition.49 Phya pa thus sets out
to show how the conclusion can nevertheless be eliminated in case (3), and hence in (4).
47
48
49
The specification pertaining to what the opponent accepts is only found in ’Od zer.
In the list of consequences that induce a reverse form in which the logical reason is of a different type
than the logical reason of the consequence, case (3) illustrates a consequence based on a logical reason
qua essential property, the reverse form of which is based on the non-apprehension of the pervader, like
(4) (see § 1.1). Cf. Mun sel 86a6: rtag pa’i sha za’i bum pa yod pa’i phyir mi rtag par thal zhes pa rang
bzhin gyi rtags kyi thal ’gyur gyis mi rtag pas stong pas na yod pas stong zhes pa khyab byed mi dmigs
pa’i rang rgyud ’phen ste gnyis so //
’Od zer 147a3–4: gzhi rtag pa’i sha za’i bum pa dang spyi gcig po nyid ma grub pas gzhi de la ’jug
pa’i tshad ma’i bsal pa myed pa bsgrub pa’i thal ba mtshan nyid myed pas de la ma khyab po zhe na.
The objection in Mun sel is more precise. For (3) it states that there is no elimination by valid cognition
because the basis is completely inaccessible (Mun sel 91b4: gzhi shin du lkog du gyur pas dam bca’ la
tshad ma’i bsal pa med pas grub (em: sgrub) pa’i thal ba de la ma khyab). For (4), it examines two
options: the conclusion is eliminated by the opponent’s belief, or it is eliminated by establishing by valid
cognition that there is no such thing as a universal. In the former case the corresponding member of the
reverse form, pakṣadharmatva, will not be established by valid cognition; in the latter case the intended
144
Universals, Demons’ Pots, and Demons’ Permanent Pots
The problem had been acknowledged by earlier scholars – in particular rNgog Blo ldan
shes rab – who came up with a solution and answered various objections.50 But Phya pa
does not agree with this solution and also claims that these scholars did not satisfactorily
answer the objections addressed to them. The presentation of his own solution to the
problem is thus to be understood against the background of this earlier debate: in addition
to presenting his own account of a valid cognition able to eliminate the conclusion, Phya
pa must show that his solution is not liable to the objections addressed to the view of his
predecessors.
Leaving the details aside, let me attempt here to provide a summary of this long and
complex discussion (Mun sel 91b3–95b5, ’Od zer 147a3–148a4):
According to Phya pa’s predecessors, to eliminate a conclusion – “S is Q” – a valid
cognition must grasp the contradiction between the derived property (Q) ascribed to the
subject and a property that is part of the subject (S), either by definition or through the
formulation of the subject (for instance “singular” in the case of the subject “singular
universal” or “permanent” in the case of the subject “demon’s permanent pot”). When
the subject is not established, the contradiction must simply be ascertained on the basis
of another instance, and can then be applied to the problematic subject. For instance, the
contradiction between “permanent” and “impermanent” can be established on the basis of
a pot, and then applied to the case of the demon’s permanent pot.
Phya pa’s solution is to appeal to appearances. The subject of (3) is “just what appears
as a demon’s permanent pot,” that of (4) is “just what appears as a singular universal.” This
appearance is established by reflexive awareness. Thus, technically speaking, the subject is
no longer unestablished. The appeal to another basis to ascertain the contradiction (which
was the solution advocated by Phya pa’s predecessors) is not needed. The ascertainment of
contradiction can be made on the basis of the appearance: The reflexive awareness that
establishes an appearance as X also establishes “aspects” or “features” (rnam pa) of this
appearance in such a way that the establishment of a feature induces the rejection of the
direct or indirect contrary feature (see § 3 for more details).
Thus in the case of the demon’s permanent pot (3), the formulation of the subject
in terms of “demon’s permanent pot” generates a conceptual cognition which reflexive
awareness can establish to be an “appearance as a demon’s permanent pot” involving,
50
thesis of the proponent would be likewise established, hence making the argument under consideration
useless. This is because negating the universal itself allows one to eliminate the connection between
the universal and multiplicity (i.e., to eliminate the conclusion), but also to eliminate the connection
between the universal and the link with multiple instances (i.e., which is the intended thesis of the
proponent). This long version of the objection runs parallel to the objection in dKa’ gnas 386.
A large portion of the dKa’ gnas (386–393) is devoted to demonstrating that cases (3) and (4) qualify as
probative and in particular that their conclusion is invalidated by valid cognition (tshad mas gnod pa).
rNgog Blo ldan shes rab presents his version of a valid cognition that enables the elimination of the
conclusion in spite of the subject’s lack of establishment and defends it against various objections. There
is little doubt that Phya pa is referring to rNgog Blo ldan shes rab in the Mun sel when he reports the
view of “previous scholars” (sngon gyi mkhas pa dag), objections to this view, and the answers of these
scholars, referred to as “the greatest of the greatest” (che ba’i che ba rnams). Phya pa’s presentation
of these previous views, of the objections and answers, are recognizably those found in the dKa’ gnas
even though the form and wording differ. If not to rNgog Blo ldan shes rab himself, Phya pa must be
referring to a faithful epigone of the latter. But note that in the Tshad ma bsdus pa (354), it is rNgog
Blo ldan shes rab who is associated with this view and not a later author.
Pascale Hugon
145
among other things, the feature of “permanence.” The establishment of the feature of
“permanence” in this appearance allows one to reject the feature of “impermanence.”
Thereby the conclusion of the consequence, “the demon’s permanent pot is impermanent”
is eliminated, and the qualification of the demon’s permanent pot as “void of impermanence”
is established. Hence both the elimination of the conclusion of the consequence and the
pakṣadharmatva of the reverse form are established by the same valid cognition: reflexive
awareness. The same method can be applied in case (3): the mental appearance of the
subject “singular universal” allows for the establishment of the feature of “singularity,”
which rejects “multiplicity” and establishes the “void of multiplicity.”
3. Method of conceptual appearance and ascertainment of features
In dealing with cases (3) and (4), Phya pa thus relies on his version of the Method of
Conceptual Subject which highlights the mental appearance. This Method of Conceptual
Appearance allows him to remove the potential failure of non-establishment of the subject
in the reverse form. Further, it allows the establishment of the pakṣadharmatva and the
elimination of the conclusion thanks to an extension brought to the function of reflexive
awareness: not only does it establish that there is a mental appearance (by which the subject
is no longer unestablished), it also establishes features (rnam pa) that are an integral part
of this mental appearance.
The notion of “feature” (rnam pa) that comes into play here appears to be related to
a distinction between appearances of properties “with a form” (rnam ldan) and “without
a form” (rnam med) (as I translated it in a previous article)51 invoked by Phya pa in his
discussion on perception to explain why opposite superimpositions can be eliminated by
perception itself for some properties of a perceived object but not for others. For instance, in
the case of the perception of an impermanent blue object, the superimposition as “non-blue”
can be eliminated but not the superimposition “non-impermanent.” This, according to
Phya pa, because the property “blue” is “endowed with a form” (rnam ldan) whereas the
property “impermanence” cannot be ascertained as such because it is “without a form”
(rnam med). One could, as in the present discussion, speak of properties being “featured”
or “not featured” in the mental appearance. Phya pa also applies this distinction in ’Od
zer when commenting on the passage of PVin 1 related to reflexive awareness, including
the cognition of “pleasure, etc.”52 It is in this context also that the terms are already found
to be used by rNgog Blo ldan shes rab. The question of a possible Indian source for this
terminology remains unanswered so far.53
The description of the exact process for eliminating the conclusion of the consequence
and establishing the pakṣadharmatva of the reverse form based on these features involves
first the establishment of a given cognition (1 in the table below). Based on the reciprocal
link between mind (blo) and the object of mind (yul), the establishment of the cognition
enables the establishment of its apprehended object as being endowed with the features
51
52
53
See Hugon 2011. I avoided the translation “aspect” for rnam pa in view of Phya pa non-aspectualist
standpoint on cognition.
’Od zer 59a2 ad PVin 1.19d etc.
See Hugon 2011: 169, n. 46 for a possible source.
146
Universals, Demons’ Pots, and Demons’ Permanent Pots
that were represented in the cognition (2). Then, the ascertainment of features of what
appears enables the negation of opposite features (3a). Phya pa adds that it also enables
what I have translated in the table below as “the negation of the [object being] true with
regard to the opposite property.”(3b) Expressions such as mi rtag par bden pa (“true with
regard to impermanence”) and du mar bden pa (“true with regard to multiplicity”) must
be understood in the framework of Phya pa’s characterization of objects and cognitions
as it is found in the section of Mun sel on apprehended objects (see in particular Mun sel
1b7–1b9): objects (yul) are characterized as “true” (bden pa) or “false” (brdzun pa), and
the cognitions that apprehend them (blo) as, respectively, “non-erroneous” (ma ’khrul ba)
and “erroneous” (’khrul ba).54 In subsequent discussions, Phya pa builds on this initial
correspondence between the characterization of objects and cognitions, notably by adding
specifications to the characterization as “true:” an object is “true” in a specific way (or with
regard to a given property) when the cognition to which it appears as such is non-erroneous,
and vice versa. Its being “true” in this specific way prevents its being true in the opposite
way. For instance, when debating whether two things X and Y are distinct or identical,
Phya pa considers whether the cognition to which they appear as distinct is erroneous or
non-erroneous. In the latter case, the object (i.e., X and Y) is qualified as “true with regard
to being distinct” (tha dad du bden pa), and its being “true with regard to being one” is
negated (gcig du bden pa khegs).55 The “negation of the feature x” (x kyi rnam pa khegs
pa) (3a) and the “negation of [the object being] true with regard to x” (x+loc. bden pa
khegs pa) (3b) are also mentioned together by Phya pa in other contexts as following from
the establishment of the feature opposite to x.56
Although Phya pa is not explicit on the issue in the context under consideration, it
would appear that step (3b) supports step (4a) – the elimination of the acceptance that the
object has the property whose feature was negated – in view of the relation between the
characterization of the object as true and of the mind that cognizes it as non-erroneous: by
negating that the object is true with regard to property x, one also negates that the mind
that apprehends this object as having the property x is non-erroneous.
Step (4b) – the establishment of the pakṣadharmatva of the reverse form – is to be
understood as deriving from 3a/3b as well insofar as the logical reason is expressed in the
form “void of property x” (e.g., void of impermanence, void of multiplicity), an explicit
negation which echoes the “negation of the feature x” (3a) and the “negation of the object
being true with regard to x” (3b) rather than the establishment of the object appearing as
being positively qualified by the opposite property (e.g., permanence, unicity) (2).
54
55
56
“True” and “false” correspond in this context to the characterization of the objects as, respectively,
“real” (dngos po) and “unreal” (dngos med). For instance, hallucinated objects are “false” and the nonconceptual cognitions in which they appear are “erroneous.” But in other contexts, such as the definition
of valid cognition in terms of “the understanding of a true object” (bden pa rtogs pa), “true” takes up
the meaning of “non-opposed” (gnod pa med pa), and “true object” is not paired with “non-erroneous
cognition,” but with “valid cognition.” See Hugon 2011.
See for instance Mun sel 3a6, where Phya pa discusses the case of the appearance of “white” and “visual
consciousness” in parallel with the case of the appearance of “pleasure” and “suffering.”
See for instance Mun sel 62b8 and the passage cited in n. 66 for the feature of “momentariness” (skad
cig gis stong pa’i rnam pa yongs gcod la grub pas / skad cig ma’i rnam pa khegs pas skad cig mar
bden pa’ang khegs pa yin no //), 63b8 for the feature of “multiplicity” (du mar bden pa), 65b9 for the
features of “momentariness” and “reality” (skad cig dang dngos por bden pa).
Pascale Hugon
147
Thus in the case of the demon’s permanent pot and, respectively, of the singular
universal, we find the following steps:57
1
2
3a
3b
4a
4b
Experience, by reflexive awareness, that
conceptual cognition (i.e., the mind) is
something that apprehends a demon’s
permanent pot
The object is established to be the appearance as a (demon’s) permanent pot
(i.e., it has the feature of a permanent
pot)
Negation of the feature of impermanence
Negation of [the object being] true with
regard to impermanence
Elimination of the acceptance of the derived consequence that [the demon’s permanent pot] would be impermanent
Establishment of the pakṣadharmatva
of [the reason] void of impermanence
Experience, by reflexive awareness, that
conceptual cognition is something that
apprehends a singular universal
The object is established to have the feature of a singular universal
Negation of the feature of multiplicity
Negation of [the object being] true with
regard to multiplicity
Elimination of the acceptance of the derived consequence that [the singular universal] is multiple
Establishment of the pakṣadharmatva
of [the reason] void of multiplicity
4. Other applications of the method of conceptual appearance in Phya
pa’s works
The method adopted by Phya pa in his commentary on the prasaṅga passage and in his
own theory of arguments by consequence is also applied in his works in other contexts.
One of those is linked with Dharmakīrti’s discussion in the svadharmin passage, which,
as pointed out in the introduction, is one of the two main contexts in which Dharmakīrti
deals with cases where he wants to rescue arguments from the fallacy of the unestablished
basis.58
In the svadharmin passage, Dharmakīrti discusses the Buddhist refutation of the
Vaiśeṣika’s permanent space and the Vaiśeṣika’s objection that the Buddhist’s argument
would be faulty because by refuting the existence of the subject the Buddhist would invalidate his own thesis.59 This passage has no direct equivalent in the PVin. But Dharmakīrti
57
58
59
Mun sel 95a3–4: rang rig pas rtog pa rtag pa’i sha za’i bum pa ’dzin pa dang spyi gcig du ’dzin pa
tsam myong pa na yul yang rtag pa’i bum par snang par grub pas mi rtag pa’i rnam pa khegs te mi
rtag par bden pa’ang khegs pas mi rtag par thal ba ’dod pa la des sel te mi rtag pas stong pa’i phyogs
chos kyang rang rig pa’i shugs la grub la / yul spyi gcig gi rnam par grub pa na’ang du ma’i rnam pa
khegs te du mar bden pa’ang khegs pas du mar thal ba ’dod pa’ang rang rig pa’i shugs la bsal la du
mas stong pa’i phyogs chos kyang rang rig pa’i shugs la grub pa yin no //
The second context is the pradhāna passage, discussed both in the PV and the PVin, which was Phya
pa’s main source. But Phya pa does not supplement his commentary on the pradhāna passage with an
excursus, and the refutation of pradhāna does not come forth in his Mun sel.
PV 4.141–143 (see Tillemans 2000: 202–205).
148
Universals, Demons’ Pots, and Demons’ Permanent Pots
hints to this issue at the end of the discussion on the word iṣṭa in the definition of the thesis
(PVin 3 18,9–19,1). It is when commenting on this part of the PVin that Phya pa introduces
in his commentary an excursus on logical reasons “refuting the nature of the subject” (chos
can gyi ngo bo ’gog pa, Skt. dharmisvarūpaviparītasādhana) (’Od zer 157b6–159b7). This
excursus has an equivalent in the Mun sel within Phya pa’s discussion of the four categories
of contradictory reasons.60
Note that in Phya pa’s discussion, the argument under consideration is not the Buddhist’s
refutation of the Vaiśeṣika’s permanent space, but the Vaiśeṣika’s proof of the permanence
of space by the logical reason “void of being produced.” Phya pa’s intention is to establish
that this proof is subject to the fault of “refuting the nature of the subject” all the while
arguing that the fallacy of the unestablished basis does not apply.61 Phya pa invokes in this
discussion the Method of Conceptual Appearance: the subject is “the (mere) appearance as
space” (nam mkhar snang pa/nam mkhar snang pa tsam), that is established by reflexive
awareness and for which the property “void of being produced” can be established.62
Hence pakṣadharmatva holds. But if the logical reason is thus not faulty by reason of nonestablishment, it is, however, faulty in that it is contradictory. This is because, according to
Phya pa, the Vaiśeṣika proponent of the proof additionally applies to the mental appearance
as space an intentional determination as “real space” via an erroneous cognition. Due to
the incompatibility of the property to be proven, “void of impermanence,” with a subject
assumed to have such a real nature, the proof ends up refuting the nature of the subject.63
A further passage worth considering occurs in the context of the establishment of
pervasion in the proof of momentariness by the logical reason “produced” or “existent.”64
This involves a secondary inference proving that what is permanent (or not momentary)
is not produced or not existent via the reason “void of instantaneous or gradual causal
efficacy.” Phya pa identifies the subject in this secondary inference as “the simple negation
‘void of momentariness’” (skad cig gis stong pa’i med dgag). This simple negation is to be
60
61
62
63
64
The four categories listed in Mun sel 54a8 are: chos kyi ngo bo ’gog pa; chos kyi khyad par ’gog
pa; chos can gyi khyad par ’gog pa; chos can gyi ngo bo ’gog pa. This distinction can be traced to
the Pramāṇasamuccaya (3.27 = Nyāyamukha 9), and the terminology to Nyāyapraveśa 3.2.3, where
one finds the terms: dharmisvarūpa°, dharmasvarūpa°, dharmiviśeṣa°, dharmaviśeṣaviparītasādhana
(see Tachikawa 1971: 125–126). Note that Dharmakīrti does not identify the fallacy occurring in the
refutation of the permanence of space in terms of dharmisvarūpariparītasādhana. On Dharmakīrti’s
silence on this fallacy see Moriyama 2019.
The proof of the permanence of space is given as an example of dharmisvabhāvaviparītasādhana by
Dharmottara (PVinṬ-Skt 24a5, PVinṬ-Tib D26a1). As Moriyama 2019 reveals, the Vaiśeṣika proof is
also given as an example of dharmisvabhāvaviparītasādhana by Jinendrabuddhi and Dharmapāla – both
argue that the fault of the unestablished subject does not apply in this case. Before presenting his own
view Phya pa makes an extensive presentation and refutation of another view ascribed to Kamalaśīla
etc. In Kamalaśīla’s works (notably Madhyamakāloka and Madhyamakālaṅkārapañjikā) one finds a
discussion which may have been the source of Phya pa’s presentation of Kamalaśīla’s position. Yet
Kamalaśīla does not discuss the faulty Vaiśeṣika proof, but the Buddhist refutation of a permanent
space, which is the case discussed by Dharmakīrti in the svadharmin passage, a case that escapes the
fault of dharmisvabhāvaviparītasādhana as well as the fault of āśrayāsiddha. Kamalaśīla appeals in
this regard to the Method of Simple Negation (see Iwata 1999: 167).
Mun sel 55a4, ’Od zer 158a9.
By invoking the “intentional determination” (zhen pa) Phya pa identifies a subject that corresponds to
Dharmakīrti’s svadharmin.
Starting at Mun sel 62b5.
Pascale Hugon
149
established for a verbal object (sgra don) or, as Phya pa rephrases, a concept (don spyi).65
More precisely, Phya pa invokes the appearance of a concept that has the feature of void of
momentariness (Mun sel 63a9: skad cig mas stong pa’i rnam pa can gyi don spyi). Only in
such a case can the void of momentariness be ascertained by mere experience (myong pa)
– this amounts to reflexive awareness.
The description of the process is the same as in the case of the demon’s permanent pot
and the singular universal: experience establishes what the apprehending mind is like – in
this case, it apprehends something that is not momentary – and thereby what the object of
the apprehending mind is like – the object has the feature of being non-momentary. This in
turns leads to the negation of the aspect of momentariness and the negation of the object as
being “true with regard to momentariness,” and to the establishment of the simple negation
“void of momentariness.”66
Phya pa specifies that the experience by which one establishes the property “void of
momentariness” does not simultaneously establish or eliminate the property to be proven,
i.e., the property “void of existence.” This is because to establish that the object (yul) is
existent or non-existent, one must be able to establish that the mind (blo) is erroneous or
non-erroneous (see above and n. 54) – but this property is not something that is featured
(rnam med) in the appearance and hence cannot be established by reflexive awareness. In
other words, introspection does not allow one to ascertain whether the mental appearance
as “something void of momentariness” is erroneous or non-erroneous with regard to the
properties of existence or void of existence, and therefore whether the object of such a
cognition is unreal or real.67
Conclusion: When does the fault of the unestablished subject actually
apply?
As mentioned at the end of § 2, the adoption of the Method of Conceptual Subject or
Method of Conceptual Appearance seems to leave no room for unestablished bases –
65
66
67
This rephrasing is worth noting since, as indicated in n. 26, Phya pa does not use the term “concept”
(don spyi) when applying the Method of Conceptual Appearance in his commentary on the prasaṅga
passage. He does not use the term either when commenting on the pradhāna passage in spite of a visible
influence of the pradhāna passage of the PVin 3 pertaining to Phya pa’s characterization of the don
spyi, something I intend to come back to on another occasion. On the other hand, in his commentary on
the Madhyamakāloka Phya pa glosses Kamalaśīla’s identification of the subject in the refutation of
pradhāna as “something imagined that exists in the mind” (nye bar brtags pa blo la yod pa) in terms of
“the feature of pradhāna that is present in the appearance of the concept to conceptual thought” (dBu
ma snang bshad 38b4: rtog pa la don spyi snang pa la yod pa’i gtso’ bo’i rnam par tsam chos can yin
no zhes bya pa’o).
Mun sel 63a9–63b1: skad cig mas stong pa’i rnam pa can gyi don spyi la skad cig mas dben par nges
pa yin te / blo skad cig ma ma yin pa ’dzin par myong pa na yul skad cig ma ma yin pa’i rnam par grub
la skad cig gi rnam pa khegs pa na skad cig du bden pa ’ang myong pa’i shugs la khegs pas skad cig
gis stong pa’i med dgag grub pas gzhi’i ldog pa grub la /
The inability to ascertain the erroneous or non-erroneous character of mind by reflexive awareness –
and thereby whether the corresponding object is false or true – is discussed by Phya pa in Mun sel
36a7–8, with the conclusion that the establishment that an object is true and that the corresponding
mind has a true object, is achieved by the cognition of an object (don rig) other than mind itself.
150
Universals, Demons’ Pots, and Demons’ Permanent Pots
anything can be established as a mental appearance since there is no restriction as to
what can be said and thought about. And if there are no unestablished bases, the fault of
āśrayāsiddha would never take place. Logical reasons would never lack pakṣadharmatva
because of the lack of establishment of the basis. Still, unestablished bases (unfounded or
not ascertained by valid cognition) are listed by Phya pa among the reasons for the absence
of pakṣadharmatva in inference and come into play in rejecting the fulfillment of the
pakṣadharmatva of the reverse form of some arguments by consequence (such as case (1)
and case (2) considered above). Failure of pakṣadharmatva is not exclusively linked with
the attribution of a positive property to a non-existent subject. Indeed, the pakṣadharmatva
of the reverse form of consequences (1) and (2) is faulty although the logical reason is
a simple negation. One could suggest that this fault is restricted to proofs formulated by
opponents of the Buddhist, while Buddhist proofs can be ‘saved’ by the application of the
Method of Conceptual Appearance. However, we have seen in § 4 that Phya pa also applies
this method to argue that the Vaiśeṣika’s proof of the permanence of space is not subject to
the fault of āśrayāsiddha.
Is there another criterion that distinguishes cases where the Method of Conceptual
Appearance applies and those where the fault of āśrayāsiddha is brought forward?68 Or
should one, as uncharitable as it may seem, conclude that the fault of the unestablished
basis is advocated whenever it is convenient for the author to do so, whereas in other cases
the Method of Conceptual Appearance is called to the rescue? Before jumping to such a
conclusion, let us consider what could be achieved by the application of this method in the
cases where Phya pa invokes this fallacy.
Consequence (2) is relevant in this regard because its formulation only differs from consequence (3) by a small detail regarding the subject, but while the argument by consequence
about the “demon’s permanent pot” (3) is probative, that about the “demon’s pot” (2) is not:
it is possible, according to Phya pa, to establish that “the demon’s permanent pot is void
of impermanence” but not to establish that “the demon’s pot is void of impermanence.”
What happens if the Method of Conceptual Appearance is applied to consequence (2)? It
would indeed remove the potential fallacy of āśrayāsiddha in the pakṣadharmatva, i.e., the
establishment that “the demon’s pot is void of impermanence.” However, consequence (2)
would remain non-probative (and even possibly not genuine) because this would still not
enable one to establish that the property “void of impermanence” qualifies the demon’s pot
or to eliminate the connection between the demon’s pot and impermanence. Indeed, the
establishment of these two features in consequence (3) is enabled by the fact that reflexive
awareness could establish the aspect of “permanence” on the basis of the mental appearance
as a “demon’s permanent pot.” Consequence (2) is different in this regard: the feature of
“permanence” cannot be established by reflexive awareness because it is not an integral
part of the mental appearance of a demon’s pot. This is because it was not explicated in the
verbal formulation “demon’s pot” and is also not a feature associated by definition with
demons’ pots.
68
Asking a similar question about Dharmakīrti, Tillemans (2000: 197) concluded that the fault of āśrayāsiddha can be invoked unproblematically unless it would involve self-refutation for the Buddhist
proponent. Only in the latter case is the method distinguishing svadharmin and keveladharmin worth
applying. This conclusion, however, concerns the fault of āśrayāsiddha as a fault of the thesis and leaves
open the question of āśrayāsiddha as a fault of the logical reason.
Pascale Hugon
151
The appearance as a demon’s pot can also not be ascertained to have the feature of
impermanence. One must not be fooled here by the potential impact of the word “pot:” it
only allows for the ascertainment of the feature of “being a pot” (one could thus refute
the claim that “the demon’s pot is not a pot”), but not of the features of “impermanence,”
“existence,” or “causal efficacy” that one typically associates with pots – an association that
is erroneous because it fails to take into account that the category of “pots” is not restricted
to real pots (it also includes dream pots, etc., that are not real).
It seems that the same analysis could be applied to the other cases where the fault of
āśrayāsiddha is invoked, that is, the pakṣadharmatva of the reverse form of consequence
(1) – “ultimate entity is void of impermanence” – and the examples adduced in the chapter
on inference – “ultimate sound is produced” or “demon’s pot is produced.” In all these cases
the Method of Conceptual Appearance would certainly remove the fault of āśrayāsiddha
but would still not enable the establishment of the pakṣadharmatva because the mental
appearance would not have the explicit features that would be relevant in this regard.
One could in conclusion propose that the problem of āśrayāsiddha has been transformed
into a failure of the establishment of the pakṣadharmatva that has nothing to do with the
ontological status of the subject, but is a matter of the features of the mental appearance
that can or cannot be ascertained by reflexive awareness. However, such an interpretation
of the examples where āśrayāsiddha is invoked is not suggested by Phya pa. One could
deplore the lack of a unifying theory on this point or, more charitably, envisage that Phya
pa is applying a principle of economy in the identification of fallacies.
References and abbreviations
bKa’ gdams gsung ’bum bKa’ gdams gsung ’bum phyogs sgrig thengs dang po/gnyis
pa/gsum pa/bzhi pa, ed. dPal brtsegs bod yig dpe rnying zhib ’jug khang. Vols.
1–30, Chengdu 2006; vols. 31–60, Chengdu 2007; vols. 61–90, Chengdu 2009; vols.
91-120, Chengdu 2015.
D sDe dge Tibetan Tripiṭaka bsTan ’gyur – preserved at the Faculty of Letters, University
of Tokyo, ed. J. Takasaki, Z. Yamaguchi, and Y. Ejima. Tokyo 1981–1984.
dBu ma snang bshad Phya pa Chos kyi seng ge, dBu ma snang ba’i ’grel pa rgya cher
bshad pa. See bKa’ gdams gsung ’bum, vol. 6, 266–428.
dKa’ gnas rNgog Blo ldan shes rab, Tshad ma rnam nges kyi dka’ ba’i gnas rnam par
bshad pa, ed. Sun Wenjing. Qinghai 1994.
Hugon 2011 P. Hugon, Phya pa Chos kyi seng ge’s Views on Perception. In: Religion and
Logic in Buddhist Philosophical Analysis. Proceedings of the Fourth International
Dharmakīrti Conference. Vienna, August 23–27, 2005, ed. H. Krasser, H. Lasic, E.
Franco, and B. Kellner. Vienna 2011, 159–176.
Hugon 2013 P. Hugon, Phya pa Chos kyi seng ge on Argumentation by Consequence
(thal ’gyur): The Nature, Function, and Form of Consequence Statements. Journal
of Indian Philosophy 41.6 (2013) 671–702 (DOI 10.1007/s10781-013-9205-4).
Hugon 2016 P. Hugon, Phya pa Chos kyi seng ge on Argumentation by Consequence
(thal ’gyur) (2): The analysis of the correspondence between a consequence and
its reverse form and the thirteenfold typology of consequences. Journal of the
152
Universals, Demons’ Pots, and Demons’ Permanent Pots
International Association of Buddhist Studies 39 (2016) 51–113 (DOI: 10.2143/
JIABS.39.0.3200523).
Iwata 1993 T. Iwata, Prasaṅga und Prasaṅgaviparyaya bei Dharmakīrti und seinen Kommentatoren. Vienna 1993.
Iwata 1997a T. Iwata, On Prasaṅgaviparyaya in Dharmakīrti’s Tradition – Prajñākaragupta and gTsang nag pa. In: Tibetan Studies. Proceedings of the Seventh Seminar
of the IATS, Graz 1995, ed. H. Krasser, M. Torsten Much, E. Steinkellner, and H.
Tauscher; gen. ed. E. Steinkellner. Vienna 1997, 427–437.
Iwata 1997b T. Iwata, Pramānaviniścaya III (2). Wiener Zeitschrift für die Kunde Südasiens 41 (1997) 207–231.
Iwata 1999 T. Iwata, On The Interpretations of The Subject (dharmin) of The Inference
Negating Invariable Entities in Dharmakīrtian Logic. In: Dharmakīrti’s Thought and
Its Impact on Indian and Tibetan Philosophy. Proceedings of the Third International
Dharmakīrti Conference. Hiroshima, November 4–6, 1997, ed. Sh. Katsura. Vienna
1999, 155–172.
Keira 2004 R. Keira, Mādhyamika and Epistemology: A study of Kamalaśīla’s Method
for Proving the Voidness of All Dharmas. Vienna 2004.
Kobayashi 1987 M. Kobayashi, Mujishosei-ronsho to shoe-fujo (āśrayāsiddha) no
mondai. Bunka 50.3/4 (1987) 41–60.
Moriyama 2019 Sh. Moriyama, On Dharmisvarūpaviparītasādhana. Wiener Zeitschrift
für die Kunde Südasiens 56–57 (2019) 37–50.
Mun sel Phya pa Chos kyi seng ge, Tshad ma yid kyi mun pa sel pa. See bKa’ gdams
gsung ’bum, vol. 8, 434–626.
N Narthang (snar thang) edition of the Buddhist canon in Tib.
NB Dharmakīrti, Nyāyabindu: Nyāyabindu with Dharmottara’s Nyāyabinduṭīkā and Durveka Miśra’s Dharmottarapradīpa, ed. Dalsukh Bhai Malvania. Patna 1955 [2nd ed.
1971].
’Od zer Phya pa Chos kyi seng ge, Tshad ma rnam par nges pa’i ’grel bshad yi ge dang
rigs pa’i gnad la ’jug pa’i shes rab kyi ’od zer. See bKa’ gdams gsung ’bum, vol. 8,
35–427.
P The Tibetan Tripiṭaka, Peking edition, ed. D.T. Suzuki. Tokyo/Kyoto 1955–1961.
PS Dignāga, Pramāṇasamuccaya, Tibetan translation in D4203, P5700.
PV Dharmakīrti, Pramāṇavārttika.
PV 1 Chap. 1 and Pramāṇavārttikasvavṛtti: The Pramāṇavārttikam of Dharmakīrti: The
First Chapter with the Autocommentary, ed. R. Gnoli. Rome 1960.
PV 2, 3, 4 Chap. 2–4 (pramāṇasiddhi, pratyakṣa, parārthānumāna), ed. Y. Miyasaka. Acta
Indologica 2 (1972).
PVin Dharmakīrti, Pramāṇaviniścaya.
PVin 1/2 Chap. 1/2, ed. E. Steinkellner. Beijing/Vienna 2007.
PVin 3 Chap. 3, ed. P. Hugon and T. Tomabechi. Beijing/Vienna 2011.
PVin-Tib Tib. translation of the third chapter of the Pramāṇaviniścaya. D4211 ce
187a6–230a7; P5710 ce 285a7–329b1; N3701 ce 299a6–347a7.
PVinṬ-Skt Dharmottara, Pramāṇaviniścayaṭīkā, copy of the manuscript (see the introduction to PVin 1 & 2 and PVin 3 for the details).
PVinṬ-Tib Dharmottara, Pramāṇaviniścayaṭīkā, Tibetan translation in D4227, P5727.
Pascale Hugon
153
PVSV Dharmakīrti, Pramāṇavārttikasvavṛtti. See PV 1.
Tachikawa 1971 M. Tachikawa, A Sixth-Century Manual of Indian Logic. (A Translation
of the Nyāyapraveśa). Journal of Indian Philosophy 1 (1971) 111–145.
Tani 1987 T. Tani, The Problem of Interpretation on Pramāṇaviniścaya III vv. 1–3 – with
the text and a translation. The Bulletin of the Kochi National College of Technology
26 (1987) 1–16.
Tillemans 1982 T. Tillemans, The ‘Neither One Nor Many’ Argument For Śūnyatā, and
Its Tibetan interpretations: Background information and source materials. Études de
Lettres, University of Lausanne 3 (1982) 103–128.
Tillemans 1984 T. Tillemans, Two Tibetan Texts on The ‘Neither One Nor Many’ Argument For Śūnyatā. Journal of Indian Philosophy 12.4 (1984) 357–388.
Tillemans 1999 T. Tillemans, Scripture, Logic, Language: Essays on Dharmakīrti and
his Tibetan Successors. Boston 1999.
Tillemans 2000 T. Tillemans, Dharmakīrti’s Pramāṇavārttika. An annotated translation
of the fourth chapter (parārthānumāna). Vol. 1 (k. 1–148). Vienna 2000.
Tillemans and Lopez 1998 T. Tillemans and D. Lopez, What Can One Reasonably Say
About Nonexistence? A Tibetan work on the problem of āśrayāsiddha. Journal of
Indian Philosophy 26 (1998) 99–129.
Tshad ma bsdus pa Anonymous (attributed to Klong chen Rab ’byams pa), Tshad ma’i
de kho na nyid bsdus pa, ed. Padma tshul khrims. Chengdu 2000.
Yoshimizu and Nemoto 2013 Ch. Yoshimizu and H. Nemoto, Zhang Thang sag pa
’Byung gnas ye shes, dBu ma tshig gsal gyi ti ka, Part I, folios 1a–26a3 on Candrakīrti’s Prasannapadā ad Mūlamadhyamakakārikā 1.1. Tokyo 2013.
Two Kinds of Causal Capacity: sāmānyā śaktiḥ and
pratiniyatā śaktiḥ1
by
Masahiro Inami
1. Introduction
Dharmakīrti and his followers define real things as those that have some causal capacity
(arthakriyāśakti), and consider those that cannot produce any effect, such as the universal
(sāmānya) and the whole (avayavin), as merely conceptual constructions. In connection
with this capacity, as is well known, they refer to the two kinds of causal capacity: sāmānyā
śaktiḥ and pratiniyatā śaktiḥ.
According to the examination of these two kinds of causal capacity by modern scholars,
Dharmakīrti and his followers consider external objects, such as jars (ghaṭa, water jar), to
have two different causal powers: sāmānyā śaktiḥ and pratiniyatā śaktiḥ. The sāmānyā
śaktiḥ is the causal power common to other objects of the same kind, and the pratiniyatā
śaktiḥ is the causal power not common to any other object. For example, a jar has the causal
power or capacity of containing water, which is common to other jars, but at the same time
it has the unique power of producing its own particular perception as well. Of these, the
sāmānyā śaktiḥ is regarded as the useful function for the fulfillment of human purposes,
and is therefore not “causal capacity” in a strict sense. In contrast, the pratiniyatā śaktiḥ is
regarded as the causal efficiency that real things have.2
1
2
The present paper is a revised English version of a Japanese publication entitled “Nishu no ingakōryoku
–sāmānyā śakti to pratiniyatā śakti–” [Two Kinds of Causal Capacity: sāmānyā śakti and pratiniyatā
śakti] (=Inami 2012). Almost all of the passages from Indian texts being treated here were translated
into Japanese in another paper entitled “Pramāṇavārttika Pramāṇasiddhi-shō no kenkyū (13)” (=Inami
2014). I would like to express my gratitude to Prof. Tom J. F. Tillemans for correcting the English of
the present paper.
Esho Mikogami was the first scholar to pay attention to the Buddhist theory of two kinds of causal
capacity. He examined PV II 100–102 and Prajñākaragupta’s commentary and concluded that the
sāmānyaśakti is the useful function possessed by things in our daily life and that it is related to the
sāmānyalakṣaṇa; the pratiniyataśakti is the causal efficiency that gives rise to the perception of color,
etc., and it is related to the svalakṣaṇa. Mikogami understood that these two correspond with the two
meanings of arthakriyā as pointed out by Nagatomi 1967/68 (Mikogami 1978, 1979). After Mikogami’s
study, Shoryu Katsura considered these two kinds of causal capacity several times. According to his
first examination of this theory (Katsura 1983: 97–100), the pratiniyataśakti is the unique capacity that
momentary existence has, and should be regarded as the primary meaning of the word arthakriyā. In
contrast, the sāmānyaśakti is, for example, a pot’s capacity for containing water, something common
to other pots, thus meaning “a useful function for the fulfillment of a human purpose.” It is related
to the conceptual cognition of the universal or of the continuum. It is not causal capacity in a strict
sense, and therefore should be regarded as the secondary meaning of the word arthakriyā. In this paper,
Katsura concluded that the pratiniyataśakti is the causal capacity to produce a direct perception and the
sāmānyaśakti is the causal capacity to produce an indirect or judgmental cognition. This understanding
also appears in a later study published in English (Katsura 1984: 218–219).
Birgit Kellner et al., eds., Reverberations of Dharmakīrti’s Philosophy: Proceedings of the Fifth International Dharmakīrti
Conference Heidelberg, August 26 to 30, 2014. Vienna 2020, pp. 155–176.
156
Two Kinds of Causal Capacity: sāmānyā śaktiḥ and pratiniyatā śaktiḥ
In addition to this, scholars have presented various other interpretations concerning
these two notions. Some scholars understand the sāmānyā śaktiḥ to be related to common
characteristics or universals (sāmānyalakṣaṇa) and the pratiniyatā śaktiḥ to be related to
unique characteristics or particulars (svalakṣaṇa).3 Some scholars explain that the sāmānyā
śaktiḥ belongs to the jar as a continuum (santāna) and the pratiniyatā śaktiḥ belongs to the
jar as a momentary thing (kṣaṇa).4 Other scholars remark that the sāmānyā śaktiḥ belongs
only to conventional existence (saṃvṛtisat), but that pratiniyatā śaktiḥ belongs to both
conventional existence and ultimate existence (paramārthasat).5 And some attribute this
theory to Prajñākaragupta, a commentator on Dharmakīrti’s Pramāṇavārttika (PV), and
not to Dharmakīrti.6
However the following questions arise: How can a jar have a causal capacity if it
is regarded as a conceptual construction such as a universal or a continuum? How can
conventional existence, such as the whole, have a causal capacity? Does the sāmānyā śaktiḥ
represent a useful function for the fulfillment of human purposes? Did Dharmakīrti not
advocate these two kinds of causal capacity? Previous studies have not yielded sufficient
results concerning these matters or, rather, they seem to contain some misunderstandings.
The aim of this paper is to clarify the Buddhist theory of the two kinds of causal capacity
by rigorously reexamining various materials.
2. Pramāṇavārttika II 100–102 and commentaries
First of all, we shall examine Dharmakīrti’s PV II 100–102, where the theory of two kinds
of causal capacity is presented. In order to prove the transmigration (saṃsāra) of living
beings against materialists, Dharmakīrti explains that the mind is not crucially dependent
on the body. Then, he examines whether the body, which is regarded as the direct cause
of the mind by materialists, is a single entity, i.e. the whole (avayavin), or an aggregate
of many parts. He denies the first alternative by pointing out that the whole never exists
distinctly from parts, and concludes that the body is nothing but an aggregate of many
parts. Next, he argues that, even if the body is an aggregate of many parts, it cannot be the
direct cause of the mind.7
In the context of his negation of the existence of the whole, Dharmakīrti refers to the
causal capacity of atoms. According to him, a jar is nothing but an aggregate of many
atoms, including color atoms. It can never be regarded as a whole. To this, the following
3
4
5
6
7
Mikogami 1978, 1979; Katsura 1983, 1984.
Katsura 1983, 1984. See n. 1. Cf. Dreyfus 1997: 361; Funayama 2012: 109.
Katsura 2012: 22–23 explains that Dharmakīrti classifies two kinds of causal capacity: one is the
capacity to produce one’s own cognition, which belongs both to ultimate existence and to conventional
existence, and the other is the capacity to fulfill human purpose, which belongs only to conventional
existence.
Katsura 2002 points out that because it was problematic whether a pot, which Dharmakīrti had explained as possessing the two kinds of causal capacity, could be regarded as an ultimate existence,
Prajñākaragupta later explained that both of these capacities belong to atoms. According to Bae 2011:
92–93, Dharmakīrti thinks that the svalakṣaṇa has a causal capacity and the sāmānyalakṣaṇa has the
ability to fulfill human purposes. Of these, Prajñākaragupta divides causal capacity into two kinds:
common capacity and specific capacity.
For an outline of PV II, see Inami and Tillemans 1986. On the details of the arguments in PV II 1–102,
see my serial studies on PV II.
Masahiro Inami
157
objection is raised: “A jar as a whole exists distinctly from its parts, such as color atoms,
because people appropriately use genitive expressions such as ‘the color and so on of a jar’
(ghaṭasya rūpādayaḥ), which indicates that there are different referents.”8 Responding to
this objection, Dharmakīrti explains that the atoms constituting a jar have two different
kinds of causal capacity.
[1] Without implying the individual causal powers (śaktibheda) of the color
and other [atoms that constitute a jar], the word “jar” (ghaṭa) [simply] serves
to express the exclusion of those things that are not the cause of their common
effects [such as containing water] (tatsamānaphala). Therefore, [despite the
fact that a jar is not distinct from its parts, including color atoms,] coreferential
expressions such as “*a color that is a jar” (rūpaṃ ghaṭaḥ) cannot be used.9
[2] The individual causal powers of the [color and other atoms], which have
their aggregate (tatsāmānya) [i.e. a jar] as a qualifier (upasarjana), are expressed by [the genitive expression] “the color and so on of a jar” (ghaṭasya
rūpādayaḥ).10
According to Dharmakīrti, when referring to aggregated atoms we use the word “jar” to
mean that they can cooperatively produce common effects such as containing water. It
8
9
10
See PVA 98,3–5: yadi tarhi nāvayavī rasādaya eva tadā na ghaṭasya rūpādaya iti bhavet / na hi bhavati
rūpādīnāṃ rūpam / nāpi ghaṭasya vā ghaṭa iti paryālocanaṃ parasyāśaṅkyāha /…; PVV(R) D341b5–6,
P195b1: gal te gzugs la sogs pa las tha dad pa’i yan lag can med na / de’i tshe bum pa dang gzugs
la ’di don gcig pa nyid du ’gyur ba ma yin nam zhe na / ’dir smras pa /… Cf. PVṬ D160b5, P187a4:
gal te bum pa’i gzugs zhes bya ba la sogs pa ldog pa ji ltar rung zhe na /…; PVSVṬ 271,15–16: yadi
rūpādaya eva kevalā ghaṭo na tu tadvyatiriktaṃ dravyaṃ, kathan tarhi ghaṭasya rūpādaya iti vyatireka
iti ced āha / … In connection with the theory that the genitive expression indicates the difference of
referents, Dharmakīrti and his followers refer to the grammatical rule vyatireke ṣaṣṭhī . PVSV 36,5–6:
na hi vyatireke ṣaṣṭhī bāhulye jasādaya ity etad api puruṣābhiprāyanirapekṣaṃ vastusaṃnidhimātreṇa
svayaṃ pravṛttam / …; PVP D43b5–43b6, P49a7–b2: gal te ldan pa la sogs pa don gzhan ma yin pa
de’i tshe / tha dad par rjod (rjod D; brjod P) par byed pa’i drug pas* tha snyad du byed par mi ’gyur
te / dper na ’di’i gcig nyid ces bya ba la sogs pa lta bu yin na yang ’gyur ba de’i phyir don gzhan
yin no // sbyor ba ni gang zhig gang las tha dad par bstan pa de ni de las don gzhan yin te / dper
na lha sbyin (lha sbyin D; lhas byin P) gyi spyi blugs lta bu’o // ldan pa la sogs pa yang bum pa la
sogs pa (sogs pa P; sogs D) las tha dad par (par D; pas P) bstan pa yin no zhes bya ba ni rang bzhin
gyi gtan tshigs yin par sems so // (*Vibh p. 46, n. 5: vyatirekaṣaṣṭhī /); PVṬ D80b3–4, P94b8–95a1:
brda sprod pa pa la sogs pa sgra de lta bu rnams brjod par bya ba’i don dag la / ’jug *pa sems pa tha
dad pa la ni drug pa’o / zhes bya ba la sogs pa gang yin pa dang / …; PVSVṬ 158,23–24: … evaṃ
bhūtānāṃ śabdānāṃ vācyeṣv artheṣu yeyam pravṛtticintā vyatireke ṣaṣṭhyādaya ityādikā / …, etc. It
can be interpreted that the word ghaṭasya refers to dravya and rūpādayaḥ to guṇas. Dharmakīrti already
criticized the Vaiśeṣika system of ontology in PV II 89–96. According to him, dravya, guṇa, karman
and so on are nothing but conceptual constructions; they do not exist distinctly from each other in the
outer world.
PV II 100–101ab: rūpādiśaktibhedānām anākṣepeṇa vartate / tatsamānaphalāhetuvyavacchede ghaṭaśrutiḥ // ato na rūpaṃ ghaṭa ity ekādhikaraṇā śrutiḥ /
PV II 102a–c: rūpādayo ghaṭasyeti tatsāmānyopasarjanāḥ / tacchaktibhedāḥ khyāpyante. The passages
[1] and [2] are translated in Mikogami 1978: 5, 1979: 81, and Katsura 2002: 269. In the Tarkarahasya
(TR 24*,2–7), PV II 100–101 are quoted together with Vākyapadīya II 155: saṃsthānavarṇāvayavair
viśiṣṭe yaḥ pravartate / śabdo na tasyāvayave pravṛttir upalabhyate //. In the Vādarahasya (also called
Udayananirākaraṇa, author unknown.) pp. 50–51, PV II 97–102 are quoted and explained in reference
to PVA.
158
Two Kinds of Causal Capacity: sāmānyā śaktiḥ and pratiniyatā śaktiḥ
should be noted that the word tatsamānaphala (PV II 100c) is said to mean the effect
common to all the constituents of the same jar, but not the effect common to all jars.
Dharmakīrti seems to apply the theory of causal complex (hetusāmagrī ) to this case.
Each constituent atom has an individual causal power or capacity (śaktibheda) as
well. The word “color” is used to define the unique power of the color atoms. Thus, each
constituent atom of the same jar has two different kinds of causal capacity: one is the
capacity common to all constituent atoms, and the other is the capacity not common to
other kinds of atoms.
The word “jar” is used to mean the capacity common to all constituents of the same jar,
without implying each individual capacity meant by a word such as “color.” Therefore, we
use genitive expressions such as “the color of a jar” (ghaṭasya rūpam) and not coreferential
expressions such as “*a color that is a jar” (rūpaṃ ghaṭaḥ).11 In this way, Dharmakīrti
insists that a genitive expression such as “the color and so on of a jar” (ghaṭasya rūpādayaḥ)
is used not because two different entities exist, viz. the whole and its parts, but because
the parts have two different causal powers. This expression means that the atoms that
can cooperatively produce common effects such as containing water also have individual
powers, such as the power to produce a visual perception.12 By the word tatsāmānya (PV
II 102b), Dharmakīrti does not mean universals such as “jarness” (ghaṭatva), but rather the
11
12
The word ghaṭa in the expression ghaṭasya rūpādayaḥ is regarded as a “term for an aggregate” (samudāyaśabda). It is also considered a “generic term” (jatiśabda) depending on context. Dharmakīrti explains
the difference between jatiśabda and samudāyaśabda in PV II 101cd: bhedaś cāyaṃ mato jātisamudāyābhidhānayoḥ // [I tentatively adopt this reading of the Sanskrit text of PV II 101cd. See PVmsH 17v1;
Ms. B of PVA 40b8.] “This difference [in implication] is found between ‘generic terms’ (jātyabhidhāna)
and ‘terms for aggregate’ (samudāyābhidhāna), as well.” In PV II 98–99, Dharmakīrti has already
explained the difference in implication between “words for property” (dharmaśabda) and “words for
property-possessor” (dharmiśabda). According to him, the non-coreferential expression aṅgulyā yogaḥ
(connection of a finger) is made because the property word yogaḥ (connection) is conventionally used
to mean one property without implying other properties of the finger and its property-possessor, i.e. the
finger. On the other hand, the coreferential expression yuktā ’ṅgulī (a connected finger) is made because
the property-possessor word yuktā (connected) is conventionally used to mean the property-possessor
with the implied properties of the finger. The same is said of samudāyaśabda and jātiśabda. The word
ghaṭa will be a samudāyaśabda in relation to the word that is used to refer to its constituents, such as
colors. In this case, it is used to mean the common capacity of the constituents without implying their
specific capacities. That is why non-coreferential expressions such as ghaṭasya rūpādayaḥ are used.
On the other hand, it will be a jātiśabda in relation to the word that is used to refer to a particular jar
and in contradistinction to words that are used to refer to other kinds of aggregates, such as trees. In
this case, since it implies particulars or other aggregates that have the same capacity, a coreferential
expression such as ghaṭaḥ suvarṇaḥ (a golden jar) can be made appropriately. Then the word ghaṭa can
also be understood to indicate the capacity of containing water, which is common to all jars. However,
it is clear that Dharmakīrti, in the context of refuting the existence of the whole, mentions the capacity
of the constituent atoms of the same aggregate. See PVP D44b6–45a5, P50b3–51a3; PVA 98, 25–28;
PVV(R) D342a4–5, P195b8–196a2; VNṬV 32,7–12.
Dharmakīrti remarks that this explanation is also applicable to other cases. PV II 102d: vācyo ’nyo ’pi
diśā ’nayā // “Other [expressions such as ‘fragrance of sandalwood’ (candanasya gandhaḥ)] should be
explained in this manner.” Cf. PVP D45b3–4, P51b1–2: phyogs ’dis gzhan la’ang brjod par bya // dper
na tsandan gyi dri zhes bya ba ’di la yang / tsandan zhes bya ba’i rdzas cung zad kyang dri la sogs pa
las tha dad pa ma yin pa lta bu’o // ’o na ci yin zhe na / dri du ma’i bdag nyid can las tsandan gyi rang
bzhin gyi dri gang yin pa des de ltar ston par ’gyur ro // (Parallel sentences are found in Haribhadra’s
Anekāntajayapatākā. See AJP II 40,9–11: etena “candanasya gandha ity atrāpi na candanaṃ nāma
gandhavyatiriktaṃ kiñcid dravyam asti / kiṃ tarhi? anekātmakasya gandhasya candanasvabhāvo yo
Masahiro Inami
159
commonness between the constituents with regard to cooperatively producing the same
single effect. He means the aggregate (samudāya) of the atoms.13
The commentators call these two kinds of causal capacity sāmānyā śaktiḥ and pratiniyatā śaktiḥ for example. Devendrabuddhi explains PV II 100 in his Pramāṇavārttikapañjikā
(PVP) as follows:
[3] The color and other [atoms] have two kinds of causal capacity [viz. the
common capacity and the specific capacity]. [Among these, the first one,
i.e.] the common [capacity of the atoms] is, for instance, the capacity to
[cooperatively] produce a [common] effect, such as containing water, which
the color and other atoms forming a jar have. [The second one, i.e.] the specific
[capacity] is, for instance, the capacity to [separately] produce a specific effect,
such as a visual cognition. … There are many substances [in this world].
[However, when one says the word ghaṭa] it serves to exclude those things
which are not the cause of a single effect, such as containing water, e.g. a
tree (vṛkṣa), among them. Why [is the word ghaṭa used in this manner]?
It is because the word ghaṭa is conventionally agreed to mean those atoms
which have the causal power to cooperatively produce a single effect [such as
containing water]. Intending to say that multiple things [i.e. color and other
atoms] have the single [common] capacity [to produce a single effect, such
as containing water], one uses the singular word ghaṭaḥ in accordance with
conventional agreement. Therefore, the singular word can be used to refer
even to multiple things without inconsistency.14
13
14
gandhaḥ sa tathā nirdiśyate” iti yad ucyate parais tad api pratyuktam eva … /). Ravigupta uses the
example “fragrance of a flower.” See PVV(R) D342b1–2, P196a6–8.
It seems to me that Manorathanandin’s explanation is of doubtful accuracy. PVV(M) 48,4–5 (=Ms
10a7–b1): tatsāmānyopasarjanā ghaṭatvasāmānyaviśeṣitās teṣāṃ rūpādīnāṃ śaktibhedā rañjanādayaḥ khyāpyante / Cf. PVP D45a7–b3, P51a6–b1: bum pa yi ni gzugs sogs shes // tha snyad byed pa
na / de’i spyi nye bar byas pa can yin / bum pa’i spyi nye bar byas pa’i bye brag de’i nus pa’i khyad
par gang la yod pa zhes bya bar tshig rnam par sbyar ro // de’i nus pa’i khyad par te / gzugs la sogs
pa de dag gi so sor nges pa’i nus pa’i khyad par ci ’dra ba de ’dra ba ston par byed do // bum pa la
sogs pa gsal ba’i khyad par la yod pa can gzugs la sogs pa mig gi rnam par shes pa la sogs pa ’bras bu
so sor nges pa can gyi bdag nyid bum pa’i khyad par gyi sgras rnams par gcod par byed de / chu ’dzin
pa la sogs pa’i khyad par can gyi rang bzhin ston par byed ces bya ba’i don to //; PVV(R) D342a7–b1,
P196a4–6: bum pa’i gzugs la sogs pa zhes brdar btags pa ’dis ni bum pa’i spyi de khyad par du dmigs
kyis ’ger ba gang yin pa’i gzugs la sogs pa de rnams kyi nus pa so sor nges pa’i bye brag mig gi rnam
par shes pa skyed par byed par ston te // gzugs la sogs pa chu len pa la sogs pa’i dgos pa can zhes bya
ba’i tha tshig go //
PVP D44b1–5, P50a5–b2: gzugs la sogs pa’i nus pa ni (pa ni D; pa’i P) rnam pa gnyis te / spyi’i ni dper
na bum pa’i dbyibs can dag chu la sogs pa ’dzin pa la sogs pa’i ’bras bu byed pa’i nus pa lta bu’o // so
sor nges pa’i ni dper na mig gi rnam par shes pa la sogs pa’i ’bras bu byed pa’i nus pa lta bu’o //* …
rdzas du ma yod na yang gzugs la sogs de dag gi (gi Corr.; gis DP) chu la sogs pa ’dzin pa la sogs pa’i
’bras bu gcig gang yin pa de’i rgyu ma yin pa shing la sogs pa gang yin pa de bzlog pa’i phyir ’jug par
’gyur ro** // ci’i phyir zhe na / ’bras bu gcig byed par nus par bstan pa’i phyir*** / de ltar brdar byas
pa nyid kyi phyir ro // gzugs la sogs pa mang po dag la yang / nus pa gcig yod pa de brjod par ’dod pa
la / brda’i rjes su byed pas bum ba zhes bya ba’i tshig gcig la ’gal ba yod pa ma yin no // (*Cf. Vibh p.
48, n. 1: dvidhā rūpādīnāṃ śaktiḥ sāmānyā yathā ghaṭāder udakāharaṇādi[-janikā?] / pratiniyatā ca
cakṣurvijñānādijanikā / **Cf. TR 23*,29–25*,1: teṣām eva rūpādīnāṃ1 sādhāraṇaṃ yat phalam uda-
160
Two Kinds of Causal Capacity: sāmānyā śaktiḥ and pratiniyatā śaktiḥ
Devendrabuddhi distinguishes between the two kinds of causal capacity with the respective
terms “the common capacity” (sāmānyā śaktiḥ) and “the specific capacity” (pratiniyatā
śaktiḥ).15 According to him, both of them belong to the color and so on (rūpādi) that
constitute a jar, and not to a jar that can be regarded as the whole or as conventional
existence. Namely the color atoms that constitute a jar together with other atoms have both
the causal capacity common to other constituent atoms and the causal capacity not common
to other kinds of constituent atoms. The common capacity is for producing a single effect,
such as containing water, cooperatively with other atoms, and therefore it is common to
other constituent atoms such as “smell” atoms.16 On the other hand, the specific capacity is
for producing a specific effect, such as visual cognition, and therefore it is not common to
other kinds of constituent atoms such as smell atoms.
It is also significant that a singular expression is applicable to a plural referent. The
assumption that a word in singular form must reflect its referent’s singularity is denied here.
According to Devendrabuddhi, the singular expression ghaṭaḥ can be used to refer to the
multiple atoms that can cooperatively produce a single effect. In this respect, the singular
form corresponds to the single capacity or to the capacity of producing a single effect, not
to a single referent.
Although a number of modern scholars have attributed the theory of two kinds of causal
capacity to Prajñākaragupta, it is clear that before Prajñākaragupta, Devendrabuddhi had
already explained Dharmakīrti’s theory using the terms sāmānyā śaktiḥ and pratiniyatā
śaktiḥ.
Next, we turn to Prajñākaragupta’s explanation in his Pramāṇavārttikālaṅkāra (PVA).
[4] Without implying the individual powers that are separately fixed (pratiniyataśaktibheda) for [each component of a jar], like color [atoms], the word “jar”
(ghaṭa) is used to mean their causal power of [cooperatively] producing one
and the same [effect], such as containing water (samānodakadhāraṇaśakti).
15
16
kāharaṇam, tatra ye hetavo na bhavanti, ghaṭasanniveśe ’navasthitā2 rūpādayaḥ, tebhyo vyavacchede
ghaṭaśrutiḥ / [1 Yaita’s correction ghaṭādīnāṃ should be disregarded. 2 -sanniveśenāvasthitā Yaita.]
***Cf. Vibh p. 47, n. 4: … ekakāryakā(a)raṇaśaktikhyāpanāya /) Interestingly, parallel sentences to
this passage are quoted in Haribhadra Sūri’s Anekāntajayapatākā. AJP Vol. II, 36, 6–9: syād etat – siddhasādhanam etad uktam eva naḥ pūrvācāryaiḥ – “dvividhā hi rūpādīnāṃ śaktiḥ sāmānyā pratiniyatā
ca / tatra sāmānyā yathā ghaṭasanniveśinām udakādyāharaṇādikāryakaraṇaśaktiḥ, pratiniyatā yathā
cakṣurvijñānādikāryakaraṇaśaktir iti” / atrocyate … Esho Mikogami focused on this passage of AJP
in his studies; however he did not notice that it has certain parallels to Devendrabuddhi’s commentary.
See Mikogami 1978: 2, 1979: 80. Similar sentences to this PVP passage are also found in Ravigupta’s
commentary. PVV(R) D341b7, P195b3: gang gi phyir bum pa la gnas pa’i gzugs la sogs pa’i nus pa ni
gnyis te / chu len par byed pa dang so sor nges pa’i mig la sogs pa’i rnam par shes pa skyed par byed
pa’o //
The word pratiniyatā śaktiḥ literally means “the causal capacity separately fixed for each [type of
cause].”
Haribhadra Sūri, in his auto-commentary on AJP, explains the Buddhist theory of two kinds of causality as follows: AJPSV 36,24–28: tatra sāmānyā śaktiḥ yathā ghaṭasanniveśināṃ rūpādīnāṃ udakādyāharaṇādikāryakaraṇaśaktiḥ, yataḥ sarvair eva ghaṭair etat kriyate / pratiniyatā tu rūpādīnāṃ
śaktir yathā cakṣurvijñānādikāryakaraṇaśaktir iti / rūpaṃ hi cakṣurvijñānam eva janayati, na rasādivijñānam; evaṃ rasādayo ’pi rasādivijñānāny eveti / According to this passage, he understands that
the common capacity of color and so on is the capacity common to all jars. Judging from the context of
PV II, it is clear that this is a misunderstanding.
Masahiro Inami
161
Therefore, a coreferential expression such as “the color and so forth are a jar”
(rūpādayo ghaṭaḥ) cannot be used.17 Thus, intending to define some causal
power of the aggregate (samudāyaśakti), we use such an aggregate-word (samudāyaśabda) [like the word “jar”]. … [Then, the following question may be
raised:] How is the [non-coreferential] expression “the color and so on of a
jar” (rūpādayo ghaṭasya) used? [We answer this as follows: It is used to] mean
that each of those things that can cooperatively produce a common effect, i.e.
containing water, can separately produce its own specific effect, such as visual
cognition.18
We should notice that Prajñākaragupta paraphrases the common capacity as “the causal
power of the aggregate” (samudāyaśakti) here. This makes it clear that the common capacity
is a causal capacity of the aggregated atoms.
As for the pratiniyatā śaktiḥ, Śākyabuddhi gives a more detailed explanation in his
Pramāṇavārttikaṭīkā (PVṬ):
[5] The sentence “The specific [capacity] is, for instance, …” [in PVP] can be
explained as follows: Color [atoms] can produce [their specific] effect, visual
cognition, and smell [atoms] can produce [their specific] effect, olfactory
cognition. In the same way, for the other [types of atoms, such as taste atoms,
their specific capacity] should be mentioned in accordance with their own
cognition.19
The pratiniyatā śaktiḥ has been understood by scholars to mean a unique power belonging
to a certain object, such as a jar, to produce its own particular perception. However, this
passage clearly shows that, in the case of a jar, the pratiniyatā śaktiḥ is the causal capacity
that is separately fixed for each type of cause; this belongs to each constituent atom. For
example, the color atoms of a jar have the causal capacity of producing a visual sensation,
which is separately fixed for the color atoms, and the smell atoms of a jar have the causal
capacity of producing an olfactory sensation, which is separately fixed for the smell atoms.
17
18
19
As pointed out by Katsura 2002: 270, Prajñākaragupta additionally remarks that expressions such as
rūpādayaḥ ghaṭāḥ and rūpādīnāṃ ghaṭaḥ are accepted by people who have learned the Buddhist system
of knowledge and have arrived at correct understandings, but not by ordinary people who blindly follow
other systems of knowledge. According to Prajñākaragupta, someone who has learned the Buddhist
theory will be convinced that a jar is nothing but the color and other atoms that have the causal capacity
of producing their common effect, such as containing water, or that a jar is the color atoms and so on,
which are arranged in a specific shape to produce such an effect. See PVA 98,17–21.
PVA 98, 13–17: rūpādīnāṃ pratiniyataśaktibhedam anākṣipya teṣu samānodakadhāraṇaśaktyākṣepeṇa
ghaṭaśrutiḥ pravartate / tato na rūpādayo ghaṭa iti samānādhikaraṇatā / ata eva samudāyaśaktivivakṣāyām ayaṃ samudāyaśabdaḥ / … kathaṃ tarhi rūpādayo ghaṭasyeti vyapadeśaḥ / udakāharaṇasādhāraṇakāryā rūpādipratyayajananasamarthāḥ pratyekam ity arthaḥ / Cf. Mikogami 1978: 6,
1979: 84. As pointed out by Mikogami, this passage of PVA is quoted together with PV II 100–102 and
criticized in NVinV ad NVin I 9.
PVṬ D110b6–7, P135a7–8: so sor nges pa ni dper na mig gi rnam par shes pa zhes bya ba la sogs pa
la / gzugs ni ’bras bu mig gi rnam par shes pa byed par nus pa yin la / dri ni sna’i rnam par shes pa
byed par nus pa yin no // de bzhin du gzhan dag gi (gi Corr.; gis DP) yang rang gi rnam par shes pa la
ltos (ltos D; bltos P) nas brjod par bya’o //
162
Two Kinds of Causal Capacity: sāmānyā śaktiḥ and pratiniyatā śaktiḥ
Accordingly, the term pratiniyatā śaktiḥ is used from the viewpoint that these different
roles are performed by different constituents of the same aggregate.20
The above examination of the commentaries makes it clear that in PV II 100–102,
Dharmakīrti introduced the theory of two kinds of causal capacity to account for noncoreferential expressions such as ghaṭasya rūpādayaḥ (the color and so on of a jar) without
admitting the existence of a whole. Dharmakīrti believes that these two words are used to
refer to the two different types of causal capacity, of which both belong to the aggregated
atoms. Namely, the word ghaṭa (jar) is used to mean their common capacity, and the word
rūpādi (color and such) is used to mean their specific capacities. The non-coreferential
expression can be made because the word ghaṭa simply means the common capacity
without implying any specific capacity. And besides, the singular word ghaṭasya can be
used to refer to the multiple aggregated atoms because they commonly have a single effect.
3. Dharmakīrti’s other works
Next, we will examine whether this concept of two kinds of causal capacity is found in
Dharmakīrti’s other works. I would like to show here that there are several passages related
to this concept.
When discussing the relationship between words and the causal capacity of objects in
PV I and in his auto-commentary (PVSV), Dharmakīrti explicates the causal capacities of
the respective causes and their complex.21 He explains that the reason why a single word
is applied to a plural referent is because manifold things share the capacity of producing
a single effect. According to him, someone who wants to simplify one’s practical action
can collectively designate such manifold things by using a single word in conformity with
verbal conventions.22 Dharmakīrti classifies such manifold things as one of three types: 1)
aggregate (samūha), 2) continuum (santāna), and 3) a particular state (avasthāviśeṣa).23
Of these, he illustrates the case of an aggregate using the example of a jar as follows:
[6] In the case where an entire set of causes produces some single effect [e.g.
containing water], a worldly person, who thinks that there is no use to express
20
21
22
23
The specific capacity of a color atom is often explained by commentators to be the causal capacity to produce visual sensation. However, in explaining the specific capacity of a color atom, a few
commentators refer to the causal capacity for coloring things such as clothes. Cf. PVV(M) 47,12: ghaṭavyapadeśabhājāṃ rūpādīnām avāntararañjanādiśaktibhedānām anākṣepeṇa …; PVV(M) 48,5–6: …
teṣāṃ rūpādīnāṃ śaktibhedā rañjanādayaḥ khyāpyante / ghaṭavyapadeśaviṣayasamudāyāntargataṃ
rañjanakṣamarūpaṃ niṣkṛṣyocyata ity arthaḥ /; TSP ad TS 323: vividhāḥ nānāprakārāḥ arthakriyāḥ
rūpādīnāṃ paryāyāṇāṃ samānāsamānabhedāt tatra samānā jalasaṃdhāraṇādilakṣaṇāḥ asamānā
vastrarāgalocanādijñānotpādalakṣaṇāḥ …; HBṬ 98,22–23: ghaṭenodakāharaṇaṃ kriyate, rūpādibhiḥ
punar vastrarāga* (Corr.; vasturāga S) iti kāryabhedaḥ / It has been explained by modern scholars that
specific capacity means the causal capacity to produce sense perception. However, this is not strictly
true. The capacity to produce sense perception must be understood as a typical example of the specific
capacity.
The passages of PV I and PVSV dealt with here have been explained in Vetter 1964: 110–112, Dunne
2004: 353–360, and Katsura 2002: 271–272, and translated in Vora and Ota 1980.
See PV I 137–138; 141–142.
As is well known, these three were argued as being non-existent by Dignāga in his Qu yin jia she lun
(取因假設論 *Upādānopādāyaprajñaptiśāstra). See Ui 1958.
Masahiro Inami
163
the differences among them because all of them have no difference in this
respect, uses a single word, e.g. “jar” (ghaṭa), to refer to all of them at once.
Though they are all [ultimately] distinct from their respective homologues and
heterologues, [we conventionally consider] there to be no difference among
them for the reason that, in terms of their causing that single effect, they are
equally differentiated from others [that do not contribute to the production
of that effect]. Therefore, all of them are understood by the single word “jar”
without distinguishing one from another.24
[7] [Question: If you insist that a jar is nothing but an aggregate of color and
other atoms and reject that a jar as the whole (avayavin) exists distinctly from
its parts, how do you explain a non-coreferential expression such as “the color
and so on of a jar” (ghaṭasya rūpādayaḥ)? Answer:] In that case, even if the
genitive expression “the color and so on of a jar” is used, [it does not indicate
that a jar exists distinctly from color and so on]. It simply means that the
color and other [atoms together] have the nature of a jar. In other words, this
means that [the color and other atoms] can [cooperatively] produce a [single]
effect such as “a specific way of containing water” (udakadhāraṇaviśeṣa).25
Those [atoms] are known to have the nature to produce an effect [e.g. visual
sensation] common (sāmānyakārya) to other atoms of the same kind, and are
established by the words “the color and so on” (rūpādi). But [when the words
are used with the modifier “of a jar” (ghaṭasya)] those atoms are specified by the
specification that indicates that they can produce a specific effect (viśiṣṭakārya)
[e.g. containing water] as well. Such things are meant by the words “the color
and so on of a jar” (ghaṭasya rūpādayaḥ). However, in the case [of the word
ghaṭa], distinctly [from the color and other atoms], there does not exist any
substance [such as a whole] having characteristics just as you described them,
because such a substance cannot be perceived at all. Moreover, the reason
why the singular word ghaṭaḥ is used [for a plurality of things] is because it
refers to their capability of cooperatively producing a single effect [such as
containing water], or, rather, just because it depends on the verbal convention
(saṃketa).26
24
25
26
PVSV 68, 7–13: ye samastāḥ kiṃcid ekaṃ kāryaṃ kurvanti teṣāṃ tatra viśeṣābhāvād apārthikā
viśeṣacodaneti sakṛt sarveṣāṃ niyojanārtham ekam ayaṃ lokaḥ śabdaṃ teṣu niyuṅkte ghaṭa iti / te
’pi sajātīyād anyataś ca bhedāviśeṣe ’pi tatprayojanāṅgatayā tadanyebhyo bhidyanta ity abhedāt tato
’viśeṣeṇa pratīyante /
It should be noted that the example of common capacity is expressed by the word udakāharaṇaviśeṣādi-.
“Containing water” regarded as the effect means a specified effect, not a general one. Namely, the
common capacity is, for example, a certain jar’s capacity for containing water in its particular way, but
not the capacity to contain water which is common to all jars. It is clear that, in this respect, the understandings of modern scholars should be corrected. Haribhadra Sūri, a Jaina philosopher, made a similar
misunderstanding. See n. 15. The expression with the word viśeṣa is also found in the commentaries on
Dharmakīrti’s works. See VNṬV 25,27–28: bahavo rūpagandha-rasasparśā udakadhāraṇaviśeṣādikāryanirvartanasamarthatvād …; PVA 98,21: yataḥ saṃniveśaviśeṣād udakadhāraṇaviśeṣaḥ /
PVSV 68, 13–18: tatra ghaṭasya rūpādayo ity api ghaṭasvabhāvā rūpādaya udakadhāraṇaviśeṣādikāryasamarthā iti yāvat / sāmānyakāryasādhanaprasiddhenātmanā rūpādiśabdaiḥ prasiddhā
viśiṣṭakārya-sādhanākhyena viśeṣeṇa viśiṣṭās ta evam ucyante / na punar atrānyat kiṃcid yathāvarṇi-
164
Two Kinds of Causal Capacity: sāmānyā śaktiḥ and pratiniyatā śaktiḥ
A jar should be regarded as a complex of multiple atoms, and not as a single whole. Each
of the constituent atoms of the same jar is differentiated from its respective homologues
and heterologues. However, the atoms are all equally differentiated from other atoms that
are not constituents of the jar, in respect to their contributions to the production of the same
single effect. Therefore, the constituent atoms are conventionally regarded as non-different,
and are referred to by the single word “jar.”
Moreover, the word rūpādayaḥ (the color and so on) refers to those atoms that have
different causal powers to produce their specific effects, such as visual perception. These
causal powers are common to other atoms that are not constituents of the jar. But in the
expression ghaṭasya rūpādayaḥ, the atoms are identified by the specification indicating
that they can produce another effect, e.g. containing water, and the atoms that are meant
are those which are the constituent atoms of a jar.27
It should be noted that Dharmakīrti mentions here another reason why the singular word
ghaṭaḥ is used. He adds that the usage of words is dependent on verbal conventions. This
solution seems preferable to him. In this respect, it is not absolute that words correspond
to the causal capacity of objects. This point will be examined again below.
The theory of the two kinds of causal capacity is clearly found in these passages of
PVSV. There, unlike what we find in PV II, an effect such as visual cognition is regarded
as a common effect (sāmānyakārya) in that it is common to atoms of the same kind that
do not constitute the jar, and an effect such as “containing water” is regarded as a specific
effect (viśiṣṭakārya) in that it is not common to other atoms that do not constitute the jar.
However, what he intends to explain here is the same as in PV II.
In the Hetubindu (HB), Dharmakīrti illustrates the two different aspects of causal
capacity using the example of the production of a jar. It should be noted that here he argues
about the causes of the production of a jar, viz. a lump of clay, a potter, strings, etc., and
not about the constituents of a jar, such as the color atoms.
[8] [Objection:] If a single effect [i.e. a visual sensation] is produced by the
cooperative causes, viz. a visual organ and so on, each of which is differentiated from others in its essence, then effects would not differ in accordance
with the difference of the cause. [Answer:] There is no such fault, because the
specific features effectuated by the contribution of causes cannot be mixed
together since the causes differ amongst themselves according to their respective essences and separately contribute to different features of the [effect].
27
talakṣaṇaṃ dra-vyam asti tasya tādṛśasyānupalambhāt / ekavacanam api tadekaśaktisūcanārthaṃ
saṃketaparatantraṃ vā /
Cf. PVṬ D160b7–161a2, P187a7–b1: ’dir don ni ’di yin te / gzugs la sogs pa’i sgra thams cad kyi
gzugs la sogs pa bsgrub par bya ba’i ’bras bu tsam gyi (gyi P; gyis D) nus pa dang ldan pa’i gzugs
la sogs pa khyad par med par bstan cing / bum pa’i sgras kyang khyad par du byas pa’i ’bras bu
dang ldan pas / snam bu la sogs pa’i rang bzhin dag las tha dad pas khyad par du byas pa’i gzugs
la sogs pa bstan to // de bas na bum pa’i gzugs la sogs pa zhes bya ba la sgra gnyis kyi bya bas spyi
dang khyad par gyi rnam pa’i blo skye ba’i [phyir] spyi dang khyad par gyi dngos po la tha dad pa’i
rten can gyi drug pa sbyor ba yin no // = PVSVṬ 271,20–24: ayam atrārthaḥ / rūpādiśabdā rūpādīn
rūpasādhyakāryamātraśaktiyuktān aviśeṣeṇa pratipādayanti / ghaṭaśabdas tu viśiṣṭakāryaśaktiyogena
paṭādisvabhāvebhyo ’pi rūpādibhyo bhedena viśiṣṭān rūpādīn āha / ato ghaṭasya rūpādaya iti śabdadvayavyāpāreṇa sāmānyaviśeṣākārabuddhyutpatteḥ sāmānyaviśeṣabhāvo vyatirekavibhaktiś ca
(-vibhaktiś ca M100b3; -vimatiś ca S) prayujyata iti /
Masahiro Inami
165
For example, a jar, made by some cooperative causal factors, viz. a lump of
clay, a potter, strings, etc., has [the following different features given by them]:
[Firstly, the jar has] the essential feature [of being made of clay], by which it
is differentiated from non-clay things such as trees. This feature is given by its
material cause, i.e. a lump of clay. [Secondly,] the jar, a product of clay, has
another essential feature, a certain characteristic form, by which it is differentiated from other [products of clay]. This second feature is given by its maker, i.e.
the potter. [Thirdly,] the jar, which is a product of clay and has a characteristic
form, has still another essential feature by which it is differentiated from other
clay things having a characteristic form, such as a jar made with a potter’s
wheel. This third feature is given by its instrumental cause, i.e. the strings.
In this way, [there is no confusion of effects]. The essential feature of being
made of clay is [given by the clay], not by the potter. The essential feature
of a characteristic form is [given by the potter], not by the clay. Moreover,
although the scope or range of the specific efficiencies of these two, i.e. clay
and a potter, are different from each other, a certain individual thing affected
by them, i.e. a certain jar, is [identical, but] not different in its essence. For
if it were not identical, a specific form and clay [of the same jar] would not
appear together because they would be essentially different from each other.28
A jar is produced by the contribution of causal factors, viz. a lump of clay, a potter, strings,
etc. Each of them contributes to the production of the same jar in one way or another. In
this respect, they all have the same single effect. However, each causal factor has its own
scope or range as well. For example, a lump of clay contributes to the jar’s feature of being
made of clay. The potter contributes to the jar’s feature of a specific form. In this respect,
every causal factor has its own specific effect as well. Thus, in the case of a causal complex,
every cause has two aspects: one is that it produces the same single effect cooperatively
with others, and another is that it produces its own specific effect. These two aspects are
also illustrated using the example of a visual perception.29
Here Dharmakīrti explains that, in the case of a causal complex, each cause has two
kinds of capacity. This is close to the explanation that each constituent atom of a jar has two
kinds of causal capacity, or rather his explanation of atoms’ two kinds of causal capacity is
based on his theory of causal complexes. It is clear that he regards aggregated atoms as a
kind of causal complex.
The common capacity of aggregated atoms is explained in the Vādanyāya (VN) as
follows:
28
29
HB 9, 13–10, 4: bhinnasvabhāvebhyaś cakṣurādibhyaḥ sahakāribhya ekakāryotpattau na kāraṇabhedāt kāryabhedaḥ syād iti cet, na yathāsvaṃ svabhāvabhedena tadviśeṣopayogatas
tadupayogakāryasvabhāva-viśeṣāsaṅkarāt, yathā mṛtpiṇḍakulālasūtrādibhyo bhavato ghaṭasya mṛtpiṇḍād amṛtsvabhāvebhyo vṛkṣādibhyo bhinnaḥ svabhāvaḥ kulālāt tasya eva mṛdātmanaḥ sataḥ saṃsthānaviśeṣātmatayā tadanyebhyo bhinnaḥ sūtrāt tasya eva mṛtsaṃsthānaviśeṣātmanaś cakrāder vibhaktaḥ
svabhāvo bhavati; tad evaṃ na kulālān mṛtsvabhāvatā, na mṛdaḥ saṃsthānaviśeṣaḥ; na ca tayoḥ
śaktiviśeṣaviṣayabhedo ’pi tajjanita-viśeṣabhedasya kāryasya svabhāvabhedaḥ, mṛtsaṃsthānayor aparasparātmatayā saṃsthānamṛdrūpābhyāṃ tayor apratibhāsanaprasaṅgāt / This passage is examined
and translated in Katsura 1983: 104–105.
HB 10,19–11,9. See Katsura 1983: 106.
166
Two Kinds of Causal Capacity: sāmānyā śaktiḥ and pratiniyatā śaktiḥ
[9] Moreover, only the multiple [atoms], including color [atoms], that cooperatively produce a single effect [e.g. containing water] should be referred to by
the single word ghaṭaḥ. There is no need to assume an object other [than the
atoms as being a referent of the word ghaṭa]. This is because even multiple
things can produce a single effect, as is seen in the case of visual organs and
so on. … In that case, if we, referring to multiple things that contribute toward
the same purpose collaboratively or separately,30 use one word in order to
express that they are such things, what kind of problem would occur? Because
an aggregate word (samudāyaśabda) is used to mean their causal capacity [to
produce a single effect], there is no inconsistency in the singular form. The
one and the same (eka) causal power belongs to the things that accompany
each other, but not to the things that are separate from each other. Therefore,
when an aggregate word is used to refer to one aggregate, the singular word
ghaṭaḥ is [used for the referent].31
The constituents of the same jar operate simultaneously and produce a single effect, such
as containing water. All of them contribute to the single effect in one way or another.
This type of causal capacity of the constituents is considered similar to the capacity of
a causal complex to produce a visual perception. To express that they are such things,
one uses the single word ghaṭa in the singular form. It is considered an aggregate word
(samudāyaśabda).32 Here again, it is clear that Dharmakīrti refers to a common capacity in
order to explain the singular expression of the aggregate, which he regards as one kind of
causal capacity of a causal complex.33
30
31
32
33
The words saha pṛthag vā are understood in various ways by the commentators. In Vinītadeva’s
understanding, these words mean the two different kinds of causal capacity, viz. the common capacity
and the specific capacity (see VNṬ D164b3–6, P59b1–4). However, in this understanding, the words
pṛthag vā seem incompatible with the word ekaprayojanāḥ. In contrast, in Śāntarakṣita’s understanding,
Dharmakīrti has added the words pṛthag vā due to his intention of saying that “every” component of a
complex contributes to producing a single effect. (See VNṬV 31,8–30. Here Śāntarakṣita mentions
two different understandings and refutes both.) In this understanding, the words seem compatible with
the word ekaprayojanāḥ, but Śāntarakṣita dares to ignore the contrasting meanings of the two words,
viz. saha and pṛthag. My tentative understanding is that Dharmakīrti intends to say here that multiple
things can collaboratively produce a single effect and can separately contribute to the different features
of that single effect.
VN 6, 19–7, 4: ghaṭa ity api ca rūpādaya eva bahava ekārthakriyākāriṇa ekaśabdavācyā bhavantu, kim
arthāntarakalpanayā / bahavo ’pi hy ekārthakāriṇo bhaveyuś cakṣurādivat / … tatra ye ’rthāḥ saha
pṛthag vā ekaprayojanās teṣāṃ tadbhāvasthāpanāya [hy] ekaśabdo* niyujyate yadi, kiṃ syāt / tadarthakriyāśaktisthāpanāya niyuktasya samudāyaśabdasyaikavacanavirodho ’pi na asty eva / sahitānāṃ
sā śaktir ekā na pratyekam iti samudāyaśabda ekasmin samudāye vācya ekavacanaṃ ghaṭa iti / [*sthāpanāyaikaśabdo (?); -sthāpanāya hy ekaśabdo Much’s ed., -khyāpanāya hi ekaśabdo Śāstri’s ed.]
See Much 1991: 17–18.
The word ghaṭa is a jātiśabda (generic term) under certain circumstances. See n. 11. In the Vādanyāya,
Dharmakīrti explains that jātiśabda is related to causal capacity as well. The plural word vṛkṣāḥ (trees)
is used for plural trees to mean the different causal powers of trees for producing different effects,
whereas the singular word vṛkṣaḥ (a tree) is used for the same trees to mean the causal power of trees
for cooperatively producing a single effect. See VN 7,5–7. Śāntarakṣita refers to ekapratyavamarśa as
the single effect produced by the plural trees. See VNṬV 32,17–18.
In the Vādanyāya, while explaining that conceptual cognition and verbal expression cannot establish
any real existence of the objects corresponding to them, Dharmakīrti points out that some words are
Masahiro Inami
167
As we have seen, the theory of two kinds of causal capacity is clearly found in Dharmakīrti’s works. He consistently argues that each cause of a causal complex has two different
kinds of capacity.34
We are now at the stage of evaluating earlier studies. The following points should be
made: First, in the case of a jar, the common capacity is the causal capacity of the constituent
atoms to produce a single effect cooperatively with each other, but not the causal capacity
of the jar, which is common to all jars.35 On the other hand, the specific capacity is the
causal capacity of each constituent atom to produce its own specific effect, such as visual
perception, but not the causal capacity of the jar to produce its own perception, which
is not common to other jars. Second, these two kinds of causal capacity both belong to
the same element of a complex, for example, in the case of a jar, to the same constituent
atoms, such as the color atoms. Therefore, it is not incorrect to say that these two belong
to the respective things. Third, when discussing the expression “the color and so on of
a jar,” Dharmakīrti introduces the theory of two kinds of causal capacity in order to
explain the usage of such expressions without admitting a whole. Therefore, he can never
say that a common capacity belongs to a conventional existent (saṃvṛtisat), such as a
whole. Fourth, it has been reported that whereas a specific capacity belongs to real things,
34
35
used to refer to non-existent things, past things, future things, and so forth, and that some singular words
are used to refer to plural things that produce a single effect, and some plural words are used to refer to
a single thing that produces different effects. See VN 6,1–8. Dharmakīrti points out that although we
can treat something as real on the basis of its efficiency, its unity or plurality is not dependant on its
efficiency. See VN 8,17–20.
Haribhadra Sūri attributes the theory to “ancient masters” (pūrvācārya). See n. 13. According to his
auto-commentary, the expression “ancient masters” refers to Dharmapāla, Dharmakīrti, and so forth.
AJPSV Vol. II, 36,23: … pūrvācāryaiḥ dharmapāladharmakīrtyādibhiḥ … Mikogami (1978: 3, 1979:
81–82) refers to the relevant passage of the Cheng Weishi Lun (成唯識論). However, it seems to me
that this passage is irrelevant to this theory. After Dharmakīrti, Śāntarakṣita and Kamalaśīla deal with
the theory of two kinds of causal capacity in criticizing the Jaina view of the two aspects of ātman,
viz., dravya (substance) and paryāya (mode). See TS 323; TSP 154,6–1. (Cf. HBṬ 98,14–102,18.)
There, Kamalaśīla explains that aggregated atoms have two different causal powers: the causal power to
cooperatively produce a common effect such as containing water, and the causal powers to separately
produce different effects such as coloring clothes. It is noted that Śāntarakṣita and Kamalaśīla use the
word arthakriyā here. In this context this word seems to be used simply to mean “producing effect.”
See Inami 2014: Appendix I.
The capacity of multiple things to cooperatively produce a single effect should be rigorously distinguished from their capacity to produce similar effects separately. The resemblance of the different effects
is conceptually constructed. (Cf. PVSV 56,12–14: yad apy udakāharaṇādikam ekaṃ ghaṭādikāryaṃ
tad api pratidravyaṃ bhedād bhidyata eveti naikaṃ bhedānāṃ kāryam asti /) Of course we may think,
as Dharmakīrti explains in PVSV, that the different effects can produce a single result, i.e. “a single
comprehensive judgement” (ekapratyavamarśa), namely, that there is no difference among them. (Cf.
PVSV 57,1–3: tad api pratidravyaṃ bhidyamānam api prakṛtyaikapratyavamarśasyābhedāvaskandino
hetur bhavad abhinnaṃ khyāti /) However, as long as the single effect is the judgement that there is
no difference among them, non-difference is judged by the conceptual cognition. Thus Dharmakīrti
mentions that each effect appears as if it were not different from other effects (abhinnaṃ khyāti). The
concept of ekapratyavamarśa is introduced to interpret the commonality or identity of different things
by pointing out that they have a common single effect (ekakaryatā). But it should be noted that the
single effect, ekapratyavamarśa, is in itself the conceptual judgement of identity. In addition, the term
ekapratyavamarśa has been understood by some scholars to mean “a judgement of sameness.” But
eka- does not mean “sameness,” but instead means “single” or “one and the same.” Here Dharmakīrti
intends to say that different things produce a single effect.
168
Two Kinds of Causal Capacity: sāmānyā śaktiḥ and pratiniyatā śaktiḥ
viz. particulars (svalakṣaṇa), momentary things (kṣaṇa) and so on, a common capacity
belongs to conceptually constructed things, viz. common characteristics or universals
(sāmānyalakṣaṇa), the continuum (santāna) and so on. However, it is clear that these both
belong to real things. Conceptual construction does not have any causal capacity.36 Fifth,
the “fulfillment of a human’s purpose,” which is reported to be the meaning of the word
arthakriyā, is not mentioned in Dharmakīrti’s explanation of common capacity. Common
capacity and specific capacity are both causal capacities to produce an effect.
4. Dignāga’s Pramāṇasamuccaya (-vṛtti) I 4cd
As is shown above, the specific capacity (pratiniyatā śaktiḥ) is, for instance, the causal
power of the color atoms of a jar to produce a visual perception, or the causal power of the
smell atoms of the same jar to produce an olfactory perception. However, the single atom
or the atoms separately situated can never produce a sense perception. Only when gathered
together can homologous atoms produce sensations, which is their single effect. Hence, in
this respect, an atom’s capacity for producing a sense perception collaboratively with other
homologous atoms can be regarded as a kind of common capacity (sāmānyā śaktiḥ).
As is well known, such commonness or similarity of the aggregated atoms is referred
to by Dignāga in the Pramāṇasamuccaya and in his auto-commentary (PS(V)). He defines
perception as cognition that is free from conceptualization and insists that its object is
the particular (svalakṣaṇa). This leads to the following two questions: (1) Why is sense
perception, which grasps aggregated atoms as one, free from conceptualization? (2) Why
is it stated in the Abhidharma treatise that sense perceptions do not have particulars
36
A continuum does not have any capacity to produce an effect because it is nothing but a conceptual
construction. Of momentary elements arising in a series, the element immediately before its effect,
affected by the preceding elements, is regarded as producing the effect. Or, rather, all momentary
elements arising in a series are thought to contribute to producing a single effect. In the latter case, the
multiple momentary elements arising in a series share the ability of producing a single effect. This leads
us to understand that these momentary elements also can be regarded as a kind of causal complex. They
are, so to speak, a complex of the multiple causes situated in different times. Such different momentary
elements share a commonality in that they collaboratively produce a single effect, not that the effects of
momentary elements resemble each other. The resemblance of different things is due to a conceptual
judgement. It should be added that the momentary elements arising in a series have a commonality in
that they all are based on the same single cause or on the same single complex of causes. Dharmakīrti
explains that the momentary rice plants arising in a series share a commonality in that all of them
are dependent on the same single seed, and therefore they are mentioned in a single word śāli (a rice
plant) in accordance with the verbal convention. Thus, many different things can have a commonality,
either in their having a single common effect (ekakāryatā) or in their having the same single complex
of causes (ekasāmagyadhīnatva). This is also true for the commonality of spatially different things.
But, care must be taken not to conclude that Buddhists regard united things as a single entity. They do
not mean that multiple things can be regarded as a single entity, but simply that such multiple things
have a single effect or a single cause. Cf. PVSV 68, 18–21: tathā ye hetuphalaviśeṣabhūtāḥ kiṃcid
ekaṃ sādhayanti sādhyante vā te ’pi sakṛt pratyayārthaṃ vrīhyādiśabdaiḥ kṛtasaṃketāḥ kathyanta iti
pūrvavad vācyam /; PVA 188,7: ekasāmāgryadhīnatvāt samudāyaprakalpane / ekakāryatvayogena
yatra tatreti kalpanā //
Masahiro Inami
169
as their object with respect to the particular characteristics of the substantial elements
(dravyasvalakṣaṇa)?37
[10] [Objection:] If you insist that aggregated atoms are not conceptually accepted as one by perception, then why is it stated [in the Abhidharma treatise]
that five groups of sensation have aggregated [atoms] as their objects? And
besides, why is it stated [in the Abhidharma treatise] that these [groups of
sensation] have particulars as their object with respect to the particular characteristics of the sense spheres (āyatanasvalakṣaṇa), but not to the particular
characteristic of the substantial elements (dravyasvalakṣaṇa)? [Answer:]
In the [above Abhidharma passages], [perception that is directed] toward
its particular object [in respect to the particular characteristic of the sense
sphere] is stated to have a sāmānya as its object [with respect to the particular
characteristic of the substantial elements], because it is produced by many
substantial elements [i.e. atoms]. (4cd)
[Perception that occurs] in its own sense sphere is stated to have a sāmānya
as its object [with respect to the particular characteristic of the substantial
elements], because it is produced by many substantial elements [i.e. atoms],
but not because the different [atoms] are conceptually accepted as non-different
from each other.38
Answering the two questions, Dignāga explains that sense perception is produced by
multiple atoms. According to him, sense perception is said to have aggregated atoms as
its object not because it conceptually grasps aggregated atoms as one, but because it is
produced by the aggregated atoms. Therefore, it is free from conceptualization. Moreover,
37
38
Cf. AKBh1 7,20–21 = AKBh2 7,22–24 ad AK I 10d: nanu caivaṃ samastālambanatvāt sāmānyaviṣayāḥ
pañca vijñānakāyāḥ prāpnuvanti na svalakṣaṇaviṣayāḥ / āyatanasvalakṣaṇaṃ praty ete svalakṣaṇaviṣayā iṣyante na dravyasvalakṣaṇam ity adoṣaḥ /; Vibhāṣā 655a26–b7: 問十一觸中極多縁幾發生身
識。有作是説、一一別縁發生身識、以十一種相用増故。有餘師言、極多縁五發生身識、謂
四大種滑等隨一。復有説者、總縁十一、亦生身識。問豈不五識唯取自相境耶。答自相有二
種、一事自相、二處自相。依事自相説縁十一種觸生於身識。依處自相説五識身取自相境。
是故無過。如是説者、縁十一事亦生身識。如縁色處二十種事亦、生眼識。此亦應爾。故五
識身通縁總別而無五識取共相過。多事自相一識能縁然不明了。; 65a12–15: 問云何身識縁共
相境以五識身縁自相故。答自相有二種、一事自相、二處自相。若依事自相説者五識身亦縁
共相。若依處自相説則五識唯縁自相。See Sakurabe 1969: 154.
PS(V) I 4cd: kathaṃ tarhi sañcitālambanāḥ pañca vijñānakāyāḥ, yadi tad ekato na vikalpayanti / yac
cāyatanasvalakṣaṇaṃ praty ete svalakṣaṇaviṣayā na dravyasvalakṣaṇam iti / tatrānekārthajanyatvāt
svārthe sāmānyagocaram // 4 // anekadravyotpādyatvāt tat svāyatane sāmānyaviṣayam uktam, na
tu bhinneṣv abhedakalpanāt / Dharmakīrti explains this passage of PS(V) as follows: “Owing to the
connection with other things, [i.e., other atoms], atoms that are different [from their own previous
moments] arise. They are said to be ‘aggregated’ (sañcita). They are causes of the production of
[sensual] cognition [unlike previous ones]. Such a special characteristic of the atoms does not occur
without other atoms [connected with them]. Therefore, since the object is not limited to a single [atom],
[but rather to aggregated atoms], the perceptual cognition is said to have a universal [in the sense of
aggregated atoms] as its object.” (PV III 194–196.) Cf. PSṬ 44,12–45,5; PSṬ 45,11–1. See Hattori
1968: 26–27, 88–91; Tosaki 1979: 294–298; Dunne 2004: 98–113; Chu 2008: 212–215; Yoshida 2011:
152–156, etc.
170
Two Kinds of Causal Capacity: sāmānyā śaktiḥ and pratiniyatā śaktiḥ
its object is said to be sāmānya in respect to dravyasvalakṣaṇa, not because it has a
conceptualized object, but because it is produced by aggregated atoms rather than by a
single atom.
The idea that multiple things cooperatively produce a single effect is basically the same
as the idea of sāmānyā śaktiḥ. Consequently, it is reasonable to suppose that the atom’s
efficiency in producing a sense perception collaboratively with other homologous atoms
can be regarded as a kind of common capacity.
It should be noted that the aggregated atoms that produce a single sense perception are
called sāmānya here. The word means neither universals nor commonness in this context.
It is being used in a similar fashion as is the word sāmānya in the compound tatsāmānya
(PV II 102b. See [2]). The aggregated atoms themselves that commonly have the sāmānya
śaktih can also be called sāmānya.39
Moreover, it is significant that Dignāga refers here to two kinds of particular (svalakṣaṇa): particulars concerning the sense spheres (āyatanasvalakṣaṇa) and particulars
concerning substantial elements (dravyasvalakṣaṇa). The aggregated atoms are regarded as
svalakṣaṇa with respect to the former, and as sāmānya, which as mentioned above means
“aggregate,” with respect to the latter. The aggregated atoms have two different aspects. In
the first aspect, for example, the color atoms are differentiated from other types of atoms
such as smell atoms. The color atoms never produce olfactory sensations. The ability of
the aggregated atoms is limited and separately defined in each type. This ability is nothing
but the specific capacity (pratiniyatā śaktiḥ) mentioned by later Buddhists. On the other
hand, in the second aspect, the aggregated atoms are differentiated from a single atom that
cannot produce any sense perception by itself.40 This aggregate’s ability of cooperatively
producing a single effect can be regarded as a kind of common capacity, as mentioned
above. In this way, Dignāga’s idea that the aggregate becomes the object of sense perception
can be interpreted in relation to the theory of two kinds of causal capacity. However, he
does not directly mention the two different kinds of causal capacity of atoms. We can safely
say that the groundwork for the theory of the two kinds of causal capacity was prepared by
Dignāga.41
In fact, Dharmakīrti’s argument concerning the two kinds of causal capacity in PV
II arises in the context of the explanation that aggregated atoms have special features for
producing their own cognition, unlike a single atom or atoms that are situated separately.42
39
40
41
42
In Tibetan logic, the word spyi is considered to have three different meanings: 1) tshogs spyi, 2) rigs
spyi, and 3) don spyi. Of these, 1) tshogs spyi seems to be related to the aggregate called sāmānya by
Dignāga and Dharmakīrti. See Dreyfus 1997: 107–109.
Jinendrabuddhi (and perhaps Dignāga as well) regards a single atom as dravyasvalakṣaṇa, unlike
Yaśomitra, who explains that the dravyasvalakṣaṇa means, for example, specific colors, such as blue.
See PSṬ 45,14–46,3; AKV 28,10–16.
Before Dignāga, it was debated in the Abhidharma tradition whether multiple things can be perceived
at once through a single cognition. See Vibhāṣā 63c22–25; 64a11–25. Dignāga insists in this respect
that aggregated atoms are grasped at once through a single perception. Following Dignāga’s insistence,
Dharmakīrti holds that even different aggregates of atoms are grasped through a single cognition. Not
only color atoms of the same kind (e.g. blue atoms) are grasped through a single cognition, but also
color atoms of different kinds (e.g. blue, yellow and red atoms). This is how Dharmakīrti develops his
argument of citrādvaita in PV III.
PV II 86cd–87: aviśeṣād aṇutvāc ca na gatiś cen na sidhyati // 86 // aviśeṣo viśiṣṭāṇām aindriyatvam
ato ’naṇuḥ /etenāvaraṇādīnām abhāvaś ca nirākṛtaḥ // 87 // See Inami 2013.
Masahiro Inami
171
The theory of causal complex, according to which multiple things can cooperatively produce
a single effect, seems to be one of the most important foundations for his epistemology and
ontology.
5. Is causal capacity absolute?
As we have seen above, Dharmakīrti additionally remarks in PVSV ([7]) that the usage of
words is dependent on verbal conventions. Judging from this, it is not absolute that words
must correspond to the causal capacity of objects. There are similar comments in his other
works. For example, in the Vādanyāya, after explaining the correspondence between words
and causal powers, Dharmakīrti states the following:
[11] However, we [Buddhists] never stick to [our explanation of the relationship
between words and causal powers] because [words simply] operate to denote
or connote their objects fixed by a verbal convention as the verbal convention
governs.43
This passage clearly shows that Dharmakīrti does not overly stress his theory of two kinds
of causal capacity. It is not absolute for him. He ultimately thinks that words are dependent
only on verbal conventions.44
In PV I and PVSV Dharmakīrti demonstrates that the relationship between words and
objects is not intrinsic but extrinsic because it is based only on the verbal conventions
established by humans. According to him, the different forms of words do not have any
relation to external objects. For example, the words dārāḥ (pl., wife), ṣaṇṇagarī (sg., six
cities) and so on do not correspond to their referents in number.45 Dharmakīrti also makes
a sarcastic comment about the false view that words are strongly connected to external
objects as follows:
[12] Words, which can produce no appearance of external things and which
can merely be causes of conceptual cognition in accordance with preceding
repeated experiences, operate to denote or connote objects, depending only
on human intentions. Nevertheless, [some linguists and philosophers seriously] consider the application of words to objects, and determine real things
separately in accordance with words. They just display their own ignorance.46
43
44
45
46
VN 7, 8–9: asmākan tu sāṃketikeṣv artheṣu saṃketavaśād vṛttir ity anabhiniveśa eva /
See VNṬV 32,25–26: saṃketavaśāc chabdānām bahuvacanāntānāṃ dārāḥ sikatāḥ pādā gurava ityādinā ’saty api bahutve ’bhidheyasya vṛttiḥ / tathā saty apy anekatve ṣaṇṇagarī ṣaṭpūlī vanam ityādinaikavacanāntānāṃ vṛttir ity anabhiniveśa eva / ko hi nāma sacetanaḥ puruṣābhiprāyamātrādhīnavṛttiṣu
śabdeṣv abhiniveśaṃ kartum utsahata iti bhāvaḥ / Cf. PV IV 116a–c: saṃketasaṃśrayāḥ śabdāḥ sa ca
icchāmātrasaṃśrayaḥ / nāsiddhiḥ śabdasiddhānām …
See PV I 67: dārāḥ ṣaṇṇagarītyādau bhedābhedavyavasthiteḥ / khasya svabhāvaḥ khatvaṃ cety atra vā
kiṃ nibandhanam //
PVSV 36, 9–12: vācyeṣu puruṣāyattavṛttīnāṃ śabdānām avastusaṃdarśināṃ yathābhyāsaṃ vikalpaprabodhahetūnāṃ pravṛtticintā tadvaśād vastuvyavasthāpanaṃ ca kevalaṃ jāḍyakhyāpanam /
172
Two Kinds of Causal Capacity: sāmānyā śaktiḥ and pratiniyatā śaktiḥ
In the ultimate sense, words themselves do not have any direct relation to external objects.
Words can be used to denote or connote objects and are dependent only on the speaker’s will.
Words can also produce no appearance of external things.47 Nevertheless, many thinkers
attempt to find the applications of words in external objects, and to determine external
objects in accordance with words. According to Dharmakīrti, they should feel deep regret
for their follies.48
Interestingly, in this context, he presents the objection that a singular word can be used
to refer to multiple things because it signifies their causal capacity to produce one and the
same effect. He refutes this objection by pointing out that capacity never exists apart from
real being.
[13] [Objection:] Such [aggregated] things [including houses, humans, and
so on] share the same capacity for [producing] some [single effect]. This
[capacity] is the grounds [for the singular word nagaram (a city)]. [Answer:]
This is not correct because capacity is not different from the things themselves.
Or, if [the capacity] were different, [it would not be dependent on the thing]
because the capacity, which cannot be assisted by anything, is not dependent
on [the thing]. Or, if the capacity (C1) could be assisted [by another capacity
(C2)], the capacity (C2) assisting the capacity (C1) would be different [from
the capacity (C1), and capacity (C2) should be assisted by still another capacity
(C3)]. In this way, a series of capacities would be infinite, and therefore we
would not understand [the capacity at all]. Or, if the [capacity (C2)] is not
different [from capacity (C1)], the first one (C1) would not be different from
the thing. Therefore, this [objection] is a matter of indifference [to us].49
We should notice that what is refuted by Dharmakīrti here is exactly the same as his own
theory of sāmānyā śaktiḥ. Here he has turned around and is criticizing the causal capacity.
According to this passage, causal capacity never exists in reality, it can only be a conceptual
construction. Dharmakīrti denies the realistic view of causal capacity.50 This reminds us of
the fact that he denies causal capacity from the viewpoint of the ultimate truth in PV III.51
The theory of two kinds of causal capacity cannot ultimately contribute to satisfactorily
explaining the application of words to referents. It is only used as a conventional theory
for the philosophical argument. Dharmakīrti presents his argument only by tentatively
47
48
49
50
51
A similar argument is found in Śāntarakṣita’s Tattvasaṃgraha. TS 570: ṣaṣṭhīvacanabhedādi vivakṣāmātrasaṃbhavi / tato na yuktā vastūnāṃ tatsvarūpavyavasthitiḥ // See Inami 2014: Appendix
II.
Prajñākaragupta also argues that there is no exact correspondence between words and external objects,
pointing out that every language has inconsistencies in grammatical rules. Furthermore, he severely
criticizes those who insist that Sanskrit is the only perfect language. Cf. PVA 99,3–12. See Inami 2014:
note 22.
PVSV 37, 9–14: tathābhūtānāṃ kvacid arthe ’bhinnā śaktir asti sā nimittam iti cet / na / śakter
vasturūpāvyatirekāt / vyatireke vānupakāryasya pāratantryāyogāt / upakāre vā śaktyupakāriṇyā api
śakter vyatireka ity anavasthiter apratipattiḥ / tadavyatireke vā ādyāyām api prasaṅga iti yat kiṃcid
etat /
As is well known, Dharmakīrti denies the realistic view of the relation in the same manner in Sambandhaparīkṣā. See SP vv. 4–5.
See Inami 2000.
Masahiro Inami
173
accepting that words have some relation to the external world. From the viewpoint of the
ultimate truth, he holds that how words are used depends merely on verbal conventions
established by humans and never reflects reality. Furthermore, he rejects the realistic view
of causal capacity. Several concepts are introduced to explain worldly affairs and these
concepts have some consistency. However, they are just conceptual constructions and are
not absolute. We should not adhere to them.52
6. Conclusion
In conclusion, I would like to state the following five points:
1. In the example of a jar, two kinds of causal capacity belong to the constituent atoms
of the jar, such as the color atoms, but neither to the jar as a whole (avayavin) nor to
the jar as a continuum (santāna).
2. In the case of the color atoms of the jar, common capacity (sāmānyā śaktiḥ) is their
causal capacity to produce a single effect, such as containing water, cooperatively
with other kinds of atoms of the same jar; specific capacity (pratiniyatā śaktiḥ) is
their causal capacity to produce their own specific effect, such as visual perception.
3. The capacity of the color atoms to produce a visual perception is regarded as the
specific capacity in comparison to other kinds of atoms, such as smell atoms. However,
it can be regarded as the common capacity in the sense that only aggregated atoms
can have this capacity.
4. The theory of two kinds of causal capacity is advocated by Dharmakīrti throughout
his works. The groundwork for this theory was prepared by Dignāga.
5. When discussing the expression “the color and so on of a jar,” Dharmakīrti introduces the theory of two kinds of causal capacity in order to explain the expression
without admitting a whole. However, he does not stick exclusively to this theory. He
finally abandons the theoretical explanation of the relationship between words and
objects and adopts the view that words are just dependent on the verbal conventions
established by humans. From the viewpoint of the ultimate truth, causal capacity is
also merely a conceptual construction.
References and abbreviations
AJP; AJPSV Anekāntajayapatākā; Anekāntajayapatākāsvopajñavyākhyā: Anekāntajayapatākā by Haribhadra Sūri. With his own commentary and Municandra Sūri’s
supercommentary, ed. Hiralal R. Kapadia. 2 vols. Baroda 1940, 1947.
AK; AKBh Abhidharmakośa; Abhidharmakośabhāṣya: Abhidharmakośabhāṣyam of
Vasubandhu, ed. Prahlada Pradhan. Patna 1 1967, 2 1975.
AKV Abhidharmakośavyākhyā: Sphuṭārthā Abhidharmakośavyākhyā. The work of Yaśomitra, ed. Unrai Wogihara. Tokyo 1936, 1971, 1989.
52
Cf. PV III 209: idam vastubalāyātaṃ yad vadanti vipaścitaḥ / yathā yathā ’rthāś cintyante viśīryante
tathā tathā // 209 //
174
Two Kinds of Causal Capacity: sāmānyā śaktiḥ and pratiniyatā śaktiḥ
Bae 2011 Bae Gyeong-a, Prajñākaragupta ni okeru kotoba no imiron [The Semantics of
Prajñākaragupta]. Journal of Indian Buddhist Studies 59-2 (2011) 90–93.
Chu 2008 Junjie Chu, On Dignāga’s Theory of the Object of Cognition as Presented in
PS(V). Journal of the International Association of Buddhist Studies 29-2 (2008)
211–253.
Dreyfus 1997 Georges B. J. Dreyfus, Recognizing Reality. Dharmakīrti’s Philosophy and
Its Tibetan Interpretations. Albany 1997.
Dunne 2004 John D. Dunne, Foundations of Dharmakīrti’s Philosophy. Boston 2004.
Funayama 2012 Toru Funayama, Ninshikiron – Chikaku no riron to sono tenkai [Epistemology: the theory of perception and its development]. In: Ninshikiron to ronrigaku.
Tokyo 2012, 91–120.
Hattori 1968 Masaaki Hattori, Dignāga, On Perception, being the Pratyakṣapariccheda
of Dignāga’s Pramāṇasamuccaya. Cambridge, Massachusetts 1968.
HB Hetubindu, Sanskrit text reconstructed in: Dharmakīrtis Hetubindu. Teil I: Tibetischer
Text und rekonstruierter Sanskrit-text, ed. Ernst Steinkellner. Wien 1967.
HBṬ Hetubinduṭīkā: Hetubinduṭīkā of Bhaṭṭa Arcaṭa with the Sub-commentary Entitled
Āloka of Durveka Miśra, ed. Pandit Sukhlalji Sanghavi and Muni Shri Jinavijayaji.
Baroda 1949.
Inami 2000 Masahiro Inami, astu yathā tathā. In: Indian Culture and logic. Dr. Hiromasa
Tosaki Felicitation Volume. Fukuoka 2000, 359–397.
Inami 2012 Masahiro Inami, Nishu no ingakōryoku –sāmānyā śakti to pratiniyatā śakti–
[Two Kinds of Causal Capacity: sāmānyā śakti and pratiniyatā śakti]. Journal of
Indian and Buddhist Studies (Indogaku Bukkyōgaku Kenkyū) 61-1 (2012) (82)–(87).
Inami 2013 Masahiro Inami, Pramāṇavārttika Pramāṇasiddhi-shō no kenkyū (12) [A
Study of the Pramāṇasiddhi Chapter of Pramāṇavārttika (12)]. Bulletin of Tokyo
Gakugei University, Division of Humanities and Social Sciences II 64 (2013) 89–114.
Inami 2014 Masahiro Inami, Pramāṇavārttika Pramāṇasiddhi-shō no kenkyū (13) [A
Study of the Pramāṇasiddhi Chapter of Pramāṇavārttika (13)]. Bulletin of Tokyo
Gakugei University, Division of Humanities and Social Sciences II 65 (2014) 71–94.
Inami and Tillemans 1986 Masahiro Inami and Tom J. F. Tillemans, Another Look at
the Framework of the Pramāṇasiddhi Chapter of Pramāṇavārttika. Wiener Zeitschrift
für die Kunde Südasiens XXX (1986) 123–142.
Katsura 1983 Shoryu Katsura, Dharmakīrti no ingaron [Dharmakīrti’s Theory of Causality]. Nanto Bukkyo 50 (1983) 96–114.
Katsura 1984 Shoryu Katsura, Dharmakīrti’s Theory of Truth. Journal of Indian Philosophy 12 (1984) 215–235.
Katsura 2002 Shoryu Katsura, Sonzai towa nanika – Dharmakīrti no shiten [On What
There Is: From Dharmakīrti’s Point of View]. Bulletin of Research Institute for
Buddhist Culture Ryukoku University 41 (2002) 263–275.
Katsura 2012 Shoryu Katsura, Bukkyō-ronrigaku no kozō to sono igi [The Structure of
Buddhist Logic and its Significance]. In: Ninshikiron to ronrigaku. Tokyo 2012,
3–48.
Mikogami 1978 Esho Mikogami, Mono ni sonawaru fuhenteki-kinō to tokushuteki-kinō
[Sāmānyā śakti and Pratiniyatā śakti: Dharmapāla, Dharmakīrti, and Prajñākara-
Masahiro Inami
175
gupta]. Bulletin of Research Institute for Buddhist Culture, Ryukoku University 17
(1978) 1–15.
Mikogami 1979 Esho Mikogami, Some Remarks on the Concept of Arthakriyā. Journal
of Indian Philosophy 7 (1979) 79–94.
Much 1991 Michael T. Much, Dharmakīrtis Vādanyāya. Teil II Übersetzung und Anmerkungen. Wien 1991.
Nagatomi 1967/68 Masatoshi Nagatomi, Arthakriyā. Adyar Library Bulletin 31–32 (1967/
68) 52–72.
NVin; NVinV Nyāyaviniścaya; Nyāyaviniścayavivaraṇa: Nyāyaviniścayavivaraṇa of Vādirājasūri, the Sanskrit Commentary on Bhaṭṭa Akalaṅkadeva’s Nyāyaviniścaya, ed.
Mahendra Kumar Jain. 2 vols. Delhi 1944.
PS I; PSV I Pramāṇasamuccaya, Chapter I; Pramāṇasamuccayavṛtti Chapter I: Sanskrit
text reconstructed by Ernst Steinkellner. https://www.oeaw.ac.at/fileadmin/Ins
titute/IKGA/PDF/forschung/buddhismuskunde/dignaga_PS_1.pdf, last visited
10-12-2016.
PSṬ Pramāṇasamuccayaṭīkā: Jinendrabuddhi’s Viśālāmalavatī Pramāṇasamuccayaṭīkā,
Chapter 1, Part 1: Critical Edition, ed. E. Steinkellner, H. Krasser, and H. Lasic.
Beijing/Vienna 2005.
PV I; PVSV Pramāṇavārttika, Chapter I (=Svārthānumāna); Pramāṇavārttikasvavṛtti: The
Pramāṇavārttikam of Dharmakīrti, the First Chapter with the Autocommentary. Text
and Critical Notes, ed. Raniero Gnoli. Roma 1960.
PV II, IV Pramāṇavārttika, Chapter II (=Pramāṇasiddhi), Chapter IV (=Parārthānumāna):
Yusho Miyasaka, Pramāṇavārttikakārikā (Sanskrit and Tibetan). Acta Indologica 2
(1972). [The Pramāṇasiddhi Chapter is treated as Chapter I in Miyasaka’s Edition.]
PV III Pramāṇavārttika, Chapter III (=Pratyakṣa). Sanskrit text in: Tosaki 1979.
PVA Pramāṇavārttikālaṅkāra: Pramāṇavārtikabhāshyam or Vārtikālaṅkāraḥ of Prajñākaragupta. (Being a commentary on Dharmakīrti’s Pramāṇa-vārtikam), ed. Rāhula
Sāṅkṛtyāyana. Patna 1953.
PVmsH A Sanskrit manuscript of Pramāṇavārttika (from the collection kept by Hemarāja
Śarman) printed in: Birgit Kellner and Francesco Sferra, A Palm-leaf Manuscript of
Dharmakīrti’s Pramāṇavārttika from the Collection kept by the Nepalese rājaguru
Hemarāja Śarman. Manuscripta Buddhica 1 Sanskrit texts from Giuseppe Tucci’s
Collection Part I, ed. F. Sferra. Roma 2008.
PVP Pramāṇavārttikapañjikā, Tibetan translation. D No. 4217. P No. 5717.
PVSVṬ Pramāṇavārttikasvavṛttiṭīkā: Pramāṇavārttikasvavṛttiṭīkā of Karṇakagomin, ed.
Rahula Sāṅkṛityāyana. Ilāhābād 1943.
PVṬ Pramāṇavārttikaṭīkā (Śākyabuddhi), Tibetan translation. D No. 4220. P No. 5718.
PVV(M) Pramāṇavārttikavṛtti (Manorathanandin): Dharmakīrti’s Pramāṇavārttika with
a commentary by Manorathanandin, ed. Rāhula Sāṅkṛtyāyana. The Journal of the
Bihar and Orissa Research Society XXIV, XXV, XXVI (1938–40).
PVV(R) Pramāṇavārttikavṛtti (Ravigupta) ad PV II, Tibetan translation. D No. 4224; P
No. 5726.
Sakurabe 1969 Hajime Sakurabe, Kusharon no kenkyū kai-kon-hon [A Study of Abhidharmakośa: the chapters on elements and faculties]. Kyoto 1969.
176
Two Kinds of Causal Capacity: sāmānyā śaktiḥ and pratiniyatā śaktiḥ
SP Sambandhaparīkṣā: Vādanyāyaprakaraṇa of Acharya Dharmakīrti with the Commentary Vipañcitārthā of Acharya Śāntarakṣita and Sambandha-parīkṣā with the
commentary of Acharya Prabhāchandra, ed. Swami Dwarikadas Shastri. Varanasi
1972.
Tosaki 1979 Hiromasa Tosaki, Bukkyō-ninshikiron no kenkyū, jōkan [A Study of Buddhist
Epistemology, vol. I]. Tokyo 1979.
TR Tarkarahasya, ed. H. Yaita. In: Hideomi Yaita, Three Sanskrit Texts from the Buddhist
Pramāṇa-Tradition. Narita 2005.
TS; TSP Tattvasaṃgraha; Tattvasaṃgrahapañjikā: Tattvasaṃgraha of Ācārya Shāntarakṣita with the Commentary ‘Pañjikā’ of Shri Kamalashīla, ed. Swami Dwarikadas
Shastri. 2 vols. Varanasi 1 1968, 2 1981.
Ui 1958 H. Ui, Jinna chosaku no kenkyū. Tokyo 1958.
Vādarahasya Vādarahasya (also called Udayananirākaraṇa, author unknown), ed. R.
Pandhey. New Delhi 1984.
Vetter 1964 Tilmann Vetter, Erkenntnisprobleme bei Dharmakīrti. Wien 1964.
Vibh Vibhūticandra’s Note to Manorathanandin’s PVV. Sanskrit text edited in the footnotes
of the Appendix to the Sāṅkṛtyāyana’s edition of PVV(M).
Vibhāṣā Abhidharmamahāvibhāṣāśāstra, Chinese translation (阿毘達磨大毘婆沙論).
Taishō Shinshū Daizōkyō (大正新修大蔵経), vol. 27, No. 1545.
VN Vādanyāya: Dharmakīrtis Vādanyāya. Teil I Sanskrit-Text, ed. Michael T. Much. Wien
1991.
VNṬ Vādanyāyaṭīkā (Vinītadeva), Tibetan translation. D No. 4240; P No. 5737.
VNṬV Vādanyāyaṭīkā Vipañcitārthā (Śāntarakṣita): Vādanyāyaprakaraṇa of Acharya
Dharmakīrti with the Commentary Vipañcitārthā of Acharya Śāntarakṣita and
Sambandhaparīkṣā with the commentary of Acharya Prabhāchandra, ed. Swami
Dwarikadas Shastri. Varanasi 1972.
Vora and Ota 1980 Pradyumna R. Vora and Shinkai Ota, A Translation of Pramāṇavārttika I and Svavṛtti (2). Bulletin of Saga Ryukoku Junior College 26 (1980) 1–19.
Yoshida 2011 Akira Yoshida, Pramāṇasamuccayaṭīkā dai-isshō (ad PS I 3c–5 & PSV)
wayaku [A Japanese Translation of the Pramāṇasamuccayaṭīkā (ad PS I 3cd–5
& PSV)]. Journal of Indian and Tibetan Studies 15 (2011) 134–164.
Jñānaśrīmitra on viparyayabādhakapramāṇa
by
Kyo Kano
Introduction
It was really a new attempt when Dharmakīrti introduced sādhyaviparyaye [hetoḥ] bādhakapramāṇa (hereafter VBP) as a methodological device with reference to svabhāvahetu,
especially with respect to the proof of momentariness of all existence, in his logicoepistemological system.1 Dharmakīrti’s followers discussed the role and logical content of
VBP in the pramāṇa system mainly in the context of the proof of kṣaṇabhaṅga.
With reference to the theory, several studies have brought into focus its impact on the
whole logico-epistemological theory of later Buddhist logicians.2 Above all Tani, in his
great work, focuses on the relation between VBP and prasaṅga/prasaṅgaviparyaya and
their roles, and discusses how Dharmakīrti’s followers evaluate them in their proof of
momentariness.3 This theory, in fact, raised new complications with respect to the structure
of the means of valid cognition and its logical and epistemic base.4
One of these complications is the relation between VBP and establishing anvaya(‑vyāpti)
as well as vyatireka(‑vyāpti). It seems unclear in both traditional and contemporary interpretations whether VBP is valid for establishing either affirmative concomitance (anvaya)
or negative concomitance (vyatireka), or whether it is valid for both at the same time. It is of
course not easy to find a direct answer to this question by an Indian logician. But in an effort
to do so, this paper will begin by focusing on Jñānaśrīmitra’s descriptions of VBP, especially
in the vyatireka section of the Kṣaṇabhaṅga chapter of the Jñānaśrīmitranibandhāvali
(JNA). At the beginning of this chapter, an opponent called viparyayabādhakapramāṇavādin appears, whose claims are criticized by Jñānaśrīmitra. By analyzing the relevant texts,
we may be able to discern where Jñānaśrīmitra stands on the issue.
As a preliminary observation, it may help us to keep in mind that anvaya-vyāpti and
vyatireka-vyāpti are accepted as logically equivalent, at least by the thinkers after Dharmakīrti. In the epistemic process, however, we cannot simply presume their equivalence, since
1
2
3
4
Cf. Steinkellner 1982, 1991.
Cf. Tani 1999, Woo 1999, Shiraishi 2005, Sakai 2014.
Tani’s understanding of the VBP’s interpretations by Dharmakīrti’s followers, such as Dharmottara and
Prajñākaragupta, are scattered throughout his work (1999), but are found in chapter 3 in particular.
For instance: First, VBP is sometimes identified with or regarded as being based on vyāpaka[dharma]anupalabdhi, at least by Dharmakīrti and some logicians after him. Is VBP an inference based on
vyāpakānupalabdhi or it cannot necessarily be identified with such an inference? Second, how is VBP
related to prasaṅga and prasaṅgaviparyaya? If VBP is different from these, what is the difference
between prasaṅga/prasaṅgaviparyaya and VBP formally, logically, and content-wise? Or is the latter
the same as the former, especially prasaṅgaviparyaya, since both of them have the same style of proof
formulation. The late Prof. Kajiyama assessed the matter as follows: “I think that Kamalaśīla was quite
right when he identified viparyaye bādhakapramāṇam with a prasaṅga. … With Kamalaśīla, I also
regard bādhakapramāṇa as prasaṅga in essence” (Kajiyama 1999: 37). He did not, however, mention
the passage upon which his estimation depends.
Birgit Kellner et al., eds., Reverberations of Dharmakīrti’s Philosophy: Proceedings of the Fifth International Dharmakīrti
Conference Heidelberg, August 26 to 30, 2014. Vienna 2020, pp. 177–189.
178
Jñānaśrīmitra on viparyayabādhakapramāṇa
it depends on what kind of independent pramāṇa or other subsidiary means of cognition
the determination of anvaya or vyatireka is related to or based upon, as will be discussed
in the following. We can suppose this is one of the reasons Jñānaśrīmitra and Ratnakīrti
treat them separately in their discussion on the proof of momentariness.
Jñānaśrīmitra’s Kṣaṇabhaṅgādhyāya consists of three pādas. In the first pāda, pakṣadharmādhikāra, the theme is that in the formula “yat sat tat kṣaṇikaṃ yathā jaladharaḥ,
santas tu bhāvā ime” (Whatever is existent is momentary, for instance a rain cloud. And
these things are existent. [Therefore, these things are momentary.]), the reason sat/sattva is
not “unestablished” (asiddha). In the second pāda, anvayādhikāra, it is not incompatible
(viruddha). And in the third pāda, vyatirekādhikāra, which I take up in the following, it is
not inconclusive (anaikāntika).
1. The relation between two kinds of pervasion, anvaya and vyatireka
The critique of the viparyayabādhakapramāṇavādin in Jñānaśrīmitra’s vyatirekādhikāra
begins with the following argument:
(A) There is also no suspicion that [the reason “being existent” might be]
inconclusive, because the proof is given based on the pervasion that comprises
all [individual instances] (sarvopasaṃhāravatyā vyāpteḥ).
〈Opponent, i.e., viparyayabādhakapramāṇavādin〉 The pervasion is established by virtue of “the means of valid cognition that annuls [the reason]
in the reverse [of that which is to be proven]” (viparyayabādhakapramāṇa).
Therefore, we do not find any additional information (vārtā) by introducing it
(i.e. the pervasion that comprises all individual instances). Then, why do you
say that the pervasion is [already] established?
…
〈Opponent〉 In this way, there must only be an affirmative concomitance
(anvayamātram). Then, how can we obtain the conclusive exclusion [of the
reason] from the dissimilar domain?
〈Jñānaśrīmitra〉 It is because the pervasion is exactly established.
〈Opponent〉 How is the pervasion of the negative exclusion itself established?5
5
JNA p. 60, 3–10: nāpy anaikāntikaśaṅkā, sarvopasaṃhāravatyā vyāpteḥ prasādhanāt. nanu viparyayabādhakapramāṇavaśād vyāptisiddhiḥ, tasya ca nopanyāsavārtā. tat kathaṃ vyāptiḥ prasādhitety
ucyate? …nanv evam anvayamātram astu, vipakṣāt punar ekāntena vyatireka iti kathaṃ labhyam iti
cet. vyāptisiddher eva. vyatireke vyāptisiddhir eva katham it cet. (Ci’ RNA p. 70, 9–11 and 15–16:
na cāyam anaikāntikaḥ, atraiva sādharmyavati dṛṣṭānte sarvopasaṃhāravatyā vyāpteḥ prasādhanāt.
nanu viparyayabādhakapramāṇabalād vyāptisiddhiḥ, tasya copanyāsavārtāpi nāsti. tat kathaṃ vyāptiḥ
prasādhiteti cet. … nanv evam anvayamātram astu. vipakṣāt punar ekāntena vyāvṛttir iti kuto labhyata
iti cet. vyāptisiddher eva. vyatirekasandehe vyāptisiddhir eva katham iti cet.) [Different readings are in
bold font.]
Kyo Kano
179
From the description above, we first notice that Jñānaśrīmitra evaluates sarvopasaṃhāravyāpti in the anvya-vyatireka context of their symmetrical roles, while the opponent,
viparyayabādhakapramāṇavādin, does not.6 This is one of the crucial points dividing
Jñānaśrīmitra’s position from that of the viparyayabādhakapramāṇavādin. I will come
back to this point again in a later section. Jñānaśrīmitra then formally differentiates and
defines the two pervasions, anvayarūpā and vyatirekarūpā.
(B) 〈Jñānaśrīmitra〉 No, the pervasion is in fact of two kinds. One is that
which has affirmative concomitance as its essential characteristic (anvayarūpā)
and is a property of the subjective element (kartṛdharma). It is the inevitable
presence of that which is to be proven in the property-possessor that has the
proving [property]. And the other is that which has negative concomitance as
its essential characteristic (vyatirekarūpā) and is a property of the objective
element (karmadharma). It is the inevitable absence of the proving [property]
when that which is to be proven is absent. This is because the following is a
sound logic: The pervasion is such that the pervader presents there (i.e., in the
pervaded) without fail, or else the pervaded presents only there (i.e., in the
pervader).7
We do not find any original element in Jñānaśrīmitra’s definition of anvaya-vyāpti and
vyatireka-vyāpti except the introduction of the expressions kartṛdharma and karmadharma.
As to these terms, taking the sādhana-sādhya relation into consideration, I tentatively
interpret kartṛ and karman as corresponding to sādhana and sādhya, respectively.8 If my
understanding is correct, then it follows that Jñānaśrīmitra regards a proof as a kriyā, of
which sādhana and sādhya are constituent elements. Then, he refers to the relation between
anvaya-vyāpti and vyatireka-vyāpti as follows:
(C) And the establishment of one of these two inevitably (niyamena) hints at the
understanding of the second (i.e., the other). This is because otherwise even one
of them cannot be established. The practical activity of the valid means of cognition is carried out only directly (sākṣāt) with reference to one [of them], and
one can gain an understanding of the other as [logically] immediate (nāntarīyakatayā). The expression “one [of the two]” is employed just as far as they are
[related to each other in this way]. And in such a case, even if suspicion about
one of these two arises before a valid means of cognition of the other functions,
the suspicion will be expelled by it afterwards. Just as for those who advocate
6
7
8
The relation between VBP and sarvopasaṃhāra has been studied recently by Sakai (2014).
JNA p. 60, 10–13: a→ na, dvividhā hi vyāptiḥ, anvayarūpā ca kartṛdharmaḥ sādhanavati dharmiṇi
sādhyasyāvaśyambhāvo yaḥ, vyatirekarūpā ca karmadharmaḥ sādhyābhāve sādhanasyāvaśyamabhāvo
yaḥ,←a vyāptir vyāpakasya tatra bhāva eva vyāpyasya vā tatraiva bhāva iti nyāyāt. (a: Ci’ RNA p.
70, 17–20: na. dvividhā hi vyāptisiddhiḥ. anvayarūpā ca kartṛdharmaḥ sādhanadharmavati dharmiṇi
sādhyadharmasyāvaśyambhāvo yaḥ, vyatirekarūpā ca karmadharmaḥ sādhyābhāve sādhanasyāvaśyamabhāvo* yaḥ.) [Different readings are in bold font.] (*Emendation. RNA p. 70, 19: sādhanasyāvaśyambhāvo.)
Another interpretation is as follows: Taking the context of the vyāpaka-vyāpya relation into consideration,
kartṛ and karman can correspond to vyāpaka (i.e., sādhya) and vyāpya (i.e., sādhana), respectively.
180
Jñānaśrīmitra on viparyayabādhakapramāṇa
“the means of valid cognition that annuls [the reason] in the reverse [of
that which is to be proven] (viparyayabādhakapramāṇavādin),” affirmative
concomitance, even though it is suspected, since the suspicion is expelled by
virtue of the conclusive exclusion [of the property] from the dissimilar domain
(vipakṣād ekāntena vyāvṛttibalāt), concludes that the “being existent,” which
never happens at all in [the things] that are not momentary, is really restricted
by [the property of] momentary perishing.
…
Accordingly, there is no difference in content (na kaścid arthato bhedaḥ)
between them; viz., the conclusive exclusion of the proving [property] in the
absence of [the property] to be proven, and the inevitable concomitance of [the
property] to be proven when the proving [property] is present [in the subject
of the thesis].9
Here Jñānaśrīmitra stresses the incorporation of the understanding of anvaya and vyatireka
into the epistemic process. In the last part of the passage cited above, he says that there
is no difference in content between them. Here, we must draw attention to the underlined
portion. According to Jñānaśrīmitra’s description, the VBP-vādin claims that the suspicion
of anvaya can be expelled by the exclusion [of the property] from dissimilar domain. That
is, VBP-vādin prioritizes vyatireka over anvaya.
Jñānaśrīmitra then adds:
(D) As to the function (vyāpāra) of the means of valid cognition, however,
its practical activity (vyavahāra) is of two kinds, principally (mukhyato) and
implicatively (arthataḥ).10
On the level of daily performance, however, we find two kinds of practical activities of the
valid means of cognition, principally (mukhyato), or directly (sākṣāt), and implicatively
(arthataḥ), which follows direct cognition as [logically] immediate (nāntarīyakatayā). For
Jñānaśrīmitra, the difference between anvaya and vyatireka lies only in the level of practical
activity (vyavahāra). In Dignāga’s system as well as Dharmakīrti’s, the relation between
anvaya and vyatireka is sometimes expressed as arthāpattyā.11 If one is proven, the other
is automatically understood by arthāpatti. That is, either of these two is logically implied
in the other. Jñānaśrīmitra’s expression arthataḥ accords with this idea.
9
10
11
JNA p. 60, 13–18 and 61, 2–3: a→ enayoś caikasiddhir niyamena dvitīyapratītim ākṣipati. anyathā
ekasyāpy asiddheḥ.←a kevalaṃ sākṣād ekatra pramāṇavyāpāro ’nyatra nāntarīyakatayā pratītir iti tāvataivaikavyapadeśaḥ. tathā ca saty ekatra pramāṇapravṛtteḥ pūrvam anyatra saṃśayaḥ pravartamāno’pi
paścāt [tanyā?] tayā nirasyate. yathā viparyayabādhakapramāṇavādinām anvayaḥ sandihyamāno ’pi
vipakṣād ekāntena vyāvṛttibalāt nirastasaṃśayaḥ paryavasyaty akṣaṇike sarvathā ’nupapadyamāneyaṃ
sattā kṣaṇabhaṅganiyataiveti, …tasmāt sādhyābhāve sādhanasyaikāntiko vyatirekaḥ, sādhane sati
sādhyasyāvaśyam anvayo veti na kaścid arthato bhedaḥ. (a: Ci’ RNA p. 70, 19–20: enayoś caikatarapratītir niyamena dvitīyapratītim ākṣipati. anyathaikasyā evāsiddheḥ.) [Different readings are in bold
font.]
JNA p. 61, 3–4: pramāṇavyāpāras tu mukhyato ’rthataś ceti dvidhā vyavahāraḥ.
NMu 3a1–3 (Katsura 1981: 71–72): 若有於此一分己成、随説一分亦成能立、若如其聲両義同
許、俱不須説、或由義准一能顕二。 PSṬ (Ms) 178b6 ad PS 4.5: arthāpattyā vetyādi. …; PVSV p.
18, 17 (=PVin2, p. 53, 10): arthāpattyā vānyatareṇobhayapradarśanād iti.
Kyo Kano
181
2. sādhakapramāṇa and viparyayabādhakapramāṇa
Then Jñānaśrīmitra presents an argument of sādhaka- and viparyayabādhaka-pramāṇa.
(E) Precisely therefore, in every reasoning, if one seeks “the means of valid
cognition that annuls the reverse of that which is to be proven” [and if there
is such a means of valid cognition], there is an affirmative concomitance [of
the reason] with that which is to be proven, in the case where the proving
[property] (i.e., the reason) is present [in the subject of the thesis]. One should
know that seeking “the means of valid cognition that establishes [that which is
to be proven] (sādhakapramāṇa),” is implied (upalakṣita) by that (i.e., seeking
VBP). Furthermore, [when one introduces VBP,] both [of these two pramāṇas,
namely, sādhakapramāṇa and bādhakapramāṇa,] are really integrated (ubhayasaṃgraha eva) by the term “annulling the reverse” (viparyayabādhaka).12
If we try to prove the pervasion by VBP, we should know that sādhakapramāṇa is implied
in VBP, and accordingly, we should know that both pramāṇas are integrated in the concept
bādhakapramāṇa. The ubhaya of ubhayasaṃgraha can also be interpreted as anvaya
and vyatireka, but it seems more likely that Jñānaśrīmitra regards sādhakapramāṇa as
corresponding to anvaya and bādhakapramāṇa to vyatireka. The above-quoted passage,
however, throws into question any simple opposition between adopting sādhakapramāṇa
and adopting bādhakapramāṇa. Though the two positions are ostensibly opposed, his
argument makes little distinction between them. Rather, he seems to take an integrative or
synthetic stance toward sādhakapramāṇa(vādin) and viparyayabādhakapramāṇa(vādin),
even if he thinks one of them is prior to the other in a particular case. Then, the question
arises what kind of pramāṇa is intended by the term sādhakapramāṇa? We can surmise
that Jñānaśrīmitra’s answer is given in the subsequent discussion, so let us examine what
follows.
3. Two kinds of viparyayabādhakapramāṇa13
Jñānaśrīmitra continues by explaining the latter position:
(F) In fact, with reference to the pervasion that has negative concomitance as
its essential characteristic, [there is a means of valid cognition which] annuls
the proving [property] in the reverse of that which is to be proven. (In the proof
of momentariness, it annuls “being existent” in those that are not momentary.)
With reference to [the pervasion] that has affirmative concomitance as its
essential characteristic, on the other hand, it annuls “the reverse of that which is
12
13
JNA p. 61, 4–6: ata eva yatra yatra sādhyaviparyaye bādhakapramāṇaparyeṣaṇā, tatra sādhane
sati sādhyānvayaḥ, sādhakapramāṇaparyeṣaṇāpi tayopalakṣitā veditavyā. viparyayabādhakaśabdena
punar ubhayasaṃgraha eva.
At the last Dharmakīrti conference in 2005, Shiraishi drew attention to this passage. Regretfully,
his paper, titled “Jñānaśrīmitra’s two interpretations of bādhakapramāṇa” was not included in the
proceedings of the conference, Religion and Logic in Buddhist Philosophical Analysis, 2011.
182
Jñānaśrīmitra on viparyayabādhakapramāṇa
to be proven” [in the proving property] in the case where the proving property is
present [in the subject of the thesis]. (anvayarūpāyāṃ tu sādhyaviparyayasya
bādhakaṃ sādhane satīti) (It annuls “not being momentary” in those that are
existent.) If these two [kinds of VBP] are absent (ubhayābhāve), the inevitable
connection of the proving [property] with that which is to be proven would be
impossible. The above is intended.14
According to Jñānaśrīmitra’s understanding, VBP proves not only vyatireka but anvaya as
well, that is, it can be introduced in two ways. Furthermore, for anvaya, VBP annuls “the
reverse of that which is to be proven” in the proving [property]. This is a new interpretation
of the viparyayabādhakapramāṇa.
As is well known, in his Hetubindu, Dharmakīrti first refers to ‘sādhyaviparyaye hetoḥ
bādhakapramāṇa’ as follows:
anvayaniścayo ’pi svabhāvahetau sādhyadharmasya vastutas tadbhāvatayā
sādhanadharmabhāvamātrānubandhasiddhiḥ. sā sādhyaviparyaye hetor bādhakapramāṇavṛttiḥ. yathā yat sat tat kṣaṇikam eva, akṣaṇikatve ’rthakriyāvirodhāt tallakṣaṇaṃ vastutvaṃ hīyate.15
From the description, it is ambiguous whether or not Dharmakīrti was conscious of the
relation between ‘sādhyaviparyaye hetor bādhakapramāṇa’ and the proof of anvayavyāpti or vyatireka-vyāpti. It seems that Dharmakīrti introduced ‘sādhyaviparyaye hetor
bādhakapramāṇa’ as a subsidiary method for supporting the establishment of the inevitable
connection of anvaya, stated as anvayaniścayo ’pi, but remained strongly conscious of the
logical equality of anvaya and vyatireka, an equality that is presupposed by the relation of
the contradiction (virodha) between sādhya and sādhya-viparyaya. Logically, it is clear
that ‘sādhyaviparyaye hetor (i.e., sādhanasya) bādhakapramāṇa’ can be understood as
expressing vyatireka, since it proves that where sādhyaviparyaya is present, that is, where
sādhya is absent, hetu is absent.
According to Jñānaśrīmitra’s understanding, however, VBP also annuls “the reverse
of that which is to be proven” in the proving [property] (sādhane sādhyaviparyayasya
bādhakapramāṇa).16 This is a new interpretation of VBP, based on the interpretation
14
15
16
JNA p. 61, 6–8: vyatirekarūpavyāptau hi sādhyaviparyaye bādhakaṃ sādhanasya, anvayarūpāyāṃ tu
sādhyaviparyayasya bādhakaṃ sādhane satīti. tadubhayābhāve sādhanasya sādhyapratibandhānupapattir ity ayam arthaḥ. Cf. Shiraishi 2005: 6ff.
HB p. 4*, 5–7. Skt. is reconstructed by Steinkellner. “Die Feststellung des gemeinsamen Vorkommens
(anvayaḥ) ferner besteht beim Eigenwesen als dem Grund in dem Nachweis, daß die zu beweisende
Beschaffenheit sich an das bloße Vorhandensein der beweisenden Beschaffenheit anschließt, weil [die
zu beweisende Beschaffenheit] der Sache nach das Eigenwesen (bhāvaḥ=svabhāvaḥ) von etwas ist
[das zugleich die beweisende Beschaffenheit zu seinem Eigenwesen hat]. Dieser [Nachweis] besteht
im Auftreten eines Erkenntnismittels (pramāṇam), das den Grund im Gegenteil des zu Beweisenden
aufhebt. Z.B.: Was seiend ist, das is ausschließlich augenblicklich; wäre es nicht augenblicklich, würde
es, da [die Nichtaugenblicklichkeit] mit der Wirksamkeit in Widerspruch steht, die Dingheit verlieren,
da [die Dingheit eben] durch diese [Wirksamkeit] bestimmt ist.” (Steinkellner 1967: 37)
We should also not overlook the expression sādhane sati. The expression is presumably intended
for those who claim that pakṣadharmatā is not necessarily required, such as Ratnākaraśānti. AVS p.
86, 2–4: bādhakāt sādhyasiddhiś ced vyartho hetvantaragrahaḥ. bādhakāt tadasiddhiś ced vyartho
dharmyantaragrahaḥ.
Kyo Kano
183
of the compound (viparyayabādhakapramāṇa) as a genitive tatpuruṣa (hereafter VBP2 ),
which is in contrast to the original interpretation of the compound as a locative tatpuruṣa
(hereafter VBP1 ). According to Jñānaśrīmitra, at least, VBP1 annuls the reason “being
existent” (i.e., hetos) in “those that are not momentary” (i.e., sādhyaviparyaye). That is,
VBP1 principally or directly proves that if something is not momentary, it is non-existent,
namely, vyatireka. It also proves that if a subject of a thesis is existent, it is momentary;
that is, it proves anvaya indirectly, because being momentary and not being momentary are
contradictory (virodha). In the second new interpretation, however, VBP2 annuls “being
not momentary” (i.e., “the reverse of that which is to be proven”) in that which “is existent”
(i.e., the proving property). It proves principally or directly that if a subject of a thesis is
existent, it is momentary, that is, anvaya. Therefore, according to his understanding, not
only vyatireka but also anvaya can and should be proven by VBP (VBP1 /VBP2 ). In the
descriptions above, however, we do not concretely find both of these bādhakapramāṇas,
such as vyāpakānupalabdhi introduced by Dharmakīrti and other logicians.17 Moreover, in
the above passage, Jñānaśrīmitra emphasizes not only the logical equivalence of anvaya and
vyatireka, but the methodological equivalence of sādhakapramāṇa and bādhakapramāṇa
as well. By ubhayābhāve does he indicate that both of these VBPs are necessary? In the
next paragraph, he states, “Those who present a proof should seek for both of these (i.e.,
anvaya and vyatireka) alternatively (vikalpena).” In other words, if one wants to prove a
pervasion by VBP, one should introduce not only VBP1 but also VBP2 , which principally
establishes the anvaya relation. It is likely that what he really wants to emphasize is the
latter, VBP2 .
He continues:
(G) Otherwise, even though one cannot directly (sākṣāt) obtain the means
of valid cognition that determines negative concomitance, if one shows a
means of valid cognition that establishes the inevitability of the affirmative
concomitance (anvayaniyama) of the proving [property] with that which is
to be proven, then what is its (=the proving property’s) inevitable connection
with reference to the establishment of that which is to be proven? It is because [the means of valid cognition that establishes the inevitability of the
affirmative concomitance] hints at the ability of negative concomitance as well
(that is, there is no inevitable connection of affirmative concomitance that is
independent from the ability of negative concomitance.)18
17
18
Cf. Steinkellner 1982: 2; 1991: 318. After Dharmakīrti, for instance: HBṬ p. 44, 24: etac ca bādhakaṃ pramāṇaṃ vyāpakānupalabdhirūpam uttaratrāvasaraprāptaṃ svayam eva vakṣyati. VNṬ p. 10,
27–28: idam uktaṃ bhavati. vyāpakānupalabdhir eva sahabhāvaṃ bādhate hetoḥ sādhyābhāvena.
RNA p. 83, 8: na ca viruddhānaikāntikate, vyāpakānupalambhātmanā viparyaye bādhakapramāṇena vyāpteḥ prasādhanāt. TBh(M) (Kajiyama 1966: 115, n. 309) … zhes pa khyab byed mi dmigs
pa’i mtshan nyid can bzlog pa la gnod pa can gyi tshad mas (… iti vyāpakānupalabdhilakṣaṇa(/
rūpa)viparyayabādhakapramāṇena…) [This part is dropped in the Sanskrit text to which we have
access. The Sanskrit above has been reconstructed by Kajiyama.]
JNA p. 61, 8–10: anyathā vyatirekaniścāyakapramāṇasya sākṣād alābhe ’pi yadi sādhanasya sādhyenānvayaniyamaprasādhakaṃ pramāṇam upadarśayet, kas tasya sādhyasiddhau pratibandho vyatirekasyāpi* sāmarthyākṣepāt. [*Corrected with Ms.; JNA: vyatirekasyānvayasyāpi.] Concerning the
last part, various readings are possible. The difference in the interpretation of the sentence lies in the
184
Jñānaśrīmitra on viparyayabādhakapramāṇa
Otherwise [that is, if the means of valid cognition that establishes the inevitability of the affirmative concomitance is only able to establish affirmative
concomitance, irrespective of the inevitability of negative concomitance], it
would ensue that the proof formulation of similarity should not really be presented, since the proof does not directly express [the inevitability of] negative
concomitance.
Accordingly, even though [an inevitable connection], the ability of which
one can understand, does not have negative concomitance as its essential
characteristic, exactly insofar as “wherever the proving [property is present],
that which is to be proven [is also present]” [is the essence of pervasion], if
there is the proving [property in the subject of the thesis], a wise person who is
provided with the expectation of that which is to be proven [being established]
achieves his purpose.
This is because, in fact, when the proving [property] is present [in the subject
of the thesis], even if [a property that is to be proven] is absent when it (i.e., the
proving property) is absent, if the [property] that is to be proven is inevitably
present when it (i.e., the proving property) is present, then it is fruitful to
accept the proof. Hence, a proponent who presents a proof should seek even
for both of these (i.e., anvaya and vyatireka) alternatively (vikalpena), and
even in either of these styles [a proof] which is established by a means of
valid cognition must be shown by a proponent. In the proofs of the opponent,
however, even an affirmative concomitance as having inevitability, is really
difficult to be obtained, as a negative concomitance is. We should know in this
way.19
From the above description, the position of the opponent, the viparyayabādhakapramāṇavādin, seems to be that VBP should be introduced for the proof of vyatireka and it is the
only way to prove the pervasion, while Jñānaśrīmitra equates VBP of vyatireka with that of
anvaya, which is the substance of his new interpretation. Here, Jñānaśrīmitra claims that it
is needless to state both of the means of valid cognition that prove anvaya and vyatireka;
that the statement of either one is enough because they are logically connected to each
other. Therefore, if the pervasion of anvaya is proven, it is unnecessary to state the proof of
vyatireka.
The final statement above is particularly noteworthy. Jñānaśrīmitra points out a disadvantage in the opponent’s view: according to the opponent’s proof by VBP, the relation
between anvaya and vyatireka is not clear. The point of his criticism lies not in VBP1 itself
19
understanding of “tasya sādhyasiddhau” [tasya: (a) sādhanasya, (b) pramāṇasya; sādhyasiddhau: (a)
when … is proven, (b) with reference to sādhyasiddhi].
JNA p. 61, 11–17: anyathā sādharmyaprayogo ’nupādeya eva syāt, sākṣād vyatirekānupadarśanāt. tad
yadi sāmarthyagamyo ’pi na vyatirekātmā, tadā ’pi yatra yatra sādhanaṃ tatra tatra sādhyam itīyataiva
sādhane sati sādhyapratyāśāprayuktaḥ kṛtī kṛtārthaḥ. tadabhāve ’bhāviny api hi sādhane sati yadi
tadbhāve ’vaśyaṃbhāvi sādhyaṃ tadā phalitaḥ sādhanasvīkāraḥ. tasmāt sādhanavādinā dvayam api
paryeṣaṇīyaṃ vikalpena, vādinā caiko ’pi prakāraḥ pramāṇasiddha upadarśayitavyaḥ, parasādhaneṣu
ca vyatirekavad anvayo ’pi niyamavān durlabha eveti veditavyam.
Kyo Kano
185
as a methodical device, but in the opponent’s position whereby the proof of vyatireka by
VBP1 is the one and only way to prove a pervasion.
According to Jñānaśrīmitra’s understanding, the difference between his position and
that of the VBP-vādin is as follows:
viparyayabādhakapramāṇavādin
anvaya
vyatireka
Jñānaśrīmitra’s interpretation of viparyayabādhakapramāṇa
sādhane sādhyaviparyayasya bādhakapramāṇa (VBP2 ) or
sādhyaviparyaye hetoḥ bādhaka- sādhyaviparyaye hetoḥ bādhakapramāṇa
pramāṇa (VBP1 )
(VBP1 )
We can say that the position of the opponent, viparyayabādhakapramāṇavādin, is a
propounder who claims VBP1 has the exclusive ability to establish pervasion only through
proving the vyatireka-vyāpti by VBP1 . Meanwhile, Jñānaśrīmitra regards VBP (VBP1 /
VBP2 ) as having the synthetic ability to prove pervasion.
3. Priority of prasaṅga and prasaṅgaviparyaya
Jñānaśrīmitra continues:
(H) However, for us, with reference to the discussed proof, affirmative concomitance with inevitability (niyamavān anvayaḥ) is really shown by prasaṅga
and prasaṅgaviparyaya prior [to other means of valid cognition] (paurastyābhyāṃ). And even if both are applied to the cloud presented as a similar
example, if one thing (A1 ) has the essential property of creating another (B1 ),
the former (A1 ) inevitably produces the latter (B1 ), and one thing (A2 ) does
not produce another (B2 ), then the former (A2 ) does not have the essential
property of creating the latter (B2 ). Thus, [anvaya and vyatireka], which are
functioning in this way, show that all [entities], indeed, that are endowed with
the ability of activity (kriyāśaktiyuktaṃ) are restricted to being momentary.
Therefore, comprising all cases (sarvopasaṃhāro) is the ground for hinting at
negative pervasion (vyatirekākṣepabījaṃ),20 such as smoke and the like.21
20
21
As to the usage of ākṣepa in the context of the relation between anvaya and vyatireka, see RNA
67.4: ākṣiptavyatirekā yā vyāptir anvayarūpiṇī / sādharmyavati dṛṣṭānte sattvahetor ihocyate // [The
pervasion that has anvaya as its essential characteristic and by which [its] vyatireka is hinted at …];
83.4: vyatirekātmikā vyāptir ākṣiptānvayarūpiṇī / vaidharmyavati dṛṣṭānte sattvahetor ihocyate // [The
pervasion that has vyatireka as its essence and has the essential characteristic that [its] anvaya is hinted
at … ] It seems that these two verses, which appear at the beginning of each chapter of Ratnakīrti’s
Kṣaṇabhaṅgasiddhi, in the anvaya and vyatireka chapters, respectively, and have symmetrical construction, were composed by the author (Ratnakīrti or someone else) (see Woo 1999: 141), being strongly
conscious of the discussion of the relation between anvaya and vyatireka, which Jñānaśrīmitra makes
here.
JNA p. 61, 18–22: asmābhis tu prakṛtasādhane niyamavān anvayaḥ prasaṅgaviparyayābhyāṃ paurastyābhyāṃ darśita eva, tau ca yady api sapakṣiīkṛte jalabhṛti pravartitau, tathāpi yo yatkaraṇasvabhāvaḥ
186
Jñānaśrīmitra on viparyayabādhakapramāṇa
Against the VBP-vādin, Jñānaśrīmitra argues for the inevitability of the affirmative concomitance (anvayaniyama) first and foremost through prasaṅga and prasaṅgaviparyaya. In
the above statement, the first point to note is the expression paurastyābhyāṃ; the second is
that the statement is not concerned with pervasion in general, including vyatireka(-niyama),
but only with anvayaniyama. As to the first point, the term paurastya means “prior to,”
“first,” or “preceding,” which suggests that he does not necessarily exclude the logic of
VBP, but insists only that prasaṅga and prasaṅgaviparyaya should be introduced before
introducing VBP. That is, for Jñānaśrīmitra, prasaṅga and prasaṅgaviparyaya are enough
for establishing affirmative concomitance with reference to the proof of momentariness,
whereas the VBP-vādin considers VBP to be the sole means for establishing the pervasion.
Moreover, the “comprising [of] all [individual instances] (sarvopasaṃhāro)” as a concept
having a connecting function between anvaya and vyatireka is highly esteemed by Jñānaśrīmitra. According to him, “if anvaya is established by prasaṅga and prasaṅgaviparyaya,”
sarvopasaṃhāra is regarded as “the grounds for hinting at negative pervasion (vyatirekākṣepabījaṃ),” whereas the VBP-vādin does not make much of this. The VBP-vādin’s low
evaluation of sarvopasaṃhāra suggests that, according to his/their view, vyatireka can be
proven independently (concerning each subject of the thesis) as having inevitability and is
sufficient for establishing a pervasion.
As mentioned at the beginning of this paper, the basic difference between Jñānaśrīmitra
and the VBP-vādin is the evaluation of efficacy of the proof based on the pervasion that
comprises all [individual instances] (sarvopasaṃhāravyāpti). The opponent, the VBPvādin, apparently takes a negative stance toward sarvopasaṃhāravyāpti. On this point it
is likely that the opponent, the VBP-vādin, is not Ratnākaraśānti, because Ratnākaraśānti
takes a positive stance toward the significant role of the sarvopasaṃhāravyāpti for the proof
of momentariness and his antarvyāpti-theory. He says, for instance, “as this pervasion
comprises all [individual instances], it depends upon the universal,”22 and, “for pervasion
comprising all [individual instances] is in fact an indispensable constituent (aṅga) of the
establishment of that which is to be proven.”23
4. Priority of anvaya over vyatireka
Furthermore Jñānaśrīmitra argues:
(I) Then, in this way, [the essence of] the pervasion [is that] “the pervading
[property] is inevitably present in those which have the [property of being]
pervaded.” And the establishment of momentary perishing cannot be repudiated on account of the efficacy itself of the reason “being existent” brought
about by such a pervasion as that which has affirmative concomitance as its
essential characteristic (anvayarūpā) and which is a property of the subjective
element. Thus, it is considered “what is the use of introducing the means of
22
23
sa taj janayaty eva, yo yan na janayati na sa tatkaraṇasvabhāvaḥ, ity evaṃ pravartamānau sarvam eva
kriyāśaktiyuktaṃ kṣaṇikatve niyataṃ darśayata iti sarvopasaṃhāro vyatirekākṣepabījaṃ dhūmādivat.
AVS p. 64, 4: sā ca sarvopasaṃhārāt sāmānyam avalambate /
AVS p. 66, 3: sarvopasaṃhāravatī hi vyāptiḥ sādhyasiddher aṅgam.
187
Kyo Kano
valid cognition that annuls [the proving property] in the reverse [of that which
is to be proven] in this case? (kim atra viparyaye bādhakapramāṇopanyāseneti)” Precisely for this reason, even the impossibility of introducing this (=the
means of valid cognition that annuls [the proving property] in the reverse [of
that which is to be proven]) does not hurt [the inference].
This is because that which is to be proven is established by [the means of valid
cognition] that annuls nothing but [the existence of ] the reverse [of that which
is to be proven in the proving property] (viparyayasyaiva bādhakena). Even
in the texts of our tenets, … , the pervasion that has affirmative concomitance
as its essential characteristic (anvayarūpā) is established in detail, appearing
as excluding the inconclusiveness [of the proving property], as with seeds and
the like…24
Here we see Jñānaśrīmitra’s efforts to defend the affirmative relation, namely, anvaya.
Viparyayabādhakapra- Jñānaśrīmitra’s interpretation
māṇavādin
of viparyayabādhakapramāṇa
anvaya
vyatireka
sādhane sādhyaviparyayasya
bādhakapramāṇa (VBP2 ) or
sādhyaviparyaye hetoḥ
bādhakapramāṇa
(VBP1 ) (independent)
Jñānaśrīmitra’s position
prasaṅga and prasaṅgaviparyaya (=?)
sādhane sādhyaviparyayasya bādhakapramāṇa
(VBP2 )
sādhyaviparyaye hetoḥ bādhakapramāṇa (VBP1 )
5. Concluding remarks
A cursory glance at the passages cited above seems to suggest that Jñānaśrīmitra criticizes
the VBP maintained by the VBP-vādin. A detailed examination of the passage as a whole,
however, leads us to the conclusion that the difference between his position and that of the
VBP-vādin lies, in fact, above all in their evaluation of sarvopasaṃhāravyāpti rather than
of VBP itself. Both of these issues are, of course, closely related to each other, at least in
Jñānaśrīmitra’s theory. The VBP-vādin’s low evaluation of sarvopasaṃhāravyāpti suggests
that the VBP-vādin is not Ratnākaraśānti, because Ratnākaraśānti takes a positive stance
toward the significance of the sarvopasaṃhāravyāpti. Secondly, with reference to VBP itself,
Jñānaśrīmitra claims the priority of prasaṅga and prasaṅgaviparyaya over VBP, whereas
for the VBP-vādin VBP1 is the only way to establish pervasion. Here Jñānaśrīmitra tries to
24
JNA p. 63, 10–14: tad evaṃ vyāptir vyāpakasya vyāpyavati bhāva eveti kartṛdharmānvayarūpavyāptisaṃpāditasāmarthyād eva sattvahetoḥ kṣaṇabhaṅgasiddhir apratihateti kim atra viparyaye bādhakapramāṇopanyāseneti samarthitam? ata eva tadupanyāsāśakyatāpi na pīḍayati, viparyayasyaiva
bādhakena sādhyasya siddhatvāt. śāstre’pi pratha[mani??]yame ca vyāptir anvayarūpā bījādivad
anekāntaparihāravyājena vistarataḥ prasādhitā. …
188
Jñānaśrīmitra on viparyayabādhakapramāṇa
interpret VBP as a more synthetic and interrelated theory by expanding his interpretation of
VBP (such that VBP is not only used for the proof of vyatireka but for that of anvaya as well).
So it would seem that in all of his argumentation on VBP examined above, Jñānaśrīmitra
seeks to establish the superiority of anvaya over vyatireka.
By this interpretation, we can easily understand and accept his positive statement
concerning VBP in other contexts. For instance, in his Vyāpticarcā, after criticizing the
bhūyodarśana of the Naiyāyikas, he says:
However, nothing but the viparyayabādhakapramāṇa, an inference, should be
inducted (unneyam). If it (i.e., viparyayabādhakapramāṇa) is absent, it would
be impossible to negate the occurrence (vṛtti) [of reason] in the dissimilar
domain (vipakṣa).25
In the above statement, where the determination of causal relations is discussed, Jñānaśrīmitra really evaluates VBP. In the Īśvara discussion as well, he states:
The means of valid cognition that establishes a pervasion is of only two kinds:
perception and non-cognition or viparyayabādhaka, because both of these
have perception and inference as their essential characteristics.26
In these passages Jñānaśrīmitra accepts the role of VBP, even if he regards it as a having a
limited secondary function.
As to the relation between VBP and prasaṅga and prasaṅgaviparyaya, Jñānaśrīmitra
gives priority to the latter over the former in the proof of the affirmative concomitance of
momentariness. Is VBP2 , as a genitive tatpuruṣa, different from prasaṅga or prasaṅgaviparyaya? If so, in which aspects is it different? And if the VBP-vādin is not Ratnākaraśānti,
who might he be? These issues require further investigation.
References and abbreviations
Primary literature
AVS Antarvyāptisamarthana (Ratnākaraśānti): See Kajiyama 1999.
HB Dharmakīrti’s Hetubinduḥ, Teil I: Tibetischer Text und rekonstruierter Sanskrit-Text.
Teil II: Übersetzung und Anmerkungen, ed. Ernst Steinkellner. Vienna 1967.
HBṬ Hetubinduṭīkā of Bhaṭṭa Arcaṭa, with the sub-commentary entitled Āloka of Durveka
Miśra, ed. Sukhlalji Sanghavi and Muni Shri Jinavijayaji. Baroda 1949.
JNA Jñānaśrīmitranibandhāvaliḥ, ed. Anantalal Thakur. Patna 1959, 2 1987.
NMu Nyāyamukha (因明正理門論). Taisho 1628.
PSṬ Pramāṇasamuccayaṭīkā.
PVin2 Dharmakīrti’s Pramāṇaviniścaya: Chapters 1 and 2, critically edited by Ernst
Steinkellner. Beijing/Vienna 2007.
25
26
VC p. 33*, 7–9: viparyayabādhakam eva tu pramāṇam anumānam unneyam, tadabhāve vipakṣavṛttiniṣedhasyāśkyatvāt,…
JNA p. 293, 1: tac ca pramāṇaṃ vyāptisādhakaṃ dvividham eva. pratyakṣānupalambhaṃ vā, viparyayabādhakaṃ vā, anayoḥ pratyakṣānumānasvabhāvatvāt,....
Kyo Kano
189
PVSV Pramāṇavārttikasvavṛtti (Dharmakīrti): The Pramāṇavārttikam of Dharmakīrti, the
first chapter with the autocommentary, text and critical notes, ed. Raniero Gnoli.
Roma 1960.
RNA Ratnakīrtinibandhāvaliḥ, ed. Anantalal Thakur. Patna 2 1975.
TBh(M) Tarkabhāṣā and Vādasthāna of Mokṣākaragupta and Jitāripāda, ed. H.R. Rangaswami Iyengar. Mysore 1952.
VC Jñānaśrīmitras Vyāpticarcā, Sanskrittext, Übersetzung, Analyse, ed. Horst Lasic.
Vienna 2000.
VNṬ Vādanyāyaprakaraṇa of Acharya Dharmakīrti with the commentary Vipanchitārthā
of Acharya Śāntarakṣita and Sambandhaparīkṣā with the Commentary of Acharya
Prabhachandra, ed. Swami Dwarikadas Shastri. Varanasi 1972.
Secondary literature
Kajiyama 1966 Yūichi Kajiyama, An Introduction to Buddhist Philosophy. An Annotated
Translation of the Tarkabhāṣā of Mokṣākaragupta. Kyoto 1966.
Kajiyama 1999 Yūichi Kajiyama, The Antarvyāptisamarthana of Ratnākaraśānti. Tokyo
1999.
Katsura 1981 Shoryu Katsura, Inmyo-Shorimonron-Kenkyu (4) [A study of Nyāyamukha
(4)]. Hiroshimadaigaku-Bungakubu-Kiyo [A Bulletin of the Faculty of Letters, Hiroshima University] (1981) 62–82.
Sakai 2014 Masamichi Sakai, Dharmottara on the viparyaye bādhakapramāṇa and trairūpya in Dharmakīrti’s sattvānumāna. Draft only, paper presented at the Fifth International Dharmakīrti Conference, Heidelberg 2014.
Shiraishi 2005 Tatsuhiko Shiraishi, Jñānaśrīmitra’s two interpretations of bādhakapramāṇa. Draft only, paper presented at the Fourth International Dharmakīrti Conference
Vienna, 24 August 2005.
Steinkellner 1967 See HB.
Steinkellner 1982 Ernst Steinkellner, Remarks on viparyaye bādhakapramāṇam – Dharmakīrti’s Development of a Theorem. Draft only, paper presented at the First International Dharmakīrti conference, Kyoto 1982.
Steinkellner 1991 Ernst Steinkellner, The logic of the svabhāvahetu in Dharmakīrti’s
Vādanyāya. In: Studies in the Buddhist Epistemological Tradition. Proceedings of
the Second International Dharmakīrti Conference Vienna, June 11–16, 1989, ed.
Ernst Steinkellner. Vienna 1991, 311–324.
Tani 1999 Tani Tadashi, Setsunametsu no Kenkyu [A Study of the Momentary Perishing
of Existence]. Tokyo 1999.
Woo 1999 Woo Jeson, The Kṣaṇabhaṅgasiddhi-anvayātmika: An eleventh-century Buddhist work on existence and causal theory. Diss., University of Pennsylvania. Pennsylvania 1999.
How Does One Cognize a Cow? A Dialogue between
Mādhava and Dignāga
by
Kei Kataoka
1. An unknown Jain ‘distinctionist,’ a Vaibhāgika
In PS 5.39–44 Dignāga defends his theory of apoha in reply to a Sāṃkhya theorist. PSV
ad 5.39 begins with the words yas tv āha; the commentator Jinendrabuddhi identifies
this theorist as Vaināśika, i.e. “the destroyer.” As Pind (2015: II Appendix 13) comments,
this theorist must be the famous Sāṃkhya theorist Mādhava, who is elsewhere often
called Sāṃkhyanāśaka, the destroyer of the Sāṃkhya system, because his unique views
often deviate from orthodox Sāṃkhya tenets.1 As Pind observes, it seems that Mādhava
criticizes the theory of apoha by quoting from a lost work of Dignāga, probably either
the Sāṃkhyaparīkṣā or the Sāmānyaparīkṣāvyāsa. The main scenario of PS 5.39 can be
depicted as follows:
1. Dignāga has criticized Sāṃkhya views in an earlier work.
2. Mādhava criticizes Dignāga’s theory of apoha.
3. Dignāga replies to Mādhava’s criticism in the Pramāṇasamuccaya.
PS 5.41ab refers to a certain view, namely, that the cognition of a cow is based on the
observation of a dewlap, and so on (sāsnādidarśanād gopratyayaḥ). Pind (2015: II 153–154,
n. 516) ascribes this view to “an unknown Jain ‘distinctionist,’ a Vaibhāgika,” on the basis
of Jinendrabuddhi’s commentary tatra hi vaibhāgikenoktam.2 Further, he ascribes the view
presented in PS 5.41d (bhinnāpohyās tu te mithaḥ) to Mādhava.3 In the following the
present author reexamines the relevant material, i.e. PS(V) and PSṬ, and shows that the
first view should not be ascribed to a Jain Vaibhāgika but to Mādhava, and the second view
not to Mādhava but to Dignāga.
PS 5.41ab: sāsnādidarśanād gopratyayaḥ
PS 5.41d: bhinnāpohyās tu te mithaḥ
1
2
3
Pind
Kataoka
Jain Vaibhāgika
Mādhava
Mādhava
Dignāga
For Mādhava, see Pind 2015: Appendix 13; and Kataoka 2011: 497–498, n. 707.
Pind 2015: II 153–154, n. 516.
Pind 2015: II 154, n. 518.
Birgit Kellner et al., eds., Reverberations of Dharmakīrti’s Philosophy: Proceedings of the Fifth International Dharmakīrti
Conference Heidelberg, August 26 to 30, 2014. Vienna 2020, pp. 191–202.
192
How Does One Cognize a Cow? A Dialogue between Mādhava and Dignāga
2. Vaibhāgika and Vaināśika
It seems that the sole evidence on which Pind ascribes the first view to a Jain Vaibhāgika is
Jinendrabuddhi’s commentary. The edited text in Pind (2015: II 153–154, n. 516) reads as
follows:
PSṬ Ms. B 233a7–233b2: tatra hi vaibhāgikenoktam. yasya darśanād yad
iti loke pratyayo bhavati, tad eva tad bhavati, tadyathā sāsnādidarśanād gopratyayo bhavati. sāsnādaya eva gaur. ātmāntarābhāvadarśanāc cātmāntare
pratyayaḥ. tasmād ātmāntarābhāva evātmāntarāṇīti.4
Here the passage vaibhāgikenoktam indicates that the subsequent paragraph quoted with iti
in the end is a quote from a Vaibhāgika. But the corresponding Tibetan translation suggests
that the original reading was not vaibhāgika but vaināśika.5
Hattori 1982: 210, 11–12: de la ’jig pa smar ba pa yis brjod pa
The Tibetan translation suggests that the original reading is tatra hi vaināśikenoktam. This
vaināśika is also mentioned previously in PSṬ ad 5.39, where the Sanskrit text reads as
follows (Pind 2015: II Appendix 13):
PSṬ Ms. 232a2: anvayavītoktisamanantaraṃ vaināśikenoktaḥ6
The same opponent is also called Sāṃkhya in the following explanation of PSṬ (Pind 2015:
II 150, n. 508, B232a6). Regarding the paragraph of PSV ad 5.39 beginning with yas tv
āha, Pind observes as follows:
This paragraph introduces a lengthy discussion, covering § 56 through § 60,
with the Sāṅkhyavaināśika Mādhava, who, as it appears, addresses Dignāga’s
criticism of his proof of the existence of pradhāna, in connection with his own
rebuttal of the apoha theory. Dignāga now answers his criticism. According
to Jinendrabuddhi, Mādhava addresses Dignāga’s objection immediately after
dealing with the direct proofs of the continuous connection of the particulars
with primordial materiality (Pind 2015: II Appendix 13).
4
5
6
The translation by Pind (2015: II 153–154, n. 516): “For in this context the Vaibhāgika has stated: ‘In
this world whatever cognition is due to the observation of whatever thing: this is such and such a thing
only. For instance, the cognition ‘cow’ is due to the observation of dewlap, etc. A cow is only dewlap,
etc. And the cognition of the nature of one thing is due to the observation of the non-existence of the
nature of other things. Therefore the nature of some things are nothing but the non-existence of the
nature of other things’.”
The Tibetan translation ’jig pa, as also shown in the next example, means perishing (vināśa) and not
dividing (vibhāga). If one wanted to support Pind’s reading vaibhāgika, one would have to explain how
’jig pa can mean vibhāga, which is usually translated as dbye ba.
vaināśikenoktaḥ at PSṬ Ms. B 232a2 is translated as ’jig pa ñid du brjod de (Hattori 1982: 208, 10–11).
Kei Kataoka
193
As Pind remarks here, Jinendrabuddhi’s expression anvayavītoktisamanantaram indicates
the location of the text quoted by Dignāga in PSV ad 5.39. It is a quote from Mādhava’s
work, in which the precise location is “immediately after the statement of anvayavīta.”
This suggests that tatra in tatra hi vaināśikenoktam in PSṬ ad 5.41 also indicates the same
context in the same text: “For in the same context it is stated by Mādhava.”
Thus, we can conclude that the quotation Pind ascribes to an unknown Jain Vaibhāgika
should be ascribed to Mādhava by correcting the reading vaibhāgikenoktam to vaināśikenoktam on the basis of the Tibetan translation.7 The main scenario of PS(V) 5.41 is the
same as that of PS(V) 5.39. The argument is between Dignāga and Mādhava in both cases.
3. The cognition of a cow due to the observation of a dewlap, etc.
It is now clear that the quote in PSṬ following vaināśikenoktam is a quote from Mādhava’s
text. In order to clarify its content, let me quote the entire PSṬ ad 5.41ab, which reads as
follows:8
A. tatra hi *vaināśikenoktam.
B. yasya *khalv api darśanād yad iti loke pratyayo bhavati, tad eva tad bhavati.
tad yathā sāsnādidarśanād gopratyayo bhavati. sāsnādaya eva gauḥ. ātmāntarābhāvadarśanāc cātmāntare pratyayaḥ. tasmād ātmāntarābhāva evātmāntarāṇīti.
C. etena yaddarśanād yatpratyayo bhavati, tad eva tad bhavati. tad yathā
sāsnādidarśanād gopratyayo bhavati. sāsnādaya eva gauḥ. ātmāntarābhāvadarśanāc cātmāntare pratyayo bhavatīti kāryam āha.
D. atra sāṃkhyena pratividhānam uktam.
E. yadi sāsnādidarśanād gopratyayo bhavati(1), evaṃ sati yad uktam ātmāntarābhāvadarśanād ātmāntare pratyayo bhavatīti(2), tad ayuktam iti.
F. ātmāntarābhāvanimittasarvapratyayābhyupagame kathaṃ sāsnādinimittatvaṃ gopratyayasyeti yāvat.
G. ātmanābhyupetahānir uktā, dṛṣṭānte svapakṣatyāgāt.
*vaināśikenoktam] Corr.; vaibhāgikenoktam Pind; vaibhāśikenokta Ms. *khalv
api] Corr.; omitted by Pind; khasvavi Ms.
7
8
According to information provided by Horst Lasic, the relevant passage of the manuscript can be read as
vaibhāśikeno° and surely not vaibhāgikeno°. The reading vaibhāgikeno° is probably a mistake arisen in
two steps: vaināśikeno° → vaibhāśikeno° → vaibhāgikeno°. First the original nā was probably mistaken
by an Indian scribe as bhā. Then the modern transcriber who prepared the transcript that Pind uses
mistakenly copied śi as gi. We can conclude that the reading vaibhāgikeno° is a modern invention.
Furthermore, the immediately following passage which Pind reads as yasya darśanād has something
in between in the manuscript. Probably the entire passage can be read as yasya khalv api darśanād,
although the manuscript seems to read kha sva vi instead of kha lva pi.
This citation is based on Pind’s edited text with slight modifications of sandhi and punctuation, etc. See
Pind 2015: II 153–154, n. 516.
194
How Does One Cognize a Cow? A Dialogue between Mādhava and Dignāga
Paragraph A (vaināśikenoktam) indicates that the subsequent paragraph B is a quote
from Mādhava’s text. Paragraph B constitutes a syllogism: udāharaṇa (vyāpti + dṛṣṭānta),
upanaya, nigamana. Paragraph C, in which the nigamana part is missing, is almost identical
to B. By adding the words etena … iti kāryam āha Jinendrabuddhi seems to classify the
reason (hetu) in the syllogism B as kāryahetu.9 Paragraph D (atra sāṃkhyena pratividhānam
uktam) indicates that the subsequent paragraph E is the Sāṃkhya’s rebuttal (pratividhāna) to
the view given in B. This Sāṃkhya theorist seems to be Mādhava, because there is no other
candidate in this context. Paragraph F restates the main point of E with the expression iti
yāvat. In order to clarify Mādhava’s intention in these paragraphs, let me start by examining
the easier paragraph F.
F. ātmāntarābhāvanimittasarvapratyayābhyupagame kathaṃ sāsnādinimittatvaṃ gopratyayasyeti yāvat.
It means: If it is accepted that all cognitions are caused by the non-existence of
non-X, how then could the cognition of a cow be caused by a dewlap, etc.?10
Here Jinendrabuddhi explains Mādhava’s intention. Mādhava is criticizing someone as
being inconsistent because he has stated something that goes against his own view. The
main view that this someone accepts is that all cognitions of X (sarvapratyaya) are caused
by the non-existence of non-X (ātmāntarābhāvanimitta). This is exactly what Dignāga
insists on as his theory of apoha. A cow is cognized by means of the exclusion of the
non-cow. This view is formulated in E2 as follows:
E2: ātmāntarābhāvadarśanād ātmāntare pratyayo bhavati.
The cognition of X is due to the observation of the non-existence of non-X.11
9
10
11
Cf. PSṬ 1 10, 6–10: yo ’nanyasattvaneyasyābhiratipūrvako hīnasthānaparigrahaḥ, sa ātmasnehavato
duḥkhasukhatyāgāptivāñchāpūrvakaḥ. tad yathā makṣikāṇām abhiratipūrvako ’śucisthānaparigrahaḥ.
ananyasattvaneyasyābhiratipūrvakaś ca garbhādihīnasthānaparigrahaḥ prāṇina iti kāryam; PSṬ 1
11, 9–11: yo yadviparītasvabhāvaḥ, sa tasya pratipakṣaḥ. tad yathā vāyuviparītasvabhāvaṃ tailaṃ
vāyoḥ. ātmadarśanaviparītasvabhāvaṃ ca nairātmyadarśanam iti svabhāvaḥ; PSṬ 1 11, 12–13: yo
yannidānaviruddhaḥ, sa tasya bādhakaḥ. yathā vātikasya vyādhes tannidānaviruddhaṃ tailam. ātmasnehādinidānātmadarśanaviruddhaṃ ca nairātmyadarśanam iti svabhāvaḥ; PSṬ 1 57, 8–9: kalpanājñānam
api nāmeti. asyāyam arthaḥ – yat svasaṃvedyam, tat svādhigamaṃ prati pratyakṣam, rāgādijñānavat.
tathā ca kalpanājñānam iti svabhāvaḥ; PSṬ 1 84, 3–4: asyāyam arthaḥ. yatra smṛtiḥ, tatrānubhavaḥ,
rūpādivat. asti ca smṛtir iti kāryam; PSṬ 1 130, 12–131, 1: kuta etat – samudāyaviṣayaṃ tu na punar
vastusadghaṭādidravyaviṣayam ity āha – rūpādyagrahe tadbuddhyabhāvād iti. yo yadagrahe saty upalabdhilakṣaṇaprāpto nopalabhyate, na sa tato vyatirikto ’sti. tad yathā kāṣṭhādibhyaḥ ṣaṇṇagarī
prāmādamālā vā. rūpādyagrahe nopalabhyate copalabdhilakṣaṇaprāptaṃ ghaṭādi dravyam iti svabhāvānupalabdhim āha; PSṬ 2 41, 6–7: siddhatvād iti. yat siddhaṃ na tat sādhyam, uṣṇo ’gnir iti yathā.
siddhau ca kevalau dharmadharmiṇāv iti svabhāvaviruddham āha; PSṬ 2 78, 15–16: saṃyogasya
cetyādi. yaḥ saṃyogāśrayaḥ sa dvitīye pratiyogini pratītihetuḥ, tad yathā dhūmaḥ. tathā cāgnir iti
svabhāvaṃ prasaṅgam āha; PSṬ 2 111, 1–4: na hītyādi. anena yat pūrvānubhūtaṃ tad evedam iti
pratyavamṛśati, tat smṛtyātmakam. yathā sa evāyaṃ dhūma iti jñānam. yathoktadharmakaṃ ca viśeṣadṛṣṭam iti svabhāvam āha; B 119a5: ekadeśatvād iti. tad anena yo yadekadeśaḥ sa tadvyapadeśam
arhati. tad yathā paṭe deśaḥ paṭavyapadeśam. pakṣaikadeśaś ca dharmīti svabhāvam āha. (I thank
Horst Lasic for these references. Orthographical modifications are given by the present author.)
My translation; cf. the translation by Pind (2015: II 153–154, n. 516).
My translation; cf. the translation by Pind (2015: II 153–154, n. 516).
Kei Kataoka
195
This view is incompatible with the view that the cognition of a cow is caused by observing
the dewlap, etc. This view is formulated in E1 as follows:
E1: sāsnādidarśanād gopratyayo bhavati.
The cognition of a cow is due to the observation of the dewlap, and so on.12
In paragraph E (yad uktam … tad ayuktam), as restated by Jinendrabuddhi in F (katham),
Mādhava criticizes Dignāga for stating the incompatible views E1 and E2. Mādhava’s
intention is summed up by Jinendrabuddhi in paragraph G as follows:
G. ātmanābhyupetahānir uktā, dṛṣṭānte svapakṣatyāgāt.
You yourself have formulated the abandonment of what you have accepted,
because you give up your own thesis in the example.13
Dignāga’s own view (svapakṣa) is E2, i.e. the view that the cognition of X is based on the
observation of the non-existence of non-X. This is what he has accepted (abhyupeta). But
Dignāga, according to Mādhava, abandons this when he states E1 as an example.
svapakṣa/abhyupeta (E2): ātmāntarābhāvadarśanād ātmāntare pratyayo
bhavati.
dṛṣṭānta (E1): sāsnādidarśanād gopratyayo bhavati.
These analyses confirm the main scenario. Dignāga first refers to E1 as an example adduced
in another work of his that is now lost. Mādhava criticizes Dignāga as being inconsistent,
because this E1 is incompatible with Dignāga’s theory of apoha, which can be summarized
as E2. But where does Dignāga state E1? A candidate is easily found in paragraph C.
C. etena yaddarśanād yatpratyayo bhavati, tad eva tad bhavati(1). tad yathā
sāsnādidarśanād gopratyayo bhavati. sāsnādaya eva gauḥ(2). ātmāntarābhāvadarśanāc cātmāntare pratyayo bhavatīti(3) kāryam āha.
With this [paragraph B] he speaks of an effect [as a reason]: If the cognition
of X arises by observing Y, X is nothing but Y. For example, the cognition of
a cow arises due to the observation of the dewlap, etc. A cow is nothing but
the dewlap, etc. And the cognition of X arises due to the observation of the
non-existence of non-X.14
The passage in C “yaddarśanād … bhavati,” i.e. C1, C2 and C3, seems to be a reformulation of B’s syllogism by Jinendrabuddhi in accordance with the Dharmakīrtian style:
udāharaṇa (vyāpti + dṛṣṭānta) and upanaya (i.e. hetu, which shows pakṣadharmatā). Here
the syllogism can be analyzed into three parts as follows:
C1 (vyāpti): yaddarśanād yatpratyayo bhavati, tad eva tad bhavati.
12
13
14
My translation; cf. the translation by Pind (2015: II 153–154, n. 516).
My translation; cf. the translation by Pind (2015: II 153–154, n. 516).
My translation. This passage is quoted but not translated in Pind 2015: II 153–154, n. 516.
196
How Does One Cognize a Cow? A Dialogue between Mādhava and Dignāga
C2 (dṛṣṭānta): tad yathā sāsnādidarśanād gopratyayo bhavati. sāsnādaya eva gauḥ.
C3 (upanaya): ātmāntarābhāvadarśanāc cātmāntare pratyayo bhavati.
C1 states an invariable concomitance (vyāpti): If X is cognized by observing Y, X is
nothing but Y. C2 gives an example (dṛṣṭānta): One cognizes a cow by observing the dewlap,
etc. Therefore, a cow is nothing but the dewlap, etc. In other words, a cow is nothing but
the aggregate of the dewlap, etc.15 C3 presents the application (upanaya) of this invariable
concomitance to his theory of apoha: One cognizes X by observing the non-existence of
non-X. The conclusion, which is not stated in C, is obvious: ātmāntarābhāva evātmāntarāṇi
(Xs are nothing but the non-existence of non-X).16 This missing part is explicitly stated in
paragraph B, which reads as follows:
B. yasya khalv api darśanād yad iti loke pratyayo bhavati, tad eva tad
bhavati(1). tad yathā sāsnādidarśanād gopratyayo bhavati. sāsnādaya eva
gauḥ(2). ātmāntarābhāvadarśanāc cātmāntare pratyayaḥ(3). tasmād ātmāntarābhāva evātmāntarāṇīti(4).
In this world, as is also well known, if the cognition “X” arises by observing
Y, X is nothing but Y. For example, the cognition of a cow arises due to the
observation of the dewlap, etc. A cow is nothing but the dewlap, etc. And
the cognition of X is due to the observation of the non-existence of non-X.
Therefore, Xs are nothing but the non-existence of non-X.17
B1, B2 and B3 are almost identical with C1, C2 and C3.
B
C
1. yasya khalv api darśanād yad iti loke
pratyayo bhavati, tad eva tad bhavati
2. tad yathā sāsnādidarśanād gopratyayo bhavati. sāsnādaya eva gauḥ.
3. ātmāntarābhāvadarśanāc cātmāntare
pratyayaḥ.
4. tasmād ātmāntarābhāva evātmāntarāṇi.
1. yaddarśanād yatpratyayo bhavati, tad
eva tad bhavati.
2. tad yathā sāsnādidarśanād gopratyayo bhavati. sāsnādaya eva gauḥ.
3. ātmāntarābhāvadarśanāc cātmāntare
pratyayo bhavati.
The main difference lies in B4, which clarifies the unstated conclusion (nigamana).
In paragraph C Jinendrabuddhi reformulates the syllogism of B in accordance with the
Dharmakīrtian style and classifies the reason as kāryahetu.
15
16
17
Cf. PSṬ B 233b5: sāsnādisamūha eva gauḥ, quoted by Pind 2015: II 154, n. 520.
With the plural form ātmāntarāṇi Dignāga intends, for example, cows in general. See, e.g. his usage in
PSV ad 5.36d (Pind 2015: I 45): śabdo ’rthāntaranivṛttiviśiṣṭān eva bhāvān āha; cf. also PSṬ Ms. B
238b5–6 quoted in Pind 2015: II 179, n. 604: yathā vṛkṣaśabdaḥ śiṃśapādīn viśeṣān abhedenābhidadhat sāmānyavācī tathā…
My translation.
Kei Kataoka
197
C. etena “C1, C2, C3 (;B1, B2, B3)” iti kāryam āha
With this [paragraph B quoted above] he speaks of an effect [as a reason, for
which the entire syllogism is reformulated as] C1, C2, C3.
But who has composed this syllogism in paragraph B? As suggested in paragraphs E,
F, G, the syllogism of B must have been originally formulated by Dignāga. Then it is
quoted by Mādhava, either literally or not literally, as a pūrvapakṣa, as Jinendrabuddhi’s
opening remark tatra hi vaināśikenoktam indicates. Thus, it is surmised that paragraph B
(which Jinendrabuddhi explains as C) is Mādhava’s quote from a lost work of Dignāga and
that Mādhava criticizes Dignāga’s view in E (which Jinendrabuddhi explains in F and G).
Recapitulating these analyses, the main scenario can be reconstructed as follows.
1. First a syllogism was stated by Dignāga in a work that is now lost.
2. Mādhava quotes Dignāga’s statement as B, which Jinendrabuddhi reformulates with
classification as C.
3. In paragraph E, which follows D (atra sāṃkhyena pratividhānam uktam), Mādhava
points out Dignāga’s inconsistency with the words yad uktam … tad ayuktam. The issue at stake raised by Mādhava is that the example Dignāga mentions does not fit with
the theory of apoha, because the cognition of a cow (gopratyaya), according to the
theory of apoha, should be based on the exclusion of the non-cow (agovyavaccheda)
and not on the dewlap, etc. (sāsnādi). By referring to, and thereby admitting the
example, Dignāga amounts to having abandoned his own tenet that the cognition of
X (e.g. a cow) is based on the observation of the non-existence of non-X (e.g. the
non-existence of the non-cow).
4. Dignāga’s intention in referring to the example
Although there are uncertainties here and there regarding the reconstruction of PSV ad
5.41, the main argument of the following part is more or less certain.18
PSV ad 5.41: yasya hy [agonivṛttagopratyayaḥ, tasya kathaṃ sāsnādidarśananimittaḥ].19
PSṬ B 233b5: yasya hīty apohavādinaḥ.
As an apoha theorist (apohavādin), it is inappropriate for Dignāga to state that the cognition
of a cow is based on the observation of the dewlap, etc., because according to the theory of
18
19
V (Hattori 1982: 142, 10–11): gaṅ la ba laṅ ma yin pa las log pa’i ba laṅ gi blo de ji ltar nog la sogs
pa mthoṅ ba’i rgyu mtshan can du smra bar byed /; K (Hattori 1982: 143, 11–12): gaṅ gi ba laṅ ma
yin pa las ldog pa’s ba laṅ gi rogs par ’gyur ba de’i ji ltar lkog śal la sogs pa mthoṅ ba rgyu mtshan
du smra bar byed /; Pind 2015: II 154: “For how could someone, to whom the cognition of a cow
(*gopratyayaḥ) as precluded from non-cows (*agonivṛttaḥ), assert that it is caused by the observation
of dewlap, etc. (*sāsnādidarśananimittaḥ)?”
Pind 2015: I 52 presents the reconstruction as “yasya hi […],” not filling in the blank. But the main
words are more or less safely reconstructed on the basis of the two Tibetan translations; Pind provides
the Sanskrit words in his translation.
198
How Does One Cognize a Cow? A Dialogue between Mādhava and Dignāga
apoha it is based on the exclusion of the non-cow. Here Dignāga seems to accept Mādhava’s
claim of inconsistency. The example sāsnādidarśanād gopratyayaḥ is indeed incompatible
with the theory of apoha. PS 5.41ab amounts to saying, using the word katham: How could
an apoha theorist accept the example? But then how can Dignāga defend his reference to
the example? The subsequent passage clarifies his strategy.
PSV ad 5.41 (reconstructed by Pind 2015: I 52): abhyupagamyāyaṃ dṛṣṭāntaḥ
svamataviruddhaḥ. śabdabhedād dhi gosāsnādiṣu bhinnam apohyam.20
PSṬ Ms. B233b5–B234a1: abhyupagamyetyādi. bhavato hi sāsnādisamūha
eva gaur iti. atas tad abhyupagamyāyaṃ dṛṣṭāntaḥ svamataviruddho ’py uktaḥ.
etad uktaṃ bhavati. yathā tava sāsnādisamūhadarśanād gopratyayas tathā mamāpy ātmāntarābhāvadarśanād ātmāntare pratyaya iti. śabdabhedād dhītyādi.
sāsnādiśabdasyāsāsnādy apohyaṃ sāsnādiṣu, gośabdasyāpy agaur gavi. yata
evaṃ bhinnam apohyam, ataḥ sāsnādiṣv asāsnādyapohena sāsnādipratyayaḥ,
gavy agovyavacchedena gopratyayaḥ. evaṃ cātrāpy ātmāntarābhāvadarśanād
evātmāntare pratyayaḥ.21
It is not easy to reconstruct the original text of PSV, because the two Tibetan translations
differ from each other. Nonetheless the main argument can be summarized as follows: the
example (dṛṣṭāntaḥ), although it is incompatible with the Buddhist view (svamataviruddho
’pi), is presented by provisionally accepting (abhyupagamya) your view, i.e. the Sāṃkhya’s
view.
Sāṃkhya: sāsnādisamūhadarśanād gopratyayaḥ (→sāsnādaya eva gauḥ)
Dignāga: ātmāntarābhāvadarśanād ātmāntare pratyayaḥ (→ātmāntarābhāva
eva ātmāntarāṇi)
It is clear from this that the view referred to by Dignāga as an example is a Sāṃkhya view.
The Sāṃkhya holds the view that the cognition of a cow is based on the observation of
the dewlap, etc. (sāsnādisamūhadarśanād gopratyayaḥ), and that a cow is nothing but
the aggregate of the dewlap, etc. (sāsnādisamūha eva gauḥ). Dignāga refers to this view
by accepting it only provisionally (abhyupagamya). Therefore, there is no inconsistency
in Dignāga’s statements, because he does not wholeheartedly accept the Sāṃkhya view.
Dignāga consistently keeps his doctrine of apoha, i.e. the view that the cognition of X (e.g.
a cow) is based on the observation of the non-existence of non-X (e.g. the non-existence of
the non-cow), i.e. ātmāntarābhāvadarśanād ātmāntare pratyayaḥ. Therefore, X is nothing
but the non-existence of non-X for Dignāga (ātmāntarābhāva eva ātmāntarāṇi). For him
the cognition of a cow is caused by the non-existence of the non-cow and not by the dewlap,
20
21
V (Hattori 1982: 142, 12–13): ba laṅ daṅ nog la sogs pa sgra tha dad pas tha dad du sel ba can yin
yaṅ khyod kyi lugs khas blaṅs nas / ’gal bźin du yaṅ de ṅes par bstan to /; K: khas blaṅs kyaṅ khyod kyi
’dod pas dpe ’di ’gal ba yaṅ yin no / ba laṅ gi lkog śal la sogs pa rnams la sgra’i khyad par gyis tha
dad pa sel ba /; Pind 2015: II 154–155: “Having assumed [this], the example is in conflict even with
your own theory (svamataviruddhaḥ). For the excluded [object] is different with regard to a cow and
the dewlap because of verbal difference (śabdabhedāt).”
For the text, see Pind 2015: I 52, n. 275, and Pind 2015: II 154–155, nn. 520, 521 and 522.
199
Kei Kataoka
etc. The two things, i.e. a cow and a dewlap, etc., have a different scope of exclusion. It
is obvious for Dignāga that the words “cow” and “dewlap, etc.” have different objects to
be excluded (apohya). The expression “dewlap, etc.” (sāsnādi) communicates the dewlap,
etc. (sāsnādiṣu) by excluding the non-dewlap, etc. (asāsnādi). The word “cow” (gauḥ)
communicates a cow (gavi) by excluding the non-cow (agauḥ). This is Dignāga’s own view.
The fundamental view of apoha is consistent.
“sāsnādi”
“gauḥ”
apohya
pratyaya
asāsnādi
agauḥ
sāsnādiṣu
gavi
The view that the two different words have different scopes of exclusion is explicitly
expressed in PS 5.41d, which runs as follows.
Pind 2015: I 51: bhinnāpohyās tu te mithaḥ.22
PSṬ: asmanmatena tu bhinnāpohyās tu te mitho gosāsnādayaḥ, bhinnam
apohyam eṣv iti kṛtvā.
Here Pind’s reconstruction of PS 5.41d is strongly supported by PSṬ. However the reconstruction and interpretation of PS 5.41abc are a bit problematic.23
PS 5.41abc, Pind 2015: I 51: sāsnādidarśanād <gopratyayo yo ’yam udāhṛtaḥ /
so> viruddho bhavanmatyā.
PSṬ: sāsnādidarśanād ityādi … viruddha iti siddhāntavirodhāt. bhavanmatyeti. bhavato hi sāsnādaya eva gaur iti matam.
Considering the meter, it would be better to change the word order of PS 5.41abc to the
following:
Kataoka: sāsnādidarśanād yo ’yaṃ gopratyaya udāhṛtaḥ / sa viruddho bhavanmatyā
It is true that viruddho bhavanmatyā can be interpreted as Pind translates, “is in conflict
with your own theory.” K’s translation supports Pind’s interpretation. But this interpretation
does not fit the entire context. Here bhavat clearly refers to the Sāṃkhya, as Jinendrabuddhi
clarifies by stating, “For it is your view that a cow is nothing but the dewlap, etc. (bhavato hi
sāsnādaya eva gaur iti matam).” As we have already confirmed, this view should be ascribed
22
23
V (Hattori 1982: 142, 6): phan tshun tha dad dag yod kyaṅ; K (Hattori 1982: 143, 0): tha dad sel la de
log pa; Pind 2009: 110: “On the contrary, they have mutually different excluded referents.”
V (Hattori 1982: 142, 7–9): nog la sogs pa mthoṅ ba las / de’i blo dper brjod ’gal ba de / khyed kyi lugs
la rten pa yin /; K (Hattori 1982: 143, 7–9): lkog śal la sogs mthoṅ phyir gaṅ / ba laṅ rtogs pa’i dper
byas pa / de ni khyod kyi ’dod pas ’gal /; Pind 2015: II 153–154: “The example [you have] adduced,
namely that the cognition of a cow is due to the observation of dewlap, and so on, is in conflict with
your own theory.”
200
How Does One Cognize a Cow? A Dialogue between Mādhava and Dignāga
to the Sāṃkhya and not the Buddhist. Then viruddho bhavanmatyā would mean that the
example is in conflict with the Sāṃkhya view. But what we expect here is the opposite: The
example is in conflict with the Buddhist view. Taking into consideration V’s translation
of bhavanmatyā as khyed kyi lugs la rten pa yin (resorting to your view), it seems more
appropriate to interpret bhavanmatyā as being separate from the preceding word viruddhaḥ.
Jinendrabuddhi’s commentary also supports this interpretation, because he comments
on viruddha separately from bhavanmatyā, and states viruddha iti siddhāntavirodhāt.
Considering that the opponent bhavat is the Sāṃkhya in this context, the opposite siddhānta
(i.e. svamata) must refer to the Buddhist view (cf. svamataviruddha in PSV ad 5.41).
Therefore, the main argument in PS 5.41abc can be reconstructed as follows:
The example (dṛṣṭāntaḥ) that the cognition of a cow (gopratyayaḥ) is based on
the observation of the dewlap, etc. (sāsnādidarśanāt) is presented (udāhṛtaḥ)
by me in my earlier work. This example is indeed incompatible with the
Buddhist view (viruddhaḥ), as you, Mādhava, claim. But it is mentioned by
me only by provisionally resorting to your Sāṃkhya view (bhavanmatyā).
Therefore, there is no fault of abandoning my thesis.
5. Positive and negative methods of cognizing a cow
The conflict of opinion between Mādhava and Dignāga is clear. Mādhava holds the view
that a cow is cognized positively, i.e. by observing the dewlap, etc., whereas Dignāga holds
the view that a cow is cognized negatively, i.e. by excluding the non-cow. For Dignāga
any X, inasmuch as it is cognized in a general form, is cognized by observing the nonexistence of non-X. A dewlap, etc. are no exception. They, too, are cognized by excluding
the non-dewlap, etc. This is explicitly stated in PSV ad 5.42 as follows:
Pind 2015: I 52: sāsnādiṣu hi <sāmānyarūpam> arthāntarābhāvanirapekṣaṃ
na bhavatīti pūrvam evopapāditam.24
Mādhava holds that X (ātmāntara) is cognized positively, without dependence on the observation of the non-existence of non-X (ātmāntarābhāvadarśana). This view of Mādhava
is criticized by Dignāga in PS 5.42ab as follows:25
PS 5.42ab, Pind 2015: I 52: so ’napekṣa <ity etat tu> svavikalpavi<nirmitam> /
24
25
V (Hattori 1982: 142, 18–19): nog la sogs pa la spyi’i ṅo bo daṅ ldan pa gźan med par mi ltos
pa ni mi srid do źes sṅar bśad zin to /; K (Hattori 1982: 143, 9–20): lkog śal la sogs pa rnams la
spyi’i ṅo bo bdag ñid gźan med pa la bltos pa med par srid pa ma yin no źes sṅar bstan pa yin no /;
Pind 2015: II 156: “For it has previously been demonstrated that the general form in a dewlap, etc.
(sāsnādiṣu), does not exist without dependence upon the non-existence of other referents (sāmānyarūpam
arthāntarābhāvanirapekṣaṃ na bhavati).”
V (Hattori 1982: 142, 14–15): de mi ltos śes pa ’di ni / raṅ gi rnam rtog spros par zad /; K (Hattori
1982: 143, 15–16): de ltos med phyir ’di yaṅ ni / raṅ gi rnam par rtog pas sprul /; Pind 2015: II 155:
“The idea, however, that this [namely the cognition of one thing (ātmāntara)] is not dependent [upon
the observation of non-existence of other things], is created out of your own imagination.”
Kei Kataoka
201
PSṬ ad 5.42: so ’napekṣa ātmāntarapratyayaḥ. kasmāt. na hi naḥ pratyayo
bhavaty ātmāntarābhāvadarśanād ātmāntare, kiṃ tarhi vidhirūpeṇaiva gaur
iti.
According to Dignāga, Mādhava holds that the cognition of X is independent (so ’napekṣaḥ),
i.e. does not depend on the exclusion of the other. A cow is cognized as such in a positive
way (vidhirūpeṇaiva). But Mādhava’s idea is a mere fancy, because a general form is
never cognized without exclusion of the other, as Dignāga has implied in PSV ad 5.34:
vyatirekamukhenaivānumānam. The individual form (svarūpa), i.e. the particular form
(svalakṣaṇa), is beyond the scope of language and therefore inexpressible (anabhilāpya).
Thus, the individual form is not the object of everyday communication (vyāvahārika). This
is stated by Dignāga in PSV ad 5.42 as follows:
PSV ad 5.42 (Pind 2015: I 52): svarūpaṃ tu ten<āvyāvahārikam> anabhilāpyatvāt.26
According to the Sāṃkhya, the individual form is denotable. Therefore, the word “cow”
refers to the aggregate of the dewlap, etc., in a positive way. For Dignāga, by contrast, the
particular form is not denotable. It is the object of perception and not inference. Words
communicate things in a general form only by excluding the other. Our cognition of a cow
is not independent but always dependent upon the non-existence of the non-cow.
6. Conclusion
1. The crucial passage in PSṬ vaibhāgikenoktam should be corrected to vaināśikenoktam.
2. The argument in PS(V) 5.41 is not between a Jain Vaibhāgika and Mādhava but
between Vaināśika Mādhava and Dignāga. The scenario is similar to that of PS(V)
5.39. The Jain ‘distinctionist’ that Pind postulates does not exist.
3. The view that the cognition of a cow is due to the observation of the dewlap, etc.
should be ascribed to the Sāṃkhya, not a Jain Vaibhāgika.
4. Dignāga refers to the Sāṃkhya view in an example in an earlier work that is now lost.
Dignāga’s text quoted by Mādhava is quoted by Jinendrabuddhi in paragraph B and
modified as in C.
5. Mādhava criticizes Dignāga’s view as being inconsistent, because Dignāga abandons
his thesis by admitting the Sāṃkhya example. Mādhava first quotes Dignāga’s earlier
work (paragraph B) and then criticizes it (paragraph E).
6. Dignāga defends his earlier statement by insisting that his mentioning of the Sāṃkhya
example that is incompatible with his thesis is not wholehearted acceptance, but only
a provisional acceptance (abhyupagamya). For Dignāga the cognition of a cow is due
to the exclusion of the non-cow (agovyavaccheda) and not due to the observation of
26
V (Hattori 1982: 142, 19–20): raṅ gi ṅo bo ni brjod par bya ba ma yin pa’i phyir de’i sgo nas tha sñad
du bya’o /; K (Hattori 1982: 143, 20–21): raṅ gi ṅo bo ci brjod par bya ba ma yin pa’i phyir de tha
dad mi bya’o /; Pind 2015: II 156–157: “The individual form, however, (svarūpaṃ tu) is not denotable
(*vyāvahārikam [sic]) in this (tena) [form] because it is inexpressible (anabhilāpyatvāt).”
202
How Does One Cognize a Cow? A Dialogue between Mādhava and Dignāga
the dewlap, etc. (sāsnādidarśana). His main thesis of apoha that the cognition of X
is based on the non-existence of non-X (ātmāntarābhāvadarśanād ātmāntare pratyayaḥ), is consistent. For him a cow is essentially the non-existence of the non-cow
and not the aggregate of the dewlap, etc. (sāsnādaya eva gauḥ; sāsnādisamūha eva
gauḥ). X is essentially the non-existence of non-X (ātmāntarābhāva evātmāntarāṇi).
References and abbreviations
Corr. Correction by the present author.
Hattori 1982 M. Hattori, The Pramāṇasamuccayavṛtti of Dignāga with Jinendrabuddhi’s
Commentary, Chapter Five: Anyāpoha-parīkṣā: Tibetan Text with Sanskrit Fragments.
Memoirs of the Department of Literature, Kyoto University 21 (1982) 101–224.
K Kanakavarman’s translation of PS(V).
Kataoka 2011 K. Kataoka, Kumārila on Truth, Omniscience, and Killing. Vienna 2011.
Ms. Manuscript.
Pind 2009 O. H. Pind, Dignāga’s Philosophy of Language. Dignāga on anyāpoha. Pramāṇasamuccaya V. Texts, Translation, and Annotation. PhD diss., Universität Wien.
Wien 2009. http://othes.univie.ac.at/8283/1/2009-12-03_0507516.pdf, last visited
15-03-2016.
Pind 2015 O. H. Pind, Dignāga’s Philosophy of Language. Dignāga. Pramāṇasamuccayavṛtti V on anyāpoha. Part I: Text. Part II: Translation and Annotation, ed. E.
Steinkellner. Vienna 2015.
PSṬ Pramāṇasamuccayaṭīkā. Jinendrabuddhi’s Viśālāmalavatī Pramāṇasamuccayaṭīkā.
Chapter 1, Part I: Critical Edition, ed. Ernst Steinkellner, Helmut Krasser, and Horst
Lasic. Beijing/Vienna 2005. Pramāṇasamuccayaṭīkā. Jinendrabuddhi’s Viśālāmalavatī Pramāṇasamuccayaṭīkā. Chapter 2, Part I: Critical Edition, ed. Horst Lasic,
Helmut Krasser, and Ernst Steinkellner. Beijing/Vienna 2012. For chapter 5, see
Pind 2015.
PS(V) Pramāṇasamuccaya(vṛtti). See Pind 2015.
V Vasudhararakṣita’s translation of PS(V).
Śāntarakṣita on Two Kinds of Arguments for Self-Awareness:
sahopalambhaniyama and saṃvedana
by
Hiroko Matsuoka
Introduction
In the last part of the Pratyakṣa chapter of his Pramāṇaviniścaya (PVin I 39,13–43,7),
Dharmakīrti (c. 600–660)1 presents two kinds of arguments to prove self-awareness (svasaṃvedana). Since the theory of self-awareness that a cognition is aware of itself presupposes
that an object and the cognition thereof are non-different; and hence the cognition of an
object is nothing but the cognition of the cognition itself, the former is established on the
basis of the fact that both are necessarily perceived together (sahopalambhaniyama);2 the
latter is established on account of its essential nature of being a cognition (saṃvedana).3
Following Iwata (1991), I am going to call these two arguments the sahopalambhaniyama
and saṃvedana arguments, respectively.
What led Dharmakīrti to develop these arguments? Concerning the sahopalambhaniyama argument, Iwata (1991: 20–24) examines how the argument could have been derived
directly from the view that is presented by Dignāga (c. 480–540) in Ālambanaparīkṣā v. 6
and the Vṛtti thereon, namely, that an object-support (ālambana) for a cognition is a form
within the cognition itself (antarjñeyarūpa). Taber (2010) argues that crucial elements of
the argument are found in the Śūnyavāda chapter of Kumārila’s (c. 600–650) Ślokavārttika
vv. 31–34. Concerning the saṃvedana argument, on the other hand, no attempts have yet
been made to determine the details of its provenance.
The aim of this paper is to explore both of these issues equally, at the same time,
namely with the help of the Bahirarthaparīkṣā chapter of the Tattvasaṅgraha. In this,
1
2
3
This is a revised and enlarged version of the paper entitled “Sho-shinjitsu-ron ni miru ukeisho-chishikiron-ronsho seiritsu no haikei” [Śāntarakṣita on Dharmakīrti’s Proof of Sākāratā], Indogaku bukkyōgaku
kenkyū, 62.2 (2014) 983–979. A draft of this paper was presented at the XVIIth Congress of the
International Association of Buddhist Studies, University of Vienna, Austria, August 18–23, 2014. I
would like to express my gratitude for comments and help to Hideyo Ogawa, Toru Funayama, Birgit
Kellner and Eli Franco, as well as to Chris Jensen and Tyler Neill for correcting the English.
For the date of Dharmakīrti I shall provisionally follow Frauwallner (1961: 137–139). This dating
has been discussed by several scholars, most notably by Krasser (2012). He pointed out the common
wordings like sattvānumāna in Bhāviveka’s (490/500–570) and Dharmakīrti’s works and an almost
identical verse in Bhāviveka’s and Kumārila’s works. In his conclusion Krasser hypothesized that
the time of activity of both Dharmakīrti and Kumārila is the middle of the sixth century. Krasser’s
hypothesis is not fully accepted by scholars, e.g. by Watanabe (2014) and by Franco (2015–2018).
This paper does not discuss the date of Dharmakīrti and Kumārila but attests to the chronological
ordering of Kumārila and Dharmakīrti, namely, in that order. For the relationship between Kumārila
and Dharmakīrti, see also Yoshimizu 2007 and Kataoka 2010.
PVin I 54ab: sahopalambhaniyamād abhedo nīlataddhiyoḥ // See also PV III 332–336; 387–397.
PVin I 42,3: saṃvedanam ity api tasya tādātmyāt tathāprathanam, na tad anyasya kasyacid ātmasaṃvedanavat / tato ’pi na tad arthāntare yuktam / See also PV III 326–329.
Birgit Kellner et al., eds., Reverberations of Dharmakīrti’s Philosophy: Proceedings of the Fifth International Dharmakīrti
Conference Heidelberg, August 26 to 30, 2014. Vienna 2020, pp. 203–220.
204
Śāntarakṣita on Two Kinds of Arguments for Self-Awareness
Śāntarakṣita (c. 725–788), while discussing the sahopalambhaniyama and saṃvedana
arguments, directly references Kumārila’s arguments aimed at proving that a cognition
has an external entity for its object. In particular, I hope to provide evidence in favor of
Śāntarakṣita’s view that Dharmakīrti was responding directly to Kumārila. In order to
do this, I will first present Kumārila’s arguments in section 1, and then, in section 2, I
will discuss the corresponding details of Śāntarakṣita’s refutation, which follows but also
builds upon the arguments of his predecessor, Dharmakīrti. With this foundation in place,
section 3 will provide one final piece of evidence in favor of Śāntarakṣita’s view, namely,
that he sees in the dual nature of Dharmakīrti’s argumentation an attempt to refute the
nirākāravādin’s view rather than to simply improve Dignāga’s theory of self-awareness.
1. Kumārila’s refutation of self-awareness
Kumārila is only one among many concerns for Śāntarakṣita. The Bahirarthaparīkṣā
chapter of the latter’s TS is meant to establish the vijñānavāda theory of vijñaptimātratā or
“mind-only.” According to Kamalaśīla (c. 740–795), author of the pañjikā commentary,
this chapter is divided into two parts: in the first part, Śāntarakṣita establishes that an
external object cannot exist independently of cognition (arthāyoga) (TS2 1964–1997); in
the second part, he proves that cognition is devoid of the characteristics of being either
grasped or grasper (grāhyagrāhakalakṣaṇavaidhurya), that is, that cognition consists in selfawareness (ātmasaṃvedana) (TS2 1998–2083). In this second part, Śāntarakṣita proceeds
in three steps: first, he posits his own arguments denying the existence of external objects
(bāhyārthaniṣedhaka) (TS2 1998–2049); secondly, he refutes arguments formulated by
various realists such as Śubhagupta, Uddyotakara and Kumārila, to establish the existence of
such objects (bāhyārthasādhaka) (TS2 2050–2077); and lastly, he presents the fundamental
(maula) argument for establishing vijñaptimātratā (TS2 2078–2083).
While refuting the realist arguments, Śāntarakṣita cites verses from the Bṛhaṭṭīkā4 of
Kumārila, corresponding to ŚV Śūnyavāda 172cd–177ab, which contain seven arguments
for proving the existence of external objects (bāhyārthasādhaka) (TS2 2063–2067). In ŚV
Śūnyavāda 172cd–177ab, after first presenting them earlier in the same chapter (verses
5–17), Kumārila then refutes vijñānavāda, more precisely, Dignāga’s arguments for selfawareness,5 by moving through the series of arguments in the same order as before.6 It is
these seven arguments that we will examine in the current section 1. They can be classified
into two groups according to the theses (pakṣa) that they support:
1.
2.
4
5
6
An object and its cognition are different (bhinna) from each other.
A cognition neither cognizes a part of itself (svāṃśa) nor is cognized by
a part of itself.
For discussions of quotations from the Bṛhaṭṭīkā in the TS(P), see Frauwallner 1962, Taber 1986–1992,
Krasser 1999: 216 and Kataoka 2011: 25–27. See also the invaluable contribution of Yoshimizu (2007)
showing that the Bṛhaṭṭīkā is earlier than the PV, thus refuting Frauwallner’s influential hypothesis.
See section 1.1.2.
See Bhatt 1962 and Teraishi 2002: 153–154 for a detailed structural analysis of the Śūnyavāda chapter.
Hiroko Matsuoka
205
These two groups closely relate to Dharmakīrti’s sahopalambhaniyama and saṃvedana
arguments, respectively, as I will demonstrate below. The first group is discussed in section
1.1, and the second in section 1.2.
1.1 Arguing for the difference between an object and its cognition
The four arguments in this first group can be labeled as follows:
1.
2.
3.
4.
The tatsaṃvitti-asaṃvitti argument;
The tatparāmarśa-aparāmarśa argument;
The itaretaraparāmarśa-itaretarāparāmarśa argument;
The aikarūpyeṇa-ajñāna argument.
These arguments will now be discussed individually in the following sections.
1.1.1 The tatsaṃvitti-asaṃvitti argument
The tatsaṃvitti-asaṃvitti argument reads as follows:7
The grasper (grāhaka) of color is different from the grasped [color] (grāhya)
because the former is not perceived when the latter is perceived (tatsaṃvittāv
asaṃvitteḥ), like the grasper of flavor, etc.
In ŚV Śūnyavāda 79,8 Kumārila explains that the tatsaṃvitti-asaṃvitti argument is based
on the Bhāṣya of Śabara (6th c.), where it is stated that it is only an object connected with the
external world (bahirdeśasambaddha) that is perceived by sense-perception (pratyakṣa).9
1.1.2 The tatparāmarśa-aparāmarśa argument
The tatparāmarśa-aparāmarśa argument is as follows:10
The grasped [color] is different from [the cognition] that grasps it because the
former is not necessarily recollected by one who recollects the latter, like the
grasper of flavor, etc.
7
8
9
10
ŚV Śūnyavāda 172cd–173ab: tasmād yad grāhakaṃ rūpe tadgrāhyāt tasya bhinnatā // tatsaṃvittāv
asaṃvitte rasādigrāhakaṃ yathā / TS2 2063: atha yad grāhakaṃ rūpe tadgrāhyāt tasya bhinnatā /
tatsaṃvittāv asaṃvitte rasādigrāhakaṃ yathā // (tasmād ŚV : atha TS; grāhakaṃ ŚVM , ŚVT , TS :
bhāsakaṃ ŚVV ; rūpe TS : rūpaṃ ŚV.) Here as well as in the following, “ŚV” is in the reporting of
variants used as a shorthand for all three editions ŚVV , ŚVT and ŚVM .
ŚV Śūnyavāda 79: sa bahirdeśasambaddha ity (ŚBh 28,17 on 1.1.4. See n. 9.) anena nirūpyate / grāhyākārasya saṃvittir grāhakānubhavād ṛte // TS2 2069: sa bahirdeśasambaddha ity anena nanūcyate /
grāhyākārasya saṃvittir grāhakānubhavād ṛte // (nirūpyate ŚV : nanūcyate TS.)
ŚBh 28,17 on 1.1.4: syād etat – evaṃ yady arthākārā buddhiḥ syāt / nirākārā tu no buddhiḥ, ākāravān
bāhyo ’rthaḥ / sa hi bahirdeśasaṃbaddhaḥ pratyakṣam upalabhyate /
ŚV Śūnyavāda 173cd–174ab: grāhyaṃ tadgrāhakād bhinnaṃ tatparāmṛśatā yataḥ // na parāmṛśyate
’vaśyaṃ rasādigrāhakaṃ yathā / TS2 2064: grāhyaṃ tadgrāhakād bhinnaṃ tatparāmṛśatā yataḥ / na
parāmṛśyate ’vaśyaṃ rasādigrāhakādivat // (-kād bhinnaṃ ŚVM : -kāc caivaṃ ŚVT TS: -kāc caiva
ŚVV ; yathā ŚV : -kādivat TS) As for TS2 2064, the reading -kāc caivaṃ attested in TSJ and TSPa (and
also chosen for the editions TS1 and TS2 ) is here emended to bhinnaṃ following ŚVM .
206
Śāntarakṣita on Two Kinds of Arguments for Self-Awareness
According to ŚV Śūnyavāda 82–85,11 the tatparāmarśa-aparāmarśa argument is also based
on Śabara’s Bhāṣya. Śabara argues that there are cases where one remembers a cognition
that occurred in the past, without at the same time remembering what one cognized.12
The tatparāmarśa-aparāmarśa argument is undoubtedly directed against a point made
by Dignāga in Pramāṇasamuccaya I v. 11cd and the Vṛtti thereon. There, Dignāga argues
that a cognition has both characteristics (dvirūpatā), namely, both the characteristic of
being the grasped and that of being the grasper, and that such a cognition is cognized by
itself (svasaṃvedyatā). He claims that these two points are to be accepted on the basis
of the empirical fact that both grasper and grasped are remembered afterwards (smṛter
uttarakālam PS I 11c).13
1.1.3 The itaretaraparāmarśa-itaretarāparāmarśa argument
The itaretaraparāmarśa-itaretarāparāmarśa argument is as follows:14
The two (i.e., the grasped and the grasper) are mutually different (bhinna) [because one is not recollected when the other is recollected (itaretaraparāmarśa
itaretarāparāmarśāt)],15 like flavor [and color],16 etc.
Pārthasārathi (10th c.) comments that each of the two is remembered by excluding the
other (parasparaparihāreṇa).17 The point is that the recollection of an object cannot
simultaneously be the recollection of its cognition and vice versa. Whatever is an object is
not a cognition, and whatever is a cognition is not an object. These two properties exclude
each other.
11
12
13
14
15
16
17
ŚV Śūnyavāda 82–85: na pūrvaṃ jñāyate buddhir ity (ŚBh 30,5 on 1.1.4. See n. 12.) atraitad vadiṣyate /
grāhakasyaiva saṃvittir lakṣyate grahaṇe kvacit // na smarāmi mayā ko ’pi gṛhīto ’rthas tadeti hi /
smaranti grāhakotpādaṃ grāhyarūpavivarjitam // tasmād abhinnatāyāṃ ca grāhye ’pi smaraṇaṃ
bhavet / grāhakasmṛtinirbhāsāt tatrāpy eṣaiva gṛhyate // tad atyantāvinābhāvān (ŚVT ; tadatyantāŚVM ŚVV ) naikākāraṃ hi jāyate // anvayavyatirekābhyāṃ siddhaivaṃ bhinnatā tayoḥ // TS2 2070–2072
correspond to 83–84 and 85cd: na smarāmi mayā ko ’pi gṛhīto ’rthas tadeti ca / smaranti grāhakotpādaṃ
grāhyarūpavivarjitam // tasmād abhinnatāyāṃ ca grāhye ’pi smaraṇaṃ bhavet / grāhakasmṛtisadbhāve
tatra tv evaiṣa gṛhyate // anvayavyatirekābhyāṃ siddhaivaṃ bhinnatā tayoḥ / evaṃ ca hetavo ’py ete
prasiddhāḥ sādhyadharmiṇi // (83b hi ŚV : ca TS; -pādaṃ grāhya- ŚVM ŚVT TS : -pādagrāhya- ŚVV ;
-smṛtinirbhāsāt- ŚVT ŚVV ; -smṛtir nirbhāsā ŚVM ; -smṛtisadbhāve TS; tatrāpy eṣaiva ŚV : tatra tv
evaiṣa TS; gṛhyate ŚVT ŚVV TS; dṛśyate ŚVM .)
ŚBh 30,5 on 1.1.4: na tu pūrvaṃ jñāyate / bhavati hi kadācid etad yaj jñāto ’rthaḥ san “ajñāta” ity
ucyate /
PSV I 5,1 on PS I 11cd (verse parts from PS are printed in bold): smṛter uttarakālaṃ ca (PS I 11c)
dvairūpyam iti sambandhaḥ / yasmāc cānubhavottarakālaṃ viṣaya iva jñāne ’pi smṛtir utpadyate
tasmād asti dvirūpatā jñānasya svasaṃvedyatā ca // kiṃ kāraṇaṃ na hy asāv avibhāvite // (PS I 11d)
na hy ananubhūtārthavedanasmṛtī rūpādismṛtivat / See the most recent studies of PS(V) I 11 by Kellner
(2010) and by Kataoka (2012).
ŚV Śūnyavāda 174cd (quoted in TS2 2065ab): dvayaṃ paraspareṇaiva bhinnaṃ sādhyaṃ rasādivat //
Kā 163,12 on ŚV Śūnyavāda 174cd: itaretaraparāmarśa itaretarāparāmarśāt… / NR 226,4 on ŚV
Śūnyavāda 174cd: parasparaparihāreṇa parāmarśāt … / TSP2 704,22 on TS2 2065: ekataraparāmarśe
saty aparasyāparāmarśanāt / (aparasyāparāmarśanāt em. following TSP1 : aparāmarśanāt TSPJ ,
TSPPa . The edition TSP2 emends to aparaparāmarśanāt, the Tibetan translation T has gzhan mi rtog
pa’i phyir ro.) Cf. TṬ 281,18 on ŚV Śūnyavāda 174cd: hetudṛṣṭāntau tāv eva /
NR 226,4 on ŚV Śūnyavāda 174cd: rūparasavad iti /
For the concept of virodha, see Bandyopadhyay 1988, Kyuma 1999 and Watanabe 2002.
Hiroko Matsuoka
207
1.1.4 The aikarūpyeṇa-ajñāna (bhedopalambhana) argument
The aikarūpyeṇa-ajñāna argument is as follows:18
[The two (i.e., the grasped and the grasper) are mutually different] because they
are not perceived as identical (aikarūpyeṇājñānāt),19 like another [mental]
continuum (santānāntara) and the cognition thereof.
Importantly, this aikarūpyeṇa-ajñāna argument is reformulated by Umbeka (8th c.) as
follows:20
Blue and the cognition thereof are different (bheda) from each other because
they are perceived separately (bhedopalambhanāt).
Here Umbeka is drawing a contrast between the aikarūpyeṇa-ajñāna argument and Dharmakīrti’s sahopalambhaniyama argument, the latter of which is formulated as follows:
PVin I 54ab:
sahopalambhaniyamād abhedo nīlataddhiyoḥ /
Blue and the cognition thereof are not different (abheda) [from each other]
because they are necessarily perceived together (sahopalambhaniyamāt).
According to Umbeka,21 Kumārila’s aikarūpyeṇa-ajñāna argument is thus a counterargument to Dharmakīrti’s sahopalambhaniyama argument. This represents the opposite
of Śāntarakṣita’s interpretation, namely that the latter argument of Dharmakīrti is actually
responding to the former by Kumārila. We will return to this point later.
1.2 Arguing against self-awareness
Kumārila goes on to formulate the following arguments of the second group to establish
that a relation between the grasped and the grasper never exists within a cognition.22
1. The jñāna-utpatti argument against a cognition grasping a part of itself;
2. The jñāna-utpatti argument against a cognition being grasped by a part of itself;
3. The jñānatva argument.
18
19
20
21
22
ŚV Śūnyavāda 175ab: aikarūpyeṇa vājñānāt santānāntarabuddhivat / TS2 2065cd: aikarūpyeṇa vājñānāt santānāntarabuddhivat // (vājñā- TS : cājñā- ŚVM : vijñā- ŚVT , ŚVV ; aikarūpyeṇa TSJ , ŚV :
ekyarūpeṇa TSPa , also adopted in the editions TS1 and TS2 .)
Cf. Kā 163,15 on ŚV Śūnyavāda 175ab: grāhyaṃ grāhakād bhinnaṃ tena sahaikarūpyeṇājñānāt / “The
grasped is not different from the grasper because the former is not perceived as identical with the latter.”
TṬ 282,9 on ŚV Śūnyavāda 175ab: nīlabuddhyor iti dharmitvam, paraspareṇa bheda iti sādhyo dharmaḥ / nīlam idam iti bhedopalambhanād iti hetuḥ /
TṬ 282,11 on ŚV Śūnyavāda 175ab: abhedasādhakasya sahopalambhasya, paroktasya vā sahopalambhaniyamād ity asyāsiddhatām anenāha /
TṬ 282,12 on ŚV Śūnyavāda 175cd–177ab: idānīm ekavijñānasambandhinor dharmayor grāhyagrāhakabhāvo nāstīti yad uktam, tatra prayogam āha – jñānaṃ svāṃśaṃ na gṛhṇātīti /
208
Śāntarakṣita on Two Kinds of Arguments for Self-Awareness
1.2.1 The two jñāna-utpatti arguments
These two arguments have a same logical reason and are thus treated as a pair.
The jñāna-utpatti argument against a cognition grasping a part of itself is as follows:23
A cognition does not grasp a part of itself (svāṃśa) because it is produced
from a cognition (jñānotpatteḥ), like a power of [cognition] itself (svaśakti).
The jñāna-utpatti argument against a cognition being grasped by a part of itself is as
follows:24
A cognition is not grasped by a part of itself because it is produced from a
cognition, like a power of [cognition] itself.25
Within the framework of the vijñānavāda,26 one may say the following: What is meant
by the word svaśakti is a latent impression (vāsanā).27 The latent impression is produced
from a cognition (jñānotpatti) and devoid of both the property of being grasped and that of
being grasper.28 Now, a cognition arises from its immediately preceding condition, which
23
24
25
26
27
28
ŚV Śūnyavāda 175cd (quoted in TS2 2066ab): jñānaṃ svāṃśaṃ na gṛhṇāti jñānotpatteḥ svaśaktivat //
ŚV Śūnyavāda 176ab (quoted in TS2 2066cd): grāhyatvapratiṣedhaś ca dvayahīnā hi vāsanā /
(grāhyatvaprati- TS; grāhyavat prati- ŚVM ; grāhyatvaṃ prati- ŚVV ; hi vā- ŚVM TS; dvivā- ŚVT
ŚVV ) “[For the same reason,] there is also the negation of [a cognition] being grasped [by a part of
itself]. For, a latent impression (vāsanā) is devoid of both [the property of being the grasped and that of
the grasper].” TṬ 282,13 on ŚV Śūnyavāda 176a: jñānaṃ svāṃśaṃ na gṛhṇāti na vā svāṃśena gṛhyate
… / NR 226,10 on ŚV Śūnyavāda 176a: jñānaṃ svāṃśaṃ na gṛhṇāti nāpi svāṃśena gṛhyate /
See TSP2 705,11 on TS2 2066c: tadyathā – jñānāṃśo na jñānagrāhyaḥ / jñānād utpannatvāt / tadvat
vāsanāvat / “A part of cognition is not grasped by the same cognition because it is produced from a
cognition, like that [power of itself], namely, like a latent impression.”
According to Sucarita (10th c.), Kumārila formulates the jñāna-utpatti arguments from the standpoint
of the “old” Buddhist who regards a latent impression (vāsanā) as distinct from a cognition (jñānātiriktavāsanā). On the other hand, the “modern” Buddhist holds that a latent impression is not
distinguished from the immediately preceding cognition (samanantarajñāna) itself. Kā 163,27 on ŚV
Śūnyavāda 175cd–176ab: etac ca cirantanabauddhābhiprāyeṇa jñānātiriktavāsanāpakṣe sthitvoktam
iti draṣṭavyam / na tūpādānāparanāmno viśiṣṭāt samanantarajñānād anyāṃ kāñcid vāsanām arvācīnā
manyante // For Prajñākaragupta (c. 750–810), a latent impression is not the power produced from
the preceding cognition (pūrvavijñānajanitā śaktiḥ), but the immediately preceding cognition itself
(samanantaravijñānātmabhūtā). PVA 356,6: vāsaneti hi pūrvavijñānajanitāṃ śaktim āmananti vāsanāsvarūpavidaḥ /; PVA 356,23: [na ca] samanantaravijñānātmabhūtā vāsaneṣyate // As Kobayashi
(2001: 327) has pointed out, Yamāri (c. 1000–1060) refers to a discrepancy in the views of Dharmottara
(c. 740–800) and Prajñākaragupta on whether a latent impression is distinct from the immediately
preceding cognition or not. PVAṬS II (D 259a4; P 347b7): de ltar na re zhig slob dpon chos mchog la
sogs pas ’dod pa’i bag chags kyi phyogs la yang pha rol pos brjod pas nyes pa bsal nas rang gi ’dod pa
brjod pa ni de ma thag pa’i zhes bya ba’o // See also PVin I 43,14–44,2 for Dharmakīrti’s theory of the
latent impression.
TṬ 282,14 on ŚV Śūnyavāda 175d: śloke svaśaktiśabdena vāsanām āha /
Kā 163,25 on ŚV Śūnyavāda 176b: na ca sādhyavikalo dṛṣṭāntaḥ / dvayahīnavāsanābhyupagamāt /
bauddhānāṃ hi svāṃśaṃ na gṛhṇāti / na ca tena gṛhyate / NR 316,17 on ŚV Śūnyavāda 176b: vāsanāvat / sā hi svāṃśaṃ na gṛhṇāti, nāpi svāṃśena gṛhyate iti / TSP2 705,12 on TS2 2066cd: katham
asminn anantare prayogadvaye (emendation following TSP1 : prayogādvaye TSPJ , TSPPa (also adopted
in TSP2 ); cf. gtan tshigs gnyis po… la T) ’pi sādhyadharmānvito dṛṣṭāntaḥ siddha ity āha – dvayahīnā
hi vāsaneti / dvayena grāhyagrāhakatvena /
Hiroko Matsuoka
209
is a cognition (samanantarapratyaya).29 Whatever arises from a cognition can neither be
grasped nor grasper, so that a cognition can neither be grasped or grasper with respect to a
part of itself (svāṃśa).
1.2.2 The jñānatva argument
The jñānatva argument is as follows:30
The cognition of Caitra does not cognize the grasped part of the cognition
occurring in [Caitra] himself31 because it is a cognition [itself] (jñānatvāt),
like [the cognition] occurring in another body [e.g., of Maitra].32
It is clear that the cognition of Maitra does not apprehend the cognition of Caitra, which
also implies that the cognition of Maitra does not apprehend a part of the cognition of Caitra.
In the same manner, the cognition of Caitra does not apprehend a part of the cognition of
Caitra himself. This is attributed to the fact that the respective cognitions of Caitra and
Maitra equally have the property of being cognitions.
In other words, this argument is based on what it means to be a cognition (jñānatva). This
will also be of importance for understanding the development of Dharmakīrti’s saṃvedana
argument and will be taken up again below.
2. Śāntarakṣita’s defense of self-awareness
Let us now examine how Śāntarakṣita rebuts all the arguments Kumārila formulated in ŚV
Śūnyavāda 172cd–177ab.
2.1 The sahopalambhaniyama argument
Śāntarakṣita commences his refutation of the first group of arguments as follows:
TS2 2068:
apṛthag vedanāt pūrvaṃ tasyaiva33 pratipāditāt /
aikarūpyāparijñānaparyanteṣu na siddhatā //
29
30
31
32
33
Kā 163,20 on ŚV Śūnyavāda 175d: jñānotpatter iti hetuḥ / tad dhi samanantarapratyayād utpadyata
iti bauddhā manyante /
ŚV Śūnyavāda 176cd–177ab (quoted as TS2 2067): caitrajñānaṃ tadudbhūtajñānāṃśagrāhyabodhakam // jñānatvān na bhaved yadvat tasya dehāntarodbhavam /
Kā 164,9 on ŚV Śūnyavāda 176cd–177ab: caitrajñānaṃ dharmi // taccaitrodbhūto yo jñānāṃśo grāhyas
tasya bodhakaṃ na bhavatīti sādhyam /
TSP2 705,15 on TS2 2067: yadvat tasya caitrajñānodbhūtajñānāṃśasya maitrādidehāntarodbhavaṃ
jñānam //
tasyaiva em. (de nyid T) : tad atra TSJ , TSPa (adopted in TS1 and TS2 ).
210
Śāntarakṣita on Two Kinds of Arguments for Self-Awareness
Since [the non-difference between an object and its cognition]34 has already
been explained as based on [the logical reason of] not being separately cognized, [all the reasons] up to the reason “not being perceived as identical” have
[the fault of] being unestablished.
According to Kamalaśīla,35 with the expression apṛthag vedanāt “because they are not
cognized separately,” Śāntarakṣita refers to the reason sahopalambhaniyamāt “because
they are necessarily perceived together,” of Dharmakīrti’s sahopalambhaniyama argument.
Since this argument establishes the non-difference between an object and the cognition
thereof, all the four reasons in Kumārila’s first group of arguments are to be considered
unestablished (asiddha).
It is to be noted, in passing, that Śāntarakṣita reformulates the sahopalambhaniyama
argument earlier in the TS as follows:36
TS2 2029–2030:
yatsaṃvedanam eva syād yasya saṃvedanaṃ dhruvam /
tasmād avyatiriktaṃ tat tato vā na vibhidyate //
yathā nīladhiyaḥ svātmā dvitīyo vā yathoḍupaḥ /
nīladhīvedanaṃ cedaṃ nīlākārasya vedanam //
If a cognition of X is necessarily a cognition of Y, X is not different from Y, or
Y does not differ from X. Just as the cognition’s own essence [is not different
from the cognition of blue, or the cognition of blue does not differ from the
cognition’s own essence], or just as the second moon [is not different from
the first one, or the first moon does not differ from the second one]. And this
cognition of the form of blue (nīlākāra) is [necessarily] a cognition of the
cognition of blue (nīladhī ). [Therefore, the form of blue is not different from
the cognition of blue, or the cognition of blue does not differ from the form of
blue.]
As noted by Umbeka, Dharmakīrti’s sahopalambhaniyama argument basically has the
reverse factors of the aikarūpyeṇa-ajñāna argument. Namely, the property to be proved
(sādhyadharma) is “being non-different [from each other]” (abheda) rather than “being
different [from each other]” (bhinna), and the logical reason is “being necessarily perceived together” (sahopalambhaniyama) rather than “not being perceived as identical”
(aikarūpyeṇājñāna) or “being perceived separately” (bhedopalambhana). In Dharmakīrti’s
treatment, this is the main argument corresponding to Kumārila’s first group, meaning
that, if Dharmakīrti’s treatment is later than Kumārila’s, then the aikarūpyeṇa-ajñāna
argument is being taken to imply or speak for the other three arguments of the first group,
34
35
36
TSP2 705,20 on TS2 2068: abhedasya nīlataddhiyoḥ prasādhitatvāt… / See n. 58.
TSP2 705,18 on TS2 2068: apṛthag vedanād iti nīlataddhiyoḥ sahopalambhaniyamāt / apratyakṣopalambhasya nārthadṛṣṭiḥ prasidhyati // (= PVin I 54cd) ity ataḥ svasaṃvitprasādhanena pratipāditād
abhedasya nīlataddhiyoḥ prasādhitatvād aikarūpyāparijñānaparyantā hetavo na siddhāḥ // (For nīlataddhiyoḥ sahopalambhaniyamāt in the quotation TSP cf. PVin I 54ab.)
See Matsuoka 2011 for the interpretation of TS2 2029–2030.
Hiroko Matsuoka
211
namely, the tatsaṃvitti-asaṃvitti, tatparāmarśa-aparāmarśa, and itaretaraparāmarśaitaretarāparāmarśa arguments.
2.2 Refutations of the tatsaṃvitti-asaṃvitti and tatparāmarśa-aparāmarśa arguments
After collectively refuting Kumārila’s first group of arguments, Śāntarakṣita discusses the
tatsaṃvitti-asaṃvitti and tatparāmarśa-aparāmarśa arguments individually.
2.2.1 Refutation of the tatsaṃvitti-asaṃvitti argument
To begin with, the logical reason of this argument is unestablished (asiddha). In order to
show this, Śāntarakṣita cites PVin I 54cd with slight modification and further argues.37
TS2 2073:
aprasiddhopalambhasya nārthavittiḥ prasidhyati /38
tan na grāhyasya saṃvittir grāhakānubhavād ṛte //39
For someone whose perception is not established, the cognition of the object
is not established. Therefore, there is no cognition of the grasped without an
experience of the grasper.
The point is that an object can be established, in Dharmakīrti’s words, as perceived only
when the cognition of an object is perceived. It cannot be the case that the cognition is not
perceived when its object is perceived.
In the tatsaṃvitti-asaṃvitti argument, Kumārila has increased the scope of the reason
that a cognition is not cognized when its object is cognized on the basis of the statement in
Śabara’s Bhāṣya that only an object as connected with the external world (bahirdeśasambaddha) is perceived. In connection with this statement, Śāntarakṣita points out that what is
perceived is not always an external object, and thereby argues that Kumārila’s reason is
inconclusive (anaikāntika).
TS2 2074:
asvasthalocanair dṛṣṭaṃ tathā pītādy avekṣyate /
vispaṣṭaṃ40 grāhakāṃśāc ca saṃvedyaṃ na tathā param //
[Unreal objects like] yellow, etc., that are seen by someone with an eye-disease
are vividly (vispaṣṭam) perceived in that way [i.e., connected with the external
37
38
39
40
See Kellner 2011 for a comparison of the two different arguments for self-awareness that are presented
in PS(V) I 11d–12 and PVin I 54cd.
This is a quotation of PVin I 54cd, with aprasiddhopalambhasya nārthavittiḥ TS : apratyakṣopalambhasya nārthadṛṣṭiḥ PVin, which reads “For someone who does not perceive perception, the perception
of the object is not established either” (trans. Kellner 2011: 420). For the further discussion with PVin I
54cd as Dharmakīrti’s argument of infinite regress, see Kellner 2011.
grāhyasya… // TS; ŚV Śūnyavāda 79cd.
vispaṣṭaṃ em. (gsal por T, vispaṣṭam TSPJ ) : niṣkṛṣṭaṃ TSJ , TSPa (adopted in TS1 and TS2 ).
212
Śāntarakṣita on Two Kinds of Arguments for Self-Awareness
world],41 and they are not perceived [separately (niṣkṛṣṭam)] from the grasping
part.42 Likewise for another [i.e. yellow as seen by someone with healthy
eyes].43
Needless to say, in the case of an erroneous cognition, someone with an eye-disease
perceives an object that is not externally existent.
2.2.2 Refutation of the tatparāmarśa-aparāmarśa argument
In attacking the tatparāmarśa-aparāmarśa argument, Śāntarakṣita begins by saying the
following:
TS2 2075:
alakṣitaviśeṣā ca grāhyarūpe44 ca sā smṛtiḥ /
sarvato bhinnarūpe tu na sābhyāsādyasambhavāt //
A recollection by which a particular is not observed occurs with respect
to an object to be grasped. But it does not occur with respect to an object
distinguished from all others since habituation (abhyāsa) and the other [causal
factors for ascertainment] are not possible.
According to Kamalaśīla,45 an individual entity cannot be remembered because recollection
(smṛti) is not capable of having an individual for its object, unlike perception. A perception
of an individual entity can cause the perceiver to determine (adhyavasāya) that the object
is an individual if one of several causal factors for ascertainment (niścayahetu) is present:
41
42
43
44
45
TSP2 706,23 on TS2 2074: vispaṣṭam ity atra chedaḥ / tatheti yathā satyābhimataṃ pītādi bahirdeśasambaddhaṃ vispaṣṭam upalabhyate tathā kāmalādyupahatanayanopalabdham api samīkṣyate /
(vispaṣṭam ity TSPJ (gsal por nges pa zhes bya ba T) : nikṛṣṭam ity TSPPa ; both editions TSP1 and TSP2
have niṣkṛṣṭam ity.)
TSP2 706,25 on TS2 2074: grāhakāṃśāc ca saṃvedyaṃ neti chedaḥ / grāhakāṃśād iti niṣkṛṣṭaṃ ity
adhyāhāryam /
TSP2 707,7 on TS2 2074: tathā param iti satyābhimatam api pītādi /
grāhyarūpe em. (gzung ba’i ngo bo la T) : bāhyarūpe TSJ , TSPa (adopted in TS1 and TS2 ).
TSP2 707,14 on TS2 2075cd: etad uktaṃ bhavati – na tāvad vikalpasya yathāvasthitavastugrahaṇasāmarthyam / tasyāvastuviṣayatvāt / kevalaṃ tathābhūtapadārthānubhavabalād yatraivārthitvādayo
niścayahetavaḥ santi tatra tadākārādhyavasāyī smārtaḥ pratyayo nirviṣaya eva / paramārthataḥ svapratibhāse ’narthe ’rthādhyavasāyena pravṛtter bhrānta eva sarvo jāyate / tasya tv adhyavasāyavaśena
viṣayavyavasthā na paramārthataḥ / na ca grāhyādhyavasāyaḥ smṛter api vidyate / kevalaṃ tathāvidhābhyāsapāṭavapratyāsattitāratamyādikāraṇābhāvād alakṣitaviśeṣā bhavati / yena smaraṇāntarād
viśiṣyate / Variant readings: yatraivā- TSPJ , adopted in TSP1 and TSP2 (gang kho na la T) : yathaivāTSPPa ; paramārthataḥ n.e. T; -ābhyāsapāṭavapratyāsattitāratamyādi- em. Cf. tasya kāraṇam abhyāsaḥ pratyāsattis tāratamyabuddhipāṭavaṃ cetyādi / TSP2 244,18 on TS2 587 : -ābhyāsapāṭavādar
apratyāsattitāratamyādi- [sic] or -ābhyāsapāṭavader apratyāsattitāratamyādi- [sic] TSPJ , TSPPa . The
editions TSP1 and TSP2 read -ābhyāsapāṭavāder apratyāsattitāratamyādi-, T reads goms pa gsal ba
gzhan myur ba dang ches myur ba la sogs pa’i. The emphasized part corresponds to PVin II 46,7:
svapratibhāse ’narthe ’rthādhyavasāyena pravartanāt. See also PVSV 31,16–32,12 on PV I 58 for
Dharmakīrti’s explanation of causal factors for ascertainment (niścayapratyaya). I referred to Kellner’s
(2004) descriptions of PVSV 32,5–12 for the translations of niścayahetu, tathāvidhābhyāsa, pāṭava,
pratyāsattitāratamya.
Hiroko Matsuoka
213
a [state of] habituation due to similar situations (tathāvidhābhyāsa), acuity [of cognition]
(pāṭava), or the [difference in] degree of proximity (pratyāsattitāratamya). There are no
such causal factors for ascertainment in the case of recollection, which means that its
objects will always be determined as universals instead of particulars.
Now, if recollection has only a universal character for its object, how could the object
be remembered?46
Śāntarakṣita answers the question by saying the following:
TS2 2076:
gṛhīta iti ko ’py evaṃ nānyathā smaraṇaṃ bhavet /
śuddhasphaṭikasaṅkāśe vidyate47 smaraṇaṃ na ca //
Otherwise, there could be no recollection as in the form “Something was
grasped.” And there is no recollection [of a grasper]48 like a pure crystal.
There is no recollection of a cognition like a colorless transparent crystal that is not marked
with the form of its object. Even if there is a case where, after cognizing a certain object, one
cannot remember the object in a specific way, it does not follow that one is remembering a
cognition which grasped no object.
Śāntarakṣita concludes that the logical reason presented in PS I 11c is established
whereas that of the tatparāmarśa-aparāmarśa argument is not.
2.3 The saṃvedana argument
Śāntarakṣita then continues on to refute the arguments of group 2 as follows:
TS2 2077:
kambupītādivijñānair hetvoḥ49 paścimayor api /
anaikāntikatā vyaktaṃ dig eṣānyatra sādhane //
In view of the cognition of a yellow conch-shell and the like, the latter two
logical reasons are clearly inconclusive (anaikāntika). This is the way to deal
with the other reasons [to prove the existence of an external object, too].50
Like in TS2 2074,51 Śāntarakṣita here gives as an example of an erroneous cognition in
order to point out that the logical reasons of the two jñāna-utpatti arguments and of the
jñānatva argument are all inconclusive.
46
47
48
49
50
51
TSP2 707,22 on TS2 2076: syād etat – katham avasīyate grāhyādhyavasāyo ’trāsti smṛter ity āha –
gṛhīta ityādi /
vidyate em. following yod pa T : vedyate TSJ , TSPa (adopted in TS1 , TS2 ).
TSP2 707,24 on TS2 2076: na cāpi kevalo grāhyākārānaṅkitamūrtitayā grāhakaḥ śuddhasphaṭikasaṅkāśaḥ smaryate /
hetvoḥ em. with TS2 (gtan tshigs… gnyis po T) : hetoḥ TSJ (adopted in TS1 ), heto [sic] TSPa .
TSP2 708,10 on TS2 2077d: eṣā dig ity anyatrāpi bahirarthasādhane paropanyaste eṣā dūṣaṇadik /
See section 2.2.1.
214
Śāntarakṣita on Two Kinds of Arguments for Self-Awareness
To explain, suppose that there occurs a cognition of yellow with reference to a white
conch-shell. The erroneous cognition cognizes the form (ākāra) of yellow, which is a part
of the cognition and which has no counterpart in the external world.52 If a continuum of
cognitions is accepted, then the erroneous cognition must be presumed to arise from an
immediately preceding cognition. Alternatively, the cognition, which is identical with the
form of yellow, is cognized by the grasping part of the erroneous cognition.53
It is to be noted that, according to Kamalaśīla,54 the other (anyat) cognition, that is, the
non-erroneous cognition, also deviates from “not cognizing a part of [cognition] itself.”
Even the cognition of a white conch-shell or that of someone with healthy eyes also cognizes
a part of the cognition, which means that every cognition cognizes itself.
This reminds us of Dharmakīrti’s saṃvedana argument which was reformulated by
Śāntarakṣita as follows:
TS2 2032:
saṃvedanam idaṃ sarvaṃ na cārthāntaragocaram /
saṃvedanasvabhāvatvāt 55 svātmasaṃvedanaṃ yathā //
Every cognition does not have for its object anything other [than the cognition
itself] because it has the essential nature of a cognition, like the cognition of
one’s own self.
According to Dharmakīrti as well as Śāntarakṣita, no cognition cognizes any object distinct
from the cognition itself because its essential nature is that of a cognition.
For both Dharmakīrti and Kumārila, the essential nature of a cognition is being that
which illuminates (prakāśaka).56 What is illuminated, however, differs for each: the cognition itself or an external object, respectively. Therefore, the same logical reason “being a
cognition itself” or “having a cognition as its essence” will yield opposite implications;
for Kumārila, a cognition never cognizes a part of the cognition (svāṃśa), whereas for
Dharmakīrti, it never cognizes anything different from the cognition.
Thus, in the same way as was the case with the aikarūpyeṇa-ajñāna and sahopalambhaniyama arguments (cf. section 2.1), here too, Dharmakīrti seems to have reversed
only the properties to be proved of the jñānatva argument and then let this imply the other
two jñāna-utpatti arguments of the second group.
52
53
54
55
56
TSP2 708,6 on TS2 2077: tathā hi – yathā yadi pītaśaṅkhādijñānaṃ jñānotpannam api sat svāṃśaṃ
pītādyākāraṃ gṛhṇāti / (tathā hi yathā TSPJ : yathā TSP2 : ’di ltar na T; n.e. TSPPa , TSP1 .)
TSP2 708,7 on TS2 2077: yathā ca jñānam api sat jñānāṃśasya pītāder grāhyasya bodhakaṃ bhavati /
TSP2 708,7 on TS2 2077: tathānyad apīti vyabhicāritā hetvoḥ /
For saṃvedanasvabhāvatvāt TSJ , TSPa (adopted in TS2 , cf. also rig pa’i rang bzhin nyid kyi phyir TSPP ,
rig pa’i rang bzhin nyid kyis rig TSPD ), TS1 emends to saṃvedanaṃ ca nīlasya.
PV III 329: prakāśamānas tādātmyāt svarūpasya prakāśakaḥ / yathā prakāśo abhimatas tathā dhīr
ātmavedinī //; TS2 2081a: vijñānatvaṃ prakāśatvam … / ŚV Śūnyavāda 187ab quoted in TS2 2017ab:
prakāśakatvaṃ bāhye ’rthe śaktyabhāvāt tu nātmani /
Hiroko Matsuoka
215
3. Against nirākāravāda
Lastly, let us consider how Śāntarakṣita regards the function of the sahopalambhaniyama
and saṃvedana arguments.
After reformulating the saṃvedana argument, he states:
TS2 2034:
śuddhasphaṭikasaṅkāśam arthākārair anaṅkitam /
yair iṣṭaṃ vedanaṃ kaiścid idaṃ tān prati sādhanam //
Both57 of these [arguments] are addressed to those who maintain that a cognition is not marked with the form of its object, being like a pure crystal.
That is, according to Śāntarakṣita (as understood by Kamalaśīla, see n. 57) both the sahopalambhaniyama and saṃvedana arguments are meant to prove self-awareness (svasaṃvid),
which, stated from a different perspective, is the same as proving that a cognition is marked
with the form of its object (sākāratā).58 Those who claim that a cognition is not marked
with the form of its object (nirākāratā) are none other than nirākāravādins represented by
Kumārila.
Thus, according to Śāntarakṣita, Dharmakīrti’s aim in including both of these arguments is not to prove svasaṃvid or sākāratā yet again after it had already been proved
(siddhasādhana) by the other of the two, as suggested by Śubhagupta;59 indeed, it would
not be appropriate, given the rules of debate, for Dharmakīrti to offer two arguments for the
same thing only for the sake of improving upon the position laid out by Dignāga. Rather,
as Śāntarakṣita and Kamalaśīla helpfully point out, these two arguments must be meant for
refuting nirākāravāda; after all, in refuting an opposing position, it is indeed permitted to
adopt as many approaches as might be necessary.
4. Concluding remarks
According to Śāntarakṣita, the reason Dharmakīrti developed his arguments is to defend
Dignāga’s theory of self-awareness against the attack by Kumārila. In order to counter
Dignāga’s arguments for dvirūpatā and svasaṃvedyatā respectively, Kumārila had argued
that an object and its cognition are different from one another on the basis of the fact that
they are perceived differently (bhedopalambhana), and, secondly, that a cognition never
cognizes a part of itself on the basis of its essential nature of being a cognition (jñānatva).
57
58
59
TSP2 696,14 on TS2 2034d: idam iti dvividham api sādhanaṃ nirākāravādinaṃ prati yatas tena na
siddhasādhyatā /
The logical reasons sahopalambhaniyama and saṃvedana are collectively called svasaṃvitprasādhana
or sākāratāsiddhisādhana by Kamalaśīla. TSP2 695,14 on TS2 2032: dvitīyam api sākāratāsiddhaye
sādhanam āha – saṃvedanam idam ityādi / TSP2 705,20 on TS2 2068: ataḥ svasaṃvitprasādhanena
pratipāditād abhedasya nīlataddhiyoḥ prasādhitatvāt… / See n. 34.
In BASK 87 Śubhagupta points out that the saṃvedana argument is not valid because it proves what
is already proved (siddhasādhana) by the sahopalambhaniyama argument. BASK 87 is quoted in
TSP2 696,11 on TS2 2034: sākārajñānapakṣe ca tannirbhāsasya vedyatā / tasyābhede ca saṃsādhye
siddhasādhanatā bhavet //
216
Śāntarakṣita on Two Kinds of Arguments for Self-Awareness
Accordingly, Dharmakīrti formulated the sahopalambhaniyama and saṃvedana arguments
as counter-arguments against these arguments by Kumārila.
Although it is unacceptable to the ŚV-commentators such as Umbeka, Śāntarakṣita’s
interpretation of the relation among the various arguments given in PS I, ŚV Śūnyavāda and
PVin I is reasonable. This is because, although it is clear that Kumārila, in his exhaustive
detail, was attempting to refute Dignāga’s arguments from multiple perspectives, it should
also be apparent from the dual nature of Dharmakīrti’s argumentation that he himself
was also responding to a challenge. To do so, Dharmakīrti simply reversed the factors
of the aikarūpyeṇa-ajñāna and jñānatva arguments in particular, implying that – from
his perspective – these two arguments would also cover the other five arguments which
Kumārila employed. If Dharmakīrti thereby successfully boiled down these discussions
to their essential issues, this could then represent a rather elegant response to Kumārila’s
detailed challenge.
References and abbreviations
Primary sources
BASK Bāhyārthasiddhikārikā: See Mikogami 1986.
D sDe dge edition of the Tibetan bstan ’gyur: sDe dge Tibetan Tripiṭaka bsTan ’gyur –
preserved at the Faculty of Letters, University of Tokyo, ed. J. Takasaki, Z. Yamaguchi,
and Y. Ejima. Tshad ma 1–20. Tokyo 1981–1984.
Kā Kāśikā: The Mīmāṃsāślokavārttika with the Commentary Kāśikā of Sucaritamiśra,
ed. K. Sāmbaśiva Śāstrī. Trivandrum 1926–1943.
NR Nyāyaratnākara: Mīmāṃsāślokavārttikam. Varanasi: 1898.
P Peking edition of the Tibetan bstan ’gyur: The Tibetan Tripiṭaka, Peking Edition, ed. D.
T. Suzuki. Tokyo/Kyoto 1955–1961.
PS I Pramāṇasamuccaya, chapter 1 (Pratyakṣa), ed. E. Steinkellner. https://www.oeaw.ac.
at/fileadmin/Institute/IKGA/PDF/forschung/buddhismuskunde/dignaga_PS_1.pdf,
last visited 09-08-2019.
PSV Pramāṇasamuccayavṛtti: See PS I.
PV I Pramāṇavārttika, ch. 1 (Svārthānumāna): See PVSV.
PV III Pramāṇavārttika, ch. 3 (Pratyakṣa): See PVV.
PVA Pramāṇavārttikālaṅkāra: Pramāṇavārtikabhāshyam or Vārtikālaṅkāraḥ of Prajñākaragupta: Being a commentary on Dharmakīrti’s Pramāṇavārtikam, ed. R. Sāṅkṛtyāyana. Patna 1953.
PVAṬS II Pramāṇavārttikālaṃkāraṭīkā Supariśuddhā, ch. 2 (Pratyakṣa): Tibetan translation, D 4226 me 1–328a7; P 5723 me 1–436a8 (Vol. 135, pp. 77–252).
PVin I, II Pramāṇaviniścaya, chapters 1 (Pratyakṣa)–2 (Svārthānumāna): Dharmakīrti’s
Pramāṇaviniścaya: Chapters 1 and 2, ed. E. Steinkellner. Beijing/Vienna 2007.
PVSV Pramāṇavārttikasvavṛtti: The Pramāṇavārttikam of Dharmakīrti, the first chapter
with the autocommentary, text and critical notes, ed. R. Gnoli. Rome 1960.
PVV Pramāṇavārttikavṛtti: Ācārya-Dharmakīrteḥ Pramāṇa-vārttikam ācārya- Manorathanandikṛtayā vṛttyā saṃvalitam, ed. R. Sāṅkṛtyāyana. Appendix to Journal of the
Bihar and Orissa Research Society 24–26 (1938–1940).
Hiroko Matsuoka
217
ŚBh Śābarabhāṣya: See Frauwallner 1968.
ŚV Ślokavārttika: ŚVV =ŚVT =ŚVM .
ŚVM Ślokavārttika: See TṬ.
ŚVT Ślokavārttika: See Kā.
ŚVV Ślokavārttika: See NR.
T Tibetan translation of TS(P) (in cases where D and P have identical readings).
TS(P) Tattvasaṅgraha/-Pañjikā: TS(P)J =TS(P)Pa =TS(P)1 =TS(P)2
TS(P)1 Tattvasaṅgraha/-Pañjikā: Tattvasaṅgraha of Śāntarakṣita with the Commentary of
Kamalaśīla, ed. E. Kṛṣṇamācārya. 2 vols. Baroda 1926.
TS(P)2 Tattvasaṅgraha/-Pañjikā: Tattvasaṃgraha of Ācārya Shāntarakṣita with the Commentary “Pañjikā” of Shrī Kamalashīla, ed. D. Śāstri. 2 vols. Varanasi 1968.
TSD Tattvasaṅgraha, Tibetan translation, D 4266 ze 1–133a6.
TSJ Jaisalmer Ms of the Tattvasaṅgraha, no. 377. In A Catalogue of Manuscripts in
Jaisalmer Jain Bhandaras, comp. Muni Jambūvijayajī. Delhi/Jaisalmer 2000, 40.
TSP Tattvasaṅgraha, Tibetan translation, P 5764 ’e 1–159a5.
TSPa Pāṭāṇa Ms of the Tattvasaṅgrahapañjikā, no. 6679. In Catalogue of the Manuscripts
of Pāṭāṇa Jain Bhaṇḍāra, comp. Muni Jambūvijayajī. 3 vols. Ahmedabad 1991, vol.
1, p. 199.
TSPD Tattvasaṅgrahapañjikā, Tibetan translation, D 4267 ze 133b1–’e 331a7.
TSPJ Jaisalmer Ms of the Tattvasaṅgrahapañjikā, no. 378. In A Catalogue of Manuscripts
in Jaisalmer Jain Bhandaras, comp. Muni Jambūvijayajī. Delhi/Jaisalmer 2000, 40.
TSPP Tattvasaṅgrahapañjikā, Tibetan translation, P 5765 ’e 159b2–ye 495a7 (Vols.
138–139, pp. 200–164).
TSPPa Pāṭāṇa Ms of the Tattvasaṅgrahapañjikā, no. 6680. In Catalogue of the Manuscripts
of Pāṭāṇa Jain Bhaṇḍāra, comp. Muni Jambūvijayajī. 3 vols. Ahmedabad 1991, vol.
1, p. 199.
TṬ Tātparyaṭīkā: Ślokavārttikavyākhyā Tātparyaṭīkā of Uṃveka Bhaṭṭa, ed. S. K. Rāmanātha Śāstrī. Madras 1971.
Secondary literature
Bandyopadhyay 1988 N. Bandyopadhyay, The Concept of Contradiction in Indian Logic
and Epistemology. Journal of Indian Philosophy 16.3 (1988) 225–246.
Bhatt 1962 G. P. Bhatt, Epistemology of the Bhāṭṭa School of Pūrva-Mīmāṃsā. Varanasi
1962.
Franco 2015–2018 E. Franco, Xuanzang’s Silence and Dharmakīrti’s Dates. Wiener Zeitschrift für die Kunde Südasiens LVI–LVII (2015–2018) 117–141.
Frauwallner 1961 E. Frauwallner, Landmarks in the History of Indian Logic. Wiener
Zeitschrift für die Kunde Süd- und Ostasiens 5 (1961) 125–48.
Frauwallner 1962 E. Frauwallner, Kumārila’s Bṛhaṭṭīkā. Wiener Zeitschrift für die Kunde
Süd- und Ostasiens 6 (1962) 78–90.
Frauwallner 1968 E. Frauwallner, Materialien zur ältesten Erkenntnislehre der Karmamīmāṃsā. Wien 1968.
Iwata 1991 T. Iwata, Sahopalambhaniyama. Stuttgart 1991.
218
Śāntarakṣita on Two Kinds of Arguments for Self-Awareness
Kataoka 2010 K. Kataoka, Ninshiki-shudan to kekka tono taisho no soui: Kumārila to
Dharmakīrti [Kumārila and Dharmakīrti on the potential problem of pramāṇa and
phala having different objects]. Indogaku bukkyōgaku kenkyū 59.1 (2010) 418–412.
Kataoka 2011 K. Kataoka, Kumārila on Truth, Omniscience, and Killing: Part 1: A Critical Edition of Mīmāṃsā-Ślokavārttika ad 1.1.2 (Codanāsūtra). Wien 2011.
Kataoka 2012 K. Kataoka, Jikoninshiki to nimen-sei: Dignāga ni okeru ninshiki no kozo
[svasaṃvedana and dvairūpya: Cognitive Structure according to Dignāga]. Indogaku
bukkyōgaku kenkyū 60.2 (2012) 1020–1013.
Kellner 2004 B. Kellner, Why Infer and not just Look? Dharmakīrti on the Psychology
of Inferential Processes. In: The Role of the Example (dṛṣṭānta) in Classical Indian
Logic, ed. S. Katsura, et al. Wien 2004, 1–51.
Kellner 2010 B. Kellner, Self-Awareness (svasaṃvedana) in Dignāga’s Pramāṇasamuccaya and -vṛtti: A Close Reading. Journal of Indian Philosophy 38 (2010) 203–231.
Kellner 2011 B. Kellner, Self-awareness (svasaṃvedana) and Infinite Regresses: A Comparison of Arguments by Dignāga and Dharmakīrti. Journal of Indian Philosophy
39 (2011) 411–426.
Kobayashi 2001 H. Kobayashi, Prajñākaragupta no senzai inshō setsu [Prajñākaragupta’s
Theory of vāsanā]. Indogaku bukkyōgaku kenkyū 50.1 (2001) 325–327.
Krasser 1999 H. Krasser, Dharmakīrti’s and Kumārila’s Refutations of the Existence of
God: A Consideration of Their Chronological Order. In: Dharmakīrti’s Thought and
Its Impact on Indian and Tibetan Philosophy, ed. S. Katsura. Wien 1999, 215–223.
Krasser 2012 H. Krasser, Bhāviveka, Dharmakīrti and Kumārila. In: Devadattīyam. Johannes Bronkhorst Felicitation Volume, ed. F. Voegeli, et al. Bern 2012, 535–594.
Kyuma 1999 T. Kyuma, bheda and virodha. In: Dharmakīrti’s Thought and Its Impact on
Indian and Tibetan Philosophy, ed. S. Katsura. Wien 1999, 225–232.
Matsuoka 2011 H. Matsuoka, Śāntarakṣita’s Interpretation of the sahopalambhaniyama
Proof Re-examined. Indogaku bukkyōgaku kenkyū 59.3 (2011) 157–162.
Matsuoka 2014 H. Matsuoka, Sho-shinjitsu-ron ni miru ukeisho-chishiki-ron-ronsho
seiritsu no haikei [Śāntarakṣita on Dharmakīrti’s Proof of Sākāratā]. Indogaku
bukkyōgaku kenkyū 62.2 (2014) 983–979.
Mikogami 1986 E. Mikogami, Śubhagupta no Bāhyārthasiddhikārikā [Bāhyārthasiddhikārikā of Śubhagupta]. Ryūkoku daigaku ronshū 429 (1986) 2–44.
Taber 1986–1992 J. Taber, Further Observations on Kumārila’s Bṛhaṭṭīkā. Journal of
Oriental Research 56–62 (1986–1992) 179–189.
Taber 2010 J. Taber, Kumārila’s Buddhist. Journal of Indian Philosophy 38.3 (2010)
279–296.
Teraishi 2002 Y. Teraishi, Ślokavārttika Śūnyavāda shō no kenkyū (6): Wayaku to Kaisetsu
[A Study of the Ślokavārttika, Śūnyavāda chapter: A Japanese Translation with
Exposition. Part 6]. Kyūshūryūkoku tankidaigaku kiyō 48 (2002) 153–174.
Watanabe 2002 T. Watanabe, Dharmakīrti no hininshiki ron: Sōhan kankei wo chūshin
ni [Dharmakīrti’s Theory of “Non-Cognition” (anupalabdhi): From the Viewpoint
of “Incompatibility” (virodha)]. Nanto bukkyō 81 (2002) 54–80.
Watanabe 2014 T. Watanabe, Buddhist Critiques of the Sāṅkhya Theory of Causality:
Dharmakīrti and His Predecessors. Paper presented at the XIIth Congress of the
219
Hiroko Matsuoka
International Association of Buddhist Studies, University of Vienna, 18–23 August,
2014.
Yoshimizu 2007 K. Yoshimizu, Reconsidering the fragment of the Bṛhaṭṭīkā on inseparable connection (avinābhāva). In: Pramāṇakīrtiḥ, Papers Dedicated to Ernst Steinkellner on the Occasion of his 70th Birthday, ed. B. Kellner, et al. Wien 2007,
1079–1103.
Concordance of passages cited
TS(P)2
TS(P)1
TS(P)J
TS(P)Pa
TS(P)P
TS(P)D
k. 2029
k. 2030
k. 2032
695,14
k. 2034
696,11
696,14
k. 2063
k. 2030
k. 2031
k. 2033
570,4
k. 2035
570,23
570,26
k. 2064
103a3
103a4
103b2
P 211a7
103b2
P 211b2
P 211b3
105a4
37b10
37b10
37b12
P 232b16
37b12
P 233a5
P 233a7
38a12
89a6
89a7
89b2
P 162a3
89b2
P 162a3
P 162b5
90b5
74a4
74a4
74a6
P 123a1
74a7
P 123a7
P 123b1
75a7
k. 2064
k. 2065
105a4
38a13
90b5
75a7
k. 2065
k. 2066
105a5
38a13
90b6
75b1
704,22
k. 2066
577,23
k. 2067
P 213a6
105a6
P 234b12
38a14
P 166b8
90b7
P 127a3
75b1
705,11
705,12
k. 2067
577,26
577,26
k. 2068
P 213a7
P 213a7
105a6
P 234b14
P 234b14
38a14
P 167a2
P 167a3
90b7
P 127a4
P 127a5
75b2
705,15
k. 2068
705,18
705,20
k. 2069
k. 2070
k. 2071
k. 2072
578,3
k. 2069
578,8
578,9
k. 2070
k. 2071
k. 2072
k. 2073
P 213a8
105b1
P 213a8
P 213a8
105b2
105b2
105b3
105b3
P 234b16
38a15
P 234b16
P 234b17
38a15
38a16
38a17
38a17
P 167a4
90b7
P 167a5
P 167a6
91a1
91a1
91a2
91a3
P 127a6
75b2
P 127a7
P 127a7
75b3
75b3
75b4
75b4
k. 2073
k. 2074
105b4
38b1
91a3
75b5
Citation of
BASK 87
ŚV Śūnyavāda
172cd–173ab
ŚV Śūnyavāda
173cd–174ab
ŚV Śūnyavāda
174cd–175ab
ŚV Śūnyavāda
175cd–176ab
ŚV Śūnyavāda
176cd–177ab
ŚV Śūnyavāda 79
ŚV Śūnyavāda 83
ŚV Śūnyavāda 84
ŚV Śūnyavāda 85cd
(=TS2 2072ab)
220
Śāntarakṣita on Two Kinds of Arguments for Self-Awareness
TS(P)2
TS(P)1
TS(P)J
TS(P)Pa
TS(P)P
TS(P)D
k. 2074
706,23
706,25
707,7
k. 2075
707,14
k. 2076
707,22
707,24
k. 2077
708,6
708,7
708,10
k. 2075
579,9
579,11
579,14
k. 2076
579,23
k. 2077
580,2
580,7
k. 2078
580,10
580,12
580,15
105b5
P 213b3
P 213b4
P 213b5
105b5
P 213b7
105b6
P 213b7
P 213b8
105b6
P 214a1
P 214a1
P 214a2
38b1
P 235a7
P 235a8
P 235a10
38b2
P 235a13
38b3
P 235a16
P 235b1
38b3
P 235b2
P 235b3
P 235b4
91a4
P 167b7
P 167b8
P 168a2
91a5
P 168a6
91a5
P 168b2
P 168b3
91a6
P 168b5
P 168b6
P 168b8
75b5
P 127b7
P 128a1
P 128a2
75b6
P 128a5
75b6
P 128a7
P 128b1
75b7
P 128b2
P 128b3
P 128b5
Citation of
Prajñākaragupta’s Argument for Two Means of Valid
Cognition
by
Patrick McAllister
1
Introduction
The1 Buddhist logico-epistemological tradition starting with Dignāga (ca. fifth to sixth century CE) accepts only two means of valid cognition (pramāṇa): perception and inference.2
Whilst many aspects of pramāṇa theories have been studied carefully by modern scholars,
the history of the arguments that are used to prove that only inference and perception are
pramāṇas has not been investigated in great detail. The analysis usually offered is that there
are, first, two means of valid cognition since there are two objects, but that, second, really
only the particular is the object of a valid cognition.
For example, Franco and Notake (2014) characterize PV III 1–63 as follows, Franco
and Notake 2014: 4:3
[The] argumentative structure is clear: There are two pramāṇas because there
are two prameyas, and there are two prameyas because of the four criteria that
distinguish between universals and particulars.
In their comments (Franco and Notake 2014: 30, n. 2) on PV III 1, they add that,
[…] in the final analysis, for Dharmakīrti and Prajñākaragupta it cannot be
said that the fact that there are two kinds of object is the reason for there being
two kinds of means of knowledge, but that two modes of cognition of the
same thing are the reason for there being two kinds of objects of knowledge
(prameya).
Such a paraphrase of the two arguments for inference and perception is in no way the
result of misunderstanding what Dharmakīrti said. He does certainly say this. The question
is, rather, how the two arguments cohere.
1
2
3
The material in this article was first presented at the XVIIth Congress of the International Association of
Buddhist Studies, 2014, in Vienna. The research was conducted in the project (226063163) “Systems of
epistemology in classical Indian philosophy: Prajñākaragupta (ca. 750-810) on the number of instruments
of knowledge (pramāṇa)”, sponsored by the DFG, German Research Foundation. I would like to thank
Birgit Kellner for her thoughtful and critical appreciation of the arguments I make here, and Markus
Viehbeck for helping me with many Tibetan issues.
The Buddha’s special role in the context of means of valid cognition is beyond the scope of this article.
See Motoi Ono’s contribution to this volume for a discussion of that issue.
The pratyakṣa (perception) chapter is here counted as the third of the Pramāṇavārttika’s four chapters,
in accordance with how Dharmakīrti can be determined to have arranged them (see Kellner 2004a),
and referred to as PV III. Prajñākaragupta comments on the pratyakṣa chapter in the second chapter of
his Pramāṇavārttikālaṅkārabhāṣya (PVABh2 ).
Birgit Kellner et al., eds., Reverberations of Dharmakīrti’s Philosophy: Proceedings of the Fifth International Dharmakīrti
Conference Heidelberg, August 26 to 30, 2014. Vienna 2020, pp. 221–250.
222
Prajñākaragupta’s Argument for Two Means of Valid Cognition
That there is a tension between the two statements can be seen by contrasting PV III 1,
the opening verse of the pratyakṣa chapter, and PV III 53:
PV III 1: pramāṇaṃ dvividhaṃ meyadvaividhyāc chaktyaśaktitaḥ |
arthakriyāyāṃ keśādir nārtho ’narthādhimokṣataḥ ||
The means of valid cognition is twofold because that to be cognized [by it]
is twofold; for [that to be cognized by it] is [either] capable [or] incapable
of fulfilling an aim;4 [an illusory object like] hair and so on is not an object
[that is to be cognized by it], because one does not apply [oneself to it] as an
object.5
This corresponds exactly to the first part of the view of Franco and Notake (2014)
quoted above: there are two means of valid cognition because there are two objects of valid
cognition. This view is already endorsed by Dignāga.6
The verse corresponding to the second statement is this:
PV III 53d: …meyaṃ tv ekaṃ svalakṣaṇam ||
There is, however, [only] one [thing] to be validly cognized, the particular.
So according to the first argument, there are two means of valid cognition because there
are two objects of valid cognition. Here, however, Dharmakīrti maintains that there is only
one object of valid cognition. Dharmakīrti continues, giving the reason that there is only
one object:
PV III 54: tasmād arthakriyāsiddheḥ sadasattāvicāraṇāt |
tasya svapararūpābhyāṃ gater meyadvayaṃ matam ||
For one examines the existence and non-existence [of the particular], since
the fulfilment of an aim is accomplished [only] due to this [particular]. [We]
think7 that there are two [objects] of valid cognition, because this [particular]
is cognized through [its] own nature and through another nature.
4
5
6
7
The term arthakriyā is here translated as referring to the usefulness something can have in conventional,
human activity. It is also a technical term in Buddhist pramāṇa theory (cf. Nagatomi 1967–1968, Dunne
2004: 256–260). In the current instance it can be taken in both ways: a thing fulfils a person’s aim, or,
in the technical sense, it is able to cause an effect. In this article, I will render it as a technical term (by
“causal efficacy”, or similar), when I think it is referring to the property of a real thing independently of
that thing’s use by another being.
For adhimokṣa, and also abhiniveśa (a term relevant below), see Kobayashi 2010: n. 23.
Cf. Hattori 1968: 24, and nn. 1.13–14.
Franco and Notake (2014: 140) take the subject of matam to be Dignāga, translating “It is held [by
Dignāga] …”, referring to PVV 132, 8–9, where an opponent asks whether PV III 54d–55b does not
contradict Dignāga’s statement that “there is no [object] to be validly cognized apart from the particular
and the universal” (cf. PS I 1, 19 and Hattori 1968: 24). I agree, but think the impersonal matam here
carries the additional notion that, at least in a revised way, this is still maintained to be the case. The
“we” I have added here is therefore meant to include Dignāga and Dharmakīrti. Also Prajñākaragupta
relates this to Dignāga, see below.
Patrick McAllister
223
So the difference of the objects of valid cognition is due to the same object being
understood in two different ways: either it is understood through the characteristic particular
to it, as it is apprehended by direct perception, or it is indirectly understood through another,
general aspect, which is how a conceptual cognition comprehends its object. This argument
broadly separates the field into two parts: it specifies perception and “something else”; this
other is conceptual cognition, of which inference is only a subclass.8
Prajñākaragupta lets an opponent use the tension between the two arguments for an
attack in the following passage, immediately after the citation of PV III 54ab:
PVABh2 213, 4–8: yadi svalakṣaṇam eva dvābhyām api viṣayīkriyate ekaviṣayatvād ekam eva mānaṃ prasaktam. athaikaviṣayatve ’pi sāmagrībhedāt pramāṇabhedaḥ. evaṃ sati prameyadvaividhyād iti virudhyate. uktaṃ cācāryeṇa
yasmāl lakṣaṇadvayaṃ prameyam iti. sāmagrībhedena ca pramāṇabhede cakṣurādivijñānānām api bhedaḥ sāmagryā iti tāvanti pramāṇāni bhaveyuḥ.
[Opponent:] If only the particular is made an object by all two, [then], because
there is [only] one object, only a single means follows.
[Proponent:] Now, even though there is only one object, there is a difference
of the means of valid cognition because of a difference of the causal complex.
[Opponent:] If that is so, then [the statement] “because there are two [objects]
to be validly cognized” (PV III 1) is contradicted, as well as what the teacher
[Dignāga] said: “[There are two means of valid cognition] because the [object]
to be validly cognized has two characteristics, [i.e., one particular to it and
one that it has in common with other things].” (PS I 2bc)
But if there were a difference of the means of valid cognition because of
the difference of the causal complex [generating a cognition], [then] also
[different types of] cognitions such as visual [cognition] and so on would be
differentiated [from each other] due to [their] causal complex. So there would
be as many means of valid cognition [as there are different causal complexes].
In this passage, the opponent addresses the most important difficulties in maintaining
both PV III 1 and PV III 53d–54ab. The core of the criticism can be explicated like this:
1. Given that there is only one object of valid cognition, the particular, and given the
argument in PV III 1 that the number of objects is the reason for the number of
means, it results that one means of valid cognition, perception, would be sufficient:
Dharmakīrti would thus be contradicting the position that there are two means,
expressed in PV III 1.
2. If the duality of the means of valid cognition is to be maintained, but for a reason
other than the number of objects that are to be validly cognized, then Dharmakīrti
8
The difference between conceptual cognition in general and inference in particular is a separate argument
and is not discussed here.
224
Prajñākaragupta’s Argument for Two Means of Valid Cognition
is contradicting both the reason he himself gave in PV III 1, “because the object of
valid cognition is twofold”,9 as well as Dignāga’s reasoning in PS I 2bc.
3. In addition to this, if the new reason for the duality of the means of valid cognition
should be that each valid cognition’s specific set of causes is different, then one will
have to assume that there are as many means of valid cognition as there are variations
in the causal complexes.10
The argument in PV III 53d–54ab is thus problematic in various respects: it contradicts
Dharmakīrti’s previous statement, and it contradicts the tradition Dharmakīrti claims to
uphold. Furthermore, if the actual reason for the differentiation of means of valid cognition
lies in the difference of their causal complexes, then this reason does not prove what it is
supposed to prove (according to this opponent, it could prove that there is either only one
or that there are very many means of valid cognition).
In the following, I will focus on Prajñākaragupta’s strategy in answering this objection.
The most relevant passages in this regard are his commentary on PV III 1–2 and on PV
III 53d–58.11
2
Prajñākaragupta on PV III 1–2: a conventional criterion of validity
Prajñākaragupta’s interpretation of the argument in PV III 1, which is clearly influenced by
the later argument found in PV III 53d–54, is as follows:
PVABhIn 36, 25–27:12 viṣayasya caikasyaiva dvaividhyaṃ pratipattiprakārasya dvaividhyāt | pratipattibhedaś ca pramāṇabhedaḥ | sa eva ca viṣayabhedaḥ |
And the object, which is only one, is twofold, because the manner of [its]
cognition is twofold. And the difference of means of valid cognition is [this]
difference of cognition. And exactly this [difference of the means of valid
cognition] is the difference of the object.
Prajñākaragupta is here interpreting PV III 1 in the light of the later argument: the
object of valid cognition is only one, but we think it is twofold because it is apprehended
in two different ways.
9
10
11
12
There is a a slight textual variation in the text as quoted here and PV III 1: (pra)meyadvaividhyāt.
The term cakṣurādivijñāna suggests that a visual and an auditory cognition would have to be taken
as different types of means of valid cognition because at least one element in each causal complex,
the sense organ involved, is different. This is contrary to the Buddhist classification of both as a single
means of valid cognition, perception.
Kobayashi 2011 has presented a concise analysis of two elements that are important to Prajñākaragupta’s
interpretation of Dharmakīrti explored in the following sections: first, Kobayashi 2011: 1257–1259
shows that, against Dharmottara, Prajñākaragupta maintains that the object of activity which any means
of valid cognition directs a person towards is always a future particular (bhāvivastu), and that this is
the single meya that Dharmakīrti is referring to; second, Kobayashi 2011: 1259–1260 illustrates that,
according to Prajñākaragupta, both perception and inference can be said to be erroneous with regard to
this future object.
Inami et al. (2002) provide a critical edition, Japanese translation, and analysis of PVABh2 ad PV
III 1–2. I do not read Japanese, but benefited a lot from the edition.
Patrick McAllister
225
Before examining the later verses in the light of Prajñākaragupta’s commentary, it is
necessary to understand a few of the programmatic points of his commentary on PV III 1–2:
for it is here, in the opening section of the pratyakṣa chapter, that he describes the purpose
and scope that he considers the chapter to have, and therefore these statements help in
interpreting his later arguments.
The commentary on these two verses, up to PVABhIn 44, 27, is too long to be discussed
here, and I will therefore limit myself to two issues that are central to Prajñākaragupta’s
interpretation of Dharmakīrti’s argument: the role that self-awareness13 and everyday activity have in establishing that there are two means of valid cognition, and Prajñākaragupta’s
characterization of the relation between inference and perception.
2.1 Self-awareness and everyday activity
Prajñākaragupta’s general answer to why one can say there are two objects of valid cognition
is as follows:14
PVABhIn 39, 6–9: atrocyate | viṣayadvaividhyaṃ pratyakṣata eva siddham |
sadṛśāsadṛśapratītir hi pratīter eva dharmaḥ | sa ca svasaṃvedanapratyakṣasiddhaḥ | na ca pratītiḥ svarūpe bhrāntisaṅgatā | tatra bhrāntiśaṅkāyām
avyavahāra eva bhaved anavatārahetur vā vādiprativādiprāśnikavacanasya |
To this [we] respond: That there are two [kinds of] object [for cognition]
is established solely from perception. For the cognition of [something as]
similar [to other things or as] dissimilar [to everything else] is a property
only of cognition; and this [property] is established by the perception [that is
cognition’s] awareness of itself. And cognition cannot be mistaken about [its]
own form. If there is a suspicion about [the possibility of] an error with regard
to that [form of cognition itself], there would be no everyday activity at all, or
there would be no reason for the talk of [either] the proponent, opponent, or
questioner to take place.
In other words, the fact that there are two objects of cognition is evident. A cognition of
something can occur in two modes: it can be cognized as similar or dissimilar to something
else—that is, it can be cognized as something that has something in common with other
things, as a universal in the broadest sense of the term; or as being dissimilar from everything
else, as a unique thing. These two modes are qualities of cognition, not of the object. As we
will repeatedly see below,15 it is the distinct (spaṣṭa) or indistinct (aspaṣṭa) appearance of
the object that differentiates the two types of cognition. This appearance as such is directly
13
14
15
The term “self-awareness”, which renders the Sanskrit svasaṃvedana, here “… refers to the idea that
all mental states and the factors like passion or feelings that accompany them are aware of themselves.”
(Kellner 2010: 204) I will below also use the phrase “cognition’s awareness of itself” for svasaṃvedana
to make it clear that it means that here the object of cognition is cognition itself.
Franco and Notake (2014: n. 2, p. 31) point out that Manorathanandin makes a similar argument.
Prajñākaragupta’s argument here is also quoted and refuted in NBhūṣ 382, 3–9.
Cf. section 4, n. 56, and n. 48.
226
Prajñākaragupta’s Argument for Two Means of Valid Cognition
perceived by the awareness that every cognition has of itself, and doubt about this aspect
of cognition would end all conventional activity, as well as make any debate impossible.
In explaining this, Prajñākaragupta emphasizes that the duality of means of valid
cognition is only conventional. This becomes especially clear in the following passage,
where he addresses the overall aim of the pratyakṣa chapter:
PVABhIn II 20–21, and 39, 20–24: pratītibheda evāstu mānabhedaḥ kathaṃ bhavet |
nanu prasiddhaṃ mānatvaṃ pūrvaṃ sāmānyalakṣaṇāt ||
tadbhedavyavahāro ’yam idānīṃ sādhyatāṃ gataḥ |
savikalpakam adhyakṣam eṣo ’gnir iti yo vadet ||
[...] vyavahārataḥ pravṛttinivṛttilakṣaṇāt prāmāṇyaṃ sāmānyalakṣaṇenaiva
prathamapariccheda eva prasiddham | bhedavyavahāramātrakam evedānīṃ
sādhyam āpannam | tatra savikalpakam ekam evedaṃ pratyakṣaṃ yad utāgnir
ayam asmād abhipretārthakriyākārīti yo vadet taṃ prati dvitayam etad iti |
pratītyākārabhedāt |
[Opponent:] There certainly may be a difference of cognitions.
[But] why would there be a difference of means of valid
cognition?
Well, the state of being a means of valid cognition was well
established earlier, from [its] general characteristic.
[Proponent:] Now, this everyday activity [of ours that treats] the
means of valid cognition as different, has come to be what
is to be established,
[as] someone might proclaim a conceptual perception, [like] “This
is a fire.”
… On the basis of everyday activity, [that is], from the characteristic of
engagement or nonengagement [with an object], what it is to be a means of valid
cognition has been well established in the first chapter [the \textitpramāṇasiddhi
chapter of the PV, but] only according to [its] general characteristic.
Now, [in this chapter, what it is to be a means of valid cognition] has become
what is to be established only [inasfar as] it is commonly treated as differentiated [into two types]; [the explanation] “this [means of valid cognition] is
twofold” [is given by Dharmakīrti] for that [person] who would say about this
[means of valid cognition] that there is only a single conceptual perception,
for example, “This is a fire, from it the attainment of a desired aim is brought
about.”; [this is said to him] because the form of cognition is differentiated.
Patrick McAllister
227
In these statements Prajñākargupta is contrasting the scope of the first (pramāṇasiddhi)
chapter and the current chapter: in the first chapter, validity was established in its general
form—as a definition applicable to all kinds of valid cognitions. Moreover, validity was
established there conventionally—that is, in relation to the engagement in some activity (or
abstention from an activity) that a means of valid cognition facilitates in a way that makes
that activity successful.16 On this background, Prajñākaragupta continues, an opponent
might object: conceptual perception is compatible with this criterion of validity, and,
in fact, it is the only (ekam eva) means of valid cognition that has to be assumed. It is
compatible with this definition of validity because, if conceptual perception is possible, it
could contain the ascertainment “This is a fire, it can fulfil my aim.” This is not acceptable
to Dharmakīrtians. Therefore, a further argument is necessary to show that there is not only
one combined means of valid cognition, a perception containing a conceptual cognition,
but rather that there are different means of valid cognition.
In other words, the pramāṇasiddhi chapter does not answer the question of which means
of valid cognition there are, but only of what a means of valid cognition is. The pratyakṣa
chapter will however deal exactly with this question: what are the different means of valid
cognition? In addition, Prajñākaragupta qualifies this with the statement that that which is
to be established now is bhedavyavahāramātraka: it consists exclusively in the conventional
talk of a division of the means of valid cognition as defined in the pramāṇasiddhi chapter.
The importance of noting this lies in the fact that Prajñākaragupta separates two elements: the capacity for leading to successful activity, the general criterion of any cognition’s
validity, and what cognitions are like (in particular, whether they are conceptual or perceptual), which is the basis for distinguishing the types of valid cognitions and is evident
in any cognition’s awareness of itself. In a conventional sense, a cognition can be considered ‘valid’ if it allows one to act successfully; it is not important for its validity what
type of cognition it is. And vice versa, the general mark of validity is not important for
distinguishing the types of these cognitions.
2.2 The relation of inference and perception
Prajñākaragupta then discusses which means of valid cognition can lead to successful
activity, or, in other words, conforms to the conventional and general criterion of validity.
First, an opponent, apparently a Buddhist,17 raises this objection:
16
17
For Prajñākaragupta’s general analysis of cognition’s validity and its relation to activity, see Ono
2000 and Franco 2004. The explicit equation of “being a means of valid cognition” with enabling or
motivating successful activity derives from PVM II 1–5, and was already elucidated by Devendrabuddhi
and Śākyabuddhi (see Dunne 2004: 253–256). We find it fully developed as one of a number of equal
definitions of validity in Dharmottara’s work, cf. Krasser 1995: 247–248, Franco 1997: 52.
Cf. Yamāri, (PVAṬSt : 103, 37): ’on te zhes bya ba la sogs pas ni rang gi phyogs pa’i rtsod pa slong ba’o/
Yamāri goes on to describe a position held by these others (explicitly excepting Bhaṭṭārcaṭa), on which
a conceptual cognition that has an object that accomplishes some end is different from perception, and
therefore not a means of valid cognition. Inami et al. (2002: 26, n. 35) give more details.
228
Prajñākaragupta’s Argument for Two Means of Valid Cognition
PVABhIn 39, 30–32: atha pratyakṣam eva pravartakaṃ nāparam | tat tu mano
’ntaraṃ18 bhavad api na pramāṇam | na hi saha tena yāvad bhavati tāvat
pramāṇam | śarīrabhūtalādīnām api prāmāṇyaprasaṅgāt |
Now, perception alone is what causes activity, nothing else. That other cognition, though it exists, is not a means of valid cognition. For, [something] is not
a means of valid cognition to the extent that [it] exists together with [a means
of valid cognition], because it would result that even the body, the earth, and
so on [which might all co-exist with a means of valid cognition] would be
means of valid cognition.
Prajñākaragupta answers that perception is not, in and of itself, capable of letting a
person act:
PVABhIn II 23: anvayavyatirekābhyām upayogītarasthitiḥ |
na ca kevalam adhyakṣaṃ tadabhāve pravartakam ||
[Something] is determined as assisting [something else] or not through the
positive and negative concomitance [of that which assists and that
which it assists].
But perception by itself is not, when that [which assists it, i.e., conceptual
cognition], is absent, what causes activity.
Without the introductory passage explaining that the topic here is the conventional
difference of perception and inference, this would of course be in stark contrast to Dharmakīrti’s well-known position that, in activity following upon perception, only the perception
is a means of valid cognition, but not the conceptual cognition that perception needs to be
followed by in order to cause a person to act.19
But, given that the discussion has here been restricted to the question about the difference
of means of valid cognition, the verse says that it is not possible that perception alone—only
a cognition with an object that is distinct (spaṣṭa)—can cause activity. The statement thus
emphasizes that the general and conventional characteristic of a means of valid cognition,
that it prompts activity, cannot be upheld if one wishes to also bind it to only one type of
cognition—the perceptual one. That would immediately land one in an untenable position.
In other words, to take perception alone as a means of valid cognition in this conventional
sense is impossible: it will only work if one grants that status also to a conceptual cognition
(or at least some conceptual cognitions).
This is then elaborated in the following, where Prajñākaragupta examines the relationship between inference and an activity following upon perception. Even in the case of
complete habituation—a state in which a being is able to act upon a perception without intervening conceptual cognition—conceptual cognition is involved: for, so Prajñākaragupta,
18
19
PVABh-msB 84a5 reads mano’ntaram, which I think is possible: it could be understood as ‘another
cognition’, which corresponds to the Tibetan translation. Inami et al. (2002) correct to mānāntaram
based on the Tibetan. Prof. Inami informs me that the tat tu is missing in his edition only due to an
unfortunate misprint, and it is represented in his Japanese translation.
See Katsura 1993 for a study of this “perceptual judgement” in Dharmakīrti’s writings.
Patrick McAllister
229
the fact that there is a habituation presupposes an inferential cognition on whose basis the
habit must initially have been formed.20 To hold otherwise, so PVABhIn II 24, would be as
clever as thinking, because one contracts a disease the first time one goes somewhere, it is
unsafe to go there the first time, but safe thereafter. In this sense, inference can be said to be
the main element (pradhāna) even in purely habitual activity, which the opponent would
like to see as the main support for the claim that perception can promote activity without
conceptual aid.21 We can now investigate Prajñākaragupta’s comments on the later verses.
3
Prajñākaragupta on PV III 53d–54: one object known in two ways
In this section of the PVABh2 , Prajñākaragupta is presenting a discussion in which, as we
will see, the following argumentative aims are intertwined:22
1. He needs to resolve and explain the possible contradiction between PV III 1 and PV
III 53d (described above).
2. He has to prepare for the proof that perception and inference are the only means of
valid cognition.
3. He has to prove that perception and inference are different means of valid cognition.
(Points 2 and 3 together make it possible to maintain that there are two, and only
two, means of valid cognition.)
4. He has to uphold cognition’s awareness of itself as the central cognitive faculty, in
the sense that in reality perception and inference are two forms of this, and can only
be conventionally separated.
Prajñākaragupta starts his commentary on PV III 53d (see above, section 1) as follows:
20
21
22
See section 5 for how this position is taken up again in showing how inference and perception depend
on each other.
See the following passage: PVABhIn II 28: uktam atra vinābhyāsān na pratyakṣe pramāṇatā | tato
’numānam evātra pradhānam iti gamyatām || (It was explained that without habituation there is no
means of valid cognition in a perception. Therefore it must be understood that inference alone is the
main element here.)
McCrea (2011) has analyzed this section. Whilst he makes many valid points, he does not take the
overall argumentative context that is carried over into this section from the beginning of the pratyakṣa
chapter into due account. This results in attributing the following positions to Prajñākaragupta, which
directly contradict Dharmakīrti:
1. that perception is not free from error (McCrea 2011: 327)
2. that it is not perception, but rather the following moment of conceptual awareness that is a
means of valid cognition (McCrea 2011: 323)
I do not agree with these attributions. Point one, though indeed Prajñākaragupta does make it, cannot
be taken out of its context (see section 6 and n. 54). Point two is a position that the opponent is being
forced into and cannot accept (see section 4).
Another issue that McCrea (2011: 323) discusses is that the conceptual awareness following perception
does not have the fault of gṛhītagrahaṇa, or that this fault is not a reason to not be a means of valid
cognition. The discussion of this point is beyond the scope of this article (see the comment in n. 40).
230
Prajñākaragupta’s Argument for Two Means of Valid Cognition
PVABh2 212, 29–30: na hi sāmānyaṃ nāma prameyam, yathākalpanam ayogāt | svalakṣaṇam eva paramārthataḥ prameyam | tasyaiva sadasattvenāvabodhasya prayojanatvāt | kuta etat |
tasmād arthakriyāsiddheḥ sadasattāvicāraṇāt |
arthakriyākāriṇo hi padārthasya sattvāsattvābhyām arthitā prekṣāvatām |
tadavabodhāya ca pramāṇam anviṣyate | anyathā pramāṇaparīkṣaṇam aprekṣāpūrvakriyaiva bhavet | tasmād arthakriyākāripadārthabhāvābhāvaviṣayīkaraṇasamartham arthavat pramāṇam | tasmāt dvābhyām api pratyakṣānumānābhyāṃ svalakṣaṇam eva viṣayīkartavyam | anyathā pramāṇatvāyogāt |
For the so-called universal is not an object of valid cognition, since it is not
possible as it is imagined. In reality, only the particular is [the object] to be
validly cognized, because the knowledge of the existence or non-existence of
it alone is of use. Why is that?
Because one examines the existence and non-existence [of the
particular], since the fulfilment of an aim is accomplished
[only] due to it. (PV III 54ab)
For, judicious beings are intent upon [whether] the object that produces the
fulfilment of [their desired] aims exists [or] not. And for [the purpose of] knowing that, [i.e., whether that objects exists or not], a means of valid cognition
has to be sought for. Otherwise, the investigation by a means of valid cognition
would be an entirely injudicious activity. Therefore, a means of valid cognition
has an object [insofar as] it is capable of making the existence or non-existence
of a thing that produces the fulfilment of an aim [its] object.
Therefore, by all two [means of valid cognition], perception and inference,
only the particular is to be made the object, because otherwise it would be
incoherent that [they should] be means of valid cognition.
So perception and inference are each a means of valid cognition only inasfar as they
direct a person to the successful accomplishment of a purpose. In order to do this, they
have to reliably let a person attain a particular, since only that is capable of causing the
desired effect. From this perspective, the particular alone is the object that is to be validly
cognized, the prameya. The opponent immediately raises the objection that, if there is only
one object of valid cognition, then one means would be enough, too.
Prajñākaragupta just remarks that this is wrong, since it is due to the differences in the
respective causal complexes that two types of valid cognition can be assumed—implying
that it is not because there are two objects.23
23
This idea of differences in the causal complex already appears in the analysis of PV III 2, most explicitly
in PVABhIn II 49, and the accompanying prose, especially this passage: PVABhIn 44, 16–18: na hi
sarvadā pramāṇadvitayaṃ prameyadvitayāt sādhyate | api tu sāmagrīsambhavād iti vayam brūmaḥ |
Patrick McAllister
231
It is after this statement that the opponent delivers the main objection, presented above in
section 1, and it should now be clearer what the exact direction of that objection is. McCrea
(2011: 322 f.) has characterized it well, along with the main defence that Prajñākaragupta
uses:
… the opponent attempts to find some way to distinguish perception from
inference as a more direct or immediate mode of awareness, so that he can
still deny the validity of inference without similarly condemning perception.
In each case, Prajñākaragupta demonstrates that the opponent’s purported
distinction is spurious, and that perception and inference are similarly indirect,
such that one could not accept one as a pramāṇa without accepting the other
as well.
In the current context, the problem concerns the causal complex of the means of valid
cognition. The opponent distinguishes the object (prameya), which is the main factor in
the causal complex, and the other, secondary factors in his criticism:
PVABh2 II 229: paramārthaprameyatve syād anantaprameyatā |
apekṣākṛtabhedatve paramārtho na lakṣaṇam ||
If the object of valid cognition were real, there would be infinite objects of
valid cognition. If [the object] is differentiated in dependence [on some other
factor], reality is not a characteristic [of the object so differentiated].
This verse sums up the two sides of the opponent’s attack: if the particulars are the
objects of valid cognition, then there would be infinitely many different objects. In other
words, if the main factor in the causal complex generating a cognition were the real thing,
there would be very many such cognitions, each a proper and separate means of valid
cognition. So the Buddhist position expressed in PV III 53d, that there is only one object
of valid cognition and that it is real, would be incoherent since this object’s reality implies
its multiplicity. Alternatively, the opponent continues, the proponent might claim that the
objects of valid cognition are differentiated in dependence on some other factor in the
causal complex (as claimed in PV III 54cd); this, however, would mean that the difference
maintained for the objects of valid cognition would not belong to the objects themselves,
but would be external to them. Again, this consequence would violate what is endorsed in
PV III 53d, namely that the object of valid cognition is real. In other words, one cannot
sāmagrīsambhavaṃ ca paścāt pratipādayiṣyāmaḥ | (“For that there are two means of valid cognition is
not always established from there being two objects of valid cognition. Rather, we say that [two means
are established] because [they each] arise from [a specific] causal complex. And we will later explain
[this] arising from a causal complex.”)
The context there is an opponent’s argument that, if one takes the two objects of valid cognition and the
two means of valid cognition as, respectively, cause and effect, the inference proving the two means of
valid cognition would be an inference from cause to effect, and therefore not certain. Prajñākaragupta’s
answer is that in this case the inference is possible, because the effect is inferred not from an individual
cause, but from the causal complex, one factor of which is the object of valid cognition.
232
Prajñākaragupta’s Argument for Two Means of Valid Cognition
maintain both that the object of valid cognition can be defined as real (paramārtho na
lakṣaṇam) and that it is only one (anantaprameyatā), both of which are asserted, however,
in PV III 53d: meyaṃ tv ekaṃ svalakṣaṇam.
The immediately following excursion leads up to PV III 55. For our purposes, a look at
its conclusion is sufficient:
PVABh2 215, 4–9: kathaṃ tarhi meyāntaram | tasyaiva pararūpeṇa pratīteḥ |
tathāpratīyamānaṃ dṛṣṭāntasādhāraṇena rūpeṇa meyāntaram | pratipattibhedena tadrūpāropān na paramārthataḥ |24 evaṃ tarhy apekṣākṛtatvān na
paramārthatā | satyam avastu sāmānyam iti pratipāditam eva | nedam apūrvam ucyate | vastusaṃvādadvāreṇa vyavahāribhir alakṣitanānātvair vastv iti
vyavahriyate | tena tadapekṣayedam ucyate prameyadvaividhyam |25
How then is there another object of valid cognition? Because of a cognition of
exactly that [same object of valid cognition] with another nature. [Inasfar as
the particular is] being cognized in this way with a form that it has in common
with the example [in an inference], there is another object of valid cognition;
[in other words, there is another object] because that form [which the particular
has in common with the example] is superimposed [on it] due to a difference
in cognitions; [but there is] not [another object of valid cognition] in reality.
[Opponent:] In that way, then, because it is made in dependence [on something
else, this other object of valid cognition] is not real.
[Answer:] That is true. Indeed it was taught that the universal is not a real
thing.26 [But] this is not said without precedent. [For], in virtue of [a universal] cohering well with a real thing, [people] engaged in everyday activity
commonly act [with regard to a universal] by [considering it] a real thing,
[inasfar as] they do not take note of the fact that [the particulars they are acting
towards] are different. Therefore, it is in dependence on this [real thing that
appears in two forms in everyday activity] that the duality of the object of
valid cognition is spoken of.
The heart of the passage rephrases a part of PV III 54, namely the cognition of a
particular in an indirect way (literally, “with another nature”; tasya svapararūpābhyaṃ gater
meyadvayaṃ matam). Prajñākaragupta is again making the notion of the duality of the
object of valid cognition depend on an everyday understanding of the matter: normal people,
when inferring something, think about the object of that cognition as “the real thing,” even
though what they are actually cognizing—the proper object of the valid cognition—is a
future real thing; and it is this which is not directly cognized, but only by means of a feature
that it shares with the example. If a fire on the hill is being inferred from smoke, it, the
24
25
26
Read paramārthataḥ with PVABh-msB 106b2, PVABh-msE 210b5 against paramārthaḥ PVABh2 .
The last line, vyavahāri° … prameyadvaividhyam, is repeated in PVABh2 . This must be a printing error,
it is not found in either PVABh-msB 106b2 or PVABh-msE 210b6.
Franco (2012) and Franco and Notake (2014: 4–17) discuss how unreal things can be the objects of
valid cognition.
Patrick McAllister
233
actual fire that could cook our food if we reach the place on the hill soon enough, is being
cognized, not as it is in itself, but as similar to a kitchen fire previously seen. This similarity
cannot be a phenomenal one, of course: the fire on the hill does not appear.
With this, the scene is set for the discussion of how inference can be a means of valid
cognition. The main points to bear in mind in reading this discussion, in addition to those
previously stressed in Prajñākaragupta’s commentary on PV III 1, are as follows:
1. That there are two means of valid cognition, as well as two objects of valid cognition,
is only a concession to everyday activity. This is, essentially, the justification for the
tension between PV III 1 and PV III 53d–54.
2. Only the real thing is the object of valid cognition.
3. There are two ways in which the real thing is known: through its own nature, and
through another nature.
4. Perception is dependent on inference because inference initiates what becomes
habitual. Without habituation, and thus without inference, perception does not cause
a person to act.
The main controversy discussed in the following passages of PVABh2 is about how to
understand item 2. The question is whether the statement in PV III 53d is true conventionally
or in reality. Depending on the answer, the consequences vary. When this statement is taken
to express a convention, the opponent’s claims are very strong: the tension between, on the
one hand, inferring “there are two means of valid cognition because there are two objects
of valid cognition, a real thing and an unreal one”, and, on the other hand, maintaining that
“there is only one object of valid cognition, the real thing” can hardly be contained if the
term “real thing” is here taken in the same sense. If the latter statement is, however, taken
as expressing a fact of reality (and the former one not), then the opponent’s attacks lose
most of their power; but this forces us to reconsider in what sense the particular is the only
object of valid cognition.27
4
Prajñākaragupta on PV III 55–58: saving inference
This section asks how inference can be considered a means of valid cognition, even though
it is mistaken because it has an imagined universal, not a real thing, as its object. This
problem is raised in PV III 55cd, and Dharmakīrti’s answer is given in PV III 56–58.28
Prajñākaragupta opens this exchange with the following objection:
PVABh2 215, 18–19: kathaṃ tarhi paramārthasya viṣayīkaraṇād bhedaḥ |
svarūpasākṣātkaraṇe hi pratyakṣataiva bhavet | tadasaṃsparśe kathaṃ tadviṣayatā |
27
28
We will see in section 4 that no present particular is the object of valid cognition: it is the future
particular, both for perception and inference. In other words, to take PV III 53d as a statement that
accords with reality means that the object of valid cognition mentioned there is neither of the two
objects in PV III 1–3, the present particular or the universal.
PV III 56 is: ayathābhiniveśena dvitīyā bhrāntir iṣyate | gatiś cet pararūpeṇa na ca bhrānteḥ pramāṇatā || abhiprāyāvisaṃvādād api bhrānteḥ pramāṇatā | gatir apy anyathā dṛṣṭā pakṣaś cāyaṃ
kṛtottaraḥ || See Franco and Notake 2014: 141–142 for a translation.
234
Prajñākaragupta’s Argument for Two Means of Valid Cognition
How, then, is there a difference [between perception and inference], since [they
both] make a real thing [their] object? For if [a cognition] directly presents the
nature [of something], then [it] is just perception. [But] if [a cognition] does
not touch that [nature], how is that [nature] the object [of that cognition]?
The opponent is here engaged in a variation on McCrea’s schema quoted above (in
section 3).29 Either one takes it seriously that there is only one object of valid cognition,
and concludes that there is no difference between cognitions of that object, or one takes
the position that one cognition has that object and the other does not. This latter position
implies that the cognition is not correct (that it is an error or bhrānti); this means, in the
eyes of the opponent, that it cannot have the status of a valid cognition.
Here, and in the following discussion, the opponent’s general position will be that the
difference between validity and invalidity hinges on the contact of cognition with reality: if
the cognition is in direct contact with a real thing, it is valid; if not, then it is invalid. In
this way, we can read the debate as not only being about what differentiates inference from
perception, but also as being about the implicit criterion that constitutes the validity of a
cognition. As we will see, Prajñākaragupta maintains quite simply that the contact with the
real thing is not what makes a cognition valid: neither perception nor inference are valid
for this reason.
Prajñākaragupta first explains (PVABh2 215, 21–23) what it means that a cognition is
erroneous in the sense that the opponent is using in his argument: it means that a cognition
does not have the actual particular as its object, and can thus not be said to conceive of
its object as it is. The opponent immediately counters that then this cognition cannot be
considered to have the status of a means of valid cognition, presenting the gist of PV
III 55cd, as follows:
PVABh2 215, 26–27: yadi pararūpeṇa gatiḥ kathaṃ tasya gatiḥ | pararūpasyaivāsau gatiḥ | tatrānyasya prāptau bhrāntir eva | bhrānteś ca na pramāṇatā |30
If there is a cognition [of something] with another nature, then how is [this]
a cognition of that [thing]? This is a cognition only of another nature. When
29
30
It would be interesting to know who the opponent is, exactly. It seems that it is someone who—like
Prajñākaragupta, and Dharmottara before him—accepts that ‘prompting activity’ (pravartaka) is a
central criterion for being a means of valid cognition, but links this capacity to the fact that the cognition
has for its object something real as it is. If one, reasonably, assumes that only the Cārvākas would argue
against inference as a means of valid cognition, then one could infer that they recognized pravartakatva
as an important criterion for being a means of valid cognition.
There are two textual difficulties here. First, for pararūpasyaiva: PVABh-msB reads pararūpasyavāsau,
but this is not completely clear, and PVABh-msE reads pararūpasyāsau. Both seem possible, and the
vā could easily have gotten lost. Since PVABh-msB is really not clearly legible here, I think one should
follow Sāṅkṛtyāyana’s reading.
The second problem is with prāptau bhrāntir eva: PVABh-msB originally read prāntireva and was then
corrected, but the marginal addition is now illegible. PVABh-msE supports prāptau bhrāntir eva. The
printed bhrāntau bhrāntir eva (PVABh2 215, 26–27) must be due to either Sāṅkṛtyāyana misreading the
correction here or a misprint. Both prāptir and prāptau seem possible. I prefer the latter, in accordance
with PVABh-msE, and probably PVABh-msB post correctionem.
Patrick McAllister
235
one attains another [thing, there is] simply an error with regard to that [thing
which was cognized]. And an error is not a valid means of cognition.
The argument of the opponent can be analyzed as follows:
1. There is a cognition of something (X) in a form other than its own (Y), e.g., the
cognition of a particular fire in the form of the general concept ‘fire’.
2. So this “cognition of X” is, in fact, a cognition only of something else, Y.
3. If one acts upon this cognition of Y, one attains X.31
4. But the cognition of Y is an error with regard to X (because X is not cognized as it
really is).
5. An error cannot be a means of valid cognition.
6. So the cognition of X in the form of Y cannot be a means of valid cognition.
Dharmakīrti answers this in PV III 56. Prajñākaragupta’s interpretation of this verse
is indeed, as McCrea (2011: 321) has noted, striking in “… the overall similarity, indeed
the virtual identity, of perception and inference” that the verse is said to imply. The first
passage runs as follows:
PVABh2 215, 30–216.2: anyatrāpi yathārthābhiniveśaḥ katham avagantavyaḥ | abhiprāyāvisaṃvādād eva | sa cātrāstīti nāyathārthābhiniveśaḥ | idaṃ
tu vāsanābalāj jñānaṃ pratibhāsabhedato bheda iti naivaṃ vyavahāriṇo vidanti | vyākhyātṛpratītir eveyam |
Even in the other [case, that of perception], how is [this] “determination
[of the object] as the object is” to be understood? Only on [the basis that
determination] does not belie [a person’s] intentions. And this is the case here
[in the case of inference as well]. So there [is, in fact], no determination that
is not according to [the] object [in the case of inference either].
But [people] engaged in everyday activity do not know that this cognition [has,
in fact, originated] in virtue of the impressions [left by previous experiences,
and] that there is a difference [of perceptual and conceptual cognition only]
due to the difference in the appearance [of an object to cognition]. This is only
the insight of those [people] who explain [things].
Prajñākaragupta is here basing his argument on the expression ayathābhiniveśena of
PV III 55a, by which Dharmakīrti means to say that inference does not determine its
object correctly (lit., “as [it] is”). It is this element that Dharmakīrti lets the opponent
criticize in PV III 55cd: if inference is not correct in that it does not determine its object
correctly—Prajñākaragupta’s opponent here taking “correctly” as “how the object really
is”—then how can it be a means of valid cognition? Prajñākaragupta turns the question
around, and simultaneously gives the more precise interpretation of “determines [its object]
as [that] object [really] is” (yathārthābhiniveśena) to Dharmakīrti’s formulation: what, he
31
It could also be that one attains some altogether different element, Z. But the important thing is that
Prajñākaragupta has to concede that one does not attain Y.
236
Prajñākaragupta’s Argument for Two Means of Valid Cognition
asks the opponent, is the meaning of the qualifier “determine something as it really is” in
the case of perception?
Prajñākaragupta adds an important qualification: normal people do not understand
that cognition—which I take here to include both perception and inference—is not due
to different causes (it is always generated by mental imprints, not external objects), and
that cognitions are different on account of how something appears in them, not, one must
understand, essentially. ‘Philosophers’, on the other hand, do know that this is the case.
It is therefore justified to say that Prajñākaragupta, as in the interpretation of PV III 1–
2 and 53d (section 2 and section 3), again distinguishes two levels on which one can
differentiate perception and inference, in this passage and in the following ones: one respect
in which they follow the everyday usage of normal people, who say that they directly
perceive some things, and indirectly cognize others (correctly, when by inference); and
one respect, corresponding to reality, in which the cognitions are distinguished only on
account of how an object appears in them.
Now, there might be a problem: if Prajñākaragupta says that ayathārthābhiniveśa is not
the case in inference, then PV III 55ab, ayathābhiniveśena dvitīyā bhrāntir iṣyate, would
seem to be contradicted; there, Dharmakīrti obviously accepts that the second kind of
cognition, inference, is an error because it does not conceive of its object as it really is.
This alerts us to a further point in Prajñākaragupta’s interpretation of PV III 55ab: it
stresses the provisional character of that statement. In order to avoid the contradiction to
Dharmakīrti, the sentence governed by the verb iṣyate (“it is assumed”) must not be understood as expressing something endorsed by Dharmakīrti, but as a general statement of fact
about what people normally take to be the case. Relying on this aspect of Prajñākaragupta’s
interpretation, one might then want to translate this sentence as “The second [kind of
cognition, inference], is [commonly] assumed to be an error since there is no determination
[of an object] as it is.”32 And the implied agent of the sentence then would be “by people
engaged in everyday activity.” In other words, that the criterion ayathārthābhiniveśa, which
makes a cognition an error, is applied to inference is something usually done by ‘normal
people’. This is an important interpretative move, because now Prajñākaragupta has created enough room to say that it is, in fact, not the case that this assumption is correct. It
leaves open three directions in which Dharmakīrti might have continued to argue: First, he
could have tried to argue that inference is in fact not an error according to the criterion of
ayathābhiniveśa (in its common sense), a path that was not chosen. Second, that inference
is, unlike perception, an error, but that this does no harm to its status as a means of valid
cognition; this is perhaps the simplest interpretation, upon which the statement “inference
is assumed to be an error” is taken as implying that perception is not. Third, that not only
is inference an error according to the criterion of ayathābhiniveśa, but also perception; on
this reading, the statement “inference is assumed to be an error” entails an insufficiency:
inference is assumed to be so by people, but they forget that perception is also like that. We
will see that, with some caveats, this is the road that Prajñākaragupta presents Dharmakīrti
as having chosen.
32
To understand iṣyate like this does not preclude that Dharmakīrti himself did indeed take inference to
be an erroneous cognition, cf., e.g., PVin 2 1cd: …bhrāntir api sambandhataḥ pramā || The point is
rather that to take inference like this has no bearing on the question of whether it is a pramāṇa or not;
likewise, for perception, its not being erroneous is irrelevant for its status as a means of valid cognition.
Patrick McAllister
237
He then continues to explain the general equality of inference and perception in terms
of their being means of valid cognition. Having said that yathārthābhiniveśa, inasfar as
it results only from abhiprāyāvisaṃvāda, is not different for them, he analyzes the next
acknowledged criterion for being a means of valid cognition, the fact that it leads to
(successful) activity:33
PVABh2 216, 2–7: sa evābhiprāyaḥ katham anyadarśānād iti cet | na vikalpānāṃ vastupratiniyamābhavāt | anādivāsanāsāmarthyam evaitat | tataḥ
katham aparicchinnatattvas tatra pravartata iti na codyam etat | dṛṣṭe ca nānupapattisambhavaḥ | pratyakṣe ’pi kathaṃ pravartate | tatrāpi naiva prāptavyarūpaparicchedaḥ sannihitamātrasya paricchedāt | pratyuta34 pratyakṣam
evāpravartakaṃ sannihitamātrasya pariprāpteḥ | tatrāpi tadekatvādhyavasāyād eva vṛttir bhāvini vastuni | tato ’numāne ’py evam eva vṛttiḥ | katham
asamānatayekṣyate |
[If one asks:] How can there be exactly this intention from seeing something
[completely] different?
[Then we answer:] It is not [because of seeing something that there is this
intention], because conceptual cognitions are not restricted to real things. Such
is simply the capacity of beginningless impressions. It must not be criticized
how, because of this [capacity, someone] who has not discerned reality can
act towards it. But if [this activity] is observed, [its] not being the case is
impossible.
Also in [the case of] perception [one could ask the same question:] how does
one act? Not only is there, in this [perception], no discerning of the nature that
is to be attained, because only that which is [immediately] present is discerned,
but perception as such [also] does not make [a person] act, because only that
which is present is completely attained [by perception].
In that case [of perception] too [a person] acts towards a future real thing only
because there is a determination [of it] as identical [with what is perceived].
Therefore activity is exactly the same also in the case of inference. [So] why
[are they] regarded as not being the same [by you]?
The explanations here underpin the main point of the previous discussion: the opponent
holds that inference cannot enable a person to act because it ‘does not discern reality’, i.e.,
it does not put the person in direct contact with the real thing that successful activity must
be directed at—contrary, so the opponent still assumes, to perception. Prajñākaragupta’s
answer to this is double sided: first, activity arising from inference can be observed, and
can therefore not be impossible. Prajñākaragupta then turns the question around: how
can one act in the case of perception? The opponent’s criticism of inference would be
33
34
A part of this passage is also translated and discussed in McCrea 2011: 321–322.
I follow McCrea (2011: 322, n. 8) in reading the marginal addition to pratyuta, kiṃ tv arthe, as a gloss
rather than as a correction or addition.
238
Prajñākaragupta’s Argument for Two Means of Valid Cognition
applicable there too: perception knows nothing of the object that is capable of fulfilling
one’s aims (since what one perceives is a present, momentary object). And as such it can
enable activity towards a future real thing only through a determination of the thing it has
directly perceived and the future thing as identical; one must think that the thing’s existence
is temporally extended. And in this respect, it would be just like inference.35
Again, the situation described by McCrea is found: Prajñākaragupta is applying the
opponent’s statement about inference to perception, and the opponent cannot accept the
consequences. If the opponent insists that the attainment of something that was not cognized
is an error (item 4), then the exact same thing has to be said for perception: it too grasps
only the present thing, yet acting upon it a person will attain a future thing; and it does, by
itself, not actually make a person act.
A concise summary of this approach is found in PVABh2 II 234 and the following prose
explanation:
PVABh2 II 234: pravartako vikalpaś ced avastugrahaṇe katham |
tathāpi vartayaty etad anumāne na kiṃ matam ||
PVABh2 216, 21–22: yadi hi vikalpajananadvāreṇa pratyakṣaṃ pravartakam |
āyātaṃ tarhi vikalpasya pravartakatvāt pramāṇatvam | tathā saty anumānasya
prāmāṇyam avyāhatam eva |
If [you say that in the case of perception] a [following] conceptual
cognition prompts [a person] to act, how [does it do that]
without grasping a real thing?
If [you say that it] prompts [a person] to act nevertheless, then
why is this not assumed in [the case of] inference?
For, if perception prompts activity by means of generating a conceptual cognition [that makes a person act], then one has arrived at [the position] that
conceptual cognition, because it prompts activity, [would be] a means of valid
cognition. [And] if it is so, it has not been rejected at all that inference is a
means of valid cognition.
This passage is exemplary for Prajñākaragupta’s defence of the difference between
inference and perception: given that perception itself does not prompt activity, it must be
the following conceptual event. But that conceptual cognition, like inference, does not
grasp anything real (being conceptual, it has a universal, not a particular, as its object).
The alternatives are dire: if the opponent were to insist on his position, not only would the
state of being a means of valid cognition be a quality of the conceptual cognition following
perception, but also the fact that inference is a means of valid cognition—the point that
the current debate is actually about—would not have been fended off. This would be a bad
defeat for the opponent.36
35
36
As McCrea (2011: 322) notes, already Dharmottara considered “this implication of the theory of
momentariness for […] perception.” See Krasser 1991, 1995; McCrea 2011; McCrea and Patil 2006.
My interpretation here diverges from the interpretation of the prose part of this passage by McCrea
(2011: 323), who takes it as expressing Prajñākaragupta’s own position. I think one has to read the
Patrick McAllister
239
The implied consequences are unacceptable to the opponent, who wishes to link what
it is to be a means of valid cognition to this criterion of knowing a real thing as it is: he has
to admit either that neither perception nor inference has a claim to being a means of valid
cognition, or that both have.
The following discussion continues in much the same way. We can skip to the end of
the discussion, and consider its summary:37
PVABh2 218, 2–8: tasmād gatir api pratyakṣābhimatā ’nyathā dṛṣṭā pararūpeṇaiva |38 na kācit pravṛttiviṣaye svarūpeṇa gatiḥ | anyatra tu vasturūpe
svarūpe vā gatir ubhayor apīti bhāvivastuni ko viśeṣaḥ |
Therefore, also [that] cognition, [namely, the one] that is considered to be
perception, is observed [to be] otherwise, [that is], only [to cognize an object]
with another nature. As regards the object of activity, [that] is not cognized at
all with its own nature; but concerning another, [be it] the nature of the real
thing or the nature of cognition, all two, [perception and inference], cognize
[it]. So what difference is there with regard to the future object?39
This shows in what sense Prajñākaragupta takes perception and inference to be parallel:
neither apprehends the future object directly, obviously impossible; but both do apprehend
something else directly: what this is, Prajñākaragupta can still leave open at this point in
the discussion, content to call it either the form of an (external) real thing (vasturūpa) or
the form of cognition itself (svarūpa).40
passage as a prasaṅga, that is, a consequence unacceptable to the opponent (and to Prajñākaragupta,
too), but not as a position that Prajñākaragupta would endorse.
37
38
39
40
Note that Kobayashi 2011: 1257–1258 identifies Dharmottara as the opponent in the statements that
follow the passage just quoted and translated.
That the following passage contains a summary was first suggested by McCrea (2011: 325). Apart
from clearly rephrasing PV III 56c, gatir apy anyathā dṛṣṭā, this is also explicitly stated by Yamāri,
PVAṬSt : 103, 279–280: de ltar rnam pa gzhan gyis kyang rtogs pa mthong zhes bya ba’i rgya cher
bshad nas mjug sdud pa ni de’i phyir zhes bya ba’o/ (“Having explained [the verse] gatir … dṛṣṭā in
detail, [Prajñākaragupta] states the summary [of this explanation] with [the word] tasmāt.”)
The bold words are lifted from PV III 56, the text this is a comment on. Read pararūpeṇaiva PVABhmsB 108a3, PVABh2 218, 2, against pararūpeṇeva PVABh-msE 213b6. Yamāri explains, PVAṬSt : 103,
280: gzhan gyi rang bzhin du zhes bya ba ni thob par bya ba las gzhan gyi rang bzhin du’o/ (“… with
another nature, [meaning] with a nature other than what is to be obtained.”)
See Kobayashi 2011: 1259–1260 for another translation and short discussion of this passage, alongside
two others in which Prajñākaragupta presents alternative interpretations of Dharmakīrti’s phrase gatir
apy anyathā dṛṣṭā. I follow Kobayashi 2011: 1259–1260 here as far as the future object is concerned.
McCrea (2011: 323, n. 13) notes in this context that Prajñākaragupta does not think that a conceptual
cognition following perception grasps what is already grasped, and therefore accepts it as a means
of valid cognition. In this claim, Prajñākaragupta would be in direct contradiction to Dharmakīrti
and Dharmottara. (For Dharmakīrti, see the summary in Kellner 2004b: 9–10, referring to these two
passages: HBS 2, 18 ff., and a prose passage to PVin 3 k. 48. For Dharmottara, see Krasser 1995: 248–
249.)
To understand Prajñākaragupta’s argument here, one has to take into account his discussion of the
second pramāṇa definition in PVM II 5c, ajñātārthaprakāśo vā, where ajñātārthaprakāśa excludes
gṛhītagrahaṇa. Franco (1997: 50) translates an important passage of it (corresponding to Ono 2000: 79,
240
5
Prajñākaragupta’s Argument for Two Means of Valid Cognition
Jewels and keyholes
In the next section of the PVABh2 , Prajñākaragupta comments on a famous example given
by Dharmakīrti. According to Prajñākaragupta’s interpretation, the example is supposed to
show that a cognition, though mistaken, can be a means of valid cognition; this is possible
because even a mistaken cognition can lead to the attainment of an intended aim.41
Dharmakīrti’s illustration of this involves two very similar situations: two people,
thinking that they have seen a jewel, proceed to get that jewel. What they have seen is,
however, not in fact a jewel, but only the shine of something: one of them has seen the shine
of an actual jewel (spilling through a keyhole, according to Prajñākaragupta), the other the
shine of a lamp. For Prajñākaragupta’s interpretation of this example, the following points
are important:
1. The cognition “A jewel!” is the same in both cases.
2. It is also wrong in both cases, since both persons only see the shine of something,
never the jewel itself.
3. There is a difference in that one person will attain a jewel, and the other will not.
4. Based on this difference, one does not say that cognitions are mistaken in the same
way.
In a first step, Prajñākaragupta explains how this example shows the possibility that
inference, though mistaken, can a means of valid cognition:42 it is actually not due to it
41
42
15–19): “This, (i.e., illumination of an unapprehended object) is the definition of the means of knowledge
relating to absolute reality, whereas the previous one (i.e., a cognition that does not belie) is [a definition]
of [the means of knowledge] relating to the conventional (sāṃvyavahārika).” It would thus not be
surprising if, in the current context of discussing the conventional difference of means of valid cognition,
the problem of gṛhītagrahaṇa were to play a secondary role. A full analysis is, however, beyond the
scope of the current article.
Cf. the introduction of the passage by Prajñākaragupta, PVABh2 218, 28: avisamvādāt pramāṇatve ’pi
bhrāntatāṃ darśayati | (“[Dharmakīrti] explains that [a cognition], even though [it] is a means of valid
cognition because it does not belie [a person’s intentions], is erroneous.”)
McCrea (2011: 321) stresses the point that Prajñākaragupta takes the example of a jewel’s and a
lamp’s shine as exemplification of both inference and perception. This is true, especially if one reads
the later passages, but the initial analysis of these verses by Prajñākaragupta does talk only about
inference (and what is not really an inference), PVABh2 219, 2: tadvad anumānatadābhāsayor api tata
eva pramāṇetarate | (“Like that [example of the jewel and the lamp], also inference and what [only]
seems to be an [inference] are, just because of that, a means of valid cognition and something else.” The
phrase ‘just because of that’ refers to what makes the difference in the case of the jewel and the lamp:
the difference in attainment of what is desired.) Prajñākaragupta does, then, at least start off with what
can be taken as Dharmakīrti’s intent in these verses. The “abrupt reversal of this emphasis” (McCrea
2011: 321), that is, of this example being only for inference, is actually not very abrupt: Prajñākaragupta
analyzes the example of the jewel and the lamp first in terms of inference, then (as we will see shortly)
makes the point that inference and perception are interdependent, and only then goes on to defend
Dharmakīrti’s position, that inference and pseudo-inference have the same relation as the cognitions of
the people in the example, by showing that, if one were not to accept this argument about inference, also
perception would be vulnerable to the same problem. Rather than seeing Prajñākaragupta break sharply
with the intent of Dharmakīrti’s example, we could thus equally well speak of his careful explication
of it, perhaps with a view to opponents that had been trying to use the example against the Buddhist
epistemologists.
Patrick McAllister
241
being a cognition, but rather to factors external to the cognition. Though everything about
the two people’s cognitions is the same, the circumstances in which they have them is not.
And this is what differentiates an erroneous and non-erroneous cognition.
A further explanation of the varying degrees of erroneousness centers on whether an
error can belong to the causes that give rise to a means of valid cognition. It is in the
context of this explanation that Prajñākaragupta again introduces everyday activity as an
important factor in determining what a means of valid cognition is, and which means of
valid cognition there are:
PVABh2 II 244–245: vinānumānaṃ pratyakṣaṃ na pravartakam
āditaḥ |
tathānumānaṃ pratyakṣaṃ vineti pratipāditam ||
viśeṣas tv ayam evātra kvacit pūrvaṃ kvacit param |
anumānāt paraṃ nākṣaṃ nākṣāt pūrvānumeṣyate ||
Without inference perception does not, at first, prompt [a person] to act; likewise inference without perception. This was explained.
[There is], however, certainly this distinction here, [that] in some cases [one
is] earlier, in some cases later. One does not assume a perception following
upon inference, nor an inference prior to perception.
In these two verses, Prajñākaragupta is stating that inference and perception are dependent on each other: perception would never allow a person to act if there were not—
temporally prior to it—an inference that ascertains the connection between what is seen and
what is acted towards. The important thing to note here is the qualifier āditaḥ, “at first” or
“the first time”, which alludes back to the point that Prajñākaragupta made at the beginning
of the chapter, namely, that habituation, the necessary prerequisite for perception’s directly
leading to activity, cannot happen without inference.43 And the same holds for inference: it
does not prompt a person to act without perception, and could hence not be considered a
means of valid cognition without perception.44
In PVABh2 II 245, Prajñākaragupta admits to a difference between the two different
types of means of valid cognition: sometimes one is earlier, sometimes one is later;45 and,
as a kind of loose reason given for this difference, he puts forward what is not commonly
held to be the case: that perception follows upon inference, or, in other words, that inference
precedes perception. Implied, we must understand, is that the usually held view is correct,
but has to be augmented by this other case. The commonly held difference of perception
and inference is made with reference to cases where perception precedes inference (as
43
44
45
Cf. n. 21, and the paraphrase of PVABhIn II 24 above, section 2.2.
I am not yet quite sure how exactly to understand this point.
Jayanta explains concisely, PVAṬ-t: 101, 97: la lar dang por zhes bya ba ni mngon sum mam rjes su
dpag par yang ngo/ (“In some [cases] earlier, [that is], in [the case of] perception or also in [the case
of] inference.”)
242
Prajñākaragupta’s Argument for Two Means of Valid Cognition
in the usual case where a person sees smoke and infers ‘There’s a fire!’), and inference
follows perception.46
6
The final position: self-awareness, perception, and inference
After the opponent states that, whereas inference does depend on perception, perception
can occur without inference, Prajñākaragupta further clarifies his position:
PVABh2 II 246–248: pratyakṣam anumānena vinā mānaṃ svavedane |
vyavahāras tathā nāsti pramāṇatve ’pi kiṃ bhavet ||
svasaṃvedanamātre ca pratyakṣe ’rthāprasiddhitaḥ |
bhedasya ca na kiṃcit syād advaitam avaśiṣyate ||
tasmād arthasya bhedasya nādhyakṣasādhakaṃ vinā |
anumānaṃ tatas tasya pramātvaṃ nānumāṃ vinā ||
Perception without inference is a means of valid cognition concerning the
awareness [any cognition has] of itself.47 In that way, [however], there is no
everyday activity. So even though it is a means of valid cognition, what should
happen?
Furthermore, in a perception [that is] merely self-awareness, an [external]
object is not established, nor a difference; so there would be nothing at all;
[only] non-duality remains.
Therefore, perception does not establish an object [or] a difference without
inference. Therefore, this [perception] is not a means of valid cognition without
inference.
These three verses for the first time present the final position that Prajñākaragupta has
had in view throughout the whole debate.48 The two main points are:
46
47
48
One of the standard uses of inference is to clear up wrong notions that arise after a perception, cf.
Kellner 2004b: 6–9.
I have taken the locative svavedane here as signifying the object of perception, or what perception
applies to. The locative could also state the condition for the main clause, “In [the case of] the [perceptual] awareness [that every cognition has] of itself, perception is a means of valid cognition without
inference.” This does not commit Prajñākaragupta to a particularly difficult position. But in the light of
Prajñākaragupta’s commentary on PV III 63 (PVABh2 223, 19–20, discussed in n. 48), I prefer to take
the locative as expressing the object of perception.
That this is his accepted position can be seen from his commentary on PV III 63, which marks the end
of the discussion of how two objects are the reason for two means of valid cognition (see Franco and
Notake 2014: 23):
PVABh2 223, 19–21: prameyaṃ pramāṇena sidhyati | pramāṇasvarūpaṃ tu svasaṃvedanākārasiddheḥ | jñānākāra eva ca svasaṃvedanaḥ svasāmānyatayopalabhyamānaḥ
pratyakṣānumānaviṣaya ity uktam | viṣayadvaividhyād ākāradvaividhyād ity arthaḥ |
svākāradvayasaṃvedane hi naikam iti yuktam ||
Patrick McAllister
243
1. Perception is a means of valid cognition in and of itself, but only as far as its own
appearance to itself is concerned.
2. However, in order to count as a means of valid cognition in the conventional sense,
i.e., as a cognition that allows a person to reliably attain a desired aim in interaction
with external objects, it has to be supplemented by inference.
It is this distinction between perception in the sense of direct awareness of consciousness
itself and the direct awareness of an external object that has been driving, and sometimes
confusing, the discussion.
Prajñākaragupta is not prepared to admit that perception, taken as a cognition that
reliably leads to activity, can be given the status of a means of valid cognition without also
granting that status to inference. The reason for this is that perception, whilst in and of
itself a means of valid cognition, does not let a person act. The only object that it can be
considered to be a means of valid cognition for is its own appearance to itself. Any other
object, or quality of an object, must be ascertained by another type of cognition. If one still
insists, as the opponent does, that perception does reliably cause successful activity, then
one must include a cognitive element which is not perception itself, and hence has lost the
argument that only perception is a means of valid cognition.
To close, let us look at how this interpretation fits in with the example of the lamp and
the jewel. Prajñākaragupta explain the errors that are involved in this situation:
PVABh2 221, 1–3: maṇipratibhāsas tu maṇau maṇiprabhāyāṃ ca samāna
eva | tatra kvacid deśabhrāntiḥ | kvacit svarūpabhrāntiḥ | kvacid ubhayam |
kvacid anubhayam | maṇiprabhāyāṃ maṇijñānasya deśabhrāntir | maṇāv eva
prāpyasvarūpabhrāntiḥ | sāmānyānumānasyobhayabhrāntiḥ | svasaṃvedanasya nobhayathāpīti prakāraḥ |
The appearance of a jewel [to cognition], however, is exactly the same in [the
case of] a jewel and in [the case of] the shine of a jewel. In this [situation],
The object of valid cognition is established by the means of valid cognition. The own
nature of the means of valid cognition, however, [is established] on [the basis of] the
establishment of the form of [cognition’s] awareness of itself. And it was stated that
[this] awareness [that cognition has] of itself, the very form of cognition [itself], is the
object of perception [or] inference, [depending on whether] it is being apprehended as
being [only] itself or common [also to other things]. [To say] “because there are two
[types of] object” means “because there are two [types of] form [of cognition].” For,
given [cognition’s] awareness of two forms of itself, it is incoherent [to say:] “There is
one [object of valid cognition].”
This passage is also discussed and partially translated in Franco and Notake 2014: 149–150 (note that
they emend °ākārasiddheḥ to °ākārasiddham).
I was unable to find an exact quote of the explanation referred to in this passage, but the differentiation
made here is closely parallel to that made repeatedly for the distinct and indistinct forms of cognition,
see n. 15, PVABhIn 39, 6–9 (discussed section 2.1), and PVABhIn 36, 25–27 (discussed section 2). The
viṣayadvaividhyāt rephrases the expression “prameyadvitvena” in tasmāt prameyadvitvena pramāṇadvitvam iṣyate (PV III 63cd), which in turn might echo the expression “meyadvaividhyāt” of PV III 1 (cf.
Franco and Notake 2014: 23).
244
Prajñākaragupta’s Argument for Two Means of Valid Cognition
there is, in some respect, an error about the place; in some respect, an error
about the proper nature; in some respect, both; in some respect, neither.
As regards the shine of a jewel, the cognition of a jewel is wrong about the
place; for the actual jewel, there is an error about the nature of the [thing]
to be obtained; the inference, [concerned with] a general property, is wrong
about both [the place and the nature of the obtained thing]; [cognition’s]
awareness of itself is wrong about neither. This is the manner [in which one
has to understand this].
There are thus quite a few different respects in which the cognition of a jewel can be
mistaken:
1. When one perceives not the jewel itself, but the shine of the jewel, the jewel cognition
is wrong about the place of the thing that will be attained on account of this cognition.
2. When one is actually perceiving a jewel, the jewel cognition is still wrong about the
essential nature of the object that will be attained (because the present jewel is not
identical with the jewel that will be attained).49
3. An inference is wrong about both the place and the nature of the attained thing.50
4. The jewel cognition’s awareness of itself is wrong about neither the place nor the
nature of what will be attained (though it is of little practical value that the gleam-ofa-jewel-appearance is of the same nature as the gleam-of-a-jewel-appearance that
will be attained).
After having presented this analysis, Prajñākaragupta can make the first part of his
closing statement:
PVABh2 221, 4–7: tato yad uktaṃ yā gatiḥ sā svarūpeṇaiva yathā pratyakṣā
gatiḥ | yat pramāṇaṃ tad abhrantaṃ yathā pratyakṣaṃ tad ayuktam | pratyakṣāpi gatir na svarūpeṇa | na cābhrāntaṃ pratyakṣam asti | svarūpe ca
yathā pratyakṣam abhrāntaṃ tathā ’numānam apy anye ca bhrāntābhimatāḥ
pratyayā iti na tathā bhrāntatā pratipādanaṃ kvacid upayogi |
49
50
If my interpretation is correct, this error would also apply to the situation described in the previous
point. It would further imply that inference has (at least) the same two errors as that perception of the
shine of a jewel: it is mistaken about the place of the inferred thing, and also about the nature of what
will be obtained. This would actually suit Dharmakīrti’s example (an inference likened to a ‘slightly
wrong’ perception). But Prajñākaragupta’s formulation of these different errors is not quite clear on this
point: is each error specific to each case (but not the only one that is applicable), or is it the only error
in each case? If the latter, it would contradict my understanding.
It is wrong about the nature because the thing to be attained is not its object; an inference does not have
particulars as its object at all, but only general concepts. It is wrong concerning the place because an
inference cannot say exactly where the inferred thing is placed: cf. PVABh2 II 249, where it is explained
that one can infer only the general existence of a fire behind a wall, but not its specific location.
Patrick McAllister
245
Therefore, what was said:51 “That is a cognition which [cognizes something]
according to [the thing’s] own nature, like perceptual cognition. What is a
means of valid cognition is not erroneous, like perception.” is incorrect.
Even if it is perceptual, a cognition is not according to [the thing’s] nature.
Nor is there a non-erroneous perception.
But as perception is non-erroneous with regard to [its] own nature, so also
inference is [non-erroneous], as well as other cognitions which are assumed
to be erroneous.52 So an explanation as being erroneous in these ways is of no
use in any case.
Prajñākaragupta is here succinctly restating the main points of his commentary on the
example of the jewel and the lamp, which in turn exemplifies the points made in the section
starting from PV III 53d: the whole section has been aimed at refuting an opponent who
claims that ‘being erroneous’ is the criterion by which to differentiate what is a means of
valid cognition from what is not.53 And as Prajñākaragupta has made abundantly clear, this
is not how means of valid cognition can be identified, at least not if one simultaneously
wants to uphold that perception is such a means of valid cognition: one would have to admit
that inference, and actually also other types of cognition which are generally held to be
erroneous, are ‘erroneous’ and ‘non-erroneous’ in the exact same way as perception. All
are wrong concerning the object to be attained (though they might be wrong in different
ways),54 and all are correct concerning their direct awareness of themselves. That they are
wrong in some respect is a necessary condition for enabling everyday activity; that they
are necessarily correct in terms of self-awareness might be a nice philosophical insight,
but one that is not useful for normal people at this point.55
51
52
53
54
55
I was not able to identify a statement that corresponds to this literally, but the same argument was made
in the context of Prajñākaragupta’s comment on PV III 55cd, PVABh2 215, 26–27, discussed above,
section 4: there too, an opponent makes non-erroneousness a criterion of validity, and maintains that
cognizing a thing with a form other than its own cannot be valid.
Cf. the quotation in TBh2 19, 7: svarūpe sarvam abhrāntaṃ pararūpe viparyayaḥ. (“Every cognition is
non-erroneous regarding its own nature, but wrong concerning another nature.”) Kajiyama (1998: 52,
n. 118) notes that the quote is not identified, and refers to PVABh2 331, 13–14 which expresses the
same idea. I thank Birgit Kellner for alerting me to this passage.
Yamāri points out that the opponent here is “Cārvāka and so on”, PVAṬSt : 103, 295: ’dis ni tshu rol
mdzes pa la sogs pa gsal ba ma yin no zhes mjug sdud pa ni/ des na zhes bya ba’o/
From the list given above (section 6), it follows that the erroneousness of perception consists in not
grasping the nature of what is to be obtained by it (prāpyasvarūpabhrānti, PVABh2 221, 2–3). The
observation by McCrea (2011: 327)—that “ […] Prajñākara is […] flatly contradicting one of the most
fundamental claims that Dharmakīrti himself made about perception” by saying that “perception is
not free from error”—has to be modified: even according to Prajñākaragupta it is non-erroneous with
regard to the present thing; but it is erroneous with regard to what is to be obtained by the activity that
it could lead to. Taken in this sense, the contradiction to Dharmakīrti vanishes. Prajñākaragupta is only
clarifying that pratyakṣa is not a means of valid cognition because it is non-erroneous.
In a soteriological context, however, cognition’s awareness of itself is a central issue, especially in
Prajñākaragupta’s discussion of PVM II 1–6, edited in Ono 2000. To illustrate this with just one passage,
cf. Franco 2004: 168: “Means of knowledge is only self-awareness, the single [type of] perception;
there is no other, because the multiplicity (prapañca) [of the means of knowledge taught by the Buddha
merely] follows [the needs of] the people to be trained.” See also Arnold 2012: 187–88.
246
Prajñākaragupta’s Argument for Two Means of Valid Cognition
In the second part of the closing statement for this section, we can see that the direct
awareness a cognition has of itself is essential for the differentiation of two, and only two,
means of valid cognition, the other central point that was at issue:
PVABh2 221, 7–10: evaṃ tarhi kathaṃ pratyakṣānumānayor lakṣaṇabhedaḥ | uktam atra spaṣṭetarapratibhāsabhedād iti | sa eva pratibhāsabhedas
tathā kuto bhavati | yasya svatantraṃ grahaṇaṃ tatra spaṣṭapratibhāsatā |
yasyānyathā tad aspaṣṭapratibhāsam atīndriyaṃ parokṣam |
[Opponent:] If it is so, how [can there be] a different definition for perception
and inference?
[Proponent:] To this [question] it was said that [it is] due to a difference in the
appearance as distinct [or] not.56
[Opponent:] Exactly this different appearance in such a way, what does it come
from?
[Proponent:] In the case of that which is grasped independently [from any form
other than its own], there is a distinct appearance. That [which is grasped] in
another way has an indistinct appearance, is beyond the senses, is remote.
With this, Prajñākaragupta has restated the main reason (as he saw it) for the differentiation of two types of valid cognitions: it is wholly ‘intrinsic’ to the content of the cognitions,
and has nothing to do with the correctness or erroneousness of the respective cognitions.57
7
Conclusion
I have tried to show that Prajñākaragupta’s commentary on PV III 53d–57 is carefully
composed and consistently argued in view of his commentary on PV III 1–2: the same
central concerns underlie many arguments, and it can sometimes be misleading to focus on
one particular segment of the text in isolation from its wider context.
The main points from Prajñākaragupta’s discussion, as I have analyzed it here, can be
restated as follows:
1. To be a means of valid cognition is here being considered as making activity according
to a cognizer’s expectations possible (pravartaka-tva is the main criterion for validity
in these passages).
2. Under that definition, there are two such means of valid cognition, perception and
another, based on the two ways in which objects appear in awareness: distinctly and
indistinctly.
56
57
There are various passages this could be referring to. The most likely candidate, in my opinion, is the
last statement before the example of the jewel and the lamp is introduced, PVABh2 218, 26: tasmāt
parokṣaviṣayapravartakatve ’pi spaṣṭāspaṣṭabhedāt pramāṇadvitayam eva | See also n. 15.
Note also that, whereas this is a sufficient criterion for the distinction of perception and non-perception,
it is certainly not a sufficient one for the distinction of inference and conceptual cognition. But Prajñākaragupta is here trying to show only that there must be perception and at least one other type of valid
cognition. For this, the argument is sufficient.
Patrick McAllister
247
3. Both means are erroneous about the object that they enable a cognizer to obtain: that
object does not appear in perception (which grasps a present thing), and the other
type operates without particulars, one of which will be obtained.
4. Using any notion of “correspondence to an object” other than making successful
activity possible, in order to distinguish what is and what is not a means of valid
cognition, is doomed to failure, because to be a means of valid cognition implies
non-correspondence to the very object to which it facilitates everyday activity; that
object is never present to it.
Prajñākaragupta thus resolves the tension between PV III 1 and PV III 53d—that the
two means of valid cognition result from two objects of valid cognition, but that there is in
fact only one such object—on the basis of a complex argument that works on two levels:
if everyday activity is to be possible, the duality of the means of valid cognition must be
accepted; perception by itself cannot support this activity. That only the particular is sought
for in this activity does not change this: perception and inference cannot be distinguished in
their erroneousness about this future thing, and the main argument of PV III 1 has therefore
to be understood to mean that there are two means of valid cognition because this particular,
the future object of successful activity, is cognized in two ways, distinctly or indistinctly.
Alternatively, it is possible to maintain that only perception is a means of valid cognition.
Only the perception of cognition by itself exists in reality: this does, however, not support
everyday activity, is not the topic of the pratyakṣa chapter, and is not acceptable to the
opponent.
Quite a few factors make the interpretation of Prajñākaragupta’s work as difficult as
it is interesting: its sheer extent and the unreliability of its edition are two major ones;
they are compounded by Prajñākaragupta’s style of writing, nearly colloquial in its flow
of arguments, and his skill in letting multiple threads run throughout long exchanges that
shift in their main focus: the question of a cognition’s correctness, the general definition of
a means of valid cognition, the distinction of the means of valid cognition acceptable to
the participants in Prajñākaragupta’s debate, the relation between cognition’s accordance
to an object and the definition of a means of valid cognition, and a number of assumptions
that in the end turn out to entail unwanted consequences and have to be discarded.
I have here presented one example of such an exchange, taking care to weigh the
arguments in respect to their dialectical context, in terms both of their content and of the
commitment with which they are made. I believe this is the best way to come to a clearer
understanding of the claims that Prajñākaragupta is actually making.
References and abbreviations
Arnold 2012 Dan Arnold, Brains, Buddhas, and Believing: The Problem of Intentionality
in Classical Buddhist and Cognitive-Scientific Philosophy of Mind. New York
2012.
Dunne 2004
John D. Dunne, Foundations of Dharmakīrti’s Philosophy. Boston 2004.
248
Prajñākaragupta’s Argument for Two Means of Valid Cognition
Franco 1997
Eli Franco, Dharmakīrti on Compassion and Rebirth. Vienna 1997.
Franco 2004 Eli Franco, Towards a Critical Edition and Translation of the Pramāṇavārttikālaṃkārabhāṣya: A Propos Two Recent Publications. Wiener Zeitschrift für
die Kunde Südasiens 48 (2004) 151–169.
Franco 2012 Eli Franco, How to Distinguish between Non-existing Entities?: Dharmakīrti and Prajñākaragupta on Universals as Objects of Knowledge. Rocznik
Orientalistyczny 65.1 (2012) 51–62.
Franco and Notake 2014 Eli Franco, Miyako Notake, Dharmakīrti on the Duality of
the Object: Pramāṇavārttika III 1–63. Berlin et al. 2014.
Hattori 1968 Masaaki Hattori, Dignāga, On Perception: Being the Pratyakṣapariccheda
of Dignāga’s Pramāṇasamuccaya from the Sanskrit Fragments and the Tibetan
Versions. Cambridge et al. 1968.
HBS
Dharmakīrti, Hetubindu. In: Dharmakīrti’s Hetubinduḥ. Vol. 1: Tibetischer Text
und rekonstruierter Sanskrit-Text, ed. Ernst Steinkellner. Vienna 1967, 29–99.
Inami et al. 2002 Masahiro Inami et al., Prajñākaragupta ni okeru nishu no taishō to
ninshiki-shudan (“Prajñākaragupta’s View on the Two Kinds of Valid Cognitions
and the Two Kinds of Objects”). Nanto Bukkyō 81 (2002) 1–53.
Kajiyama 1998 Yuichi Kajiyama, An Introduction to Buddhist Philosophy: An Annotated
Translation of the Tarkabhāṣā of Mokṣākaragupta, ed. Ernst Steinkellner. Reprint
with corrections in the author’s hand. Vienna 2 1998.
Katsura 1993 Shōryū Katsura, On Perceptual Judgement. In: Studies on Buddhism in
Honour of Professor A.K. Warder, ed. Narendra K. Wagle, Fumimaro Watanabe.
Toronto 1993, 66–75.
Kellner 2004a Birgit Kellner, First logic, then the Buddha? The Controversy about
the Chapter Sequence of Dharmakīrti’s Pramāṇavārttika and the Soteriological
Relevance of Inference. Hōrin 11 (2004) 147–167.
Kellner 2004b Birgit Kellner, Why Infer and not just Look? Dharmakīrti on the Psychology of Inferential Processes. In: The Role of the Example (dṛṣṭānta) in Classical
Indian Logic, ed. Ernst Steinkellner, Shōryū Katsura. Vienna 2004, 1–51.
Kellner 2010 Birgit Kellner, Self-Awareness (svasaṃvedana) in Dignāga’s Pramāṇasamuccaya and -vṛtti: A Close Reading. Journal of Indian Philosophy 38 (2010)
203–231.
Kellner et al. 2020 Birgit Kellner et al., eds., Reverberations of Dharmakīrti’s Philosophy: Proceedings of the Fifth International Dharmakīrti Conference Heidelberg,
August 26 to 30, 2014. Vienna 2020.
Kobayashi 2010 Hisayasu Kobayashi, Self-Awareness and Mental Perception. Journal
of Indian Philosophy 38.3 (June 2010) 233–245.
Patrick McAllister
249
Kobayashi 2011 Hisayasu Kobayashi, Prajñākaragupta’s Interpretation of svalakṣaṇa.
Journal of Indian and Buddhist Studies 59.3 (Mar. 2011) 1256–1261.
Krasser 1991 Helmut Krasser, Dharmottaras kurze Untersuchung der Gültigkeit einer
Erkenntnis Laghuprāmāṇyaparīkṣā: Materialien zur Definition gültiger Erkenntnis in der Tradition Dharmakīrtis. Vol. 2: Übersetzung. Vienna 1991.
Krasser 1995 Helmut Krasser, Dharmottara’s Theory of Knowledge in his Laghuprāmāṇyaparīkṣā. Journal of Indian Philosophy 23.2 (1995) 247–271.
McCrea 2011 Lawrence J. McCrea, Prajñākaragupta on the pramāṇas and their objects.
In: Religion and Logic in Buddhist Philosophical Analysis: Proceedings of the
Fourth International Dharmakīrti Conference; Vienna, August 23–27, 2005, ed.
Helmut Krasser et al. Vienna 2011, 319–328.
McCrea and Patil 2006 Lawrence J. McCrea, Parimal G. Patil, Traditionalism and Innovation: Philosophy, Exegesis, and Intellectual History in Jñānaśrīmitra’s Apohaprakaraṇa. Journal of Indian Philosophy 34.4 (2006) 303–366.
Nagatomi 1967–1968 Masatoshi Nagatomi, Arthakriyā. The Adyar Library Bulletin
31–32 (1967–1968) 52–72.
NBhūṣ
Bhāsarvajña, Nyāyabhūṣaṇa. In: Śrīmadācārya-Bhāsarvajña-praṇītasya Nyāyasārasya Svopajñaṃ Vyākhyāṇaṃ Nyāyabhūṣaṇam, ed. Svāmī Yogīndrānandaḥ.
Vārāṇasī 1968.
Ono 2000 Motoi Ono, ed., Prajñākaraguptas Erklärung der Definition gültiger Erkenntnis: Pramāṇavārttikālaṃkāra zu Pramāṇavārttika II 1–7. Vienna 2000.
PS I
Dignāga, Pramāṇasamuccaya 1. Reconstruction in Steinkellner 2005.
PV III
Dharmakīrti, Pramāṇavārttika: Chapter on perception. In: Bukkyō ninshikiron no
kenkyū: Hōshō “Pramāṇavārttika” no genryōron. Bukkyō ninshikiron no kenkyū,
ed. Hiromasa Tosaki. 2 vols. Tōkyō 1979–1985.
PVABh-msB Prajñākaragupta, Prajñākaragupta’s Pramāṇvārttikabhāṣyam: Manuscript
B. In: Sanskrit Manuscripts of Prajñākaragupta’s Pramāṇvārttikabhāṣyam: Facsimile Edition, ed. Shigeaki Watanabe. Patna, Narita 1998, 1–81.
PVABh-msE Prajñākaragupta, Prajñākaragupta’s Pramāṇvārttikabhāṣyam: Manuscript
E. In: Sanskrit Manuscripts of Prajñākaragupta’s Pramāṇvārttikabhāṣyam: Facsimile Edition, ed. Shigeaki Watanabe. Patna, Narita 1998, 83–101.
PVABh2
Prajñākaragupta, Pramāṇavārttikālaṅkārabhāṣya. In: Pramāṇavārtikabhāshyam
or Vārtikālaṅkāraḥ of Prajñākaragupta (Being a Commentary on Dharmakīrti’s
Pramāṇavārtikam), ed. Rāhula Sāṅkṛtyāyana. Patna 1953.
250
Prajñākaragupta’s Argument for Two Means of Valid Cognition
PVABhIn Prajñākaragupta, Pramāṇavārttikālaṅkāra ad PV III 1–2. In: Prajñākaragupta
ni okeru nishu no taishō to ninshiki-shudan (“Prajñākaragupta’s View on the Two
Kinds of Valid Cognitions and the Two Kinds of Objects”), ed. Masahiro Inami
et al. 2002, 1–53.
PVAṬ-t
Jayanta, tshad ma rnam ’grel gyi rgyan gyi ’grel bshad (Pramāṇavārttikālaṅkāraṭīkā). In: dpe bsdur ma, ed. krung-go’i bod rig pa zhib ’jug lte gnas kyi bka’
bstan dpe sdur khang gis dpe bsdur zhus. Vol. 100–101. Beijing 2009, 100.759–
101.1592.
PVAṬSt
Yamāri, tshad ma rnam ’grel rgyan gyi ’grel bshad shin tu yongs su dag pa
(Pramāṇavārttikālaṅkāraṭīkā Supariśuddhā). In: dpe bsdur ma, ed. krung-go’i bod
rig pa zhib ’jug lte gnas kyi bka’ bstan dpe sdur khang gis dpe bsdur zhus. Trans.
by Sumatikīrti, rngog blo ldan shes rab. Vol. 102–103. Beijing 2009, 102.471–
103.932.
PVin 2
Dharmakīrti, Pramāṇaviniścaya 2. In: Dharmakīrti’s Pramāṇaviniścaya: Chapters 1 and 2, ed. Ernst Steinkellner. Beijing, Vienna 2007, 45–101.
PVin 3
Dharmakīrti, Pramāṇaviniścaya 3. In: Dharmakīrti’s Pramāṇaviniścaya: Chapter
3, ed. Pascale Hugon, Toru Tomabechi. Beijing, Vienna 2011.
PVM
Dharmakīrti, Pramāṇavārttika-kārikā. In: Pramāṇavārttika-kārikā (Sanskrit and
Tibetan), ed. Yūsho Miyasaka. Quotes of chapters are by number, in the following
order: pramāṇasiddhi quoted as “2”; pratyakṣa as “3”; svārthānumāna as “1”;
parārthānumāna as “4”. Narita 1971–1972, 1–206.
PVV
Manorathanandin, Pramāṇavārttikavṛtti. In: Dharmakīrti’s Pramāṇavārttika with
a Commentary by Manorathanandin, ed. Rāhula Sāṅkṛtyāyana. Patna 1938–1940.
Steinkellner 2005 Ernst Steinkellner, ed., Dignāga’s Pramāṇasamuccaya, Chapter 1: A
hypothetical reconstruction of the Sanskrit text with the help of the two Tibetan
translations on the basis of the hitherto known Sanskrit fragments and the linguistic
materials gained from Jinendrabuddhi’s Ṭīkā. Apr. 2005. https://www.oeaw.ac.at/
fileadmin/Institute/IKGA/PDF/forschung/buddhismuskunde/dignaga_PS_1.pdf
(last visited 09-2020).
TBh2
Mokṣākaragupta, Tarkabhāṣā. In: Tarkabhāṣa and Vādasthāna of Mokṣākaragupta
and Jitāripāda, ed. H.R. Rangaswami Iyengar. Mysore 2 1952, 1–71.
Balancing the Scales: Dharmakīrti Inside and Out
by
Lawrence McCrea
Recent years have witnessed much debate about the best way to make sense of Dharmakīrti’s
apparent diversity of philosophical positions regarding the status of mind-independent
objects: at some times he appears to argue in ways that presuppose the existence of such
objects, while at others he argues against them. Several approaches have been proposed
to account for this apparent contradiction in Dharmakīrti’s various statements regarding
ontology. Most famously John Dunne (2004), building on earlier interpretive strategies
suggested by Georges Dreyfus (1997) and Sara McClintock (2003), has suggested that
we should see such positions as ranged along a “sliding scale” of hierarchically arranged
stances, in which “more accurate descriptions of what we perceive and think supersede less
accurate ones.” He finds in the Pramāṇavārttika four such levels of analysis: “The Views
of Ordinary Persons,” “the Abhidharma Typology,” “External Realism” and “Epistemic
Idealism.” Each of these levels is said to give way to the next through a specific kind of
“mereological” transition argument, which shows that certain entities accepted as real on a
given ontological level cannot be accounted for either as unitary or as multiform – as “one
or many” – forcing one to abandon this ontology and ascend to the next, higher level of
analysis.
Dunne’s views have provoked much discussion and several significant critiques. I
will mention here specifically only Kellner (2011) Arnold (2008). Kellner challenges the
uniformity of the “mereological” model for ascending the levels, among others discussing
a specific idealist argument of Dharmakīrti, showing that there is no way to satisfactorily
distinguish definitionally between the supposedly external object of a cognition and the
immediately preceding cognition (samanantara-pratyaya) that gives rise to it, since both
are causes of the awareness, and resemble it in form. Arnold questions whether the third,
external realist level is philosophically viable, even as a theoretical alternative to the
“epistemic idealist” level, to which he believes it inevitably reduces. But neither of them,
or any other critic of the sliding scale so far as I know, questions the hierarchy itself
– in particular the idea that the idealist or Yogācāra position set forth at certain points
in the Pramāṇavārttika and Pramāṇaviniścaya is meant to be presented as clearly and
unambiguously preferable to the external realist, “Sautrāntika” or bāhyārthavāda level. It
is this widespread assumption I wish here to call into question.
On Dunne’s view, the external realist or “Sautrāntika” arguments Dharmakīrti often
relies upon are there only as preliminary, conditional positions—as stepping stones which
invariably give way through a specific kind of “transition argument” to the more accurate
and more soteriologically beneficial idealist or Yogācāra position which represents Dharmakīrti’s real view. Dunne characterizes the interpretive strategy that leads him to this
conclusion as one of a “hermeneutics of charity.”1 As Dunne explains this,
1
See Dunne 2004: 11, 68–69, 239.
Birgit Kellner et al., eds., Reverberations of Dharmakīrti’s Philosophy: Proceedings of the Fifth International Dharmakīrti
Conference Heidelberg, August 26 to 30, 2014. Vienna 2020, pp. 251–265.
252
Balancing the Scales: Dharmakīrti Inside and Out
If… a well articulated passage or position clearly contradicts some other,
equally well articulated passage or position, then the general principle of
hermeneutical charity we have adopted will prompt us to rank these positions
along a hierarchical scale; the lower position, while contextually expedient, is
superseded by the higher position that, by virtue of passing a rational test that
the lower position fails, conforms more closely to Dharmakīrti’s version of
ultimate truth (Dunne 2004: 239, n. 24).
But should we take it as a given that there can be no reason why a philosophical author
might choose to present multiple, genuinely alternative points of view, or alternative
explanations of the phenomena he is attempting to account for, without falling prey to
charges of incoherence? It is not in fact difficult to imagine motives that might prompt
such an approach. For example, Vincent Eltschinger (2010, 2014) has recently stressed
the importance of the Buddhist epistemological tradition as foregrounding for apologetic
purposes a “supersectarian” identity “aimed at defending Buddhism as a whole against
Brahmanical hostility” (2010: 399; see also Eltschinger 2014: 174ff.). Taking this into
account, one can easily see how it might have seemed desirable to formulate a broadbased, “big tent” defense of Buddhism which could accommodate defenders of somewhat
different Buddhist positions within a largely common epistemological framework. I will
later mention one more way (apart from the “sliding scale”) in which we might make sense
of Dharmakīrti’s apparent accommodation of multiple ontological stances.
At this point I would like, instead, to point out one obvious counterexample to Dunne’s
line of thinking on this question—one which has not, so far as I know, been discussed
in connection with the vexed question of Dharmakīrti’s ontology and his argumentative
method and goals. I am thinking of Dharmakīrti’s own primer on Buddhist logic and
epistemology, the Nyāyabindu. It has long been a commonplace among commentators and
subcommentators on this work (stretching back at least to Vinītadeva in the 8th century) to
view the Nyāyabindu as a work designed to conform to both a realist and an idealist ontology
(usually labelled as “Sautrāntika” and “Yogācāra,” respectively).2 Certain commentators
observe that, at a few specific points in the text, these two contrary ontological perspectives
necessarily diverge, and the question is sometimes raised as to whether some point advanced
in the text is incompatible with either the Yogācāra or the Sautrāntika positions. But the
idea that the text on the whole is meant to accommodate both ontologies is widely shared,
and nowhere, I think, seriously disputed.
What I would like to suggest, simply put, is that both of Dharmakīrti’s major systemic
treatises on Buddhist logic and epistemology, the Pramāṇavārttika and the Pramāṇaviniścaya, are, like the Nyāyabindu, designed to accommodate and to accord, as seamlessly
as possible, both with an idealist, Yogācāra, “mind-only” ontology and with that specific
variety of external realism which later doxographers designate as “Sautrāntika” – what
is sometimes known as the anumeya-bāhyārthavāda, the “inferred external object” view.
There may well be – indeed there certainly is – reason to believe that, between these
two views, Dharmakīrti regarded the Yogācāra position as philosophically (and perhaps
2
See Vinītadeva ad Nyāyabindu 1.1 and 1.3 (Nyāyabinduṭīkā, pp. 4 and 7): iṣṭaś ca prakaraṇārambhaḥ
sautrāntikayogācārobhayanayānudhāvanārtham. See Dharmottarapradīpa, pp. 42–44 and 61 for additional references.
Lawrence McCrea
253
soteriologically) preferable, but, as we shall see, he nevertheless goes out of his way on
multiple occasions to present these two positions as viable alternative philosophical paths.
Once we abandon the presupposition that discerning Dharmakīrti’s own preferred
position on the question of objects and discerning what he intends to argue for in his works
are necessarily equivalent, it becomes much easier to account for certain aspects of the way
he juxtaposes the vijñānavāda and bāhyārthavāda positions at several key points in the
Pramāṇavārttika and the Pramāṇaviniścaya. One important reason it is difficult to read
Dharmakīrti’s key statements of the vijñānavāda position as an unambiguous ontological
step up from the bāhyārthavāda position is that it is often after having developed what
seem to be, and have generally been accepted as, his definitive vijñānavādin arguments
that Dharmakīrti advances his most fully developed bāhyārthavādin arguments. Indeed, it
seems to be Dharmakīrti’s regular, and perhaps even invariable, practice (as I will demonstrate), to follow his fullest and seemingly most conclusive arguments for the vijñānavādin
position with a corresponding and parallel argument defending a specific version of the
bāhyārthavādin position. It is difficult to see why he would adopt this approach if his
ultimate philosophical and soteriological strategy were designed to bring his readers up to
the highest, idealist, level and leave them there.
Pramāṇaviniścaya 1.58
The most striking passage of this kind, and the one that seems to have garnered the most
attention from later chroniclers and critics of the Sautrāntika position, is one for which
the original Sanskrit text has only recently become fully available, and has not yet, I
think, received the attention it deserves – that is, Pramāṇaviniścaya 1.58 (the final kārikā
in the chapter on perception), along with its accompanying vṛtti. The passage occurs
just after Dharmakīrti has presented his famous sahopalambha-niyama (“necessary coapprehension”) argument – that “there is no distinction between blue and the awareness of
blue, because they are necessarily co-apprehended.”3 Having advanced this claim as part
of his defense of the vijñānavāda position, Dharmakīrti now turns to consider its relevance
to the bāhyārthavādin position:
Because of this, even if there is an external object, there is no difference
between the appearing object and the awareness of it. (1.58ab)
Even if an external object does exist, the non-differentiation between the
appearing object (“blue,” or the like) and the awareness of it is established by
the co-apprehension and co-awareness of the two.
Therefore, the awareness has a double form.
Therefore it is established that awareness has a twofold form, since it bears
the form of both object and awareness. [Pūrvapakṣin:] If the appearing object
is not differentiated from the awareness, then how can one think that it is
something different?
3
PVin 1.54 (p. 39): sahopalambhaniyamād abhedo nīlataddhiyoḥ /
254
Balancing the Scales: Dharmakīrti Inside and Out
There can be proof of an external object on the basis of disjunction [vyatireka]. (1.58cd)
The fact that awareness [sometimes] does not arise when all other causes
capable of producing it exist indicates the lack of some other cause; this could
be the external object.4
Even if we accept that all that is phenomenally apparent to us in any awareness is the
awareness’s own form, we may still be able to infer an external object as a cause of what
appears to us. We do not, for example, see “blue” at all times and, in the absence of any
evident factor in our immediately prior awareness that determines whether we do or not
see blue at any particular moment, we may legitimately infer an additional, extra-mental
cause for the occasional appearance of, e.g., blue in our visual awareness. This is clearly
presented as a positive argument in favor of the bāhyārthavādin position, and has invariably
been interpreted as such by later authors, both Buddhist and non-Buddhist (for whom
Pramāṇaviniścaya 1.58 and its attendant vṛtti become, indeed, the classic statement of the
Buddhist bāhyārthavādin or Sautrāntika position).5
It is hard to see why, if the defense of bāhyārthavāda (conditional or otherwise) is
meant to be a stepping stone on the path to Dharmakīrti’s ultimate stance of vijñānavāda,
he should choose, after directly presenting his most fully developed argument in defense
of this vijñānavāda, to then deliberately step backwards, or downwards, to the “lower,”
supposedly philosophically less tenable and soteriologically less useful level of external
realism. Moreover, even if we were able to find some motive for this return to defense of the
bāhyārthavādin position after the supposed transition to “epistemic idealism” has already
been effected, there is no way what we see in Pramāṇaviniścaya 1.58 can be read as a
step backward to a more primitive, more naive, or less fully developed position than the
vijñānavādin view that precedes it. On the contrary, it is presented explicitly as a sequel to,
and an outgrowth of, the purportedly vijñānavādin analysis that immediately precedes it. It
builds on, and entirely presupposes, the dvairūpya of perceptual cognitions that Dharmakīrti
has just argued for, and the validity of the sahopalambhaniyama argument that formed its
centerpiece. Indeed, in the light of the way Dharmakīrti constructs the bāhyārthavādin
argument in 1.58, and the way he connects it with the preceding argument, one must in
fact recognize that the sahopalambhaniyama argument, despite the way it has so often
been characterized by both premodern and modern expositors of Dharmakīrti’s thought,
is in fact not an idealist argument at all. That is to say, it is fully consistent with the final
versions of both the vijñānavādin and bāhyārthavādin positions as Dharmakīrti develops
them in both the Pramāṇaviniścaya 1.58 and the parallel passage of the Pramāṇavārttika
4
5
PVin 1.58 (p. 43): bāhye ’py arthe tato ’bhedo bhāsamānārthatadvidoḥ / (1.58ab) saty api bāhye
’rthe sahopalambhavedanābhyāṃ bhāsamānasya nīlādes tatsaṃvidaś cāvivekaḥ siddhaḥ. dvairūpyaṃ
tad dhiyaḥ (1.58c) viṣayajñānarūpābhyām ato dvirūpā buddhiḥ siddhā bhavati. yadi bhāsamāno
viṣayākāro buddher abhinnaḥ, tato bhinnam astīti kutaḥ? bāhyasiddhiḥ syād vyatirekataḥ // (1.58cd)
satsu samartheṣu anyeṣu hetuṣu jñānakāryāniṣpattiḥ kāraṇāntaravaikalyaṃ sūcayati. sa bāhyo ’rthaḥ
syāt.
For quotations of the passage, see for example the following: bāhyasiddhiḥ syād vyatirekataḥ: NM, vol.
2, p. 492; ĪPVV, vol. 2, p. 129; TBh, p. 35 (as “Sautrāntika” view); Jinendrabuddhi (PSṬ, p. 68). satsu
samartheṣu… sa bāhyo ’rthaḥ syāt: Kāśikā ad Ślokavārttika, Śūnyavāda 20 (Śāstrī 1927–1943, vol. 2,
p. 100), Jinendrabuddhi (PSṬ, p. 68).
Lawrence McCrea
255
discussed below. Rather than an argument for the non-reality of mind independent objects,
the sahopalambhaniyama argument is designed to establish only that the phenomenal
content of perception, its grasping and grasped aspects, are artificially abstracted parts
of a single cognitive image. Establishing this is a necessary prerequisite for defending a
vijñānavādin position, but does not require one to adopt such a position, as Dharmakīrti’s
own discussion in Pramāṇaviniścaya 1.58 explicitly shows.6
While the final kārikā of the Pramāṇaviniścaya chapter on perception thus presents us
with a viable avenue for the defense of the existence of extra-mental objects, Dharmakīrti
does not leave us with a definitive endorsement of this position over the vijñānavādin stance
developed earlier. This new sahopalambhaniyama-consistent argument for extra-mental
objects is immediately followed by a further remark that rehabilitates the vijñānavādin position but, again, as a viable alternative path, not as a decisive victor over the bāhyārthavādin
position:
Unless one can say that this disjunction of effect is produced by the absence
of a specific material cause [upādāna, i.e. the samanantara-pratyaya].7
The idealist might try to account for the occasional appearance of images such as blue in our
awareness by positing, instead of an extra-mental object, a specific (but unperceived) factor
in the immediately prior awareness (samanantara-pratyaya – which is the material cause of
any given awareness-moment) that causes us to see blue on some occasions and not others.
So, we are left with two possible ways of explaining the occasional appearance of specific
images in our perceptual awareness, both of which require us to postulate something not
itself phenomenally evident – either a putative extra-mental object, or an unperceived causal
factor in our prior awareness – whose presence or absence accounts for our perception
containing a given image at some times but not others. But there is, notably, no explicit
indication of whether either of these views is to be preferred over the other, nor even the
suggestion of an argument or a criterion that would enable one to choose between them.
Far from attempting to compel us to move “up” from the external realist to the epistemic
idealist level, Dharmakīrti seems almost to be going out of his way to display ambivalence:
there “can be proof” of the external object, “unless someone were able to say” that the
disjunction in effect is produced by a difference in the samanantara-pratyaya. He does not
say explicitly whether the object is proven in this way, or whether the vijñānavādin is or
is not able to explain the occasional appearance of the effect in this way. We are simply
left with two ways of accounting causally for the appearance of specific images in our
perceptual awareness, with no indication of how (or whether) we should decide between
them.
Having set forth this pair of parallel accounts, the Pratyakṣa chapter of the Pramāṇaviniścaya concludes with one further objection against and response in support of the
vijñānavādin view:
6
7
I am not in fact the first to take note of the fact the sahopalambha-niyama argument is consistent
with both the vijñānavādin and bāhyārthavādin positions as Dharmakīrti ultimately develops them.
Matsumoto (1980: 26) makes the same point and Alex Watson (2010: 311, n. 46) has recently drawn
attention to this point as well.
PVin ad 1.58cd (p. 43): yady atra kaścid upādānaviśeṣābhāvakṛtaṃ kāryavyatirekaṃ na bruyāt.
256
Balancing the Scales: Dharmakīrti Inside and Out
[Objection:] But how then could this person [i.e. the vijñānavādin], while
denying objects for all awarenesses, nevertheless say that a confused awareness is not a pramāṇa while an unconfused one is, since there would be no
difference between them? [Reply:] Seeing that, even for someone who is not
fully awakened, activity sometimes proves to be untrustworthy, due to the
fault of being connected with karmic traces of confusion, one can say that one
sort of awareness is not a pramāṇa; seeing that, in other cases, this activity
has dependable consequences which persist as long as saṃsāra does, being
based on firmly established karmic traces, one can say that this other sort
of awareness is a pramāṇa, in consideration of its non-disparity with [other]
worldly activity. What is stated here is the form of conventional pramāṇas.
Others are confused even about this, and draw people into disagreements.
Those who devote themselves to the wisdom consisting in thought (cintāmayī
prajñā), however, turn their attention toward the ultimate pramāṇa, which is
faultless through the discrimination of error and free from decay. This too is
slightly hinted at.8
This explanation of how pramāṇa and non-pramāṇa can be distinguished in an object-free
world does redress the opponent’s objection, and rehabilitates vijñānavāda, but only as a
possibility, not as a position to which we are compelled. It leaves in place the unresolved
ambiguity of Dharmakīrti’s prior treatment of the two as alternatives. Furthermore, it
rehabilitates the idealist position even as a possible stance only by bracketing both the
vijñānavādin and bāhyārthavādin positions as less than the ultimate truth. Again, at least
within the realm of “conventional pramāṇas,” we are left not with a decisive argument that
forces us to move from an external realist to an idealist stance (or vice versa); these two
basic ontological stances are both left on the table as Dharmakīrti ends his discussion of
perception.
Pramāṇavārttika, Pratyakṣa 391–398
The basic features of this ambiguous back and forth between the vijñānavādin and bāhyārthavādin positions in PVin 1.58 and its accompanying vṛtti are all in fact prefigured in
what is clearly the parallel passage in the Perception chapter of the Pramāṇavārttika:
Therefore it is unavoidable that the object, which appears at the same time
as the awareness, is not distinguished from the awareness. There could be an
inference of a difference in the cause [of one’s awareness],
8
PVin ad 1.58 (pp. 43–44): so ’pi kathaṃ sarvajñānānāṃ viṣayaṃ vyatirecayann upaplavetarayoḥ
pramāṇetaratāṃ brūyād, viśeṣābhāvāt? upaplavavāsanāvisandhidoṣād aprabuddhasyāpy anāśvāsikaṃ
vyavahāram utpaśyann ekam apramāṇam ācakṣīta, aparam āsaṃsāram aviśiṣṭānubandhaṃ dṛḍhavāsanatvād iha vyavahārāvisaṃvādāpekṣayā pramāṇām. sāṃvyavahārikasya caitat pramāṇasya rūpam
uktam. atrāpi pare mūḍhā visaṃvādayanti lokam iti. cintāmāyīm eva tu prajñām anuśīlayanto vibhramavivekanirmalam anapāyi pāramārthikapramāṇam abhimukhīkurvanti. tad api leśataḥ sūcitam
eveti.
Lawrence McCrea
257
due to the absence of [specific] perceptual awarenesses even when all the other
causes are present. Unless one can state a restriction based on the immediately
preceding awareness.9
Here too we have, immediately following an argument for the nondifferentiation of the
awareness from its object, the claim that one may nevertheless infer an extra-mental object
as a cause for the occasional occurrence of specific images in our awareness. And here
again we have the possibility of an unrecognized causal factor in our immediately prior
awareness suggested as an alternate explanation of these occasional images. The phrasing is
extremely close to that of Pramāṇaviniścaya 1.58, and again we have the same juxtaposed
optative construction: “There could be an inference… unless one can say…,” leaving
us with a similar ambiguity on the external object question. But there is one significant
difference between this passage and the (later) vṛtti on Pramāṇaviniścaya 1.58. There
Dharmakīrti presented the vijñānavādin and bāhyārthavādin as hypothetical alternative
stances without ultimately indicating a preference for one over the other. Here, on the other
hand, Dharmakīrti does at least indirectly indicate such a preference (PV, Pratyakṣa 398):10
This [vijñānamātra view] is the doctrine of the wise; but this dual nature of
awareness [as grāhya and grāhaka] is [also] established in reliance on an
external object, due to the rule of co-awareness.11
Labelling the vijñānavādin view as “the doctrine of the wise” does certainly suggest that
Dharmakīrti considers this view to be preferable (whether philosophically or soteriologically). But it is important to note that this offhand indication of preference does not come
coupled with any sort of argument that would justify it. Here too, as in Pramāṇaviniścaya
1.58, we are left with two philosophically viable accounts of perceptual images, one which
posits the reality of extra-mental objects, and one which denies them, but with no clear
argument for preferring one over the other.
If, then, as it seems, Dharmakīrti does not wish to present any compelling argument that
forces his readers to abandon the external realist position in favor of an idealist one, what is
he trying to do here? If, as the label “doctrine of the wise” would at least seem to suggest,
the vijñānavādin position is Dharmakīrti’s preferred view, why should he present it and the
parallel bāhyārthavādin account in such careful equipoise (as he later does, without a similar
indication of preference, in the Pramāṇaviniścaya)? Both available Sanskrit commentaries
on this passage of the Pramāṇavārttika address the question directly. Prajñākaragupta
comments on Pratyakṣa 398 as follows:
[Pūrvapakṣin:] Why then did the Teacher [Dignāga] explain that [the awareness has] a dual form even if the external object exists? Vijñānavāda alone is
the correct position. In response to this, he [Dharmakīrti] says:
9
10
11
PV, Pratyakṣa 391–392: tasmād arthasya durvāraṃ jñānakālāvabhāsinaḥ / jñānād avyatirekitvaṃ
hetubhedānumā bhavet // 391 abhāvād akṣabuddhīnāṃ satsv apy anyeṣu hetuṣu / niyamaṃ yadi na
brūyāt pratyayāt samanantarāt // 392
The intervening verses (PV, Pratyakṣa 393–397) parallel the objection against the vijñānavādin account
of prāmāṇya and the vijñānavādin response given in the vṛtti on PVin 1.58, but in more detail.
PV, Pratyakṣa 398: asty eṣa viduṣāṃ vādo bāhyaṃ tv āśritya varṇyate / dvairūpyaṃ sahasaṃvittiniyamāt
tac ca sidhyate // 398
258
Balancing the Scales: Dharmakīrti Inside and Out
This [vijñānamātra view] is the doctrine of the wise; but this dual nature
of awareness [as grāhya and grāhaka] is [also] established in reliance on
an external object, due to the rule of co-awareness.
The dual form [of awareness] is explained with provisional acceptance of the
external object, in consideration of others [who hold opposing views]. But
wise people accept vijñānavāda alone.12
Manorathanandin, whose comment on the verse closely mirrors that of Prajñākaragupta,
explains further:
This is the doctrine “of the wise,” i.e. the Yogācārins who understand reasoning, which explains that there is nothing but consciousness in all conditions.
The teacher [Dignāga] has [also] explained the duality of form of awareness
relying on the external object accepted by the Sautrāntikas. And this duality of
form is established “due to the rule of co-awareness,” i.e. due to the rule of
co-apprehension [sahopalambha-niyama] since, even if there is a difference
[between the awareness and the object], this [rule of co-apprehension] still
exists.13
As Prajñākaragupta and Manorathanandin both rightly point out, Dignāga in his investigation of the objects of perception in his Pramāṇasamuccaya (on which the Pramāṇavārttika
is ostensibly a commentary) similarly offers parallel vijñānavādin and bāhyārthavādin
accounts without indicating a preference for one over the other:
If the awareness together with its content is the object, then what one apprehends is an object in the form of one’s own self-awareness, which may be
either desirable or not. If, however, the thing to be known is an external object,
then
The pramāṇa for this is simply that it has the form of the object.
For then, setting aside the [awareness’ own] form, even though this is what is
cognized by the awareness, the pramāṇa for this [awareness] is simply that it
has the form of the object, since that object
is known through this. 9 (emphasis mine)14
12
13
14
Pramāṇavārttikabhāṣya ad Pratyakṣa 398 (PVA, p. 416): kathaṃ tarhi bāhye ’py arthe ācāryeṇa
dvirūpatā vijñānasyoktā? vijñānavāda eva yuktaḥ. atrocyate. asty eṣa viduṣām vādo bāhyaṃ tv āśritya varṇyate / dvairūpyaṃ sahasaṃvittiniyamāt tac ca sidhyati // 398 // bāhyārthābhyupagamena
parāpekṣayā dvirūpatā varṇyate, viduṣāṃ tu vijñānavāda eva.
Pramāṇavārttikavṛtti on Pratyakṣa 398 (PVV, p. 142): asty eṣa sarvavyavasthāsu vijñaptimātratāpratipādako viduṣāṃ nyāyadarśināṃ yogācārāṇāṃ vādaḥ. sautrāntikair iṣṭaṃ bāhyam artham āśritya
jñānasya dvairūpyam ācāryeṇa varṇyate. tac ca dvairūpyaṃ sahasaṃvedaniyamāt sahopalambhaniyamāt sidhyati bhede ’pi sati tadbhāvāt [corr. tadabhāvāt].
Pramāṇasamuccaya 1.9cd (PS, p. 4): yadā hi saviṣayaṃ jñānam arthaḥ, tadā svasaṃvedanānurūpam
arthaṃ pratipadyata iṣṭam aniṣṭaṃ vā. yadā tu bāhya evārthaḥ prameyaḥ, tadā viṣayābhāsataivāsya
pramāṇaṃ tadā hi jñānasvasaṃvedyam api svarūpam anapekṣyārthābhāsataivāsya pramāṇam. yasmāt
so ’rthaḥ tena mīyate // 9 //
Lawrence McCrea
259
The connection these commentators posit between Dharmakīrti’s practice and Dignāga’s
thus makes sense,15 though it simply pushes the basic question back one step further: It
is plausible to suggest that Dharmakīrti’s parallel presentation of the vijñānavāda and
bāhyārthavāda as viable alternatives is meant to conform to Dignāga’s practice, but this
does not explain why Dignāga himself would have adopted this approach in the first place. A
basic gesture of inclusiveness towards others in the Buddhist camp is presumably intended
by Prajñākaragupta’s “in consideration of others” (parāpekṣayā) and Manorathanandin’s
reference to “the external object accepted by the Sautrāntikas.” In any case, both clearly
acknowledge that both Dharmakīrti and Dignāga, while themselves personally inclined
toward the vijñānavādin stance, are seeking in these passages to accommodate (and precisely
thereby not to supersede) the bāhyārthavādin/Sautrāntika position.
Pramāṇavārttika, Pratyakṣa 223–224
I will close by briefly examining one further passage in which Dharmakīrti appears to
balance the vijñānavādin and bāhyārthavādin positions in this way: verses 194–224 of
the Pratyakṣa chapter of the Pramāṇavārttika. This passage deals with the problem of
citratā – “multicoloredness,” or more generally “multiplicity” of any kind in a cognitive
image. When we perceive a multicolored object, or any object that appears to have multiple
properties of any kind (such as a spatially extended object that appears to exist at more
than one point), how is it that we have an impression of the “oneness”? Does it exist only
in our minds, or is there some way that external, atomic objects can be aggregated into
what is in some sense a “single” object? This is an absolutely key passage for both Dunne’s
and Dreyfus’s analyses, as both see it as marking Dharmakīrti’s crucial transition from
the external realist to the idealist level of analysis (i.e. from the third to the fourth level in
Dunne’s “sliding scale”). It is with reference to this passage that Dunne says:
Finally, in moving from the third to the fourth level of analysis, the inability to
specify whether the image in perception is single or multiple is the primary
argument against the existence of extra-mental objects (Dunne 2004: 63).
In response to the intractable problem of the cognitive image’s variegated
singularity, Dharmakīrti abandons External Realism in favor of Epistemic
Idealism (Dunne 2004: 112).
Dreyfus similarly sees Dharmakīrti’s exploration of the problems posed by the multiformity
of perceptual images as setting the stage for a definitive transition from external realism to
idealism. As he says:
This denial of the reality of external objects is where Dharmakīrti finds a
solution to the dilemma created by an impression of a solid extended object
produced by atoms, which do not have any extension by themselves. The
15
The link between Dharmakīrti’s treatment of the external object question and Dignāga’s is reinforced
by Jinendrabuddhi, who, in explaining the version of the bāhyārthavādin view set forth by Dignāga in
the Pramāṇasamuccaya, quotes PVin 1.58cd and closely paraphrases the following vṛtti (PS, p. 68).
260
Balancing the Scales: Dharmakīrti Inside and Out
problem is solved by rooting out its source, the assumption that objects exist
external to consciousness as a result of atomic aggregation. The Yogācāra
view that objects exist only as reflections of consciousness is Dharmakīrti’s
answer to the problem created by extended objects (Dreyfus 1997: 103).
The problem with both Dunne’s and Dreyfus’s reading of this passage of the Pramāṇavārttika is that here too, as in Pramāṇaviniścaya 1.58, we find, as a coda to what they
take to be Dharmakīrti’s conclusive argument driving the shift from the “external realist”
to the “epistemic idealist” stance, a return to the defense of the external realist (or, at
any rate, an external realist) position. Dharmakīrti’s response to the problem of citratā
(Pramāṇavārttika, Pratyakṣa 209–224) is too complex to analyze in detail here. He offers in
fact there three different ways to account for multiform appearance in awareness, apparently
presented as parallel, alternative lines of argument: one which denies that multiformity
can really exist either externally or within awareness (209–219), a second that accepts the
existence of multiformity in cognitive images (ākāras) but denies that it can exist externally
(220–222), and a third that, accepting the stricture that multiformity can exist only in
awareness, nevertheless asserts that multiple, real external objects can give rise to a single,
multiform awareness (223–224).16 Obviously it is the third position here that is of most
interest for my argument, as it represents yet another retrospective rehabilitation of the
external realist position. The relevant verses run as follows:
On the other hand, what contradiction is there if many [atoms] which have this
special causal capacity [collectively] do not separately cause the awareness,
just as is the case with the self, the sense organ, and so on.17
To be “grasped” is nothing other than to be a cause. When the awareness has
the same appearance as something [that causes it], one says that this thing is
grasped by the awareness.18
16
17
18
The commentator Prajñākaragupta says that these three alternative lines of argument are offered
“in accordance with the reasoning of the Mādhyamikas, the Yogācāras, and the Bāhyārthavādins”
(mādhyamikayogācārabāhyārthavādinayena – PVA, p. 288). Steinkellner, with good reason, questions
Prajñākaragupta’s use of the label “Mādhyamika” here (1990: 76–78 and p. 86, n. 35). The first
position seems closer to what later becomes known as the nirākāravādin position (that cognitions
are in reality “without form;” or, in Dreyfus’s terms, the “False Aspectarian” position) – Dreyfus in
fact straightforwardly labels it as such, plainly recognizing the divide between the nirākāravādin view
argued for in Pramāṇavārttika, Pratyakṣa 210–219 and the sākāravādin view advanced in 220–222
(but not, as far as I can see, taking heed of the third, bāhyārthavādin, position on offer in 223–224 – see
Dreyfus 1997: 103–104 and 490).
On the Nyāya view of perception a group of factors – self, sense organ, mind, object, etc. – are said
to collectively cause perceptual awareness, while any of these factors individually will not do so. In
the same way, there is no contradiction in asserting that multiple “blue” atoms in appropriate spatial
proximity can collectively cause an awareness of blue, even though no single atom could cause such an
awareness by itself.
Pramāṇavārttika, Pratyakṣa 223–224: ko vā virodho bahavaḥ saṃjātātiśayāḥ pṛthak / bhaveyuḥ kāraṇaṃ buddher yadi nātmendriyādivat // 223 hetubhāvād ṛte nānyā grāhyatā nāma kācana / tatra
buddhir yadābhāsā tasyās tad grāhyam ucyate // 224 Dunne does recognize that these verses form
part of Dharmakīrti’s discussion of the citratā problem, and translates them (2004: 411), but does not
appear to consider their implications for his “sliding scale” analysis.
Lawrence McCrea
261
That Dharmakīrti is here transitioning to an alternative line of argument is clearly marked in
the text (“On the other hand”) and, as in Pramāṇaviniścaya 1.58, there is nothing to indicate
whether this line of argument or the one that precedes it is to be preferred for any reason. On
Dharmakīrti’s own explicit analysis, then, the problem of citratā explored in this passage
does not compel us, or even forcefully nudge us, toward an idealist stance. He offers the
reader answers to the citratā problem that work just as well whether one accepts external
objects or rejects them. It is true that the alternative resolution of the citratā question
presented in the final two verses of this passage only opens the door to bāhyārthavāda
as a possibility; the claim is that multiple atoms can act as external causes for a single
multicolored mental image, not that such external causes must exist. Nevertheless, taken
seriously, these verses undo much of the supposed work done by the earlier portion of the
passage, on both Dunne’s and Dreyfus’s accounts. The external realist position is certainly
not definitively abandoned here, and the move from realism to idealism cannot reasonably
be described as “Dharmakīrti’s answer” to the citratā problem, but (at best) one of several
possible answers.
Conclusion
In the light of the treatment of the bāhyārthavādin position in all of the passages discussed
above, it seems difficult to maintain that Dharmakīrti was seeking in his major epistemological works to push readers to ultimately adopt a vijñānavādin stance. Whatever his personal
position on the status of extra-mental objects, he seems deliberately to have written in
such a way that either vijñānavādin or bāhyārthavādin readers could comfortably adopt his
epistemology. He in fact developed what came to be some of the most influential arguments
in support of Buddhist bāhyārthavāda. This helps to explain why later doxographers, when
seeking to delineate the “Sautrāntika” view, so often turned precisely to Dharmakīrti’s
statements, and specifically to PVin 1.58, as the classic textual formulation of this position. The specific variety of “external realism” set forth in PVin 1.58 is not a preexisting
philosophical position that Dharmakīrti (or Dignāga) adopted as a heuristic device or a
stepping stone to draw readers along to his “real” idealist position; it is, rather, explicitly
presented as an outgrowth of his own analysis of the phenomenology of perception. It is
not only that one can be a “Sautrāntika” in the sense recognized by the later doxographers
and still be a Dharmakīrtian. Rather, one can be a Sautrāntika, in the doxographers’ sense
of this term, only if one is a Dharmakīrtian. That is, the externalist argument as set forth in
Pramāṇaviniścaya 1.58 and, more briefly, in Pramāṇavārttika, Pratyakṣa 223–224 and
398 grows specifically out of the analysis of the dvirūpatā of all perceptual cognitions – the
artificial division what is really a unitary cognitive image into a “grasping” and “grasped”
form (grāhaka and grāhya ākāra) – and fully presupposes the validity of the arguments on
which this analysis is based. That is, the epistemology and phenomenology of perception
elaborated by both Dharmakīrti and Dignāga is entirely shared between the vijñānavādin
and the bāhyārthavādin position as they elaborate them.
Arnold is entirely right to stress this in his discussion of Dunne’s “sliding scale.”19
Whether he is right in suggesting that such a phenomenology (which takes the only “object”
19
See Arnold 2008, especially pp. 5–6.
262
Balancing the Scales: Dharmakīrti Inside and Out
directly accessible to awareness to be its own image) tends by its nature to force one into
an idealist rather than a bāhyārthavādin position is less obvious. This is in fact precisely
the question at issue between the bāhyārthavādin and the vijñānavādin as Dharmakīrti
finally represents them: given that we have immediate cognitive access only to the ākāras
of our own awareness, can we or can we not legitimately infer an extra-mental cause
for the occasional appearance of particular ākāras? Obviously if Arnold were right in
this claim, and if Dharmakīrti held a similar view of the question, this would tend to
support Dunne’s model of the sliding scale, as it would leave us with a Sautrāntika position
deliberately designed to collapse into a Yogācāra one. But, as I have already said, this makes
Dharmakīrti’s regular return to defense of the bāhyārthavādin position after presenting his
seemingly ultimate vijñānavādin arguments hard to explain.
So, the same general pattern recurs in Dharmakīrti’s works, seemingly whenever he
presents his ultimate versions of the vijñānavāda and bāhyārthavāda positions in explicit
contrast to one another. We first find an argument for how the phenomenology and epistemology of perception function on the vijñānavāda view (in one case only, PV, Pratyakṣa
398, accompanied by a fairly clear suggestion that this is Dharmakīrti’s preferred view), followed by a defense of the bāhyārtha position, adopting precisely the same phenomenology
of perception, but taking it to imply the existence of an extra-mental cause for the images
that appear in our awareness.
Is Dharmakīrti, then, contradicting himself? Is there anything problematic in developing
a set of views regarding perception that accommodates and accords with two conflicting
ontologies? Must we, in charity, seek to forcibly extract from his works one definitive stance
on the external object question, even when he himself seems to be trying carefully to avoid
doing so? I think Dreyfus himself makes a very important observation, when considering
the question of Dharmakīrti’s possible self-contradiction in his seemingly inconsistent
position on bāhyārthavāda. To quote:
Suffice to say that I take Dharmakīrti’s essential preoccupation to be epistemology not metaphysics. His interest in defining the nature of reality is to
ontologically ground his epistemology… The exact nature of real entities
(whether, for example, they are ultimately mind independent or not) is a lesser
concern (Dreyfus 1997: 105).
I think this quite nicely sums up and makes sense of Dharmakīrti’s method here. The
real matter Dharmakīrti wishes to stress, and does stress, is precisely the account of the
epistemology and phenomenology of perception that is common to his final versions of both
the vijñānavādin and the bāhyārthavādin positions. This is what he argues for consistently,
forcefully, and without ambiguity.
An interesting side-question here is whether, and if so how, Dharmakīrti sees the final
bāhyārthavādin argument in PVin 1.58 as escaping the earlier samanantara-pratyaya
argument already discussed by Kellner. She notes that this argument is directed against
what she describes as “the theory that Dharmakīrti first adopts and then abandons: that
external objects produce a perception which has their form (ākāra), or which resembles
them” (Kellner 2011: 294 – emphasis mine). The samanantara-pratyaya argument basically
hinges on the problem of discriminating exactly what part of the causal complex producing
Lawrence McCrea
263
an awareness would count as the “object.” If one defines the object as that which both
causes an awareness and bears the same or a similar form, then one will in many cases
have to consider the samanantara-pratyaya, the awareness immediately preceding the one
in question, to be its object; if one has a continuous awareness of, say, blue, then any later
moment of this stream will have as part of its causal complex its samanantara-pratyaya,
which also contains a blue image, and hence will meet the test for being an “object.”
The consequent inability to define an “object” in a way that distinguishes it from the
samanantara-pratyaya is presented as a crippling flaw in this version of the external realist
theory. In connection with this, it is interesting to note that the positive arguments for the
bāhyārtha position presented in Pramāṇaviniścaya 1.58 (after the samanantara-pratyaya
argument has been presented) and in the parallel passage of the Pramāṇavārttika make no
reference to any resemblance of form between the inferred extra-mental object and the image
contained in the cognition it produces. In PV, Pratyakṣa 224, Dharmakīrti still refers to the
awareness having the appearance (ābhāsa) of the aggregated atoms which are “grasped”
by it, and it seems that this position should therefore fall prey to the samanantara-pratyaya
argument. But the later treatments of the bāhyārtha position say nothing about a shared
image or appearance between the atomic external objects and the macroscopic image they
produce in our awareness (which, perhaps intentionally, renders the samanantara-pratyaya
argument moot). Instead, the external object is postulated purely as an extra-mental cause
which explains the appearance of an image such as “blue” at some times but not at others.
This is then something like a Kantian Ding an sich, the “thing in itself,” something we
can know of as the cause of particular manifestations in our own awareness, but of which
we can know nothing more that it causes these manifestations. On this understanding blue
atoms would be “blue” only in the sense that they cause in us an awareness containing a
blue image. There is no sense in which the atoms can be thought of as “blue” in and of
themselves, when not producing an awareness, and consequently there could be no question
or resemblance between the atoms and the image they produce in our awareness.
The avoidance of this question of resemblance in the presentation of the Sautrāntika
position in Pramāṇaviniścaya 1.58 and the parallel portion of Pramāṇavārttika may or
may not represent a deliberate departure from the discussion in Perception 223–224 in
the Pramāṇavārttika, but in any case we can still see a basic continuity of approach to
the question of external objects in both of Dharmakīrti’s major works, and for that matter
in Dignāga’s Pramāṇasamuccaya as well. In all of these works, as in the Nyāyabindu, it
seems that, even though Dignāga clearly held vijñānavādin views, and Dharmakīrti seems
to indicate (at least in Pramāṇavārttika, Pratyakṣa 398) that this was his preferred position
as well, both authors sought to produce an account of Buddhist epistemology that was
consistent with either a vijñānavādin or a bāhyārthavādin stance, and hence could be
adopted by people of varied ontological persuasions.
There is a larger context to the claim I am making here about Dharmakīrti’s treatment
of the Sautrāntika or bāhyārthavādin view in the Pramāṇavārttika and Pramāṇaviniścaya.
As I have argued elsewhere,20 there is actually quite strong evidence to indicate that some
followers of Dharmakīrti were in fact bāhyārthavādins, most notably the highly influential
commentator Dharmottara. It was precisely Dharmakīrti’s consistently guarded and careful
20
See McCrea and Patil 2006: 332–333; McCrea and Patil 2010: 19–20, 139, 142–143; and McCrea 2016.
264
Balancing the Scales: Dharmakīrti Inside and Out
balancing of the vijñānavādin and bāhyārthavādin positions that made it possible for his
followers to adopt either of these ontological stances while still presenting themselves as
loyal adherents of the Dharmakīrtian tradition.
In the light of Dharmakīrti’s own seemingly ambivalent treatment of the status of
mind-independent objects in his major epistemological works, a full understanding of the
Buddhist pramāṇa tradition as it developed in India in the wake of Dharmakīrti’s work
would seem to call for a significant reevaluation of the role and status of the bāhyārthavādin
position within it.
References and abbreviations
Arnold 2008 D. Arnold, Buddhist Idealism, Epistemic and Otherwise: Thoughts on the
Alternating Perspectives of Dharmakīrti. Sophia 47 (2008) 3–28.
Dharmottarapradīpa Durvekamiśra, Dharmottarapradīpa, ed. D. Malvania. Patna 2 1971.
Dreyfus 1997 G. Dreyfus, Recognizing Reality: Dharmakīrti’s Philosophy and its Tibetan
Interpretations. Albany 1997.
Dunne 2004 J. Dunne, Foundations of Dharmakīrti’s Philosophy. Boston 2004.
Eltschinger 2010 V. Eltschinger, Dharmakīrti. Revue Internationale de Philosophie 253
(2010) 397–440.
Eltschinger 2014 V. Eltschinger, Buddhist Epistemology as Apologetics. Vienna 2014.
ĪPVV Abhinavagupta, Īśvarapratyabhijñāvivṛtivimarśinī, ed. M. Kaul Shastri. 3 vols. Bombay 1938–1943.
Kāśikā Sucaritamiśra, Kāśikā. See Śāstrī 1927–1943.
Kellner 2011 B. Kellner, Dharmakīrti’s Criticism of External Realism and the Sliding
Scale of Analysis. In: Religion and Logic in Buddhist Philosophical Analysis. Proceedings of the Fourth International Dharmakīrti Conference, ed. Helmut Krasser,
Horst Lasic, Eli Franco, and Birgit Kellner. Wien 2011, 291–298.
Matsumoto 1980 S. Matsumoto, Sahôpalambha-Niyama. Sōtōshū Kenkyūin Kenkyū Kiyō
(Journal of Sōtō Sect Research Fellows) 12 (1980) 1–34.
McClintock 2003 S. McClintock, The Role of the “Given” in the Classification of Śāntarakṣita and Kamalaśīla as Svātantrika-Mādhyamikas. In: The Svātantrika-Prāsaṅgika
Distinction: What Difference Does a Difference make?, ed. G. Dreyfus and S. McClintock. Boston 2003, 125–171.
McCrea 2016 L. McCrea, Abhinavagupta as Intellectual Historian of Buddhism. In:
Around Abhinavagupta: Aspects of the Intellectual History of Kashmir from the Ninth
to the Eleventh Century, ed. Eli Franco and Isabelle Ratié. Berlin 2016, 263–286.
McCrea and Patil 2006 L. McCrea and P. Patil, Traditionalism and Innovation: Philosophy, Exegesis, and Intellectual History in Jñānaśrīmitra’s Apohaprakaraṇa. Journal
of Indian Philosophy 34 (2006) 303–366.
McCrea and Patil 2010 L. McCrea and P. Patil, Buddhist Philosophy of Language in
India: Jñānaśrīmitra on Exclusion. New York 2010.
Nyāyabinduṭīkā Vinītadeva’s Nyāyabindu-ṭīkā: Sanskrit original reconstructed from the
extant Tibetan version, with English translation and annotations, tr. Mrinalkanti
Gangopadhyaya. Calcutta 1971.
Lawrence McCrea
265
NM Jayantabhaṭṭa, Nyāyamañjarī, ed. K. Varadacharya. 2 vols. Mysore 1969–1983.
PS Dignāga’s Pramāṇasamuccaya: A hypothetical reconstruction of the Sanskrit text with
the help of the two Tibetan translations on the basis of the hitherto known Sanskrit
fragments and the linguistic materials gained from Jinendrabuddhi’s Ṭīkā, ed. E.
Steinkellner. Wien 2005. https://www.oeaw.ac.at/fileadmin/Institute/IKGA/PDF/fo
rschung/buddhismuskunde/dignaga_PS_1.pdf, last visited 13-08-2019.
PSṬ Jinendrabuddhi, Viśālāmalavatī Pramāṇasamuccayaṭīkā, Chapter 1, Part 1: Critical
Edition, ed. E. Steinkellner, H. Krasser, and H. Lasic. Beijing/Vienna 2005.
PV Dharmakīrti, Pramāṇavārttika, ed. Ram Chandra Pandeya. Delhi 1989.
PVA Prajñākaragupta, Pramāṇavārttikabhāṣyam or Varttikālaṅkāra, ed. Rahula Sankrityayana. Patna 1943.
PVin Dharmakīrti’s Pramāṇaviniścaya: Chapters 1 and 2, ed. E. Steinkellner. Beijing/
Vienna 2007.
PVV Manorathanandin, Pramāṇavārttikavṛtti. In PV.
Śāstrī 1927–1943 Mīmāṃsāślokavārttika with the Commentary Kāśikā of Sucaritamiśra,
ed. K. Sāmbaśiva Śāstrī. 3 vols. Trivandrum 1927–1943.
Ślokavārttika See Śāstrī 1927–1943.
Steinkellner 1990 E. Steinkellner, Is Dharmakīrti a Mādhyamika. In: Earliest Buddhism
and Madhyamaka, ed. D.S. Ruegg and L. Schmithausen. Leiden 1990, 72–90.
TBh Tarkabhāṣā of Mokṣākaragupta, ed. E. Krishnamacharya. Baroda 1942.
Watson 2010 A. Watson, Bhaṭṭa Rāmakaṇṭha’s Elaboration of Self-Awareness (svasaṃvedana), and How it Differs from Dharmakīrti’s Exposition of the Concept. Journal of
Indian Philosophy 38 (2010) 297–321.
The Concept of vyava√sthā in Dharmottara’s and
Prajñākaragupta’s Explanation of pramāṇa and
pramāṇaphala
by
Mai Miyo
Introduction
My paper will examine the concept of vyava√sthā (“to differentiate, establish, fix”) in
the context of the theory of the non-distinction between pramāṇa and pramāṇaphala.
This theory is characteristic of the Buddhist logico-epistemological school and criticized
mainly by Brahmanical philosophers. Dharmakīrti expounds the relevant notions in the
chapters on perception (pratyakṣa) of his Pramāṇavārttika (PV III), Pramāṇaviniścaya
(PVin I) and Nyāyabindu (NB I), more specifically, at PV III 301–319, at PVin I 34–37
with prose,1 and NB I 18–21. My focus, however, are the divergent interpretations of
vyava√sthā given by Dharmottara and Prajñākaragupta. Firstly, by way of an introduction
I shall briefly present Dharmakīrti’s theory of the non-distinction between pramāṇa and
pramāṇaphala, how the concept of vyava√sthā is introduced in this context, and how it was
interpreted by his followers. Secondly, I will investigate the interpretations of Dharmottara
and Prajñākaragupta, which differ from each other in quite obvious ways; this confirms the
conclusions of other scholars that these thinkers are adversaries on many topics.2 The central
point of their controversy in the present context is whether or not the function of vyava√sthā
involves ascertainment (niścaya) or conceptual construction (vikalpa). Dharmottara affirms
this, while Prajñākaragupta denies it.
1
2
I would like to express my thanks to Prof. Takashi Iwata, Prof. Masahiro Inami, and the members of the
Prajñākaragupta Study Group for their valuable comments on my ideas, to Prof. Birgit Kellner who gave
me crucial suggestions, and to Allison Aitken who kindly corrected my English. Moreover, I would like to
express my sincere regret at Dr. Helmut Krasser’s passing. His discussion of Dharmottara’s interpretation
of perception and conceptual construction, in his great study on the Laghuprāmāṇyaparīkṣā, inspired
me in many ways.
For an edition of the Sanskrit text of PV III and a critically annotated Japanese translation, see Tosaki
1979: 394–413. For German and Japanese translations of PVin I based on the Tibetan translations,
see Vetter 1966: 79–87 and Tosaki 1991, respectively. An edition of the Sanskrit text of the first two
chapters of PVin was made by Ernst Steinkellner (see PVin I, II).
See, above all, Ono 1995.
Birgit Kellner et al., eds., Reverberations of Dharmakīrti’s Philosophy: Proceedings of the Fifth International Dharmakīrti
Conference Heidelberg, August 26 to 30, 2014. Vienna 2020, pp. 267–281.
268
The Concept of vyava√sthā in Dharmottara and Prajñākaragupta
1. Dharmakīrti’s theory of the non-distinction between pramāṇa and
pramāṇaphala
1.1 The meaning of the terms pramāṇa and pramāṇaphala
As is well known, the term pramāṇa has two main meanings in epistemological discourse.3
The first meaning is “means of valid cognition,” according to the nirvacana-analysis4 of
the term as deriving from the verb pra√mā (“to cognize”), and formed with the suffix -ana
in the meaning of an instrument (karaṇa). The second meaning is “valid cognition,” that is,
not the means of valid cognition, but the cognition itself. In this respect pramāṇa is used
synonymously with samyagjñāna.5 We can detect this second meaning in the so-called
first definition of pramāṇa in the first verse of the pramāṇasiddhi chapter of PV (PV II),
according to which pramāṇa is non-belying cognition (avisaṃvādi jñānam).6 The same
meaning is also at work in the definitions of perception and the two kinds of inference
(i.e., svārthānumāna and parārthānumāna), which are the subcategories of pramāṇa.7
Dignāga attempts to resolve the confusion arising from this dual meaning by arguing that
the valid cognition is metaphorically designated, by way of upacāra, as the means of
valid cognition.8 Dharmakīrti seems to use the term pramāṇa in a broader sense which
encompasses both meanings. Regardless, we can ascertain that in the cases where the two
terms pramāṇa and pramāṇaphala are used as a pair pramāṇa means “means of valid
cognition,” and pramāṇaphala means “valid cognition.”
The word pramāṇaphala is normally analyzed as a genitive tatpuruṣa, i.e. as pramāṇasya phalam. Given that pramāṇa here refers specifically to a means of valid cognition,
3
4
5
6
7
8
Yoshida 2011 also treats this subject.
I use the term nirvacana-analysis according to Kahrs 1998. Nirvacana is the analysis of a word in
accordance with its meaning, and one of the fundamental features of it is that all nouns are regarded as
related to an activity expressed in language by a verbal form. Although “nirvacana” is often rendered
as “etymological analysis,” it is indeed not simply etymological in the sense that it does not reflect the
findings of historical linguistics which probably is how we understand “etymology” today. See Kahrs
1998: 23, 27, and 35.
Cf. PVin I 1,8–10, including PVin I 1a (printed in bold): tad dvividhaṃ samyagjñānaṃ pratyakṣam
anumānaṃ ceti. It is worth noting that Dharmakīrti explicitly uses the word samyagjñāna in the NB,
whereas in the PV and PVin he mainly uses pramāṇa to convey the same meaning. It is possible that he
tried to avoid confusing the novice, intended reader of the NB with regard to the two meanings of the
word.
“non-belying cognition” and “illumination of a [previously] unapprehended object” (ajñātārthaprakāśa)
in PV II 5 are often assumed to be two definitions of pramāṇa especially according to Devendrabuddhi’s
commentary, but Dharmakīrti himself does not declare them to be formal definitions. In addition,
commentators have different opinions on how these two definitions relate to each other. For this topic,
see Franco 1997: 45–66, Krasser 2001, and Ono 2012: 166–178.
See PVin I 7,3–4: timirāśubhramaṇanauyānasaṅkṣobhādyanāhitavibhramam avikalpakaṃ jñānaṃ pratyakṣam. PVin II 46,3: trilakṣaṇāl liṅgād yad anumeye ’rthe jñānam, tat svārtham anumānam. PVin III
1,2–4: yathaiva hi svayaṃ trirūpāl liṅgāl liṅgini jñānam utpannam, tathā paratra liṅgijñānotpādayiṣayā
trirūpaliṅgākhyānaṃ parārtham anumānam, kāraṇe kāryopacārāt.
See PS(V) I 3,22–4,1: savyāpārapratītatvāt pramāṇaṃ phalam eva sat // (PS I 8cd) na hy atra bāhyakānām iva pramāṇād arthāntaraṃ phalam. tasyaiva tu phalabhūtasya jñānasya viṣayākāratayotpattyā
savyāpārapratītiḥ. tām upādāya pramāṇatvam upacaryate nirvyāpāram api sat. For English translations, see Hattori 1968: 28, and (partially) Kellner 2010: 219.
Mai Miyo
269
pramāṇaphala literally translates as “the result of a means of valid cognition,”9 and by
consequence refers to the valid cognition itself, as a resulting cognition. Dharmakīrti does
not distinguish the action of cognizing (i.e. the action of knowing) from the resulting
cognition (i.e., knowledge); rather, he denies the action of cognizing which has active sense
and argues that only a resulting momentary cognition arises. This becomes most evident in
PV III 307cd–308, the passage that is central for the concept of vyava√sthā:
dadhānaṃ tac ca tām ātmany arthādhigamanātmanā //
savyāpāram ivābhāti vyāpāreṇa svakarmaṇi /
tadvaśāt tadvyavasthānād akārakam api svayam //
Moreover, the [cognition] which possesses the [property of having the form of
the object] (meyarūpatā) in itself appears as if it performs an activity toward
its object through the activity whose nature is the understanding of the object,
because the [understanding of the object] is differentiated (vyavasthāna) by
the [property of having the form of the object] even if [the cognition] does not
act by itself.
In PV III 307–309, Dharmakīrti mainly expounds Dignāga’s point in PS I 8cd and the
prose thereon, using the same metaphor, but he also adds the new idea of vyava√sthā to
characterize the relationship between pramāṇa and pramāṇaphala.
1.2 Specific referents of pramāṇa and pramāṇaphala
What are the specific referents of pramāṇa and pramāṇaphala? The theory of the means
and the result of valid cognition is not only applicable to perception, but also in the case
of inferential cognition (anumāna),10 but we shall limit the discussion to perception here.
Moreover, we may also confine ourselves to how Dharmakīrti explains means and result
on the assumption of an external object (bāhyārtha). According to Dharmakīrti, we do
not perceive external objects directly. Rather, the object projects its own form onto the
cognition and we perceive the form or image of the cognition.11 Dharmakīrti changes his
position from this type of representationalism (Sautrāntika) to idealism (Yogācāra) in the
latter part of the pratyakṣa chapters of the PV and PVin, according to a “sliding/ascending
scale of analysis,”12 where self-awareness (svasaṃvedana) plays a key role. The resulting
complications are, however, outside the scope of this paper.13 In Dharmakīrti’s representa9
10
11
12
13
For example, Katsura 2007: 409 and Moriyama 2010: 263 use this translation.
See NB II 4: pramāṇaphalavyavasthātrāpi pratyakṣavat // (“The distinction between a means of valid
cognition and [its] result is here too [i.e., in the case of inference for oneself] the same as in the case
of perception.”) PVin II 46,4: pratyakṣavad asya phalavikalpo vijñeyaḥ (“The differentiation of the
result of this [i.e., inference for oneself] should be known just like in the case of perception.”) The latter
passage is also translated in Steinkellner 1979: 26.
PV III 247: bhinnakālaṃ kathaṃ grāhyam iti ced grāhyatāṃ viduḥ / hetutvam eva yuktijñā jñānākārārpaṇakṣamam // (“[Objection:] How could [an object] which exists at a different time [from its
cognition] be grasped [by that cognition]? [Answer:] People who know reasoning understand that to be
grasped is to be a cause which is able to project [its own] form onto the cognition.”) Also translated in
Tosaki 1979: 346.
For this issue, see Dreyfus 1997: 83, Kellner 2011, and others.
We may leave the details to Murakami 2008, Kobayashi 2009, Kataoka 2011, Kellner 2011, and others.
270
The Concept of vyava√sthā in Dharmottara and Prajñākaragupta
tionalist theory, pramāṇa refers to the cognition’s property of having the form of the object
(jñānasya meyarūpatā),14 or its similarity to the object (arthasārūpya),15 and pramāṇaphala refers to the cognition’s function of understanding the object (prameyādhigati16 or
arthapratīti).17
1.3 The relationship between pramāṇa and pramāṇaphala
Both meyarūpatā and prameyādhigati belong to one and the same perceptual cognition.
Accordingly, they are distinguished only as properties (dharma) or aspects of the cognition,
and not as real entities (vastu).18 This is the main point of Dharmakīrti’s theory of the
non-distinction between pramāṇa and pramāṇaphala.
There are many ways in which the relationship between pramāṇa and pramāṇaphala is
characterized in Dharmakīrti’s works. Most prominently, they are referred to as sādhana
and sādhya, what establishes and what is established, and as karaṇa and kriyā, instrument
and function. Moreover, Dharmakīrti does not limit himself to the general understanding
of pramāṇa given by Dignāga, which does not go beyond the commonly used grammatical
derivation.19 Instead, Dharmakīrti maintains that pramāṇa is something which “differentiates” (√bhid, vi√bhaj, vyava√sthā) pramāṇaphala. This means that a cognition of an object
like blue or yellow is differentiated by the cognition’s own property of having the form of
the object, e.g., blue or yellow.
The differentiating function of pramāṇa is stressed in various places in PV and PVin:
“[only the factor] which is uninterrupted (antyaṃ) toward and differentiating (bhedakam)
for the [function (kriyā)] is regarded as the most effective establishing factor (sādhakatama,
i.e., pramāṇa),”20 “[In a cognition,] there should be its own nature by which [the cognition]
14
15
16
17
18
19
20
See PV III 306ab (= PV I 35ab): tasmāt prameyādhigateḥ sādhanaṃ meyarūpatā // (“Therefore, what
establishes the understanding of the object of valid cognition is [the cognition’s] property of having the
form of the object.”)
See NB I 20: arthasārūpyam asya pramāṇam // (“The means of valid cognition is the [cognition’s]
similarity to the object.”) PVin I 31,10–11: na ceyam arthaghaṭanārthasārūpyād anyato jñānasya
sambhavati (“And this connection of the cognition to the object cannot be performed by anything
other than its similarity to the object.”) Dharmakīrti demonstrates why meyarūpatā is pramāṇa more
elaborately in this part of the PVin than in the PV. See Miyo 2008.
See n. 14.
See NB I 18–19: tad eva ca pratyakṣaṃ jñānaṃ pramāṇaphalam // arthapratītirūpatvāt // (“And this
perceptual cognition itself is the result of the means of valid cognition because it has the property of
understanding the object.”)
See PV III 318: kriyākaraṇayor aikyavirodha iti ced asat / dharmabhedābhyupagamād vastv abhinnam
itīṣyate // (“[Objection:] It is contradictory that action and instrument are one. [Reply:] [That is] not
true because a distinction of properties is admitted. We assent that the real entity is not distinguished.”)
See PS(V) I 4,11–12: ... arthābhāsataivāsya pramāṇam. yasmāt so ’rthaḥ tena mīyate (9d’) (“... the
means of valid cognition is exactly its [i.e., the cognition’s] having the appearance of the object, for the
object is cognized by means of it [i.e., cognition’s having the appearance of the object].”) The same
kind of interpretation is found in Vātsyāyana’s Nyāyabhāṣya 91,2–3 on NS 1.1.3: upalabdhisādhanāni
pramāṇāni samākhyānirvacanasāmarthyād boddhavyam. pramīyate ’neneti karaṇārthābhidhāno hi
pramāṇaśabdaḥ (“It should be known that, by a nirvacana-explanation of the term, pramāṇas are what
establishes cognition, because the word pramāṇa expresses the meaning of instrument by means of
which [the object] is cognized.”)
PV III 311cd: yad antyaṃ bhedakaṃ tasyās tat sādhakatamaṃ matam //
Mai Miyo
271
is differentiated (vibhajyate) according to the object,”21 or “the understanding of the object
(arthādhigama as pramāṇaphala) is differentiated (vyavasthāna) by the property of having
the form of the object (meyarūpatā as pramāṇa).”22
1.4 Vyava√sthā in the relationship between pramāṇa and pramāṇaphala
Commentators on Dharmakīrti’s works use words derived from vyava√sthā to interpret
his definition of pramāṇa, referring to pramāṇa as differentiator (vyavasthāpaka) and
pramāṇaphala as that which is differentiated (vyavasthāpya).23 The relationship between
differentiator and differentiated (vyavasthāpyavyavasthāpakabhāva) is contrasted with the
causal relationship between producer and product (janyajanakabhāva) as which certain
opponents analyze the relationship between pramāṇa and pramāṇaphala. For example, the
contact between the senses and the object (indriyārthasannikarṣa) and the cognition of
the object (arthavijñāna) have such a causal relationship, and Naiyāyikas assume them
to be related as pramāṇa and pramāṇaphala. Dharmakīrti’s successors attempt to justify
the theory of non-distinction by keeping these two relationships clearly separate. If one
supposes pramāṇa and pramāṇaphala to be related as producer and product, then the two
factors must be distinct from each other, because producer and product are two different
entities occupying different moments in time; they are non-contemporaneous. But if it is
accepted that pramāṇa and pramāṇaphala are related as vyavasthāpaka and vyavasthāpya,
so that the means differentiates the result and, at the same time, is the result’s own nature,
then both belong to one and the same cognition, and they cannot be distinguished as real
entities. Of course, they are then also simultaneous.
Dharmottara is thought to be the most authoritative commentator on these two relationships by modern scholars, but their distinction is also found in other commentaries, such
as Manorathanandin’s Pramāṇavārttikavṛtti.24 Moreover, Dharmottara’s remarks in the
Nyāyabinduṭīkā (NBṬ) are actually not as simple as one might expect. As Oki 1993 pointed
out, his explanation includes not only the two aspects of perceptual cognition, but also, as
a third factor, an ascertaining cognition (niścayapratyaya) that arises immediately after the
perceptual cognition. Moreover, as we shall see, Prajñākaragupta criticizes Dharmottara’s
view in his Pramāṇavārttikālaṅkāra (PVA). It is to this controversy that we now turn.
21
22
23
24
PV III 302cd: bhāvyaṃ tenātmanā yena pratikarma vibhajyate //
PV III 308c: tadvaśāt tadvyavasthānād ... //
For example, Tosaki 1979: 397 abstracts this understanding from the Nyāyabinduṭīkā.
See PVV 213,15–16 on PV III 315: nāsty atra kāryakāraṇatātmakaḥ (PVVMS ; kāryakāraṇātmakaḥ
PVV) kriyākaraṇabhāvaḥ, kiṃ tu vyavasthāpyavyavasthāpakabhāvaḥ (“In this case, the relationship
between instrument and action does not have the nature of the relationship between producer and
product, but is the relationship between differentiator and what is differentiated.”) Devendrabuddhi also
refers to the relationship between janaka and janya, but does not use the words vyavasthāpaka and
vyavasthāpya. See PVP D 216a3–4/P 253a8–253b2 on PV III 301: ’o na ci yin zhe na / gang las ’bras
yin / gang yin pa’i las de bya ba de ni de’i sgrub byed / de sgrub par byed pa las chod pa med par tha
snyad la brten pa’i sgo nas rab tu ’grub pa thob par ’gyur gyi / bskyed par bya ba skyed par byed pa’i
dngos po (D; pos P) ni ma yin te / de’i bya ba ni de nyid yin no //
272
The Concept of vyava√sthā in Dharmottara and Prajñākaragupta
2. Dharmottara’s explanation of vyava√sthā as involving an ascertaining cognition
2.1 Two kinds of relationships and vyavasthāpaka as vyavasthāpanahetu
After commenting on the individual words of NB I 21,25 Dharmottara justifies the nondistinction between pramāṇa and pramāṇaphala in the way we have outlined above. The
text reads as follows:
na cātra janyajanakabhāvanibandhanaḥ sādhyasādhanabhāvaḥ, yenaikasmin
vastuni virodhaḥ syāt, api tu vyavasthāpyavyavasthāpakabhāvena,26 tata ekasya vastunaḥ kiñcid rūpaṃ pramāṇaṃ kiñcit pramāṇaphalaṃ na virudhyate.
vyavasthāpanahetur hi sārūpyaṃ tasya jñānasya, vyavasthāpyaṃ ca nīlasaṃvedanarūpam.
In the case of [pramāṇā and pramāṇaphala], the relationship between what
establishes (i.e., pramāṇa) and what is established (i.e., pramāṇaphala) is not
based on the relationship between producer and product. If it were, [then the
relationship would result in a] contradiction in one and the same real entity
(i.e., a momentary cognition). Rather, it is [given] in terms of the relationship
between differentiator and differentiated. Therefore, it is not contradictory
that a certain property of one real entity [i.e., cognition] is the means of valid
cognition and a certain other [property of the same cognition] is the result of
the means of valid cognition. For, the cause for differentiation (vyavasthāpanahetu) is the similarity (sārūpya) of cognition [to the blue object], and what
is differentiated is the property of the awareness of blue (nīlasaṃvedanarūpa).
Dharmottara attempts to counter the objection that there would be a contradiction if pramāṇa and pramāṇaphala belonged to the same cognition. In addition to introducing the
contrast between the two kinds of relationships, he characterizes cognition’s similarity to
the object (sārūpya) as cause for the differentiation of the property of the awareness of the
object (nīlasaṃvedanarūpa). The point I wish to emphasize is that the word vyavasthāpaka
is here explained as vyavasthāpanahetu. This word vyavasthāpaka is problematic, and
in fact, Malvania changes this reading, which appears in all the manuscripts, to vyavasthāpana, apparently following Durvekamiśra’s suggestion in the Dharmottarapradīpa
(DhPr) subcommentary. However, the reading vyavasthāpaka that all manuscripts used
by Malvania contain should be kept, because Durvekamiśra himself confesses that the
reading vyavasthāpana contradicts what will be stated, by which he probably refers to
vyavasthāpyavyavasthāpakabhāva in NBṬ 83,2. Moreover, it seems more reasonable that
Dharmottara would explain vyavasthāpaka as vyavasthāpanahetu, whereas it appears less
plausible that he would apply the same explanation to vyavasthāpana. Durvekamiśra’s commentary is nevertheless helpful for understanding the meaning of the word: vyavasthāpaka
25
26
NB I 21: tadvaśād arthapratītisiddher iti // (“For, the understanding of the object is established through
the force of that [i.e., cognition’s having the form of the object].”)
-vyavasthāpaka- MSS. (reported in NBṬ); -vyavasthāpana- NBṬ. See below.
Mai Miyo
273
is the cause for differentiating, with the -aka suffix having the sense of instrument (karaṇa),
and vyavasthāpanahetu is the cause for the function of differentiation.27
2.2 The ascertaining cognition as the third factor
Dharmottara next goes on to explain the relationship between vyavasthāpaka, vyavasthāpanahetu, and vyavasthāpya in more detail.
NBṬ 83,2–84,1: vyavasthāpyavyavasthāpakabhāvo28 ’pi katham ekasya jñānasyeti cet. ucyate. sadṛśam29 anubhūya tadvijñānaṃ yato nīlasya grāhakam
avasthāpyate niścayapratyayena. tasmāt sārūpyam anubhūtaṃ vyavasthāpanahetuḥ. niścayapratyayena ca tajjñānaṃ nīlasaṃvedanam avasthāpyamānaṃ
vyavasthāpyam. ... vyavasthāpakaś ca vikalpapratyayaḥ pratyakṣabalotpanno
draṣṭavyaḥ.
[Question:] How does the relationship between differentiator and differentiated,
for its part, exist for one and the same cognition?
[Answer:] We say: having experienced (anubhūya) a likeness [to the external
blue object] ([nīla]sadṛśa, i.e., the blue form of the cognition), the [perceptual] cognition is ascertained (avasthāpyate)30 to be grasping blue by the
[subsequent] ascertaining cognition (niścayapratyaya) [“I perceive blue”].31
27
28
29
30
31
DhPr 83,14–17: vyavasthāpyate viśiṣṭenātmanā niyamyate ’neneti vyavasthānimittaṃ vyavasthāpanam
abhipretam. vyavasthāpanabhāvenety ayaṃ pāṭho vakṣyamāṇavirodhī . yadā tu vyavasthāpakabhāveneti pāṭho dṛśyate, tadā karaṇe kartṛbhāvavivakṣayā tathā draṣṭavyam. sādhvasiś chinattīti yathā
(“[In NBṬ according to Durvekamiśra’s reading: vyavasthāpyavyavasthāpanabhāvena,] vyavasthāpana
is intended to be the factor for differentiating, by means of which [the property of the awareness of
blue] is differentiated, that is, limited by [its] distinct nature. This reading vyavasthāpanabhāvena
contradicts what will be stated [by Dharmottara in NBṬ 83,2]. When on the other hand the reading
vyavasthāpakabhāvena is observed, [the word vyavasthāpaka] should be taken in the same way [as the
word vyavasthāpana] with the intention to express agency (-aka suffix) in the sense of instrument (-ana
suffix); for example, an excellent sword cuts”); DhPr 83,24–25: vyavasthāpanaṃ vyavasthākāraṇam,
vyavasthāyāṃ prayojakavyāpāra iti yāvat, tasya hetur nimittam (“vayvasthāpana [in the phrase, vyavasthāpanahetu] is the function of differentiating, to be precise, the activity causing differentiation. Its
[i.e., vyavasthāpana’s] cause [means] factor.”)
Oki 1993: 136, n. 23 suggests to change vyavasthāpaka to vyavasthāpana in accordance with Malvania’s earlier emendation. But now that Malvania’s emendation has been rejected, this change is also
unnecessary. It is true that Oki’s modification solves the problem of two meanings in one word in this
passage, which seems to be mentioned by Durvekamiśra, but it produces the same problem for the word
vyavasthāpana in the earlier passage.
sadṛśam MSS. (reported in NBṬ); nīlasadṛśam NBṬ. Malvania adds nīla according to DhPr, but
this seems to be an addition by Durvekamiśra. I have not incorporated it into Dharmottara’s text, in
accordance with all the manuscripts reported in NBṬ and the Tibetan translation.
ava√sthā, in causative, is a function of the ascertaining cognition and explained using the verbs ni√yam
(DhPr 84,15) and niś√ci (DhPr 84,17), so I translate it as “to ascertain.” Nevertheless, I am not sure
about the difference between the meanings of ava√sthā and vyava√sthā – both in causative – here. They
seem to be almost the same, so the meaning “differentiate” for avasthāpayati is also possible.
See DhPr 84,10: niścayapratyayeneti. niścayātmakajñānenottarakālabhāvinā (“[Answer:] By the
ascertaining cognition. [This means] by the cognition whose nature it is to ascertain, and which arises
at a subsequent time.”)
274
The Concept of vyava√sthā in Dharmottara and Prajñākaragupta
Therefore, the similarity which has been experienced is the cause for differentiation, and the [perceptual cognition] is what is differentiated, being
ascertained, by the ascertaining cognition, to be the awareness of blue. ...
Moreover, it is to be observed that the conceptual cognition (vikalpapratyaya,
i.e., the ascertaining cognition), which arises due to the [preceding] perception,
is the differentiator (vyavasthāpaka, i.e., the agent of differentiation).32
In this section Dharmottara uses the term vyavasthāpaka in two different senses. In the
question, it is used in the same sense as before, i.e. as the cause for differentiating (vyavasthāpanahetu), and it indicates pramāṇa understood as similarity. On the other hand, in
the answer, the term vyavasthāpaka is used in the sense of “agent of differentiation,”33
referring to the ascertaining cognition which arises immediately after the perception.
The argument, now, runs as follows. First of all, Dharmottara presupposes that the
cognitive process consists of two stages in different moments: In a first moment the perceptual cognition (pratyakṣaṃ jñānam, anubhava) experiences a likeness to the external
blue object (nīlasadṛśa, i.e., the blue form of the cognition), and in a second moment
there arises the ascertaining cognition, “I perceive this very blue” (nīlam evānubhavāmi).34
The preceding perceptual cognition’s similarity to the blue object here is the cause for
differentiating (vyavasthāpanahetu, vyavasthāpaka 1) because the succeeding ascertaining
cognition carries out the function of ascertaining (avasthāpayati) due to the fact that the
preceding cognition experiences (anu√bhū) the blue form in cognition. The preceding
perceptual cognition – the awareness of blue (nīlasaṃvedana) – is what is ascertained
(vyavasthāpya) because it is ascertained to be grasping blue by the succeeding cognition.
The succeeding ascertaining cognition, for its part, is the agent of differentiation (vyavasthāpaka 2). This ascertaining cognition is also characterized as a conceptual cognition
(vikalpapratyaya) because it accomplishes its function of ascertaining by the exclusion of
others (anyavyāvṛtti).35
3. Prajñākaragupta’s refutation of Dharmottara’s interpretation
Prajñākaragupta explicitly denies the intervention by the ascertaining cognition that
Dharmottara assumes. A dialogue at the end of the PVA on PV III 31136 clearly illustrates
32
33
34
35
36
This vyavasthāpaka does not mean the cause for differentiating, but the agent of differentiation. See
below.
See DhPr 84,25: vyavasthāpayatīti vyavasthāpakaḥ (“It differentiates [the perceptual cognition], so [it
is] the differentiator.”)
See DhPr 83,29–30: na tu nīlasadṛśam anubhavāmīti niścayo ’sti, api tu nīlam evānubhavāmīti nīlasya
grāhakam avasthāpyate.
See NBṬ 83,5–6: tasmād asārūpyavyāvṛttyā sārūpyaṃ jñānasya vyavasthāpanahetuḥ. anīlabodhavyāvṛttyā ca nīlabodharūpatvaṃ vyavasthāpyam (“Therefore, the [perceptual] cognition’s similarity [to the
object] is the cause for differentiating by the exclusion of something which is not the similarity, and the
[cognition’s] property of awareness of blue is what is ascertained by the exclusion of that which is not
the awareness of blue.”)
This numbering of the verse follows Tosaki 1979. Sāṅkṛtyāyana’s numbering in PVA is mistaken.
Mai Miyo
275
this point.37 The opponent, whom the subcommentator Yamāri identifies as Dharmottara,38
first says:
PVA 345,9f. on PV III 311: tadākāratāyām api niścayaḥ paraḥ pratiniyatavyavasthāyām apekṣyata iti cet.
[Objection:] Even if [a cognition] has the form (ākāra) of that [object], ascertaining (niścaya),39 which is different [from the cognition’s property of having
the form], is required for the differentiation [of the cognition/object] as being
limited in each case (pratiniyatavyavasthā).40
According to the opponent’s view, ascertaining (niścaya) intervenes between the cognition’s
property of having the form of the object (i.e., pramāṇa) and cognition’s differentiation
according to the object (i.e., pramāṇaphala). We can see how this opponent’s view corresponds to that of Dharmottara as discussed above.
Prajñākaragupta refutes this position as follows:
PVA 345,10–12: na, pratyakṣaprāmāṇyaprastāvāt. abhyāsasambhave hi pratyakṣaṃ pramāṇam. tadākāramātrād eva ca tadā pravartanaṃ41 niścayam
antareṇāpi. yadā tu niścayāpekṣā, tadānumānaṃ pramāṇam. tasyāpy ākāramātrād eva pravṛtteḥ nāparāpekṣā. tasmād ākārān nāparaṃ karaṇam.
No. For, the relevant subject is perception’s being a valid cognition. To explain:
perception is the pramāṇa when there is habituation (abhyāsa). And then,
[perception] prompts [the cognizing person] to act (pravartana), only due
to its form (ākāra), even without ascertaining (niścaya). On the other hand,
inference (anumāna) is the pramāṇa when [the person] depends on ascertaining
[in order to act]. [However, he/she] also acts only due to the form of that
[inference], so the [inference] does not depend on anything other [than its
form for prompting him/her to act]. Accordingly, there is no other instrument
(karaṇa, i.e., pramāṇa) apart from the [object’s] form [in direct perception or
inference].
Prajñākaragupta introduces a new topic here regarding pramāṇa in the broader sense,
i.e., valid cognition and human action, thus broadening the scope of the discussion. As
Dharmakīrti claims in PV II 3cd, valid cognition is the main cause (pradhāna) for human
37
38
39
40
41
See Miyo 2013 for a critical edition of the whole Sanskrit text of PVA on PV III 311, a Japanese
translation, and additional information.
Y (D 251a3, P 337b1–2): da ni slob dpon chos mchog gi lugs sun phyung ba’i don du nye bar ’god
pa ni / de’i rnam pa yang zhes bya ba’o // (“In order to criticize the teacher Dharmottara’s theory,
[Prajñākaragupta] states here that even if [the cognition] has the form of the [object,] and so on.”)
It is also possible to interpret niścaya here as “ascertaining cognition,” as in the NBṬ.
The interpretation of this compound involves some difficulties. We have a synonym for it, pratiniyatārthavyavasthā, cited in n. 46. I tentatively interpret it according to PVA 346,23–24: yadi sa tathābhūta
ākāro na syāt, na kaścit pratiniyatam arthaṃ vyavasthāpayet (“If such a form did not exist, nothing
would differentiate the object as being limited in each case.”)
pravarttanaṃ PVAMS [’jug pa yin no T (D 15a4, P 18a7)]; pravarttana- PVA.
276
The Concept of vyava√sthā in Dharmottara and Prajñākaragupta
action (pravṛtti) toward an object that is to be obtained or avoided.42 If one interprets this
statement strictly speaking, perception as valid cognition, or more precisely, its object-form,
should directly prompt the cognizing person to act without ascertaining. But if perception
could cause the action only through ascertaining, then the ascertaining cognition would
be the pramāṇa.43 It is important to note that Prajñākaragupta here adopts the concept of
“habituation” (abhyāsa) as a powerful device that guarantees perception’s ability to cause
action.44 He clearly separates perception and inference in accordance with the presence
or absence of habituation, so the range of perception as a pramāṇa is limited to those
particular cases that directly cause human action based on habituation.45 Consequently,
perception causes human action only through its object-form without ascertaining. On
the other hand, inference is the pramāṇa if the person acts depending on ascertainment,
but inference does not depend on subsequent ascertaining; rather, inference itself is the
ascertaining cognition. Therefore, the object-form is the means of valid cognition in both
cases, perception and inference, and the ascertaining cognition which is different from
them does not have any influence on them.
A detailed discussion of Prajñākaragupta’s views on the relationship between pramāṇa
and pramāṇaphala, as well as an examination of his unique terminology for explaining
the aspects of cognition that are involved,46 must be left to another occasion. But for the
present context we may note that according to Prajñākaragupta, a cognition’s aspect of
understanding (i.e., pramāṇaphala) is completely identical with its aspect of having the
object’s form (i.e., pramāṇa), so there is no room for a third factor that might intervene to
differentiate the understanding;47 rather, it must be differentiated by another aspect of the
very same perception.
42
43
44
45
46
47
See PV II 3b’–d: dhīpramāṇatā / pravṛttes tatpradhānatvād dheyopādeyavastuni // (“Cognition is
the pramāṇa because it is the main cause for [human] action toward an object that is to be obtained or
avoided.”) The verse is also translated in Dunne 2004: 262. As Katsura 1989: 541 points out, it is a
distinctive feature of Dharmakīrti’s thought, in contrast to Dignāga’s, to formulate pramāṇa theory as
concerning the whole of the cognitive process including the action toward the object.
However, Dharmakīrti accepts that some other factors intervene between perception and the action that
it prompts. Dharmottara justifies this intervention by paraphrasing “what prompts to act” (pravartaka)
as “what indicates the object of the action” (pravṛttiviṣayapradarśaka) and Durvekamiśra interprets it
introducing the concept of “capable of action” (pravṛttiyogya). Therefore, we can say Prajñākaragupta
makes a radical interpretation here. On this subject, see Miyo 2014.
The concept of atyantābhyāsa is already used by Śākyabuddhi in the same context, as seen, among
others, in Inami 1993: 96 and Krasser 2003. This issue may relate to the topic of yogic perception
(yogipratyakṣa).
For other instances demonstrating the same idea, see PVAO 55,15–56,1 on PV II 3b’cd: na ca cakṣurādikāt pravartate jñānam antareṇa, vikalpam antareṇāpi tv abhyāsāt pravartate. tato heyopādeyaviṣaye
dhīr eva pūrvikā pravartanāt pramāṇam, na vikalpādayaḥ. yatra tu nābhyāsaḥ, tatrānumānam eva
pratyabhijñānādayaḥ, dealt with in Ono 1993: 110–111; PVA 218,6–8 on PV III 56: yatra bhāvigatis
tatrānumānaṃ mānam iṣyate / vartamāne ’timātreṇa vṛttāv adhyakṣamānatā //240// yatrātyantābhyāsād avikalpayato ’pi pravartanam, tatra pratyakṣaṃ pramāṇam. anyathā vikalpasya pramāṇāntaratā
prāptā. yadi yatra vikalpas tatraiva pratyakṣaṃ pravartakaṃ pramāṇaṃ ceti.
Some examples for his unique terminology are: pratiniyatākāratā (PVA 344,31), tadākāraviśeṣa (PVA
345,3), and ākāraniyama (PVA 345,7) for pramāṇa; and arthaparicchedavyavasthā (PVA 345,1),
niyatārthavyavasthā (PVA 345,2), arthavyavasthiti (PVA 345,6), and pratiniyatārthavyavasthā (PVA
345,4–5) for pramāṇaphala.
PVA 345,6: ākāraniyamaḥ siddho yadi sārthavyavasthitiḥ (em.; -vyasthitiḥ PVA) / akāraṇaṃ paraḥ
(PVAMS ; śaraḥ PVA) siddhe prāptaḥ kim iti poṣyate // (618) (“If the limitation of the form has been
Mai Miyo
277
4. Conclusion
As we have shown, for Dharmottara the ascertaining cognition (niścayapratyaya), which
arises immediately after a perceptual cognition, plays an important role even in his discussion of perception. This ascertaining cognition works as the agent of differentiation
(vyavasthāpaka), and the perception’s similarity to the object is merely the cause for differentiating (vyavasthāpanahetu). Prajñākaragupta, on the other hand, denies this kind of
intervention by an ascertaining cognition. He excludes it from the discourse on perception
by taking recourse to the idea of habituation (ābhyāsa). According to Prajñākaragupta, a
cognition’s aspect of understanding the object (i.e., pramāṇaphala) is completely identical
with its aspect of having the object’s form (i.e., pramāṇa), so one aspect of cognition is
simply differentiated by another aspect of the very same cognition. This difference in the
two commentator’s interpretations also results in a difference in their usage of derivatives
of vyava√sthā. Dharmottara uses these only in cases where the function of ascertaining, or
conceptual construction, is involved. On the other hand, Prajñākaragupta uses them even in
cases where conceptual construction is not concerned.
Why do Dharmottara and Prajñākaragupta present us with such divergent opinions on
the influence of the conceptual construction on perception? While this topic calls for further
in-depth studies, I would like to highlight three points that seem to be of special interest.
The first is the position of the form or image in a cognition. Prajñākaragupta emphasizes
the unity of the form and cognition, which easily leads to a negation of the intervention by
conceptual construction,48 while Dharmottara does not.49 A second point worthy of note is
the kind of cognition they are thinking about when discussing perception as a pramāṇa.
Dharmottara thinks mainly about mundane cognition, such as perception by the sense
faculties, but Prajñākaragupta seems to think more about supramundane cognition, such
as yogic perception up to a Buddha’s cognition.50 This soteriologically relevant cognition
should be free from the influence of conceptual construction. As a last and final point,
I would like to highlight that already the earlier commentaries on the Pramāṇavārttika
by Devendrabuddhi and Śākyabuddhi show a faint indication of the same controversy.51
48
49
50
51
established, that is [exactly] the differentiation of the object. Something other [than the limitation, such
as the soundness of the sense faculty,] is not a cause [for differentiation]. Why would a thing which
has [already] become established be increased [by something else]?”) Prajñākaragupta’s identification
of cognition’s form of the object and understanding of the object affects his interpretation of the
sahopalambhaniyama inference as well. On this subject, see Iwata 1991.
See also the passage discussed in n. 47.
These two different views on cognition’s form are the basis of the two sub-schools of Yogācāra, i.e.,
sākāravāda and nirākāravāda/alīkākāravāda. See, among others, Iwata 1991: 211–231, Funayama
2007.
As Ono 2012: 166; 170–173 points out, Dharmottara takes mundane cognition into account in his
discussion of the definition of pramāṇa, but on the other hand Prajñākaragupta regards the Buddha
as the supreme pramāṇa (paraṃ pramāṇam) whose nature is perception. PVAO 84,1–2 on PV II 7a:
tathāgato hi bhagavān tadvān kṛtvā pratyakṣarūpa eva bhagavān pramāṇam, translated in Ono 1993:
84.
Devendrabuddhi probably assumes that perception prompts the cognizing person to act after it brings
forth the ascertaining cognition. However, Śākyabuddhi divides human action based on perception into
two types, the first instance and habituated instances, and he states that the fully habituated perception
with a vivid form ascertains its validity by itself. This could mean that habituated perception prompts
278
The Concept of vyava√sthā in Dharmottara and Prajñākaragupta
Thus, a satisfactory understanding of the dissimilarity of views held by Dharmottara and
Prajñākaragupta will emerge only after comprehensive investigation into its epistemological
and historical roots.
References and abbreviations
DhPr Dharmottarapradīpa (Durvekamiśra): Paṇḍita Durveka Miśra’s Dharmottarapradīpa: Being a sub-commentary of Dharmottara’s Nyāyabinduṭīkā, a commentary of
Dharmakīrti’s Nyāyabindu, ed. Paṇḍita D. Malvania. Patna 1955.
Dreyfus 1997 G. B. J. Dreyfus, Recognizing Reality: Dharmakīrti’s Philosophy and Its
Tibetan Interpretations. Albany 1997.
Dunne 2004 J. D. Dunne, Foundation of Dharmakīrti’s Philosophy. Somerville 2004.
Franco 1997 E. Franco, Dharmakīrti on Compassion and Rebirth. Wien 1997.
Funayama 2007 T. Funayama, Kamalaśīla’s distinction between the two sub-schools
of Yogācāra. A provisional survey. In: Pramāṇakīrtiḥ: Papers Dedicated to Ernst
Steinkellner on the Occasion of His 70th Birthday, ed. B. Kellner et al. Wien 2007,
187–202.
Hattori 1968 M. Hattori, Dignāga, On Perception, being the Pratyakṣapariccheda of
Dignāga’s Pramāṇasamuccaya from the Sanskrit fragments and the Tibetan versions.
Cambridge, MA 1968.
Inami 1993 M. Inami, Bukkyō ronri gakuha no shinriron: Devendrabuddhi to Śākyabuddhi
(Theory of Truth in the Buddhist Epistemological School: Devendrabuddhi and
Śākyabuddhi). In: Genshi bukkyō to daijō bukkyo: Watanabe Fumimaro hakushi
tsuitō kinen ronshū. Vol. 2. Kyoto 1993, (85)–(118).
Iwata 1991 T. Iwata, Sahopalambhaniyama. Stuttgart 1991.
Kahrs 1998 E. Kahrs, Indian Semantic Analysis: The Nirvacana Tradition. Cambridge
1998.
Kataoka 2011 K. Kataoka, Dharmakīrti ni yoru Jyuryōron I 9 no kaishaku: Ryōhyōshaku
III 320–352 no bunseki (Dharmakīrti’s Interpretation of Pramāṇasamuccaya I 9: An
Analysis of Pramāṇavārttika III 320–352). Annual of Philosophy (Kyushu University)
70 (2011) 43–75.
Katsura 1989 S. Katsura, Chikaku handan, giji chikaku, sezoku chi (On Perceptual Judgement). In: Fujita Kōdatsu hakushi kanreki kinen ronshū. Kyoto 1989, 533–553.
Katsura 2007 S. Katsura, Dharmakīrti’s Proof of the Existence of Other Minds. In: Pramāṇakīrtiḥ: Papers Dedicated to Ernst Steinkellner on the Occasion of His 70th Birthday,
ed. B. Kellner et al. Vol. 1. Wien 2007, 407–421.
human action directly because of its vivid form of the object. PVP D 5a5/P 5b7: ’on kyang bzung ba’i
rnam pa gang yin pa dag la yang nges pa skyed par byed pa de la / ’jug par byed pa’i phyir tshad ma
nyid du ’dod kyi / gzhan du ni ma yin no //; PVṬ(Ś) D 72a5–7/P 87a5–87b2: mngon sum ni rnam par
rtog pa med pa nyid kyi phyir nges pa ma yin mod kyi ’on kyang der snang bar skyes pa tsam yin no // ...
mngon sum gyi rten can gyi ’jug pa ni rnam pa gnyis te dang po nyid dang goms pa can no // de la
goms pa dang ldan pa gang yin pa de la shin tu goms pa gsal (D; bsal P) ba can gyi mngon sum skyes
pa na ji lta ba bzhin du goms pa’i rnam par ’khrul pa’i rgyu mtshan spangs pa can nyid kyis yongs su
bcad nas skye ba dang / de lta bur gyur pa’i phyis ’byung ba’i nges pa skyed par byed pa’i phyir de la
skyes bu ’jug par byed do // This passage is translated in Inami 1993: 95–96.
Mai Miyo
279
Kellner 2010 B. Kellner, Self-Awareness (svasaṃvedana) in Dignāga’s Pramāṇasamuccaya and -vṛtti: A Close Reading. Journal of Indian Philosophy 38 (2010) 203–231.
Kellner 2011 B. Kellner, Dharmakīrti’s Criticism of External Realism and the Sliding
Scale of Analysis. In: Religion and Logic in Buddhist Philosophical Analysis: Proceedings of the Fourth International Dharmakīrti Conference, Vienna, August 23–27,
2005, ed. H. Krasser et al. Wien 2011, 291–298.
Kobayashi 2009 H. Kobayashi, Ninshiki kekka to shiteno jiko ninshiki (Self-awareness
as the Result of Cognition (pramāṇaphala)). Report of the Japanese Association for
Tibetan Studies 55 (2009) 121–130.
Krasser 2001 H. Krasser, On Dharmakīrti’s understanding of pramāṇabhūta and his definition of pramāṇa. Wiener Zeitschrift für die Kunde Südasiens 45 (2001) 173–199.
Krasser 2003 H. Krasser, On the Ascertainment of Validity in the Buddhist Epistemological Tradition. Journal of Indian Philosophy 31 (2003) 161–184.
Miyo 2008 M. Miyo, Dharmakīrti no ryō ryōka hi bettai setsu: Naze gaiteki yōin ha ryō
de nainoka (An Interpretation of Dharmakīrti’s Negation of the Difference between
Pramāṇa and Pramāṇaphala). Journal of Indian and Buddhist Studies 57-1 (2008)
(314)–(410).
Miyo 2013 M. Miyo, Kettei chi ni kansuru Prajñākaragupta no Dharmottara hihan: Pramāṇavārttikālaṅkāra ad PV II 311 yakuchū kenkyū (Prajñākaragupta’s Criticism of
Dharmottara’s Theory of niścayapratyaya: A Study of the Pramāṇavārttikālaṅkāra
ad PV III 311). Bulletin of the Graduate Division of Letters, Arts and Sciences of
Waseda University 58 (2013) 93–107.
Miyo 2014 M. Miyo, Controversy between Dharmottara and Prajñākaragupta Regarding
Pravartaka. Journal of Indian and Buddhist Studies 62-3 (2014) (223)–(228).
Moriyama 2010 S. Moriyama, On Self-Awareness in the Sautrāntika Epistemology. Journal of Indian Philosophy 38 (2010) 261–277.
Murakami 2008 A. Murakami, Rang rig pa ni kansuru mKhas grub rje no kaishaku (The
Interpretation of rang rig pa by mKhas grub rje). Report of the Japanese Association
for Tibetan Studies 54 (2008) 17–31.
NB Nyāyabindu (Dharmakīrti): see DhPr.
NBṬ Nyāyabinduṭīkā (Dharmottara): see DhPr.
NS Nyāyasūtra (Akṣapāda Gautama): see NBh.
Nyāyabhāṣya Nyāyabhāṣya (Vātsyāyana): Nyāyadarśanam with Vātsyāyana’s Bhāṣya,
Uddyotakara’s Vārttika, Vācaspati Miśra’s Tātparyaṭīkā & Viśvanātha’s Vṛtti,
ed. Taranatha Nyāya-Tarkatirtha, Amarendramohan Tarkatirtha and Hematakumar
Tarkatirtha. 2 vols. Calcutta 1936–44.
Oki 1993 K. Oki, Dharmottara no ryō ryōka hibettairon: Nyāyabinduṭīkā ni okeru
(Dharmottara’s Commentary on Nirvikalpaka Pratyakṣa in the First Chapter of
the Nyāyabinduṭīkā). In: Genshi bukkyō to daijō bukkyō: Watanabe Fumimaro hakushi tsuitō kinen ronshū. Vol. 2. Kyoto 1993, (119)–(136).
Ono 1993 M. Ono, Prajñākaraguptas Erklärung der Definition gültiger Erkenntnis
(Pramāṇavārttikālaṃkāra zu Pramāṇavārttika II 1–7). Dissertation, Universität
Wien. Wien 1993.
280
The Concept of vyava√sthā in Dharmottara and Prajñākaragupta
Ono 1995 M. Ono, Bukkyō ronri gakuha no ichi keifu: Prajñākaraguputa to sono kōkeishatachi (A Lineage in Buddhist Logico-epistemological School: Prajñākaragupta
and his successors). Tetsugaku shisō ronshū 21 (1995) 142–162.
Ono 2012 M. Ono, Shinri ron: puramāna toha nanika (Chapter five: What is pramāṇa).
In: Ninshiki ron to ronri gaku (Epistemology and Logic), ed. S. Katsura et al. Tokyo
2012, 155–188.
PS I Pramāṇasamuccaya, chapter I (Dignāga): see PS(V) I.
PS(V) I Pramāṇasamuccaya and -vṛtti, chapter I (Dignāga), ed. E. Steinkellner. Wien 2005.
https://www.oeaw.ac.at/fileadmin/Institute/IKGA/PDF/forschung/buddhismuskund
e/dignaga_PS_1.pdf, last visited 20-07-2019.
PV Pramāṇavārttika (Dharmakīrti).
PV II Pramāṇavārttika, chapter II (Dharmakīrti): see PVV.
PV III Pramāṇavārttika, chapter III (Dharmakīrti): edited in Tosaki 1979.
PVA Pramāṇavārttikālaṅkāra (Prajñākaragupta): Pramāṇavārttikabhāshyam or Vārttikālaṅkāraḥ of Prajñākaragupta: Being a commentary on Dharmakīrti’s Pramāṇavārtikam, ed. R. Sāṅkṛtyāyana. Patna 1953.
PVAMS Sanskrit Manuscript (Manuscript B) of Prajñākaragupta’s Pramāṇavārttikālaṅkāra:
The Sanskrit Commentaries on the Pramāṇavārttikam from the Rāhula Sāṅkṛtyāyana’s Collection of Negatives. 3 vols. Vol. I: Sanskrit Manuscripts of Prajñākaragupta’s Pramāṇavārttikabhāṣyam. Patna/Narita 1998.
PVAO Pramāṇavārttikālaṅkāra (Prajñākaragupta): Prajñākaraguptas Erklärung der Definition gültiger Erkenntnis (Pramāṇavārttikālaṃkāra zu Pramāṇavārttika II 1–7),
Teil I, ed. M. Ono. Wien 2000.
PVin Pramāṇaviniścaya (Dharmakīrti).
PVin I, II Pramāṇaviniścaya, chapter I, II (Dharmakīrti): Dharmakīrti’s Pramāṇaviniścaya, Chapters 1 and 2, ed. E. Steinkellner. Beijing/Vienna 2007.
PVin III Pramāṇaviniścaya, chapter III (Dharmakīrti): Dharmakīrti’s Pramāṇaviniścaya,
chapter 3, ed. P. Hugon and T. Tomabechi. Beijing/Vienna 2011.
PVP Tibetan translation of Pramāṇavārttikapañjikā (Devendrabuddhi): D 4217 (che), P
5717(b) (che).
PVṬ(Ś) Tibetan translation of Pramāṇavārttikaṭīkā (Śākyabuddhi): D 4220 (nye), P 5718
(nye).
PVV Pramāṇavārttikavṛtti (Manorathanandin): Dharmakīrti’s Pramāṇavārttika with a
Commentary by Manorathanandin, ed. R. Sāṅkṛtyāyana. Appendix to Journal of the
Bihar and Orissa Research Society 24-3. Patna 1938.
PVVMS Sanskrit Manuscript of Manorathanandin’s Pramāṇavārttikavṛtti: The Sanskrit
Commentaries on the Pramāṇavārttikam from the Rāhula Sāṅkṛtyāyana’s Collection
of Negatives. 3 vols. Vol. III: Sanskrit Manuscripts of Manorathanandin’s Pramāṇavārttikavṛttiḥ. Patna/Narita 1998.
Steinkellner 1979 E. Steinkellner, Dharmakīrti’s Pramāṇaviniścayaḥ, Zweites Kapitel:
Svārthānumānam. Teil II: Übersetzung und Anmerkungen. Wien 1979.
Tosaki 1979 H. Tosaki, Bukkyō Ninshikiron no Kenkyū. Tokyo 1979.
Tosaki 1991 H. Tosaki, Hossyō cho Pramāṇaviniścaya dai isshō genryō (chikaku) ron no
wayaku (7) (An Annotated Translation of Pramāṇaviniścaya, Chapter I (7)). Annual
of Philosophy (Kyushu University) 50 (1991) 1–10.
Mai Miyo
281
Vetter 1966 T. Vetter, Dharmakīrti’s Pramāṇaviniścayaḥ, 1 Kapitel: Pratyakṣam: Einleitung, Text der tibetischen Übersetzung, Sanskritfragmente, deutsche Übersetzung.
Wien 1966.
Y Tibetan translation of Pramāṇavārttikālaṅkāraṭīkā Supariśuddhi (Yamāri): D 4226 (me),
P 5723 (me).
Yoshida 2011 A. Yoshida, Bukkyō ninshikiron ni okeru pramāṇa no ryōgisei (The Ambiguity of pramāṇa in the Buddhist Epistemology). Bukkyōgaku kenkyū (Ryūkoku
University) 67 (2011) 31–48.
Dharmakīrti’s Apoha Theory: Perceptual Judgment and the
Lack of Superimposition (samāropaviveka)
by
Miyuki Nakasuka
Introduction
In the Apoha section in the Svārthānumāna chapter of his Pramāṇavārttika, Dharmakīrti starts by characterizing apoha as the difference (bheda) among real entities from an
ontological point of view (PV I.40–42), and then as the exclusion of superimposition
(samāropavyavaccheda) from an epistemological point of view (PV I.43–58). He defines
as a determinate cognition (niścayajñāna) an inference (anumāna) and a perceptual judgment, of which the latter is the conceptual cognition that follows a perceptual cognition
(pratyakṣapṛṣṭhabhāvivikalpa), and argues that both have the exclusion of superimposition
as their object. It is important to note that what Dharmakīrti means by the term samāropavyavaccheda “exclusion of superimposition” differs for inference and perceptual judgment.
In the case of an inference, the term in question signifies the function of excluding superimposition (samāropavyavaccheda) that is carried out by the inference when there is an
erroneous cognition with respect to the subject of the inference. In the case of a perceptual
judgment, on the other hand, it means “lack of superimposition” (samāropaviveka). This
latter point is made by Dharmakīrti in PV I.48.
PV I.48: kvacid dṛṣṭe ’pi yaj jñānaṃ sāmānyārthaṃ vikalpakam /
asamāropitānyāṃśe tanmātrāpohagocaram //
Katsura (1984, 1989, 1993) and Kellner (2004), when dealing with Dharmakīrti’s concept
of perceptual judgment, refer to the exclusion of superimposition. Katsura (1984: 226),
relying on PV I.48–49, says: “Apart from provoking a practical activity, another important
function is to prevent wrong judgement from arising.” In addition, Kellner (2004: 41)
explicitly states that “ascertainment which follows immediately after perception, serves to
keep error and doubt away and for this reason also has an ‘exclusion’ as its object.” Their
view of perceptual judgment is quite acceptable. But to my knowledge, no studies have ever
tried to show how perceptual judgment is linked to the exclusion of superimposition. The
aim of this paper is thus to show how Dharmakīrti establishes that a perceptual judgment
has “exclusion” as its object.
I wish to express my gratitude to Professor Kei Kataoka for valuable advice. I would also like to thank
Mr. Suguru Ishimura for helpful suggestions. This work was supported by JSPS KAKENHI Grant
Number 263284.
Birgit Kellner et al., eds., Reverberations of Dharmakīrti’s Philosophy: Proceedings of the Fifth International Dharmakīrti
Conference Heidelberg, August 26 to 30, 2014. Vienna 2020, pp. 283–299.
284
Dharmakīrti’s Apoha Theory
1. Arguments in PV I.43–47: samāropavyavaccheda
Before entering into the main task, let us look at the preceding arguments Dharmakīrti
brings forward in PV I.43–47. Dharmakīrti introduces PV I.43 by saying the following:
PVSV ad PV I.43 (G25.26–26.1, M17.1–2): kathaṃ punar etad gamyate –
vyavacchedaḥ śabdaliṅgābhyāṃ pratipādyate vidhinā na vasturūpam eveti
pramāṇāntarasya śabdāntarasya ca pravṛtteḥ /
[Objection] But how can this be understood: it is exclusion (vyavaccheda)
and not a real thing in a positive form that is understood from a word and an
inferential mark?
[Answer] [This is reasonable] because another word occurs [with respect to
one and the same real thing] and because another valid cognition occurs [with
respect to one and the same real thing].
Dharmakīrti here cites Dignāga’s well-known statement: vyavacchedaḥ śabdaliṅgābhyāṃ
pratipādyate vidhinā na vasturūpam eva.1 According to Dharmakīrti, the fact that different
words or different valid cognitions can occur with reference to one and the same real thing
shows that what is understood from a word or an inferential mark is not the real thing
itself but exclusion (vyavaccheda). In order to demonstrate this point, Dharmakīrti starts to
develop arguments.
In PV I.43–44, to begin with, Dharmakīrti says the following:
PV I.43: ekasyārthasvabhāvasya pratyakṣasya sataḥ svayam /
ko ’nyo na dṛṣṭo bhāgaḥ syād yaḥ pramāṇaiḥ parīkṣyate //
PV I.44: no ced bhrāntinimittena saṃyojyeta guṇāntaram /
śuktau vā rajatākāro rūpasādharmyadarśanāt //
When the single essential nature of an object is perceived as it is, what part
could be left unseen separately from it? To be sure, if a part is left unseen, it is
examined through valid cognitions.
[But the condition must be imposed:] “On the condition that a quality foreign
to the object is not connected with it by cause of an erroneous cognition
(bhrāntinimitta), unlike the case where the form of a piece of silver is connected
with mother of pearl due to the perception of the similarity in form between
them.”
PVSV ad PV I.44 (G26.18–27.2, M17.16–21): tasmāt paśyan śuktirūpaṃ
viśiṣṭam eva paśyati / niścayapratyayavaikalyāt tv aniścinvan tatsāmānyaṃ
1
This translation of Dignāga’s statement is based on Śākyabuddhi’s interpretation. According to Karṇakagomin, the statement is interpreted as follows: It is exclusion in a positive form and not a real thing
that is understood from a word and an inferential mark. On the different interpretations, see Kellner
2004: 5, n. 3. On this statement of Dignāga’s, see Pind 2015: II 127, n. 431.
Miyuki Nakasuka
285
paśyāmīti manyate / tato ’sya rajatasamāropaḥ / tathā sadṛśāparāparotpattyālakṣitanānātvasya tadbhāvasamāropāt sthitibhrāntiḥ / yāvanto ’sya parabhāvās tāvanta eva yathāsvaṃ nimittabhāvinaḥ samāropā iti tadvyavacchedakāni
bhavanti pramāṇāni saphalāni syuḥ / teṣāṃ tu vyavacchedaphalānāṃ nāpratītavastvaṃśapratyāyane pravṛttis tasya dṛṣṭatvāt / anaṃśasya caikadeśena
darśanāyogāt /
Therefore, a person, when seeing the form of mother of pearl, is to see it
precisely as what is differentiated [from a piece of silver]. But, due to the
deficiency of causes of a determinate cognition,2 the person, without determining the form, conceives: “I see a property common to them.” Consequently
the person superimposes [the image of] silver [on mother of pearl]. Similarly,
to a person who does not recognize the difference between moments since
there arise similar moments in sequence, there arises the erroneous cognition:
“This is [temporally] persistent” because of the superimposition of identity
(tadbhāvasamāropa).3
If an entity has a certain number of foreign properties, that number of properties
is what arises from causes [of erroneous cognitions] according to each case
and is superimposed on the entity.
Therefore, valid cognitions which arise to exclude what is superimposed must
be fruitful.
But, it is not the case that the [valid cognitions] bringing about the exclusion
take place in order to make known the part of a real thing which has not
been understood. For the [part for which valid cognitions take place] has been
perceived. Indeed, it is not proper to say that something indivisible is partially
perceived.
It is important to note that Dharmakīrti accepts the following points:
1.
2.
3.
2
3
On the condition that no erroneous cognition (bhrānti) arises with respect
to a real thing, when a real thing is perceived, all its parts are perceived.
An erroneous cognition can arise with respect to a real thing.
When a person does not recognize the difference between moments,
the person superimposes identity between them, and erroneously cognizes “this is permanent.” (The non-recognition of the difference between
According to Dharmakīrti, the causes of a determinate cognition in the form of a perceptual judgment
are: acuity of a cognition (buddhipāṭava), a state of habituation to the imprint of this (tadvāsanābhyāsa), situation-context (prakaraṇa), and others. PVSV ad PV I.58 (G32.3–8, M20.29–21.1): yady
apy aṃśarahitaḥ sarvato bhinnasvabhāvo bhāvo ’nubhūtas tathāpi na sarvabhedeṣu tāvatā niścayo
bhavati / kāraṇāntarāpekṣyatvāt / anubhavo hi yathāvikalpābhyāsaṃ niścayapratyayān janayati / yathā rūpadarśanāviśeṣe ’pi kuṇapakāminībhakṣyavikalpāḥ / tatra buddhipāṭavaṃ tadvāsanābhyāsaḥ
prakaraṇam ityādayo ’nubhavād bhedaniścayotpattisahakāriṇaḥ / On the concept of the causes of a
determinate cognition, see Kellner 2004: 19–32.
Karṇakagomin glosses tadbhāva as sattā “the property of continuing to exist.” PVSVṬ 123.28: tadbhāvasamāropāt sattāsamāropāt. But the term tadbhāva, synonymous with tattva, can be taken as meaning
“identity,” as opposed to “difference” (nānātva).
286
Dharmakīrti’s Apoha Theory
4.
5.
moments or the observation of the similarity between moments [bhrāntinimitta] → superimposition of identity [samāropa] → an erroneous
cognition [bhrānti].)
A valid cognition (pramāṇa), i.e., an inference, is resorted to in order to
exclude the superimposition.
This superimposition must take place with reference to the part which
has been perceived.
Dharmakīrti here explains the basic structure in which a valid cognition brings about the
exclusion of superimposition.
In PV I.45,4 which is a recapitulating verse (saṃgrahaśloka), he goes on to argue
that when a real thing is perceived, all its qualities are perceived; but when an erroneous
cognition arises with reference to that real thing, these qualities are not determined. Thus
one must resort to an inference in order to determine them.
Moreover, in PV I.465 he states that if an inference were to grasp a real thing, then
it would follow that once one property of it is determined, all properties are determined;
and that if an inference has exclusion (apoha) as its object, the undesirable consequence
does not follow. It is in commenting this verse that Dharmakīrti makes an important point:
an inference brings about the exclusion of superimposition (samāropavyavaccheda). He
states:
PVSV ad PV I.46 (G27.13–15, M18.1–3): yadā punar anumānena samāropavyavacchedaḥ kriyate tadā naikasamāropavyavacchedād anyavyavacchedaḥ
kṛto bhavatīti tadartham anyat pravartate /
However, when [it is assumed that] the exclusion of superimposition is brought
about by an inference, then [it can be said that] since the exclusion of a
different [superimposition (non-B)] is not brought about just because the
superimposition (non-A) is excluded by one [inference (A)], another [inference
(B)] takes place in order to [exclude the different superimposition (non-B)].6
Here Dharmakīrti assumes that, with reference to sound, two superimpositions arise: (nonA) the superimposition of the property of not being produced (akṛtakasamāropa), and
(non-B) that of the property of being permanent (nityasamāropa). Also, two inferences take
place: (A) the inference to prove the property of being produced (kṛtakānumāna) and (B)
that to prove the property of being impermanent (anityatvānumāna). He supposes that when
a real thing is perceived, a property which the real thing does not bear is superimposed on
it and that an inference is performed in order to exclude such a superimposition.
4
5
6
PV I.45: tasmād dṛṣṭasya bhāvasya dṛṣṭa evākhilo guṇaḥ / bhrānter niścīyate neti sādhanaṃ saṃpravartate //
PV I.46: vastugrahe ’numānāc ca dharmasyaikasya niścaye / sarvadharmagraho ’pohe nāyaṃ doṣaḥ
prasajyate //
PVSVṬ 125.23–25: tadaikena kṛtakānumānenaikasyākṛtakasamāropasya vyavacchedād anyasya nityasamāropasya vyavacchedaḥ kṛto na bhavatīti kṛtvā tadartham anyasamāropavyavacchedārtham anyad
ity anityatvādyanumānaṃ pravartate /
Miyuki Nakasuka
287
In PV I.47,7 which is also a saṃgrahaśloka, he states that it is due to the reason stated
in PV I.46 that Dignāga declares an inference to have exclusion as its object; if an inference
did not have exclusion as its object, it would follow that when the subject of an inference is
known, it would be knowns in its entirety, and therefore there would be no part left to be
inferred.
All these points make the following clear: In PV I.43–47 Dharmakīrti develops an
argument to establish that an inference has for its object exclusion and not a real thing. It is
in this connection that, in PV I.48, Dharmakīrti introduces an argument to establish that a
perceptual judgment also has exclusion as its object.
2. PVSV on PV I.48
2.1 Now let us look closely at what Dharmakīrti says in PVSV on PV I.48. He comments
on the verse as follows:
A: PVSV G28.8–9, M18.15–16: yad rūpādidarśanānantaram aliṅgaṃ niścayajñānaṃ bhavati, tat katham asati samārope bhavad vyavacchedaviṣayaṃ
bhavati /
[Objection] A determinate cognition that arises immediately after the perception of an entity like color is not an inferential cognition (aliṅga). Therefore,
how is it that such a determinate cognition, arising when superimposition is
absent (asati samārope bhavat), has exclusion for its object?8
B: PVSV G28.10–11, M18.16–17: samāropaviṣaye tasyābhāvāt 9 / yatra hy
asya samāropo yathā sthiraḥ sātmaka iti vā, na tatra bhede10 niścayo bhavati /
[Answer] For the very reason that the [determinate cognition] does not arise
with respect to an object of superimposition (samāropaviṣaya), [it is proper
to say that it has for its object the exclusion of superimposition]. Indeed, a
determinate cognition does not arise with respect to the difference (bheda) on
which something is superimposed by the cognizer (asya), as in the case where
[one has an erroneous cognition such as] “[this is] permanent” or “[this has]
an Ātman.”11
7
8
9
10
11
PV I.47: tasmād apohaviṣayam iti liṅgaṃ prakīrtitam / anyathā dharmiṇaḥ siddhāv asiddhaṃ kim ataḥ
param //
PVSVṬ 129.21–24: yad ityādinā ślokaṃ vyācaṣṭe / ādiśabdāc chabdādiparigrahaḥ / nāsya liṅgam
astīty aliṅgaṃ rūpādikam (aliṅgaṃ rūpādikam Ms; iliṅgarūpādikam S.) etad iti niścayajñānam / asati
samārope bhavati / na hi pratyakṣadṛṣṭe rūpādau tadānīm viparītākārasamāropo ’sti / tat kathaṃ
vyavacchedaviṣayaṃ bhavati / iyatā ślokasya pūrvārddho vyākhyātaḥ /
samāropaviṣaye tasyābhāvāt G; samāropaviṣaye ’bhāvāt M. PVSVt and PVṬ: sgro ma btags pa’i yul
la yod pa’i phyir ro (*asamāropitaviṣaye bhāvāt). G’s reading is supported by PVSVṬ: samāropaviṣaye
tasya niścayajñānasyābhāvāt (see note 11). In PVṬ ad PV I.48 we find a passage parallel to G’s reading:
de ni sgro ’dogs pa’i yul la med pa’i phyir (see 2.3.1).
bhede G; bhedo M.
PVSVṬ 129.25–28: uttarārddhaṃ vyākhyātum āha / samāropaviṣaye tasya niścayajñānasyābhāvāt
tadvyavacchedaviṣayaṃ bhavatīti prakṛtena sambandhaḥ / etad evāha / yatra bhede ’sya puṃsaḥ
samāropo na tatra bhede samāropaviṣaye niścayo bhavaty asthiro nirātmaka iti vā //
288
Dharmakīrti’s Apoha Theory
In A Dharmakīrti makes two points. First, immediately after the perception of color there
arises a determinate cognition (niścayajñāna) which is not an inference. Needless to say,
this determinate cognition is what we call perceptual judgment. Second, a perceptual
judgment arises when there is no superimposition (asati samārope bhavad), and thus,
unlike an inference, the perceptual judgment cannot have the exclusion of superimposition
as its object. This second point is extremely important because it is a strong indication
that a perceptual judgment arises when superimposition is absent. Let us note here that
Dharmakīrti holds that if the “exclusion” (vyavaccheda) is that of superimposition as in
the case of an inference, then the exclusion certainly does not obtain; but if it is not, then it
does obtain.
In B Dharmakīrti argues that perceptual judgment does have the exclusion of superimposition as its object, because a perceptual judgment does not arise with respect to
an object of superimposition (samāropaviṣaye tasyābhāvāt). To explain. First of all, an
object of superimposition (samāropaviṣaya) is bheda “difference.” With respect to the
bheda on which (yatra) something is superimposed, there is no determinate cognition.
The superimposition leads to the erroneous cognition “this is permanent” (sthira). Thus
the superimposition in question is that of the property of being permanent. We should,
however, have the determinate cognition “this is impermanent.” Accordingly, what is meant
by the term bheda here must be the property of being impermanent. We must note that
Dharmakīrti holds the property of being permanent to be superimposed on the property
of being impermanent. A real thing has differences (bheda) from all other things, such
as the property of being impermanent. Suppose that a person superimposes the property
of being permanent on the property of being impermanent, resulting in the erroneous
cognition “this is permanent.” In this case the person has no determinate cognition with
respect to the property of being impermanent. In contrast, suppose that a person has a
determinate cognition with respect to the latter property. In this case, the person clearly
has no superimposition with respect to the property of being impermanent. This is how the
determinate cognition in the form of a perceptual judgment is said to have the exclusion of
superimposition as its object. According to Dharmakīrti, one is justified in saying that the
exclusion of superimposition forms the object of a perceptual judgment only insofar as the
perceptual judgment arises only when a superimposition is absent.
2.2 In the latter half of the verse Dharmakīrti says: asamāropitānyāṃśe tanmātrāpohagocaram. It is important to note here that what Dharmakīrti means by the word aṃśa “part”
in PV I.48 is a difference (bheda) such as the property of being impermanent. Naturally
it follows that that which has parts is an object to be perceived, referred to by the phrase
kvacid dṛṣṭe in the verse.
If we accept that the phrase asamāropitānyāṃśe tanmātrāpohagocaram can be paraphrased as asamāropitānyāṃśe niścayajñānam, the expression asamāropitānyāṃśa must
refer to the difference (bheda) that is the object of a determinate cognition. If the difference
is the property of being impermanent, the term anyāṃśa “a different part” must refer to the
property of being permanent, which is different (anya-) from the property of being impermanent. According to Dharmakīrti, the property of being permanent can be superimposed
on the property of being impermanent. Therefore, the expression asamāropitānyāṃśa, as a
Miyuki Nakasuka
289
bahuvrīhi, must refer to the property of being impermanent on which the property of being
permanent is not superimposed. When the property of being permanent is superimposed
on the property of being impermanent, the latter property is concealed by the former and
goes beyond the scope of determination.
2.3 At this point it may be useful to consider how the commentators interpret this compound.
2.3.1 Śākyabuddhi
PVṬ ad PV I.48cd (D63b6–64a1, P75a4–7): rnam pa gang la gzhan gyi cha
ste / zla bo rnam pa sgro ma btags pa de ni gzhan gyi cha sgro ma btags
pa’o // de la ’jug pa de yang de tsam sel ba’i spyod yul can cha sgro ma btags
pa gang yin pa de tsam gyi rnam par gcod pa’i yul can yin no // ’di skad du
mngon sum gyi rjes la ’byung ba’i shes pa rnam par rtog pa can gyis sgro
’dogs pa sngar zhugs pa sel bar mi byed mod kyi / ’on kyang de ni sgro ’dogs
pa’i yul la med pa’i phyir de rnam par gcod pa’i yul can kho na yin no zhes
bya bar shes so // gal te yang rnam par rtog pa can gyi shes pa de dngos po
la12 bsgrub par ’jug par ’gyur na ni / rnam pa thams cad la nges par ’gyur ro
zhes bshad pa yin no //
(a) The compound asamāropitānyāṃśa is a bahuvrīhi, referring to “an image
(x) on which a different part, i.e., an image (non-x) contradictory to the image
(x), is not superimposed.”
(b) The [cognition] which arises with respect to the [image (x)], too, has for
its object the exclusion only of that, i.e., the exclusion only of the part (non-x)
which is not superimposed.
(c1) [What Dharmakīrti means to say] is this: It is not the case that a conceptual
cognition arising immediately after perception excludes a superimposition
which occurred before [this conceptual cognition].
(c2) On the contrary, it is seen that the [conceptual cognition] does have
exclusion as its object because it does not arise with respect to the object of
superimposition.
(c3) If the conceptual cognition arose as what establishes a real thing, then it
would follow that [the conceptual cognition] determines all forms [of the real
thing].
This is what was explained by the teacher.
The compound asamāropitānyāṃśa is a bahuvrīhi formed from asamāropita and anyāṃśa,
referring to an image (rnam pa, *ākāra) (image x). The latter part of the compound,
anyāṃśa, signifies “a different part,” which refers to “an image contradictory to the image”
(zla bo rnam pa, *pratiyogyākāra) (image non-x). Thus the compound in question means
12
la D; om. P.
290
Dharmakīrti’s Apoha Theory
“an image (x) on which an image (non-x) contradictory to it is not superimposed.” The
demonstrative tad in tanmātra refers to the image which is not superimposed (image non-x).
2.3.2 Karṇakagomin
PVSVṬ ad PV I.48cd (128.4–12): (A) asamāropitaḥ anyāṃśaḥ pratiyogyākāro yasmin viṣaye sa tathā / tatra pravarttamānaṃ tad api tanmātrāpohagocaram / tenāyam artho bhavati / samāroparahitaṃ svalakṣaṇaṃ svākārabhedena gṛhṇad 13 vikalpakaṃ jñānaṃ bhrāntatvāt tatsamāroparahitabāhyādhyavasāyakam eva na tu bāhyasvarūpagrāhakam / atas tanmātram eva
niyatabāhyāvasāya evānyasya samāropasyāpohagocaraṃ vikalpakaṃ jñānam /
(B) yadvā asamāropitaś cāsāv anyāṃśaś ca tasmin sati vikalpakaṃ jñānaṃ
pravarttamānaṃ tanmātrāpohagocaram / yo ’sāv asamāropito ’nyāṅśas tanmātravyavacchedaviṣayaṃ bhavati /
(A) [First interpretation:]
(a) The compound asamāropitānyāṃśa is a bahuvrīhi, denoting “an object
(viṣaya) on which a different part, i.e., an image contradictory to it (pratiyogyākāra), is not superimposed.”
(b) The [cognition], which arises with respect to the [object], too, is spoken of
as tanmātrāpohagocara “the cognition whose part is of that [i.e., an external
object] and whose object is exclusion.”
(c1) Therefore, [by the present verse] is meant the following: Since [a conceptual cognition] grasps an individual (svalakṣaṇa) that is devoid of superimposition (samāroparahita) through its own image, it is an erroneous cognition.
Therefore, the conceptual cognition is a mere judgment about the external
object that is devoid of superimposition and does not grasp the external object
in itself.
(c2) Accordingly, [the cognition] is a conceptual cognition in the form of [a
cognition] that has a part related to the [external object] (tanmātram eva), i.e.,
in the form of a judgment about a specific external object (niyatabāhyāvasāya
eva), whose object is the exclusion of what is other, i.e., superimposition.
(B) [Second interpretation:]
(a) Or, the compound asamāropitānyāṃśa is a karmadhāraya, meaning “that
part (non-x) which is different [from what is to be determined] and which is
not superimposed.”
(b) When there is such a part (non-x), a conceptual cognition arises. Accordingly, the conceptual cognition has for its object the exclusion only of such a
part (non-x). That is, [the conceptual cognition has] for its object the exclusion
only of such a part (non-x) which is different [from what is to be determined]
and which is not superimposed.
13
gṛhṇad Ms; gṛhṇan S.
Miyuki Nakasuka
291
Karṇakagomin offers two interpretations for the compound asamāropitānyāṃśa: as a
bahuvrīhi and as a karmadhāraya. In the bahuvrīhi interpretation the compound refers to
an object (viṣaya) on which an image contradictory to the object is not superimposed. The
object is an individual (svalakṣaṇa), which is said to admit of no superimposition. The
expression tanmātra refers to adhyavasāya “judgment,” which is said to be that whose part
(mātrā), i.e., an image, is related to the object.
In the karmadhāraya interpretation the compound refers to a different part that is not
superimposed, i.e., the part (non-x) that is different from the part (x) to be determined and
that is not superimposed on the latter part (x). The demonstrative tad refers to this putative
part (non-x). When such a different part is present (sati), that is, when a different part is not
superimposed on the object with respect to which determination takes place, a perceptual
judgment arises as what excludes only the different part. Accordingly, perceptual judgment
is said to have the exclusion only of the different part as its object.
2.3.3 Manorathanandin
PVV ad PV I.48 (305.11–15): nanu kvacin nīlādāv asamāropito ’nyo viparītāṃśo yasmin tasmin pratyakṣeṇa dṛṣṭe yaj jñānam aliṅgajaṃ vikalpakaṃ
sāmānyaviṣayaṃ bhavati, tad āropābhāvāt katham anyāpohaviṣayaṃ / āha
tanmātrasyānīlamātrasyāpoho vijātīyād vyāvṛttir vyavacchedaḥ sa gocaro
yasya tat tathā / nīlavikalpasyānīlavyavaccheda eva viṣaya ity arthaḥ //
[Objection] When a certain [entity], say, blue, on which a part foreign to it, i.e.,
a part contradictory to it, is not superimposed is cognized through perception,
there arises a cognition (jñāna) which is not based on an inferential mark,
which is conceptual, and which has a universal as its object. How could it
be that such a cognition has exclusion as its object, since no superimposition
occurs?
[Answer] [Dharmakīrti] answers [this objection by stating]: The cognition is
spoken of as “that whose object is the exclusion only of that, i.e., the exclusion
only of non-blue; exclusion means the exclusion (vyāvṛtti = vyavaccheda) of
an entity of a dissimilar class.” The conceptual cognition “this is blue” has as
its object only the exclusion of non-blue. This is what is meant by the verse.
According to Manorathanandin, the compound asamāropitānyāṃśa refers to a perceived,
real blue on which a part foreign to it is not superimposed. The expression tanmātrāpoha
means the exclusion of non-blue. We must say that his interpretation strays quite far from
the earlier interpretations. Manorathanandin fails to grasp the point made by Dharmakīrti
that in the context of a perceptual judgment, the exclusion of superimposition forms the
lack of superimposition.
3. PV I.49 and PVSV
In PVSV B, Dharmakīrti states that a determinate cognition does not exist with respect
to an object of superimposition (samāropaviṣaye tasyābhāvāt). This means, as shown
292
Dharmakīrti’s Apoha Theory
in 2.1, that a determinate cognition does not arise with respect to the property of being
impermanent on which the property of being permanent has been superimposed. In PV
I.49ab, the reason for this point is given.
3.1 bādhyabādhakabhāva
Dharmakīrti states the following:
C (=PV I.49ab):
niścayāropamanasor bādhyabādhakabhāvataḥ /
Because the relation of what is to be blocked and what does the blocking
obtains between a determinate cognition and a superimposing cognition.14
According to Dharmakīrti, the relation of what is to be blocked and what does the blocking
obtains between the determinate cognition “this is impermanent” and the erroneous cognition “this is permanent,” which is based on superimposition. Because of this relationship,
it is said that a determinate cognition does not arise with respect to the property of being
impermanent on which the property of being permanent is superimposed. This point is
explained in PVSV as follows:
D: PVSV G28.13–14, M18.19–20: na hi sarvato bhinno dṛṣṭo ’pi bhāvas
tathaiva pratyabhijñāyate / kvacid bhede vyavadhānasaṃbhavāt / yathā śukteḥ
śuktitve /
Indeed, even if an entity has been perceived as something differentiated from
all others, it is not recognized as it really is. For there can be an impediment
to a certain difference [being cognized], just as there is [an impediment to the
cognition of] mother of pearl’s property of being mother of pearl[, if mother
of pearl is erroneously cognized].15
E: PVSV G28.15–16, M18.20–21: yatra tu pratipattur bhrāntinimittaṃ nāsti
tatraivāsya taddarśanāviśeṣe ’pi smārto niścayo bhavati /
With respect only to the [difference] about which a cognizer has no cause
to have an erroneous cognition, a determinate cognition[, i.e., a perceptual
judgment,] arises for the cognizer based on remembrance even if [all] the
[differences] have been equally perceived.16
14
15
16
PVSVṬ 130.9–10: kiṃ kāraṇam / niścayāropamanasor bādhyabādhakabhāvataḥ (bhāvataḥ Ms; bhāvaḥ
S.) / niścayajñānasya tadviparītasamāropajñānasya ca bādhyabādhakabhāvataḥ /
PVSVṬ 130.10–15: bādhyabādhakabhāvam eva sādhayann āha na hīti / sarvataḥ sajātīyād vijātīyāc
ca bhinno dṛṣṭo ’pi bhāvas tathaiveti yathādṛṣṭena sarveṇākāreṇa pratyabhijñāyate niścīyate na hīti
sambandhaḥ / kiṃ kāraṇam / kvacid bhede kṣaṇikatvādike vyavadhānasambhavāt / *bhrāntinimittagataḥ (read; bhrāntinimittasambhavāt) / yathā śukteḥ sarvato vyāvṛttāyā darśane ’pi śuktikāditve
rajatasādharmyasya bhrāntinimittasya sambhavān na niścayaḥ / *PVṬ: ’khrul pa’i rgyu mtshan srid
pa’i phyir ro.
PVSVṬ 130.15–17: yatra tv ākāre bhrāntinimittaṃ nāsti tatraivāsya pratipattur anubhavottarakālabhāvī
smārto niścayo bhavati / taddarśanāviśeṣe ’pi sarvasvākāreṣu pratyakṣasyāviśeṣe ’pi / smārta iti
smṛtirūpaḥ /
Miyuki Nakasuka
293
D. An entity that is differentiated from all others is a real thing (svalakṣaṇa). A real thing
has differences relative to all other things. Such a real thing, even if it has been perceived,
is not recognized as it really is through a perceptual judgment. This is because a certain
difference of the real thing can be blocked from being cognized.
E. Recall that, according to Dharmakīrti, when a person does not recognize the difference between moments, the person superimposes identity between them and so has the
erroneous cognition “this is permanent.” That a cognizer has no cause of an erroneous
cognition with respect to a difference of a real thing means that the cognizer recognizes
the difference as it is. Therefore, when the cognizer has no cause of an erroneous cognition
with respect to a difference, there is no chance of superimposition arising with respect to
the difference. The point Dharmakīrti makes here is that a determinate cognition arises
only with respect to the difference on which nothing is superimposed.
Now we can see the following. A real thing has differences (bheda) from all other things,
which can also be conceptually posited as parts (aṃśa) of that real thing. On the condition
that no erroneous cognition arises with respect to a real thing, when the real thing is
perceived, all its differences are perceived. A determinate cognition arises only with respect
to a difference that is not erroneously cognized. For instance, the determinate cognition
“this is impermanent” arises only with respect to the property of being impermanent, which
is not the object of an erroneous cognition. When the property of being permanent is
superimposed on the property of being impermanent, the latter property is said to be the
object of an erroneous cognition. Thus it is clear why Dharmakīrti says that a determinate
cognition does not arise with respect to an object of superimposition (samāropaviṣaye
tasyābhāvāt).
3.2 samāropaviveka
3.2.1 In PV I.48d Dharmakīrti speaks of a determinate cognition in the form of a perceptual judgment as tanmātrāpohagocara. The next step is to consider how this expression
should be interpreted.
Consider the following passage:
F1 : PVSV G28.16–19, M18.22–24: samāropaniścayayor bādhyabādhakabhāvāt / niścayasya
samāropaviveke ’sya pravṛttir iti gamyate // [49cd]
tadviveka eva cānyāpohaḥ /
Because there obtains the relation of what is to be blocked and what does the
blocking between superimposition and a determinate cognition, it is understood
that the determinate cognition
occurs when there is a lack of superimposition (samāropaviveka). [49cd]
And this very lack of [superimposition] is the exclusion of others (anyāpoha).17
17
PVSVṬ 130.20–24: yataś ca pratyakṣāviśeṣe ’pi samāroparahita eva viṣaye niścayo bhavati tasmāt
samāropaniścayayor bādhyabādhakabhāvo gamyate / tato bādhyabādhakabhāvāt kāraṇāt samāropavi-
294
Dharmakīrti’s Apoha Theory
According to Dharmakīrti, a determinate cognition arises when there is a lack of superimposition (samāropaviveka). The reason for this is that a determinate cognition and
superimposition are incompatible with each other.
3.2.2 The term samāropaviveka can be paraphrased as samāropābhāva “absence/negation
of superimposition.”18 It is proper to say that viveka is an equivalent for abhāva “nonexistence, absence.” In PV I.56–57a Dharmakīrti, having introduced an opponent’s position
that, after a real thing is grasped in its entirety by a conceptual cognition, a valid cognition
arises to remove an erroneous cognition which prevents one from determining the real
thing as it is, says the following:
PV I.56–57a: yadi bhrāntinivṛttyarthaṃ gṛhīte ’pi anyad iṣyate /
tad vyavacchedaviṣayaṃ siddhaṃ tadvat tato ’param //
asamāropaviṣaye vṛtteḥ
If it is accepted that even if an entity is grasped in its entirety [by a determinate cognition (C1)], another [valid cognition (C2)] occurs to remove an
erroneous cognition [that prevents one from determining the real thing as
it is],19 it is established that the [valid cognition (C2)] has exclusion as its
object. Like the [valid cognition (C2)], [it is established that] the [cognition
(C1)] different from the [valid cognition (C2) has exclusion as its object]. For
[cognition (C1)] occurs with respect to an object that has no superimposition
(asamāropaviṣaya).20
In PVSV Dharmakīrti explains:
F2 : PVSV ad PV I.56cd–57a (G31.12–15, M20.14–16): tat tarhi bhrāntinivṛttyarthaṃ pravṛttaṃ pramāṇam anyasamāropavyavacchedaphalam21 iti
siddham anyāpohaviṣayam / tadvad anyad api / asamāropaviṣaye vṛtteḥ /
yatrāsya samāropo na tatra niścaya iti samāropābhāve22 vartamāno ’nyāpohaviṣayaḥ siddhaḥ /
(a) In that case, it is established that the valid cognition which occurs to remove
an erroneous cognition, bringing about the exclusion of the superimposition
18
19
20
21
22
veke samāropavirahe niścayasyāsya (samāropavirahe niścayasya… Ms; samāropavirahaniścayasya…
S.) pravṛttir iti gamyate / bhavatu nāma samāropaviveke pravṛttis tathāpi nānyāpohaviṣayatvam vidhirūpeṇa pravṛtter ity āha / tadviveka evānyāpohaḥ (read: eva cānyāpohaḥ) samāropaviveka eva
cānyāpohaḥ /
In PVSV Dharmakīrti states that existence is viveka of non-existence, while non-existence is viveka of
existence. PVSV ad PV I.291 (G154.10–11): bhāvābhāvayor anyonyavivekarūpatvāt /
PVSVṬ 139.18–19: yadītyādinā parābhiprāyam āśaṃkate / ekena niścayajñānena sarvātmanā gṛhīte
’pi vastuni bhrāntinivṛttyartham anyad iti pramāṇāntaram /
PVSVṬ 139.23–26: yady evam ityādinā siddhāntavādī / yat tad bhrāntinivṛttyartham uttaraṃ pramāṇam iṣyate tad vyavacchedaviṣayam anyāpohaviṣayaṃ siddhaṃ pūrvoktena nyāyena / tadvad uttarapramāṇavat / tata uttarakālabhāvi pramāṇād (read: uttarakālabhāvipramāṇād) aparam api pūrvakālabhāviniścayajñānaṃ tad api vyavacchedaviṣayam / kiṃ kāraṇam / asamāropaviṣaye vṛtteḥ /
anyasamāropavyavacchedaphalam G; anyavyavacchedaphalam M.
samāropābhāve G; samāropābhāve ca M.
Miyuki Nakasuka
295
of a different [image] (vyavacchedaphala), has as its object the exclusion of
others (anyāpoha).
(b) In the same manner, it is established that, like the [valid cognition], another
[cognition], [i.e., the preceding determinate cognition,] also [has as its object
the exclusion of others]. For it occurs with respect to an object which is devoid
of superimposition.
(c) Since a determinate cognition does not arise with respect to an object
on which something is superimposed by a cognizer, it is established that the
determinate cognition, occurring when there is an absence of superimposition
(samāropābhāva), has as its object the exclusion of others.23
An opponent makes the following points. First, a conceptual cognition grasps a real thing in
its entirety. Second, a valid cognition can arise in order to remove an erroneous cognition
that is assumed to occur with respect to the real thing. According to Dharmakīrti, in this
case it can be said that the valid cognition has exclusion as its object, since the valid
cognition brings about the exclusion of the superimposition of something foreign to the
real thing. What is important is that Dharmakīrti points out that, like a valid cognition, a
perceptual judgment can also be said to have exclusion as its object. It is imperative to
note here that the reason the perceptual judgment is said to have exclusion as its object is
held to be that it occurs with respect to an object which is devoid of superimposition, or
that it occurs when there is absence of superimposition with respect to the object. In this
connection, let us recall the passage in B: samāropaviṣaye tasyābhāvāt. There it was said
that no determinate cognition arises with respect to an object of superimposition which
consists in a difference. In the above-cited passage (F2 b), on the other hand, it is stated that
a determinate cognition arises with respect to an object which is devoid of superimposition.
It is clear that the reason a perceptual judgment is described as having exclusion as its
object must be found in the way the object of the perceptual judgment should be: the object
is far from being what something is superimposed on. To put it in another way, it is because
its object lacks superimposition that the perceptual judgment is defined as having exclusion
as its object.
3.3 tanmātra
In PV I.48 Dharmakīrti has defined a determinate cognition as tanmātrāpohagocara instead
of defining it as anyāpohaviṣaya. A question arises: What is meant by the term tanmātra?
The following passage offers a key to solving this problem.
G: PVSV ad PV I.49cd (G28.19–22, M18.24–26): tasmāt tad api tanmātrāpohagocaram / na vastusvabhāvaniścayātmakam / tathā hi kasyacin niścaye ’py
anyasyāpratipattidarśanāt / tatsvabhāvaniścaye ca tasyāyogāt /
23
PVSVṬ 140.11–15: tat tarhītyādinā ślokaṃ vyācaṣṭe / anyasyākārasya yaḥ samāropas tad vyavacchedaphalam iti kṛtvā siddham anyāpohaviṣayam utpitsusamāropaniṣedhadvāreṇa / tadvad anyad api
pūrvam api niścayajñānam anyāpohaviṣayam / kiṃ kāraṇam / avidyamānasamārope viṣaye vṛtteḥ /
etad evāha / yatrākāre ’sya *pratipatteḥ (read: pratipattuḥ) / iti hetoḥ samāropābhāve varttamānaḥ
paurastyo niścayo ’nyāpohaviṣayaḥ siddhaḥ // *PVṬ: ’di’i zhes bya ba ni rtogs par byed pa po’i’o.
296
Dharmakīrti’s Apoha Theory
Therefore, the [determinate cognition in the form of a perceptual judgment]
also has as its object the exclusion only of that [putative specific difference]. It
does not consist in a determinate cognition which determines the real thing
in itself. For [the determinate cognition is defined] in that manner because it
is seen that, even if a certain difference is determined, another difference is
not understood and because, when the real thing in itself is determined, it is
improper to say that [another difference is not determined].24
It is clear that the term mātra makes a restriction on the differences the real thing has. In B
it has been stated that a determinate cognition does not arise with respect to a difference on
which something is superimposed, and in E that a determinate cognition arises only with
respect to a difference which does not fall under the scope of an erroneous cognition. In
PV I.48c it is stated that a determinate cognition occurs with respect to the part on which a
part different from it is not superimposed.
All this makes the following clear: Suppose that a given real thing has the property of
being impermanent as its part or difference. When a determinate cognition occurs with
respect to this property, the property is that on which the property of being permanent
is not superimposed, that is, what has no superimposition of this latter property. Thus
the absence of superimposition must be that of the superimposition only of this property.
What is meant by the expression tanmātrāpoha is the absence (viveka, abhāva) of the
superimposition only of that part or difference which is different from the part or difference
to be determined. There is no doubt that the demonstrative tad in tanmātra- refers to the
part which is different from the part to be determined (anyāṃśa).
4. PV I.48: reinterpretation
Now we are in a good position to interpret PV I.48.25
24
25
PVSVṬ 130.24–29, 131.7–9: tasmāt samāroparahite vṛttivaśāt / tad apīti na kevalam anityaḥ śabda iti
niścayajñānaṃ pūrvoktena nyāyena tanmātrāpohagocaram, tad api pratyakṣapṛṣṭhabhāviniścayajñānam (-niścayajñānam Ms; niściyajñānam S.) api tanmātrāpohagocaraṃ na vastusvabhāvaniścāyakaṃ
(read: -niścayātmakaṃ) svarūpeṇa / kiṅ kāraṇam / tathā hi kasyacid ākārasya rūpatvāder niścaye ’py
anyasya kṣaṇikatvādyākārasyāpratipattidarśanāt / yadi tu pratyakṣapṛṣṭhabhāvinā niścayena (niścayena Ms; niṣcayena S.) vastusvabhāvasya niścayaḥ kriyate tadā tatsvabhāvaniścaye ca niraṃśatvād
vastunas tasyāyogād anyasyākārāntarasyāniścayāyogāt //
This verse has already been interpreted as follows: Mookerjee and Nagasaki 1964: 108: “Now when in
respect of a perceived datum a determinate judgement arises (in its trail) referring to a conceptualized
construction without the obtrusion of the opposite misconception, it also serves to eliminate the
unperceived misconception (or doubt) of the opposite of that very datum.” Steinkellner 1971: 199, n. 71:
“Die vorstellende Erkenntnis, welche ein Allgemeines zum Gegenstand hat, richtet sich, wenn irgendein
[Ding] wahrgenommen wurde, ohne daß [auf dieses Ding] ein Teil eines andern [Dinges] übertragen
worden ist, bloß darauf, diesen [fremden Teil] fernzuhalten …” Zwilling 1976: 96: “Even in the absence
of an opposing misconception the determinate cognition which follows the direct cognition of some real
and whose prime object is a [constructed] general nature nevertheless has for its sphere of operation
the elimination of just that [absent misconception.]” Okada 2007: 4, 6: [シャーキヤブッディ解
釈]知覚された何らかのものに関しても、およそ(直接知覚の後に生じる)認識は、普遍
を対象としており、分別を有している。他の部分(=対立項の形象)が付託されていない
(形象)に対して(働くその認識は)、それ(=付託されていない他の部分)のみを排除
Miyuki Nakasuka
297
PV I.48: kvacid dṛṣṭe ’pi yaj jñānaṃ sāmānyārthaṃ vikalpakam /
asamāropitānyāṃśe tanmātrāpohagocaram //
Even when a certain [real thing] has been perceived, a cognition which has a
universal as its object and which is conceptual arises; [this cognition, arising]
with respect to the [real thing’s part (x)] on which a part (non-x) different
[from it] is not superimposed, [is characterized as] what has as its object the
exclusion only of that [part non-x].
Dharmakīrti is saying this: A conceptual cognition arising immediately after the perception
of a real thing, that is, a perceptual judgment, has for its object only one part or difference
of that real thing, say, the property of being impermanent (x); if this part or difference is
determined, then it is one on which a putative part or difference contradictory to it, say,
the property of being permanent (non-x), is not superimposed. In this case the perceptual
judgment is said to have as its object the exclusion of the superimposition only of the
putative part or difference, or, in other words, the exclusion only of the putative part or
difference.
Thus the reason that a perceptual judgment is described as that which has exclusion as
its object lies in the fact that the object of the perceptual judgment is what is devoid of the
superimposition that gives rise to an erroneous cognition with respect to that object. We
may say that Dharmakīrti establishes that perceptual judgment has exclusion as its object
の活動領域とする。 [カルナカゴーミン解釈]知覚された何らかのものに関しても、およ
そ(直接知覚の後に生じる)認識は、普遍を対象としており、分別を有している。付託さ
れていない他の部分がある時、(分別知は)それ(=付託されていない他の部分)のみを
排除の活動領域とする。 [シャンカラナンダナ解釈]他の部分が未だ付託されていない何
らかの知覚された(対境)に関しても、およその分別知は、(知覚されたものに限定され
る)普遍を対象としており、「それのみ(=未来に付託される可能性のある他の部分)」
を排除の活動領域としている。“[Śākyabuddhi’s interpretation:] With respect to something that
has been perceived, too, any cognition [that arises immediately after perception] has for its object an
universal and has conceptualization. [Such a cognition, which functions towards] an image on which the
part different from the image (= an image conflicting with it) is not superimposed, has for its domain of
exclusion only that (= the different part which is not superimposed). [Karṇakagomin’s interpretation:]
With respect to something that has been perceived, too, any cognition [that arises immediately after
perception] has for its object an universal and has conceptualization. When there is a different part
which is not superimposed, (a conceptual cognition) has for its domain of exclusion only that (= the
different part which is not superimposed). [Śaṅkaranandana’s interpretation:] With respect to a certain
(object) which has been perceived and on which the part different from it is yet to be superimposed,
too, the conceptual cognition, whatever it be, has for its object an universal (qualified by the perceived
object) and has for its domain of exclusion ‘only that (= the different part that has the possibility of being
superimposed in the future)’.” (English translation mine.) Fukuda 2011: 62: ある[対象が現量によっ
て]知覚された場合であっても、普遍を対象とした分別知が他の属性(aṃśa)の増益され
ていない[対象]に対して、tanmātrāpohaを活動領域(gocara)として[生じる]。 (“Even
in the case where a certain [object] has been perceived [through perception], a conceptual cognition
whose object is a universal [arises], having for its domain tanmātrāpoha, with reference to an [object]
on which another property (aṃśa) has not been superimposed.”) (English translation mine.) Eltschinger
et al. 2018: 43: “The conceptual cognition [arising] with reference to something [just] perceived on
which an extraneous element (anyāṃśa) has not been superimposed [and] which has a universal for its
object, also pertains to an exclusion[, namely, the exclusion] of just this [non-superimposed extraneous
element].”
298
Dharmakīrti’s Apoha Theory
from the point of view of the object of a perceptual judgment. By contrast, he establishes
that inference is said to have exclusion as its object from the point of view of how one can
remove an erroneous cognition which can occur with reference to a real thing.
5. Conclusion
In PV I.48 Dharmakīrti tells us that in the case of a perceptual judgment, the occurrence of
determination implies the absence of superimposition. To put it in another way, perceptual
judgment and superimposition are incompatible with each other. Unlike in the case of an
inference, in the case of a perceptual judgment there is no chance of a prior superimposition.
Dharmakīrti defines a perceptual judgment as having exclusion as its object in the sense
that the perceptual judgment arises with respect to the difference which a real thing is
supposed to bear and on which no putative contradictory difference has been superimposed.
According to Dharmakīrti, a perceptual judgment is described as “what has exclusion as
its object” because its object is devoid of superimposition. In Dharmakīrti’s theory of
perception, perceptual judgment plays a critical role in sustaining the practical efficacy of a
perceptual cognition. By defining perceptual judgment as having as its object the absence
of superimposition, he tries to show that perceptual judgment, distinguished from a mere
erroneous cognition, is a trustworthy cognition (avisaṃvādijñāna). The great significance
Dharmakīrti attaches to perceptual judgment is clearly seen in his interpreting apoha as
“lack of superimposition.”
References and abbreviations
Eltschinger et al. 2018 V. Eltschinger, J. Taber, M.T. Much, and I. Ratié, Dharmakīrti’s
Theory of Exclusion (apoha). Tokyo 2018.
Fukuda 2011 Y. Fukuda, Dharmakīrti to anyāpoha. Indo Ronrigaku Kenkyū 2 (2011)
57–71.
Katsura 1984 Sh. Katsura, Dharmakīrti’s theory of truth. Journal of Indian Philosophy
12 (1984) 215–235.
Katsura 1989 Sh. Katsura, Chikaku handan・giji chikaku・sezoku chi. In: Fujita
Koutatshu Hakase Kanreki Kinen Ronshū Indo Tetsugaku to Bukkyō. Kyoto 1989,
533–553.
Katsura 1993 Sh. Katsura, On perceptual judgement. In: South Asian Studies Papers, No.
5, Studies on Buddhism in Honour of Professor A. K. Warder, ed. N.K. Wagle and F.
Watanabe. Toronto 1993, 66–75.
Kellner 2004 B. Kellner, Why infer and not just look? Dharmakīrti on the Psychology of
inferential processes. In: The Role of the Example (Dṛśṭānta) in Classical Indian
Logic, ed. Sh. Katsura and E. Steinkellner. Vienna 2004, 1–51.
Mookerjee and Nagasaki 1964 S. Mookerjee and H. Nagasaki, The Pramāṇavārttikam of
Dharmakīrti, An English Translation of the First Chapter with the Autocommentary
and with Elaborate Comments. Patna 1964.
Okada 2007 K. Okada, Chokusetsu chikaku go no kettei chi no haijotaishō ni tuite. Nihon
Seizou Gakkai Kaihō 53 (2007) 3–10.
Miyuki Nakasuka
299
Pind 2015 O.H. Pind, Dignāga’s Philosophy of Language. Dignāga. Pramāṇasamuccayavṛtti V on anyāpoha. Part I: Text. Part II: Translation and Annotation, ed. E.
Steinkellner. Vienna 2015.
PV I Pramāṇavārttika, Chapter 1 (Svārthānumāna): See PVSV.
PVSV(G) Pramāṇavārttikasvavṛtti: The Pramāṇavārttikam of Dharmakīrti, ed. R. Gnoli.
Roma 1960.
PVSV(M) Pramāṇavārttikasvavṛtti: Svārthānumāna-pariccheda by Dharmakīrti, ed. D.
Malvania. Varanasi 1959.
PVSVt Pramāṇavārttikasvavṛtti (Tib.): D 4216, P 5717.
PVSVṬ Pramāṇavārttikasvavṛttiṭīkā: Karṇakagomin’s Commentary of Pramāṇavārttikavṛtti of Dharmakīrti, ed. R. Sāṅkṛtyāyana. Allahabad 1943 (repr. 1982).
PVṬ Pramāṇavārttikaṭīkā (Tib.): D 4220, P 5718.
PVV Pramāṇavārttikavṛtti: Dharmakīrti’s Pramāṇavārttika with a Commentary by Manorathanandin, ed. R. Sāṅkṛtyāyana. Patna 1938–1940.
Stcherbatsky 1932 Th. Stcherbatsky, Buddhist Logic. Vol. 1. St. Petersburg 1932 (repr.
1994).
Steinkellner 1971 E. Steinkellner, Wirklichkeit und Begriff bei Dharmakīrti. Wiener Zeitschrift für die Kunde Südasiens 15 (1971) 179–211.
Zwilling 1976 L. Zwilling, Dharmakīrti on apoha, the ontology, epistemology and semantics of negation in the Svārthānumānapariccheda of the Pramāṇavārttikam. PhD
dissertation, University of Wisconsin-Madison. Madison 1976.
Dharmakīrti’s Notion of Permanence and Its Impact on the
Tibetan Buddhist Doctrine of Buddhahood
by
Hiroshi Nemoto
Introduction
The purpose of this study is to examine Dharmakīrti’s notion of permanence (nitya), as
found in his Pramāṇavārttika II 204cd, and its impact on the Tibetan Buddhist doctrine
of Buddhahood.1 As is well known, Dharmakīrti holds the view that whatever exists is
momentary (yat sat tat kṣaṇikam eva), which literally means that everything is impermanent.
This view is accepted by later Indian commentators, as well as by many of his Tibetan
successors like Sa skya paṇḍita (1182–1251) and Glo bo mkhan chen (1456–1532), who
belong to the Sa skya pa.2
However, Tsong kha pa blo bzang grags pa (1357–1419), the founder of the Dge lugs
pa, and his followers have a different opinion. They consider that Dharmakīrti himself
admits the existence of the permanent when he says: “Wise men speak of the thing that
itself does not disintegrate as the permanent” (PV II 204cd: nityaṃ tam āhur vidvāṃso
yaḥ svabhāvo na naśyati). Moreover, the Dge lugs pa scholars assert that to be permanent
(rtag pa) does not necessarily mean to be always existing (dus thams cad pa), and hence
that there are permanent entities that exist only temporarily (res ’ga’ ba). Such an idea of
permanence is peculiar to Tsong kha pa and his followers, and it plays an important role
especially in their analysis of Buddhahood or the Nature Body (svābhāvikakāya; ngo bo
nyid sku) of a Buddha.
In what follows, I would like to discuss how Dharmakīrti’s notion of permanence
has survived in the Dge lugs pa’s exposition of Buddhahood. After a short remark about
the Indian concept of permanence, I will give an analysis of the Dge lugs pa’s idea of
permanence, together with the interpretation of Pramāṇavārttika II 204cd as found in the
Dbu ma rgyan zin bris, Tsong kha pa’s memorandum on the Madhyamakālaṃkāra.3 I will
then move on to examine the Dge lugs pa’s analysis of the Nature Body, which is given in
their commentaries on Chapter VIII of the Abhisamayālaṃkāra.
1 The concept of permanence in India
The concept of permanence (nitya) is found in ancient Indian thought. For instance, Patañjali
(2nd cent. BCE), in his Mahābhāṣya, defines permanence in several ways. Patañjali says that
1
2
3
The Dge lugs pa’s concept of “permanence” (rtag pa) is briefly discussed in Nemoto 2009. The aim of
this paper, then, is to show how the concept plays an important role in their analysis of Buddhahood.
See Dreyfus 1997: 79ff.; Nemoto 2011: 368f.
The title of the text is given as: Dbu ma rgyan gyi zin bris rjes rang gis gnang ba, which means “A
memorandum on the Madhyamakālaṃkāra given by the Lord [Tsong kha pa] himself,” so that we may
tentatively assume that the text was written by Tsong kha pa himself.
Birgit Kellner et al., eds., Reverberations of Dharmakīrti’s Philosophy: Proceedings of the Fifth International Dharmakīrti
Conference Heidelberg, August 26 to 30, 2014. Vienna 2020, pp. 301–311.
302
Dharmakīrti’s Notion of Permanence
the term nitya is applied to immovable (kūṭastha), or invariable things (avicālin); and he
says that it also means continued repetition (ābhīkṣṇya).4 Furthermore, he remarks: “That
in which its identity is not destroyed is also permanent” (tad api nityaṃ yasmiṃs tattvaṃ na
vihanyate).5 We notice here that what is expressed by the last definition is almost the same
as what Dharmakīrti says in Pramāṇavārttika II 204cd. It is highly likely that Dharmakīrti
was fully aware of the definition of permanence as found in the Mahābhāṣya.
In Buddhism, specifically in the Abhidharma tradition, factors (dharma) are classified into two categories, the conditioned (saṃskṛta) and unconditioned (asaṃskṛta), of
which unconditioned factors are considered to be permanent.6 Vasubandhu’s Abhidharmakośa teaches that there are three unconditioned factors: space (ākāśa), extinction through
discernment (pratisaṃkhyānirodha), and extinction not through discernment (apratisaṃkhyānirodha). According to Yaśomitra’s commentary, these three factors are “permanent
since they do not pass through the [three] time periods [of future, present, and past]”
(adhvasaṃcārābhāvāt nityāḥ).7
In the Mahāyāna tradition, the concept of permanence plays an important role, especially in the doctrine of Buddhahood and the Buddha Body. The Uttaratantra says
that Buddhahood (buddhatva) is permanent and already present in every sentient being.8
Concerning this, the Vyākhyā explains that Buddhahood is “permanent because it is free
from origination” (utpādavigamān nityam).9 In the same vein, Vasubandhu’s Sūtrālaṃkārabhāṣya states that a Buddha’s Nature Body is essentially permanent (prakṛtyā nityatā
svābhāvikasya).10 According to Sthiramati’s commentary, this is because the Nature Body
4
5
6
7
8
9
10
MBh I, 6.17f.: nityaparyāyavācī siddhaśabdaḥ / kathaṃ jñāyate / yat kūṭastheṣv avicāliṣu bhāveṣu
vartate / (“The word siddha is a synonym of nitya. How is it understood? It is used to refer to things
which are immovable and invariable”); MBh I, 7.3f.: ayaṃ khalv api nityaśabdo nāvaśyaṃ kūṭastheṣv
avicāliṣu bhāveṣu vartate / kiṃ tarhi / ābhīkṣṇye ’pi vartate / (“Furthermore, the word nitya is not
always used to refer to things which are immovable and invariable. Then what does it mean? It is also
used to refer to continued repetition”).
MBh I, 7.22.
AK I 48b: nityā dharmā asaṃskṛtāḥ / (“Unconditioned factors are permanent”).
AKV I, 90.23: asaṃskṛtā eva nityā ity avadhāraṇam / adhvasaṃcārābhāvāt nityāḥ / (“It is ascertained
that only unconditioned [factors] are permanent. [They are] permanent since they do not pass through
the [three] time periods [of future, present, and past]”).
UT 84.7ff. (II 29): acintyaṃ nityaṃ ca dhruvam atha śivaṃ śāśvatam atha praśāntaṃ ca vyāpi vyapagatavikalpaṃ gaganavat / asaktaṃ sarvatrāpratighaparuṣasparśavigataṃ na dṛśyaṃ na grāhyaṃ śubham
api ca buddhatvam amalam // (“Buddhahood is inconceivable, permanent, stable, quiescent, everlasting,
calm, all-pervading, free from conceptualization, and akin to space; it has neither attachment nor hindrance in all respects and is devoid of rough sensation; it cannot be perceived or grasped; furthermore,
it is auspicious and immaculate”).
UT 84.20 (II 34).
MSABh 46.12ff. (ad MSA IX 66): teṣu ca triṣu kāyeṣu yathākramaṃ trividhā nityatā viditavyā yena
nityakāyās tathāgatā ucyante / prakṛtyā nityatā svābhāvikasya svabhāvena nityatvāt / asraṃsanena
sāṃbhogikasya dharmasaṃbhogāvicchedāt / prabandhena nairmāṇikasyāntardhāya* punaḥ punar
nirmāṇadarśanāt / (“It is to be understood with respect to these three bodies, respectively, that there are
three types of permanence, on account of which Tathāgatas are said to have permanent Bodies. [1] The
Nature Body is essentially permanent because it is permanent by nature. [2] The Enjoyment Body is
[permanent] in terms of non-cessation because its enjoyment of good qualities is devoid of interruption.
[3] The Manifestation Body is [permanent] in terms of continuity because, after having hidden itself, it
displays manifestation repeatedly.”) *-ntarvyaye; read -ntardhāya in accordance with Lévi’s note.
Hiroshi Nemoto
303
is free from origination and cessation.11 Thus, apart from the fact that the Buddha teaches
the doctrine of impermanence of all conditioned phenomena, these Buddhist thinkers hold
that there do exist unconditioned factors, which are permanent.
2 Dignāga and Dharmakīrti on permanence
But turning to the pramāṇa school, we find that both Dignāga and Dharmakīrti take a
different position on the issue. It is true that Dignāga speaks of permanent things, such as
space, when he gives examples of various formulations. This, however, does not mean that
Dignāga himself acknowledges the existence of the permanent. For example, let us look at
the following formulation given by Dignāga:12
(Proposition:) Sound is impermanent.
(Reason:) Because it is produced by human effort.
(Similar example:) Whatever is produced by human effort is impermanent,
just like a pot.
(Dissimilar example:) Whatever is permanent is not produced by human effort,
just like space.
One may suspect that the last statement presupposes the existence of permanent entities. But
Dignāga clearly says that the sentence showing a dissimilar example should be interpreted
as a non-affirming negation (prasajyapratiṣedha), and not as an implicative negation
(paryudāsa), so that the sentence does not imply the existence of permanent entities.13
Therefore, he says that a dissimilar example is established without needing to accept the
existence of permanent entities. It seems that Dharmakīrti takes a similar position on this
issue.14 He says:
11
12
13
14
VBh D 138a7f.: de la rang bzhin rtag pa ni chos kyi sku ste / chos kyi sku ni rang bzhin gyis skye ’gag
med pa’i rang bzhin yin pa’i phyir ro // (“Among them, what is essentially permanent refers to Nature
Body, since Nature Body is an entity that is free from origination and cessation by nature”).
See Katsura 2004: 143, n. 16: sādharmyeṇa tāvad anityaḥ śabdaḥ prayatnāntarīyakatvāt / yad dhi
prayatnāntarīyakaṃ tad anityaṃ dṛṣṭaṃ yathā ghaṭa iti / vaidharmyeṇa nityam aprayatnāntarīyakaṃ
dṛṣṭaṃ yathākāśam iti /.
PSVK 148b2f. (cf. Iwata 2004: 99; Katsura 2004:153): de lta na snga ma la ni ma yin la phyi ma la ni
med par dgag pa yin no zhes smras pa yin no // de ltar na rtag pa khas ma blangs kyang chos mi mthun
pa’i dpe grub pa yin no //. See Katsura 2004: 153, n. 26: evaṃ ca pūrvatra paryudāsaḥ uttaratra tu
prasajyapratiṣedha ity uktaḥ / evaṃ ca nityānabhyupagamasyāpi vaidharmyadṛṣṭāntaḥ siddhaḥ /.
See Sa paṇ’s comments on this issue. Rigs gter rang ’grel 100.26ff.: chos kyi grags pas skabs ’ga’ zhig
tu nam mkha’ rtag par gsungs pa ni / ṭī ka byed pa kha cig na re rdul phran dang / nam mkha’ rtag par
gsungs pa la sogs pa rang nyid mi bzhed pa de dag phal che ba gzhan gyi bsam pa la dper brjod pa yin
no // zhes gsungs la / de ltar yang ’gal ba med kyi ’on kyang mi rtag pa log pa tsam la rtag par sgro
btags nas gsungs kyi rtag pa nyid ni ma yin te / (“Concerning the fact that Dharmakīrti in some places
speaks of the permanence of space, a certain commentator [lit. a certain author of the ṭīkā] explains as
follows: ‘Most of the statements about the permanence of atoms, space, and so forth, which are not
acknowledged by himself, are what illustrate [permanence] in accordance with the thoughts of others.’
Although there is no contradiction in such an explanation, [we think that] he speaks of them by reifying
permanence upon the mere absence of impermanence, and that he never [speaks of] permanent things
themselves”).
304
Dharmakīrti’s Notion of Permanence
nityaṃ tam āhur vidvāṃso yaḥ svabhāvo na naśyati //
tyaktvemāṃ hrepaṇīṃ dṛṣṭim ato nityaḥ sa ucyatām / (PV II 204cd–205ab)15
Wise men speak of the thing that itself does not disintegrate as the permanent.
For this reason, after rejecting this shameful view, you must say that it [i.e.,
pudgala] is permanent.
Dharmakīrti here refutes the Vātsīputrīya’s notion of the pudgala, the personal entity that
performs action and that receives pleasure and pain. The Vātsīputrīya school asserts that
the pudgala is neither permanent nor impermanent.16 Dharmakīrti then argues that, if the
pudgala were not impermanent, it would be free from disintegration, which implies that
the pudgala is permanent since wise men consider that which does not disintegrate to be
permanent.17 Thus, Dharmakīrti applies hypothetical reasoning to refute the Vātsīputrīya’s
view. He never questions whether permanent entities really exist or not.
3 Tsong kha pa on permanence
Tsong kha pa, however, regards the passage in question as conveying Dharmakīrti’s own
view on permanence. According to Tsong kha pa, Dharmakīrti not only affirms the existence
of permanent entities but also gives the clear definition as approved by wise men (mkhas
pa rnams), which is contrasted with the one accepted by foolish men (skye bo blun po). Let
us first examine the foolish men’s view on permanence. Tsong kha pa summarizes their
view as follows:
de yang rtag par ’dod phyin chad dus snga ma {na}* yod tshad gang yin dus
phyi ma na’ang yod par ’dod pa ni / mu stegs kyi ’dod pa’i rtag pa’i don yin
la / skye bo blun po’i rtag par ’dzin tshul yang de yin te rtag pa’i don ’di ltar
byed pa ni rang sde la bye brag tu smra ba ma gtogs pa / mdo sde pa dang
sems tsam pa dang / dbu ma pa su yang mi ’dod do // (Dbu ma rgyan zin bris
13b4ff.) *ni Zhol; read na.
15
16
17
Cf. Vetter 1984: 104.
Cf. Thar lam gsal byed 187a4f.: gnas ma bu pa na re / kho bo cag la nyes pa med de / gang zag de
rtag mi rtag gang du yang brjod du med par ’dod pas so zhe na / (“The Vātsīputrīya says: There is
no fault in our opinion, since we assert that the pudgala cannot be expressed as either permanent or
impermanent”).
Cf. Thar lam gsal byed 187a5ff.: gang zag gang de ni chos can / […] rtag pa yin par thal / bcings grol
gyi gzhi gang zhig ’jig med yin pa’i phyir / khyab par thal / rang gi rang bzhin ’jig pa med pa’i chos
de la mkhas pa rnams rtag pa zhes brjod pa’i phyir / ’dis ’jig pa log pa’i rang bzhin gcig rtag par
’thad pa bstan nas rtag pa yin na dngos pos khyab pa dang / gzhi ma grub rtag par ’dod pa legs par
bkag go // (“It follows that the subject, the pudgala, is permanent because it is the support for bondage
and liberation and because it is devoid of disintegration. It follows that there is entailment because
wise men speak of the factor that itself does not disintegrate as the permanent. Through this [statement,
Dharmakīrti], by saying that it is proper to consider the nature of the absence of disintegration to be
permanence, correctly refutes the assertion that if something is permanent, it must necessarily be a
functioning thing, and the assertion that what is not existent [lit. ‘what is not established as a basis’] is
permanent”).
Hiroshi Nemoto
305
Furthermore, with respect to whatever is considered to be permanent, [some
people] assert that whatever existed formerly exists at a later time, too. But
[what is stated here] is the meaning of “permanence” as asserted by nonBuddhist schools. And this is nothing but the way foolish men apprehend
[the meaning of] “permanence.” Apart from the Vaibhāṣika, others in our
[Buddhist] schools – the Sautrāntika, Cittamātra, and Madhyamaka schools –
never assign such a meaning to “permanence.”18
What Tsong kha pa has in mind here is Śāntarakṣita’s argument in the Madhyamakālaṃkāra 2–3,19 where the unity of permanent entities, as postulated by the Vaiśeṣika and
the Vaibhāṣika, is negated. The Vaiśeṣika school asserts īśvara to be a single permanent
entity. The Buddhist Vaibhāṣika school asserts that unconditioned factors are permanent
and indivisible. Both the Vaiśeṣika and Vaibhāṣika schools consider that to be permanent
means to be always existent without losing unity or singularity, as represented by Tsong kha
pa’s phrase: “whatever existed formerly exists at a later time.” However, according to Tsong
kha pa, this is nothing but the way foolish men understand the meaning of “permanence.”
He says that such a view on the permanent is not acceptable to the Sautrāntika, Yogācāra,
and Madhyamaka schools.
Then Tsong kha pa moves on to explain the wise men’s view on permanence. He says
as follows:
des na rnam ’grel las 〈/〉 gang gi rang bzhin ’jig med pa // de la mkhas
rnams rtag ces brjod / ces gsungs pa ltar ’jig pa med pa’i chos ni rtag pa’i
don yin gyi sngar ltar mi ’dod do // (Dbu ma rgyan zin bris 13b6f.)
Therefore, the meaning of “permanence” is identified with the property of not
being subject to disintegration, as stated in the Vārttika: “Wise men speak of
the thing that itself does not disintegrate as the permanent.” [The meaning of
that] stated before is, on the contrary, not accepted [by wise men].20
Here Tsong kha pa, citing the passage from the Pramāṇavārttika, gives another definition
of permanence, which he says is accepted by the Sautrāntika, Yogācāra, and Madhyamaka schools. In short, Tsong kha pa says that to be permanent means to not be subject
to disintegration. To make this point clear, let us consider the example of “emptiness”
(stong nyid). By definition, emptiness is immutable and unchangeable. It is not subject
18
19
20
Cf. Thub bstan lhun po’i mdzes rgyan [cha–ka] 66b1f., 79a2 (Klein 1991: 140); Gcig du bral gyi rnam
bzhag 21b4f.
MAl vv. 2–3: ’bras bu rim can nyer sbyor bas // rtag rnams gcig pu’i bdag nyid min // ’bras bu re re tha
dad na // de dag rtag las nyams par ’gyur // bsgoms las byung ba’i shes pa yis // shes bya ’dus ma byas
smra ba’i // lugs la’ang gcig min de dag ni // rim can shes dang ’brel phyir ro // (“Permanent things
cannot have a single nature, since they bring about effects in succession. Suppose that their effects
existed at different moments respectively, then the [Vaiśeṣika’s] assertion that they are permanent would
be inappropriate. Even those objects known by the cognition arisen from meditation, which are [said to
be] unconditioned factors according to the [Vaibhāṣika’s] doctrine, cannot be unitary since they are
related to a series of cognitions arising in succession”).
Cf. Thub bstan lhun po’i mdzes rgyan [cha–ka] 80b2f. (Klein 1991: 144); Gcig du bral gyi rnam bzhag
21a1f.
306
Dharmakīrti’s Notion of Permanence
to disintegration, so that it is permanent. To be sure, the emptiness possessed by p exists
only when the property-possessor p exists; and the emptiness of p disappears when the
property-possessor p disappears. But, according to Tsong kha pa, it does not follow that
the emptiness possessed by p is impermanent. He remarks:
rang bzhin rnam dag gi dbang du byas na chos can snga ma’i stong nyid chos
de log pa na ldog pas / chos de med pa’i dus na med pas dus res ’ga’ ba yin
kyang mi rtag par mi ’gyur te / dgag bya rnam par bcad tsam gyi med dgag yin
pa’i phyir dang / khyad gzhi log pa na khyad chos ldog pa’i phyir ro // (Dbu
ma rgyan zin bris 14a3ff.)
With reference to [the emptiness, which is] innately pure, [we can point out
the following things]: The emptiness of a property-possessor [p] that existed
before disappears when the factor [p] disappears, and it is nonexistent when
that factor [p] is nonexistent; therefore, it exists only temporarily. However, it
does not follow that it is impermanent because it is a non-affirming negation
that is [recognized] merely by eliminating an object of negation, and because
[it is natural to say that] the attribute should disappear when the basis of the
attribute disappears.21
To sum up, Tsong kha pa says that the emptiness possessed by p is permanent because it is a
non-affirming negation (med dgag). But what does it mean to be a non-affirming negation?
Tsong kha pa uses the term “non-affirming negation” to refer to a thing that is posited as
present only by way of eliminating an object of negation (dgag bya), without affirming the
existence of other factors.22 The emptiness of p is recognized only by eliminating an object
of negation, that is, p’s intrinsic existence. Nothing else is affirmed to be present through
that process of negation. It is for this reason that emptiness is said to be a non-affirming
negation.
This brings us to the second point. The emptiness of p as such cannot be perceived directly, but it is cognized only by means of eliminating another factor, p’s intrinsic existence;
it has a secondary existence imputed by conceptual consciousness. In other words, the
emptiness of p is a conceptual construct devoid of origination and disintegration. It is true
that the attribute, emptiness, disappears when the basis of the attribute p disappears. But it
21
22
Cf. Thub bstan lhun po’i mdzes rgyan [cha–ka] 81a1f. (Klein 1991: 145); Gcig du bral gyi rnam bzhag
21a3f.
’Jam dbyangs bzhad pa defines dgag pa and med dgag, respectively, as follows. Bsdus chen gyi rnam
bzhag 28b5 (cf. Sras bsdus grwa 295.7ff.): rang dngos su rtogs pa’i blos rang gi dgag bya dngos su
bcad nas rtogs par bya ba’i chos de dgag pa’i mtshan nyid / (“A negative phenomenon [p] is defined as
a factor that is recognized only after having directly eliminated its object of negation by the mind that
directly knows p”); Bsdus chen gyi rnam bzhag 31b5f. (cf. Sras bsdus grwa 295.14ff.): rang dngos su
rtogs pa’i blos rang gi dgag bya dngos su bcad nas rtogs par bya ba gang zhig / rang dngos su rtogs
pa’i blos rang gi dgag bya bkag shul du chos gzhan ma yin dgag dang sgrub pa gang rung mi ’phen pa
de med dgag gi mtshan nyid / (“A non-affirming negation [p] is defined as the thing that is recognized
only after having directly eliminated its object of negation by the mind that directly knows p, and that
does not make known another factor – whether it is an affirming negative or a positive phenomenon –
through the process of negating its object of negation by the mind which directly knows p”). See also
Klein 1991: 107ff.
Hiroshi Nemoto
307
is not the case that the emptiness loses its identity over the course of time. Consequently,
Tsong kha pa argues, the emptiness of p is permanent irrespective of the fact that it exists
only at a particular moment.
4 The Dge lugs pa’s exposition of the Nature Body
Tsong kha pa applies the same reasoning to other unconditional factors such as nirvāṇa (mya
ngan las ’das pa), extinction through discernment (so sor brags ’gog; Skt. pratisaṃkhyānirodha), the truth of cessation (’gog bden; Skt. nirodhasatya), the true nature of things
(chos nyid; Skt. dharmatā), and so on, all of which pertain to religious attainments. Tsong
kha pa holds that these factors are permanent even if they may occur only at a particular
moment. By now it is not surprising that Tsong kha pa’s disciple, Rgyal tshab Dar ma rin
chen (1364–1432), offers a similar analysis with respect to a Buddha’s Nature Body, which
is permanent but not always existent. In his commentary on the Abhisamayālaṃkāra, Dar
ma rin chen says as follows:
dag pa gnyis ldan gyi sku de nyid thog ma med pa nas sems can gyi rgyud la
med par sangs rgyas pa’i dus ’ba’ zhig tu ’byung na mi rtag par ’gyur dgos so
snyam pa ni / mkhas rmongs la grags pa’i rtag pa’i khyad par dang / dngos
po res ’ga’ ba dang ’dus ma byas res ’ga’ ba’i khyad par ma phyed pas nongs
so // (Rnam bshad snying po rgyan 311b6ff.)
Someone might claim: “If [a Buddha’s Nature] Body endowed with twofold purity is not present in the continuum of sentient beings since beginningless time
and will occur only at the moment of enlightenment, then it must necessarily
be impermanent.” But this is an error due to confusion between [the concept
of] permanence accepted by wise men and that accepted by the foolish, and
also it is due to confusion between functioning things occurring at a particular
moment and unconditioned things occurring at a particular moment.
Dar ma rin chen interprets the eighth chapter of the Abhisamayālaṃkāra in terms of the
fourfold Buddha Bodies, i.e., the Nature Body, the Gnostic Dharma Body, the Enjoyment
Body, and the Manifestation Body.23 He identifies the Nature Body with the twofold purity.
The first is the innate purity (rang bzhin rnam dag), which again is identified with a
Buddha’s mind and body being empty of intrinsic existence. The second is the purity from
adventitious stains (glo bur rnam dag), i.e., the cessation of all obstructions including
afflictive obstructions (nyon sgrib) and obstructions to the knowable (shes sgrib).24 The
Nature Body characterized as such is the quintessence of Buddhahood.
23
24
Tsong kha pa, as well as other Dge lugs pa scholars including Dar ma rin chen, accepts Haribhadra’s
four-kāya interpretation, instead of the three-kāya interpretation by Āryavimuktisena, Ratnākaraśānti,
and Abhayākaragupta. For further details, see Makransky 1997: 289ff.
Rnam bshad snying po rgyan 311b5: gzugs kyi sku dang ye shes chos kyi sku’i glo bur rnam dag gis
khyad par du byas pa’i rang bzhin rnam dag gi cha dang / de’i steng gi glo bur rnam dag gi cha ni
’dus ma byas kyi sku dang / ngo bo nyid kyi sku zhes bya la / (“The following elements are said to
be the Unconditioned Body or Nature Body: the element of innate purity that is characterized by the
purity from adventitious stains on the Form Body and Gnostic Dharma Body; and that of purity from
adventitious stains on them”).
308
Dharmakīrti’s Notion of Permanence
The point to observe is that the Nature Body is embodied only at the moment of supreme
enlightenment; for, otherwise, it would absurdly follow that every sentient being is already
enlightened without requiring any effort. Therefore, it must be said that a Buddha’s Nature
Body is a thing occurring at a particular moment (res ’ga’ ba). But this fact does not
invalidate the assumption that the Nature Body is permanent. In this connection, let us look
at the argument given by ’Jam dbyangs bzhad pa ngag dbang brtson ’grus (1648–1721):
kho na re / ngo bo nyid sku de sems can gyi rgyud la sangs rgyas pa dang dus
mnyam par thal / de de’i rgyud la ’byung dus yod pa gang zhig / thog ma med
pa nas med pa’i phyir / […] rtsa bar ’dod na / de res ’ga’ bar thal / ’dod pa’i
phyir / ’dod na / mi rtag par thal / ’dod pa’i phyir na ma khyab / khyab par
thal / res ’ga’ ba yin na rang rgyu dang bcas pas khyab pa’i phyir te / rnam
’grel las / res ’ga’ ba nyid yin pas na // sdug bsngal ’di rgyu can nyid grub //
ces gsungs pa’i phyir na ma khyab ste / res ’gag’ ba’i dngos po la rang rgyu
dang bcas pa’i don yin pa’i phyir te / (Mi pham zhal lung [skabs brgyad pa]
39b4ff.)
He [namely, a certain opponent] says: “It follows that the Nature Body occurs in
the continuum of sentient beings simultaneously with enlightenment, because
there is the moment in which it occurs in their continuum, and because it is
not the case that it exists since beginningless time […] If you accept the root
thesis (rtsa ba), then it follows that it occurs at a particular moment because
you have accepted that [it occurs simultaneously with enlightenment]. If you
accept this, it follows that it is impermanent because you have accepted that
[it occurs at a particular moment].”
[We reply:] “There is no entailment.”
[The opponent says:] “It follows that there is entailment because if something
occurs at a particular moment, it must necessarily be a thing which has a cause,
for the Vārttika says: ‘It is established that this suffering has a cause since it
occurs [only] at a particular moment’.”25
[We reply:] “There is no entailment because what is meant here is that, if
something is an effective thing occurring at a particular moment, it must be
something that has a cause.”
We notice that ’Jam dbyangs bzhad pa tries to modify the meaning of PV II 179, which
explicitly says that suffering has a cause since it occurs only at a particular moment (res
’ga’ ba). The exact meaning of the verse, according to ’Jam dbyangs bzhad pa, is that
suffering has a cause and hence is impermanent since it is an effective thing occurring only
at a particular moment (res ’ga’ ba’i dngos po). Instead of saying that a thing occurring
at a particular moment is impermanent, ’Jam dbyangs bzhad pa suggests the idea that an
effective thing occurring at a particular moment is impermanent, which implies that there
exist permanent, non-effective, and causeless things which occur at a particular moment.
What he means to say is that the Nature Body of the Buddha is a permanent (non-effective
25
PV II 179cd (cf. Vetter 1984: 82): kādācitkatayā siddhā duḥkhasyāsya sahetutā //.
Hiroshi Nemoto
309
and causeless) thing, despite the fact that it is present only at the moment of enlightenment
and absent prior to that time. The wise men’s definition of the permanent must be recalled
here. ’Jam dbyangs bzhad pa says:
res ’ga’ mi ’byung la res ’ga’ ’byung ba ’di mi rtag pa’i don ma yin par thal /
[…] mkhas pa rnams res ’ga’ ba yin min tsam la mi byed par chos gang gi
rang bzhin la ’jig pa yod med kyis mi rtag par dang rtag par ’jog pa’i phyir
te / (Mi pham zhal lung [skabs brgyad pa] 40a3ff.)
It does not follow that the meaning of “impermanence” is defined as being
present at a certain time and absent at another […] because wise men posit
[a factor] either as impermanent or permanent in accordance with whether or
not the factor itself is subject to disintegration, without needing to consider
whether or not it is present only at a certain time.
This is a paraphrase of Pramāṇavārttika II 204cd. A thing that itself is not subject to
disintegration is permanent irrespective of its presence or absence at a certain time. The
Nature Body, then, is just the mere absence of intrinsic existence and obstructions; it is a
conceptual construct that never arises from causes, and that never undergoes disintegration.
Hence, it is concluded that the Nature Body is permanent. This of course does not imply
that the Nature Body is always existent, nor does it mean that it is a positive and independent
reality (sgrub pa rang dbang ba’i bden grub), as conceived of by the Jo nang pa school.
Rather, that the Nature Body is permanent means that it is something that is ascertained
through a simple negation of disintegration.26
5 Concluding remarks
This is how the Dge lugs pa scholars interpret the meaning of “permanence” especially
within the context of the theory of Buddhahood. They identify the quintessence of Buddhahood with the Nature Body, which again is characterized as the mere absence of intrinsic
existence and obstructions. The Nature Body so identified is a conceptual construct imputed
on the mere absence of the object of negation. It is free from origination and disintegration
so that it satisfies the condition of being permanent.
The important point to note here is that these Dge lugs pa scholars refer to Dharmakīrti’s
hypothetical definition of permanence in order to justify their theory of Buddhahood. They
assume that Dharmakīrti himself admits the existence of the permanent and defines it as
26
See Mi pham zhal lung [skabs brgyad pa] 37b1ff.: kun mkhyen jo nang na re / rang bzhin rnam dag gi
char gyur pa’i ngo bo nyid sku de sgrub pa rang dbang ba’i bden grub yin pa rgyud bla ma dang dus
’khor lugs zer na / ’o na / rtag pa rnams ’jig pa rnam par bcad tsam la mi ’jog par thal / dam bca’
’thad pa’i phyir / ’dod mi nus te / mdo sde pa yan chad kyi lugs la rtag dngos med pa’i phyir te / (“The
omniscient scholar of the Jo nang [i.e., Dol po pa,] says: ‘The Nature Body which is the element of
innate purity is a positive independent reality; this is the doctrine of the Uttaratantra and also that of
the Kālacakra.’ Then it follows that the mere absence of disintegration is not posited as the permanent,
because [according to you] your thesis is true. But you cannot accept this because, according to the
higher doctrine of the Sautrāntika [Yogācāra and Mādhyamika schools], there does not exist a thing
which is both permanent and functioning”).
310
Dharmakīrti’s Notion of Permanence
“the thing that itself does not disintegrate” in its own right. Such a concept of permanence,
which goes back to ancient Indian thought as seen in the Mahābhāṣya, functions forcefully
to explain why the Nature Body is permanent in spite of the fact that it is not always
manifested in an individual being.
It is also interesting to note that the Dge lugs pa carefully avoid the two extreme positions
held by other Tibetan thinkers. First, unlike Sa skya paṇḍita and Glo bo mkhan chen, Tsong
kha pa and his successors strongly assert that permanent things do exist, as they think that
otherwise the state of emptiness and the attainment of Buddhahood would be inexplicable.
Secondly, the Dge lugs pa reject the Jo nang pa position that the innate purity of the Nature
Body is a positive and independent reality (sgrub pa rang dbang ba’i bden grub); the Dge
lugs pa say instead that the innate purity is a non-affirming negation imputed on the mere
absence of disintegration. Thus, we see that Dharmakīrti’s notion of permanence survived
in the Dge lugs pa’s theory of Buddhahood and enabled them to give the most plausible
explanation of the Nature Body without falling into the two extreme positions.
References and abbreviations
Works in Sanskrit and Tibetan
AK Abhidharmakośa (Vasubandhu). See AKBh.
AKBh Abhidharmakośabhāṣya (Vasubandhu): Abhidharmakośabhāṣyam of Vasubandhu,
ed. P. Pradhan. Patna 1975.
AKV Abhidharmakośavyākhyā (Yaśomitra): Sphuṭārthā Abhidharmakośavyākhyā by
Yaśomitra, ed. U. Wogiwara. 2 vols. Tokyo 1936 (repr. 1971).
Bsdus chen gyi rnam bzhag Bsdus chen gyi rnam bzhag rigs lam gser gyi sgo ’byed lung
dang rigs pa’i gan mdzod blo gsal yid kyi mun sel skal ldan dad pa’i ’jug ngogs
(’Jam dbyangs bzhad pa ngag dbang brtson ’grus), Bkra shis ’khyil ed. Ba.
D Derge (sde dge) edition of the Buddhist canon (tripiṭaka) in Tibetan.
Dbu ma rgyan zin bris Dbu ma rgyan gyi zin bris rje rang gis gnang ba (Tsong kha pa
blo bzang grags pa), Zhol ed. Ba. Tohoku No. 5407.
Gcig du bral gyi rnam bzhag Gcig du bral gyi rnam bzhag legs bshad rgya mtsho las btus
pa’i ’khrul spong bdud rtsi’i gzegs ma (A lag sha bstan dar lha rams pa), Collected
Gsung ’bum of Bstan-dar Lha-ram of A-lag-sha. (Ka–Na). New Delhi 1971.
MAl Madhyamakālaṃkāra (Śāntarakṣita): Madhyamakālaṃkāra of Śāntarakṣita with his
own Commentary or Vṛtti and with the subcommentary or Pañjikā of Kamalaśīla,
ed. M. Ichigo. Kyoto 1985.
MBh Mahābhāṣya (Patañjali): The Vyākaraṇa-mahābhāṣya of Patañjali, ed. F. Kielhorn.
3 vols. Bombay 1880–1885.
Mi pham zhal lung Shes rab kyi pha rol tu phyin pa’i mtha’ dpyod ’khrul sel gang ga’i
chu rgyun mi pham zhal lung (’Jam dbyangs bzhad pa ngag dbang brtson ’grus),
Bkra shis ’khyil ed. Nya.
MSA Mahāyānasūtrālaṃkāra (“Maitreya”). See MSABh.
MSABh Mahāyānasūtrālaṃkārabhāṣya (Vasubandhu): Mahāyāna-sūtrālaṃkāra: Exposé
de la doctrine du Grand Véhicule, selon le système Yogācāra, ed. S. Lévi. Paris 1907.
Hiroshi Nemoto
311
PSVK Pramāṇasamuccayavṛtti (Dignāga), Tibetan Peking ed. gtan tshigs rig pa Ce. Otani
No. 5702.
PV Pramāṇavārttika (Dharmakīrti): Pramāṇavārttika-kārikā (Sanskrit and Tibetan), ed. Y.
Miyasaka. Acta Indologica 2 (1971–72) 1–206.
Rigs gter rang ’grel Tshad ma rigs pa’i gter gyi rtsa ba dang ’grel pa (Sa skya paṇḍita
kun dga’ rgyal mtshan). Beijing 1989.
Rnam bshad snying po rgyan Shes rab kyi pha rol tu phyin pa’i man ngag gi bstan bcos
mngon par rtogs pa’i rgyan gyi ’grel pa don gsal ba’i rnam bshad snying po’i rgyan
(Rgyal tshab rje dar ma rin chen), Zhol ed. Kha. Tohoku No. 5433.
Sras bsdus grwa Tshad ma’i dgongs ’grel gyi bstan bcos chen po rnam ’grel gyi don gcig
tu dril ba blo rab ’bring tha ma gsum du ston pa legs bshad chen po mkhas pa’i mgul
rgyan skal bzang re ba kun skong (Thugs sras ngag dbang bkra shis). Beijing 1985.
Thar lam gsal byed Tshad ma rnam ’grel gyi tshig le’ur byas pa’i rnam bshad thar lam
phyin ci ma log par gsal bar byed pa (Rgyal tshab rje dar ma rin chen), Zhol ed. Cha.
Tohoku No. 5450.
Thub bstan lhun po’i mdzes rgyan Grub pa’i mtha’ rnam par bzhag pa thub bstan lhun
po’i mdzes rgyan (Lcang skya rol pa’i rdo rje), Lcang skya rol pa’i rdo rje’i gsung
’bum. Cha and Ja. Beijing 1995.
UT Mahāyānottaratantraśāstra: The Ratnagotravibhāga Mahāyānottaratantraśāstra, ed. E.
H. Johnston. Patna 1950.
VBh D Sūtrālaṃkāravṛttibhāṣya (Sthiramati), Tibetan Sde dge ed. sems tsam Mi. Tohoku
No. 4034.
Works in western languages
Dreyfus 1997 G. B. J. Dreyfus, Recognizing Reality. Albany 1997.
Iwata 2004 T. Iwata, The Negative Concomitance (vyatireka) in the Case of Inconclusive
(anaikāntika) Reasons. In: The Role of the Example (dṛṣṭānta) in Classical Indian
Logic, ed. Sh. Katsura and E. Steinkellner. Vienna 2004, 91–134.
Katsura 2004 Sh. Katsura, The Role of dṛṣṭānta in Dignāga’s Logic. In: The Role of the
Example (dṛṣṭānta) in Classical Indian Logic, ed. Sh. Katsura and E. Steinkellner.
Vienna 2004, 135–173.
Klein 1991 A. C. Klein, Knowing, Naming & Negation, A Sourcebook on Tibetan Sautrāntika. Ithaca 1991.
Makransky 1997 J. J. Makransky, Buddhahood Embodied, Sources of Controversy in
India and Tibet. Albany 1997.
Nemoto 2009 H. Nemoto, On the dGe lugs pa Concept of Permanence. Journal of Indian
and Buddhist Studies/Indogaku Bukkyogaku Kenkyu 57-3 (2009), 126–130.
Nemoto 2011 H. Nemoto, The Proof of Impermanence in the dGe lugs pa’s pramāṇa
Theory. In: Religion and Logic in Buddhist Philosophical Analysis (Proceedings of
the Fourth International Dharmakīrti Conference, Vienna, August 23–27, 2005), ed.
H. Krasser, H. Lasic, E. Franco, and B. Kellner. Vienna 2011, 363–374.
Vetter 1984 T. Vetter, Der Buddha und seine Lehre in Dharmakīrtis Pramāṇavārttika:
Der Abschnitt über den Buddha und die Vier Edlen Wahrheiten im PramāṇasiddhiKapitel. Vienna 1984.
Dignāga on the View of a Generic Term as Denoting a
Relation1
by
Hideyo Ogawa
Introduction
Like grammarians, Dignāga bases his linguistic theory, which is well known as the apoha
theory, on actual usage of language. His observation of the fact that the word go “cow” is not
found to be used in actual usage to convey the meaning of the word aśva “horse” leads him to
build up the apoha theory, which is constructed out of three factors: a word’s own meaning
(svārtha), the meaning to be conveyed by another word (arthāntara, anyaśabdārtha), and
the exclusion of the latter (apoha, nivṛtti, vyāvṛtti, vyavaccheda). These three factors are
connected with one another in such a way that the word go brings about the understanding of
its meaning as qualified by the exclusion of the meaning of another word, say, the word aśva
(arthāntaranivṛttiviśiṣṭabhāva). Dignāga assumes that from the word go one understands
nothing else but the cow, and not simply the cow. When he says that anyāpoha “exclusion
of others” is the meaning of a word, he wishes to imply that anyāpoha is a meaning of a
word as a qualifier (viśeṣaṇa) of the word’s own meaning, or as the occasioning ground for
the use of the word (pravṛttinimitta).
In the Apoha chapter of his Pramāṇasamuccaya Dignāga argues that a generic term
(jātiśabda) like sat “being, existent” cannot denote an individual (bheda), a generic property
(jāti), a relation between the two (yoga, sambandha), or an individual qualified by a generic
property (tadvat). In the latter half of the second kārikā, he, relying on the observation of
utterances such as sad dravyam “A substance is existent,” in which the item sat is supposed
to be co-referential with a co-occurring item for an individual, states his view that the
generic term cannot denote a relation.
PS V.2cd: [na jātiśabo] vācako yogajātyor vā bhedārthair apṛthakśruteḥ /
Nor does the generic term denote a relation or a generic property, since it
refers to the same entity (apṛthakśruti) as [words] signifying individuals.2
The third kārikā of the Apoha chapter is devoted to adducing the conclusive reason for the
view. The kārikā goes as follows:
1
2
I deeply respect Dr. Ole Holten Pind, whose epoch-making achievement is represented by Pind (2009,
2015). The present paper owes much to the work, without which I could not even have stood on the
threshold of Dignāga’s apoha theory. Editor’s note: This contribution contains changes that the author
has left to the editor’s discretion.
Pind 2015: II 14 (§ 4): “(vācakaḥ) neither the general property nor the inherence relation, because it is
not ‘heard apart’ (apṛthakśruteḥ) from [words] whose referents (bhedārthaiḥ) are particular [general
properties].”
Birgit Kellner et al., eds., Reverberations of Dharmakīrti’s Philosophy: Proceedings of the Fifth International Dharmakīrti
Conference Heidelberg, August 26 to 30, 2014. Vienna 2020, pp. 313–330.
314
Dignāga on the View of a Generic Term as Denoting a Relation
PS V.3: sambandhaś cātra sambandhidharmeṇa vācya ucyate /
tathā hi bhāvaḥ kṛtvoktaḥ, bhāvaś cānyena yujyate //
Pind 2015: II 18–19 gives the following translation of the kārikā:
And in this context it is explained that the connection is denotable through
the property of the relatum (sambandhidharmeṇa vācya ucyate). That is, it
[viz. the connection] is denoted on the assumption that it is a state of action
(bhāvaḥ kṛtvoktaḥ); and a state of action is connected with the other [relatum]
(bhāvaś cānyena yujyate).
Pind’s interpretation of the kārikā is well grounded. But I have to say that, resting even on
Pind’s interpretation, it is almost impossible to fathom what Dignāga intends to say in the
kārikā.
Dignāga here simply brings out the point Bhartṛhari makes about the denotation of
a relation. According to Bhartṛhari, there is no nominal (nāman) that denotes a relation
qua relation; a nominal such as sambandha “relation” cannot denote a relation in its own
property (svadharmeṇa) but simply as a substance (dravya), because an act (bhāva) denoted
by an action noun (bhāvasādhana) is treated like a substance (dravyavat).
The aim of this paper is to give a plausible interpretation of the kārikā by taking into
consideration arguments Bhartṛhari brings forward about the denotation of a relation in
his Vākyapadīya and thereby to show clearly the reason for Dignāga’s argument that a
generic term cannot denote a relation. It will be shown that the same approach to everyday
speech (lokavyavahāra) as Dignāga takes for the purpose of establishing anyāpoha as a
word meaning (padārtha) is found in his arguments about the denotation of a relation.
1 Before turning to a closer examination of the question at issue, a few remarks should be
made concerning Dignāga’s apoha theory. It is extremely important to note that Dignāga accepts the abstraction (apoddhāra) theory which is known as forming the pivot of Bhartṛhari’s
linguistic theory.
Consider PS V.46, where Dignāga introduces the concept of the abstraction of words
from a sentence.
[A] PS V.46: apoddhāre padasyāyaṃ vākyād artho vikalpitaḥ /
vākyārthaḥ pratibhākhyo ’yaṃ tenādāv upajanyate //
When a word is abstracted from a sentence, this [i.e., anyāpha] is conceptually
posited (vikalpita) as the meaning of the word. This sentence meaning called
pratibhā “a flash of knowledge” is brought about by means of [the grasping
of] that [word meaning] at the outset [when one has not familiarized oneself
with the sentence].3
3
Pind 2015: II 166 (§ 61): “the referent of the syntactical word (padasya) is imagined (vikalpitaḥ) when
abstracted (apoddhāre) from the sentence (vākyāt). Yet the referent of the sentence which is called
intuition (pratibhā) is in the beginning (ādau) produced by that [namely the syntactic word].” According
to Jinendrabuddhi, the pronominal ayam “this” in pāda a refers to anyāpoha. PSṬ Ms B 236b1 (Pind
Hideyo Ogawa
315
Dignāga comments on the first half of the kārikā as follows:
[B] PSV on PS V.46 (Pind 2015: I 55–56): (a) padasyāsato 〈’pi〉 vākyād
apoddhṛtasya yathāgamaṃ utprekṣayārtho vyavasthāpyate kevalasyāprayogāt prakṛtipratyayavat.4 (b) sā cotprekṣānyeṣv āgameṣv ayuktārthagrahaṇī .
tasmād idam arthāntaram utkṣiptam,5
(a-1) A word or a syntactic unit (pada) is abstracted from a sentence (vākya).
(a-2) The word is unreal (asat) because it is not used in isolation [in the realm
of actual communication], just like a base (prakṛti) and an affix (pratyaya)
that are abstracted from a word. (a-3) Nonetheless, [a certain entity is] posited
as its meaning through invention (utprekṣā) in accordance with tradition (yathāgamam). (b-1) And the invention based on other traditions is that through
which a [word] meaning that is logically invalid (ayuktārtha) is conceived of.
(b-2) Therefore, this [word] meaning [called anyāpoha], which is different
from entities posited as [word] meanings by other traditions, has been brought
forward here.6
As Pind (2015: II Appendix 14) has pointed out, [B](a) presents a striking parallel to
Bhartṛhari’s Vṛtti.7 Let us take note of (a-3): A word meaning is an invented entity (padasya … utprekṣayā … artho vyavasthāpyate). The invention is conditioned by tradition
(yathāgamam). There are different inventions according to different traditions. As mentioned earlier, regarding the question of what a generic term denotes Dignāga tries to reject
four views: an individual, a generic property, a relation between the two, and an individual
4
5
6
7
2015: II 166, n. 554): ayam ity anyāpohaḥ. I shall waste no words on Dignāga’s view that pratibhā is the
sentence meaning. This view, needless to say, is borrowed from Bhartṛhari. The details of Bhartṛhari’s
concept of pratibhā are to be left to Ogawa (forthcoming).
Pind 2015: II 167–168 (§ 61): “Even though the syntactical word is unreal (asat) as abstracted from
the sentence, its referent is determined by invention (utprekṣayā) according to the [grammatical]
tradition, because it is not used in isolation (kevalasyāprayogāt) in the same way as a stem and an affix
(prakṛtipratyayavat) [are not used in isolation].”
Pind 2015: II 169–170 (§ 61): “And this invention apprehends a referent that is not justified (ayuktārthagrahaṇī ) in other traditions. Therefore this different referent (arthāntaram) has been brought forward
(utkṣiptam).”
The phrase arthāntara (lit. “another meaning”) refers to a word meaning that is reasonable and that is
called anyāpoha. PSṬ Ms B 237a7: yuktimad … anyāpohākhyam. See Pind 2015: II 169, n. 566.
Vṛtti on VP 1.24 (65.1–4): tatrāpoddhārapadārtho nāmātyantasaṃsṛṣṭaḥ saṃsargād anumeyena parikalpitena rūpeṇa prakṛtavivekaḥ sann apoddhriyate / pravivaktasya hi tasya vastuno vyavahārātītaṃ
rūpam / tat tu svapratyayānukāreṇa yathāgamaṃ bhāvanābhyāsavaśād utprekṣayā prāyeṇa vyavasthāpyate / In this passage Bhartṛhari makes the following points. Of the eight topics to be dealt with in the
Vākyapadīya (aṣṭaka), what is called apoddhārapadārtha “abstracted word meaning” is absolutely fused
in a single whole meaning (atyantasaṃsṛṣṭa). The word meaning is abstracted from a putative complex
(saṃsarga) on the basis of a form that is to be inferred and that is conceptually assumed, so that it is
distinguished from other abstracted word meanings. Such a partial word meaning which is distinguished
from other partial word meanings has a form that is outside the realm of actual communication. Grammarians posit (vyavasthāpyate) such an assumed form (parikalpita) through invention (utprekṣayā), in
accordance with what they have understood (svapratyayānukāreṇa) following traditions (yathāgamam)
and exposing themselves to them so repeatedly as to have their latent traces (bhāvanābhyāsa). See
Cardona 1999: 96–98 for a full account of the points made by Bhartṛhari here.
316
Dignāga on the View of a Generic Term as Denoting a Relation
qualified by a generic property. Entities posited as the word meaning in these views are
inventions according to traditions.8
It is important to note [B](b-1) and [B](b-2), which clearly show that anyāpoha is
posited as a word meaning through invention in accordance with the Buddhist tradition.
According to Dignāga, the assumption that anyāpoha is a word meaning is more reasonable
than the others. That is, this assumption is not only based on the Buddhist tradition (āgama)
but also is well grounded on logical reasoning (yukti). But in terms of what is the assumption
reasonable? Consider the following Vṛtti.
[C] PSV on PS V.37b (Pind 2015: I 46): (a) tasmād asmābhir api 〈lokavyavahārā naimittikā vā〉 pāribhāṣikā 〈vā〉 bhūtārthatvena na mṛśyante,
(b) lokavad evānugamyante. (c) siddhaś ca rūpaśabdo loke nīlādiṣv eva, na
rasādiṣu.
(a) Therefore, we, too, do not cling (mṛśyante) to everyday speech (lokavyavahāra),9 whether it has a basis for application or comes from a scientific field, as
related to a real entity (bhūtārtha). (b) Just as the world follows (anugamyante)
everyday speech, so do we. (c) And in everyday usage it has been established
that the word rūpa “color” denotes only (eva) blue and the like and not taste
and others.
Let us take note of [C](b): Dignāga makes the point that he follows everyday speech, just
as does the world. Clearly what this implies is that he holds that his apoha theory can most
reasonably account for everyday speech. Moreover, when he states [C](c), he intends to
imply that the apoha theory is constructed on the basis of the fact found in everyday speech
such that the word rūpa denotes only (eva) blue and the like and not taste and others. It may
be said without much exaggeration that Dignāga’s theory of apoha is meant for reasonably
explaining the fact of speech that a certain linguistic item occurs only in the domain of a
certain meaning and not in the domain of another meaning.
In [C](a) it is stated that Dignāga does not suppose that everyday speech must be related
to a real entity. Naturally this suggests that he stands far apart from realism according to
which language reflects reality.
Interestingly and importantly, Dignāga’s approach to established everyday usage, which
is shown in [C](b), is not different from Bhartṛhari’s. Bhartṛhari asserts that he accepts
things as they are spoken of, without making an ontological commitment to them.10 Relevant
to this point is Bhartṛhari’s following remark:
8
9
10
Jinendrabuddhi says: PSṬ Ms B 237a1–2 (Pind 2015: II 168, n. 561): yasya (scil. padasya) ya āgamaḥ:
keṣāṃ cid bhedā vācyāḥ, pareṣāṃ jātir, anyeṣāṃ sambandhaḥ, keṣāṃ cit tadvad iti padasyārtha ity
āgamaḥ. (“That which is the tradition as to a word is the tradition such that this is the meaning of a
word: some have a tradition that individuals (bheda) are to be denoted by words; some have a tradition
that a generic property is to be denoted by a word; others have a tradition that a relation between a
generic property and its bearer is to be denoted by a word; some have a tradition that an individual
qualified by a generic property is to be denoted by a word.”)
On the meaning of mṛśyante, see Pind 2015: II 142, n. 468.
In the Sādhanasamuddeśa Bhartṛhari says of the power of functioning as agent (kartṛtva) the following:
VP 3.7.38: tattve vā vyatireke vā vyatiriktaṃ tad ucyate / śabdapramāṇako lokaḥ sa śāstreṇānugamyate //
(“The property of being an agent, whether it be identical with its bearer or different from the latter,
we declare that it is different from its bearer. The world holds usage as their standard. The science of
Hideyo Ogawa
317
[D] VP 3.3.88: vyavahāraś ca lokasya padārthaiḥ parikalpitaiḥ /
śāstre padārthaḥ kāryārthaṃ laukikaḥ pravibhajyate //
The world (loka) achieves verbal communication by resorting to conceptual
word meanings. In the science [of grammar] the word meanings commonly
known to the world (laukika) are divided [into actions (kriyā), substances
(dravya), qualities (guṇa), and others]11 (pravibhajyate) for the sake of [grammatical] operations.12
Thus, if we accept that in constructing his linguistic theory Dignāga takes the same position
as Bhartṛhari, namely that the usage common in the world is taken as the standard that is to
be followed, we may say the following: Dignāga conceptually posits anyāpoha as the word
meaning in order to establish a linguistic theory that holds up in the universe as the world
speaks of it and believes it to be real. The universe as such is not spoken of, it is only a
conceptual construct.
2 Now let us consider how Dignāga argues against the assumption that a relation between
a generic property and its bearer is denoted by a generic term. Here I will consider the Vṛtti
on PS V.3. To begin with, I will give the text of the Vṛtti and its translation by Pind.
[E] PSV on PS V.3: sambandhanaṃ hi sambandhaḥ. 〈so ’nyena yujyate
rāgādivat〉. tasmāt sambandhidharmeṇa sambandho vācya iti kṛtvāśaṅkitaṃ
svadharmeṇa tu nāsti sambandhasya vācakaḥ śabda ity idaṃ tat prati nāsti.
ato naivasya (sic) jātiśabdena vācyatvam upapadyate.
For connection means “state of connecting;” it [namely the state of connecting]
is connected to the other [relatum] in the same way as the state of colouring
(rāgādivat), etc. Therefore, assuming (iti kṛtvā) that the connection is denotable
through the property of the relatum, doubt (āśaṅkitam) about (prati) the claim
11
12
grammar follows them.”) Iyer 1971: “As to whether it (power) is identical with or different from (its
substratum), it is held that it is different. For the world, the word is the authority and the śāstra (that
is, Grammar) follows it too.” The point made here is that, without examining the ontological question
of whether the agent-power is identical with or different from its bearer, grammarians favor the view
commonly accepted by the world that the agent-power is different from its bearer. I need not elaborate
on Bhartṛhari’s concept of power here. See Ogawa 2009.
Prakāśa on VP 3.3.88 (181.11–12): -śāstrīyakāryaprasiddhyarthaṃ laukika eva kriyādravyaguṇādilakṣaṇo ’poddhārapadārthaḥ pravibhajyata iti … /
Iyer 1971: 119: “Verbal communication in the world is done with meanings of words fashioned by the
mind and in the science of grammar, it is the meanings adopted in the world on the basis of which the
work (of explaining the forms) is done.” Houben 1995: 321; 424: “But verbal usage of everyday life
[takes place] by means of conceptualized meanings of words. In grammar (śāstra) the word meaning
of everyday usage is separated [from the sentence meaning] for the sake of grammatical operations.”
(Houben’s interpretation of pravibhajyate “… is separated [from the sentence meaning]” is far-fetched.)
According to Helārāja, that which is framed out of word meanings (padārthaprakriyā) and which is
well established in the world is to be accepted as it is in the science of grammar. Prakāśa on VP 3.3.88
(181.9–10): tad atra lokaprasiddhaiva padārthaprakriyā samāśrayaṇīyā … / Note that the word laukika
is derived from loka with the taddhita ṭhañ by A 5.1.44 lokasarvalokāṭ ṭhañ, signifying “known in the
world” (“known in X” [tatra viditaḥ]).
318
Dignāga on the View of a Generic Term as Denoting a Relation
(idan tat) that a word, on the other hand, does not denote the connection by its
own property (svadharmeṇa) is meaningless (nāsti). Consequently (ataḥ) its
[viz. the connection’s] denotability (vācyatvam) by a general term is not at all
(naiva) justified (upapadyate). (Pind 2015: II 20)
2.1
Let us first look at the following introductory sentence.
[F] PSV on PS V.3: (a) sambandhanaṃ hi sambandhaḥ /
2.1.1 This statement is meant to give a vigraha, i.e., “an utterance which paraphrases the
meaning of an integrated form (vṛtti),”13 of the term sambandha, which shows that this
term is bhāvasādhana, that is, the item that is formed from a verb with a kṛt affix signifying
bhāva.14 The term sambandha is derived from the verb bandh preceded by the preverb
(upasarga) sam with the kṛt affix ghañ,15 meaning “connecting, connection, relation.” The
term sambandhana is derived from the same string with the kṛt affix lyuṭ.16 Jinendrabuddhi
properly notes that the term sambandha is an item that denotes bhāva, or what is called an
action noun.17
In Pāṇini’s grammar, bhāva as signified by a kṛt affix is to be distinguished from bhāva as
signified by an L-affix (as in āsyate devadattena “Devadatta is sitting:” A 3.4.69 laḥ karmaṇi
ca bhāve cākarmakebhyaḥ): the former is an act in the abstract, something that has been
brought to accomplishment (siddha), termed bāhyabhāva “external bhāva,” while the latter
is an act in process, which is the significand of a verbal base itself (prakṛtyartha) and hence
something that is to be brought to accomplishment (sādhya), termed ābhyantarabhāva
“internal bhāva.”18
13
14
15
16
17
18
SK, sarvasamāsaśeṣaprakaraṇa (II.215): kṛttaddhitasamāsaikaśeṣasanādyantadhāturūpāḥ pañca vṛttayaḥ / parārthābhidhānaṃ vṛttiḥ / vṛttyarthāvabodhakaṃ vākyaṃ vigrahaḥ / MBh on A 1.3.1 (I.256.20):
kaḥ punar bhāvaḥ / bhavateḥ svapadārtho bhavanaṃ bhāva iti / Pradīpa on MBh to A 1.3.1 (II.186):
bhavanam bhāva iti vigrahaḥ kartṛsādhananivṛttyarthaḥ /
To adduce a typical example of a vigraha for a bhāvasādhana term, I will cite from the Bhāṣya. MBh
on A 3.1.22 (II.28.13–14): samabhihāra iti ko ’yaṃ śabdaḥ / samabhipūrvād dharater bhāvasādhano
ghañ / samabhiharaṇaṃ samabhihāraḥ / The term samabhihāra, used in A 3.1.22, signifies the act of
performing repeatedly, intensively, or repeated, intensive perfomance.
Here and in the following, bold letters are used to signify the anubandhas.
sam-bandh-ghañ (A 3.3.18); sam-bandh-lyuṭ (A 3.3.115) → sam-bandh-ana (A 7.1.1).
PSṬ Ms B 195a5 (Pind 2015: II 20, n. 44): bhāvatvam asya darśayati. Grammatical rules we have to
take into consideration here are the following: A 3.3.18 bhāve and A 3.3.115 lyuṭ ca. A 3.3.18 and A
3.3.115 provide respectively for introducing ghañ and lyuṭ after a verb on condition that an act in the
abstract (bhāva) is to be signified.
MBh on A 3.3.19 (II.145.16–18): nanu coktaṃ vihitaḥ pratyayaḥ svārthe bhāve ghañ itīti / anyaḥ
sa bhāvo bāhyaḥ prakṛtyarthāt / anenedānīm ābhyantare bhāve syāt / kaḥ punar etayor bhāvayor
viśeṣaḥ / ukto bhāvabhedo bhāṣye // See PWT 297. In the Kriyāsamuddeśa Bhartṛhari states: VP
3.8.47: ākhyātaśabde bhāgābhyāṃ sādhyasādhanavartitā / prakalpitā yathā śāstre sa ghañādiṣv api
kramaḥ // (“In grammar, just as, with reference to a finite verb form, its divisions are fictitiously
assumed respectively to denote something to be brought to accomplishment and something to bring
it to accomplishment, so, with reference to items ending in affixes like ghañ also, the same analysis
is given”); VP 3.8.48: sādhyatvena kriyā tatra dhāturūpanibandhanā / sattvabhāvas tu yas tasyāḥ sa
ghañādinibandhanaḥ // (“In those items ending in affixes like ghañ, an action is conveyed as something
to be brought to accomplishment on the basis of a verb, while the action’s status of being a substance is
Hideyo Ogawa
319
2.1.2 The external bhāva is denoted by an item ending in a kṛt affix (kṛdanta), while
the internal bhāva is denoted by an item ending in a verbal ending (tiṅanta). Both bhāvas
have in common the property of being an action. They have the following distinguishing
characteristics:19
1. kṛdabhihito bhāvo dravyavad bhavati /
2. The external bhāva behaves like a substance (dravya). It functions as what brings an
action to accomplishment (sādhana, kāraka).
3. tiṅabhihitena bhāvena kālapuruṣopagrahā abhivyajyante /
4. Time (kāla), person (puruṣa), and aspect (upagraha) are manifested by the internal
bhāva.
5. tiṅabhihito bhāvaḥ kartrā samprayujyate /
6. The internal bhāva is connected with an agent.
7. ayam api viśeṣaḥ syāl liṅgakṛtaḥ saṅkhyākṛtaś ceti /
8. The external bhāva is capable of being connected with gender and number.
Thus what Dignāga means by the statement in question ([F]) is clear: The sambandha in
question is a relation as signified by the action noun sambandha.
2.2
Next let us take up the second sentence.
[G] PSV on PS V.3: (b) 〈so ’nyena yujyate rāgādivat〉.
2.2.1
In mentioning the term rāga, Dignāga might have had the following rule in mind.
[H] A 6.4.27 ghañi ca bhāvakaraṇayoḥ //20
By this rule we have rāga “coloring, dye” as follows: ranj-ghañ → raj-a → rāj-a (A
7.2.116) → rāga (A 7.3.52).21 Rule A 3.3.121 halaś ca accounts for raṅga “paint, place of
19
20
21
conveyed on the basis of affixes such as ghañ”); Iyer 1974: 27–28: “Just as Grammar divides a verb
into two parts, one expressive of a process (sādhya) and the other of a thing (sādhana), the same can be
done to a word ending in a primary suffix like ghañ;” “The expression of action as a process depends
upon the root and its aspect as a thing depends for its expression on suffixes like ghañ.”
In the Bhāṣya on A 3.1.67 Patañjali states: MBh on A 3.1.67 (II.57.7–17): asti khalv api viśeṣaḥ
kṛdabhihitasya bhāvasya tiṅabhihitasya ca / kṛdabhihito bhāvo dravyavad bhavati / kim idaṃ dravyavad
iti / dravyaṃ kriyayā samavāyaṃ gacchati / kaṃ samavāyam / dravyaṃ kriyābhinirvṛttau sādhanatvam
upaiti / tadvac cāsya bhāvasya kṛdabhihitasya bhavati / pāko vartata iti / kriyāvan na bhavati / kim idaṃ
kriyāvad iti / kriyā kriyayā samavāyaṃ na gacchati / pacati paṭhatīti / tadvac ca asya kṛtabhihitasya
na bhavati / pāko vartata iti / asti khalv api viśeṣaḥ kṛdabhihitasya bhāvasya tiṅabhihitasya ca /
tiṅabhihitena bhāvena kālapuruṣopagrahā abhivyajyante / kṛdabhihitena punar na vyajyante / asti
khalv api viśeṣaḥ kṛdabhihitasya bhāvasya tiṅabhihitasya ca / tiṅabhihito bhāvaḥ kartrā samprayujyate/
kṛdabhihitaḥ punar na samprayujyate / yāvatā kiṃcit sāmānyaṃ kaścit viśeṣo yuktaṃ yad ayam api
viśeṣaḥ syāl liṅgakṛtaḥ saṅkhyākṛtaś ceti /
This rule provides that the penultimate n of ranj “dye, color” is deleted in a stem followed by ghañ
introduced on condition that an action in the abstract (bhāva) or an instrument (karaṇa) is to be signified
(A 3.3.18–19).
See PWT 446.
320
Dignāga on the View of a Generic Term as Denoting a Relation
public amusement,” in which ghañ signifies an instrument (karaṇa) or a locus (adhikaraṇa)
and which is used as an appellation (saṃjñā) to denote a particular thing. Thus we may say
that the item rāga is an apt example of an action noun.
2.2.2 The expression so ’nyena yujyate must be separated from the phrase rāgādivat,
which has to be construed with the previous sentence. The expression corresponds to PS
V.3d: bhāvaś cānyena yujyate, which Jinendrabuddhi glosses as bhāvaś cānyena sambandhinā yujyate.22 The act of connecting in the abstract, which is denoted by the nominal
sambandha that is an item ending in the kṛt affix ghañ, is to be treated like a substance.
Thus what Dignāga means by the expression is that the entity relation denoted by the
nominal sambandha is related to the other relatum through another relation, which implies
that a relation (sambandha) itself becomes a relatum (sambandhin): to be precise, a relation
behaves like a relatum.
2.2.3 In the situation in which a relation behaves like a relatum, the term which denotes
such a relation is a sambandhiśabda “relative term.” At this point, it is important to note
what a sambandhiśabda is. According to Patañjali, the item called sambandhiśabda implies
(ākṣipati) a relatum. One uses the utterance mātari vartitavyam “One should obey a mother”
instead of svasyāṃ mātari vartitavyam “One should obey one’s own mother.” The word
mātṛ “mother” is a sambandhiśabda. By virtue of a relation, the mother is understood to
be the mother of a certain person.23 Similarly, when the term sambandha is used, a relation
signified by this term behaves like a relatum, so that it implies a relatum on the basis of
another relation.24
22
23
24
PSṬ Ms B 68b7 (Pind 2015: II 18, n. 41): sambandhanaṃ hi sambandha iti bhāvarūpeṇa sambandhaśabdenābhidhānāt. bhāvaś cānyena sambandhinā yujyata iti. sambandhyantarākāṅkṣā jāyate: kasya
sambandha iti. tataś ca sambandhy eva sa bhavatīti na svarūpeṇa sambandho ’bhidhīyate.
MBh on A 3.1.19 (II.27.14–16): sambandhiśabdāś ca punar evamātmakāḥ yaduta sambandhinam
ākṣipanti / tadyathā / mātari vartitavyaṃ pitari śuśrūṣitavyam iti / na cocyate svasyāṃ mātari svasmin
vā pitarīti / sambandhāc caitad gamyate yā yasya mātā yaś ca yasya piteti /
Jinendrabuddhi illustrates this point by taking as an example the term samuccaya “connection” which
is supposed to convey a meaning of the particle (nipāta, A 1.4.57) ca “and.” PSṬ Ms B 195a1–3 (Pind
2015: II 19, n. 43): iha kaścic chabdārthaḥ kenacic chabdenābhidhīyamānaḥ sambandhyantarākāṅkṣopajanahetuḥ. tathā hi samuccayaḥ samuccayaśabdābhidhāne parākāṅkṣopajanahetur bhavati,
sa eva cādyupādānaḥ kasyeti ākāṅkṣāṃ nopajanayati (em: ākāṅkṣām upajanayati Pind). bhāvaś ca
bhāvasādhanena śabdenābhidhīyamāno niyatam anyaviṣayām ākāṅkṣāṃ janayatīti: (“In this everyday
speech, a certain meaning of an item, when being conveyed by a certain other item, becomes the cause
of awakening expectations of the other relatum. To explain. Connection (samuccaya), when being
conveyed by the term samuccaya, is the cause of awakening expectations of something other than it. But
the very same entity connection, when being referred to by a particle such as ca ‘and,’ does not awaken
expectations as ‘connection of what?’ The act [of connecting], when being conveyed by an action noun,
necessarily awaken expectations regarding something different from it.”) Pind 2015: II 19–20, n. 43: “In
this case when a certain word referent is being denoted by a certain word it is the cause of generating
expectation [of the complementation of] the other relatum. Like, for instance, an accumulation is the
cause of generating expectation [of the complementation] of the correlate when being denoted by a
word denoting accumulation. The [word denoting accumulation] comprising [the word] ‘and,’ and so
on, with certainty generates expectation [of complementation] at the thought ‘[accumulation] of what’?
And a state that is denoted by a word having a state [of action] as its means of realisation necessarily
generates expectation concerning the other relatum.” Suffice it to give the utterances ghaṭaś ca paṭaś ca
Hideyo Ogawa
2.3
321
The next step is to consider the remaining passage.
[I] PSV on PS V.3: (c) tasmāt sambandhidharmeṇa sambandho vācya iti
kṛtvāśaṅkitaṃ svadharmeṇa tu nāsti sambandhasya vācakaḥ śabda iti idaṃ
tat prati nāsti. ato naivasya jātiśabdena vācyatvam upapadyate.
2.3.1 Let us note the statements sambandhidharmeṇa sambandho vācyaḥ and svadharmeṇa nāsti sambandhasya vācakaḥ śabdaḥ. Strangely enough, Dignāga and Jinendrabuddhi
do not specify what the terms sambandhidharma and svadharma mean, which seems to be
self-explanatory for them.25 In this connection it will be useful to consider what Bhartṛhari
says in the Sambandhasamuddeśa.
[J] VP 3.3.4: nābhidhānaṃ svadharmeṇa saṃbandhasyāsti vācakam /
atyantaparatantratvād rūpaṃ nāsyāpadiśyate //
There is no term which denotes a relation qua relation [lit. “a relation as
characterized (upalakṣita) by its own property”].26 Since a relation is absolutely
dependent, its form is not spoken of [by any particular nominal term].
A relation’s own property (svadharma) is that of being absolutely dependent (atyantaparatantratva). According to Bhartṛhari, a power (śakti) and a quality (guṇa) are of a dependent
nature; a relation is a power that even powers have and a quality that even qualities have.27
25
26
27
“a pot and a piece of cloth” and ghaṭasya paṭena samuccayaḥ “the connection of a pot with a piece of
cloth.”
As to the term svadharma Pind (2015: II 20, n. 48) says: “The term applies to any given noun that is
not subject to the grammatical operation of introducing the sixth triplet that denotes the relation …” I
cannot understand his point here.
Helārāja takes the third-triplet ending of svadharmeṇa to denote a characteristic that makes known
something as having acquired a certain aspect (A 2.3.21 itthaṃbhūtalakṣaṇe). Prakāśa on VP 3.3.4
(128.10–11): tatra svena asādhāraṇena dharmeṇa svabhāvenopalakṣitasya vācakaṃ pratyāyakam,
abhidhānaṃ ṣaṣṭhīvyatiriktaṃ nāsti, idaṃtayā svarūpānavadhāraṇāt /
Bhartṛhari states the following kārikā. VP 3.3.5: upakārāt sa yatrāsti dharmas tatrānugamyate / śaktīnām api sā śaktir guṇānām apy asau guṇaḥ // (“Where, by virtue of one thing rendering service
(upakāra) to another, there is the relation, there the property of being absolutely dependent is recognized.
It [i.e., the relation] is a power even of powers; it is a quality even of qualities”); Iyer 1971: 81: “Where
there is service rendered, there an attribute (that is, relation) is understood. It is the power of powers, it
is the attribute of attributes”; Houben 1995: 170; 341: “Where this [relation] is, because some service
is rendered [from one thing to another, or: from signifier to signified and vice versa], there one arrives
at a property (viz. dependence). Even for capacities [dependent on that which has the capacity] it is a
capacity, even for qualities [dependent on that which has the quality] it is a quality [so it is extremely
dependent].” According to Bhartṛhari, when a certain entity is seen to render service to, i.e., help
(upakāra), another, there obtains a relation between them. A power and a quality are what renders
service to an entity: a power helps the entity to bring about a certain result and a quality helps the entity
to be differentiated from other entities. A power is also to be served by a certain thing in order to be
determined both to abide in a locus and to bring about a specific result. Such a thing is a relation that
the power has with respect to its locus (āśrayāśrayibhāva) and to its result (kāryakāraṇabhāva), and
the relation can consequently be viewed as a power in that it helps the power to abide in a locus and to
bring about a specific result. The reason that, with regard to qualities, a relation is called a quality is
that it is viewed as meant for a quality (parārtha) because it serves to determine the quality abiding in a
specific substance. On these points, see Ogawa 2009.
322
Dignāga on the View of a Generic Term as Denoting a Relation
2.3.2 It is proper to say that an entity cannot at the same time both be a relation (sambandha) and not be a relation, or be a relatum (sambandhin). We must say accordingly that
in a certain context a relation has a property which is like that of a relatum.28 In addition,
we have seen that a relation’s own property is that of being absolutely dependent. What is
meant by the term sambandhidharma is just opposite to this property. In order to make
these points clear, it is useful to consider the Vṛtti on VP 2.439. On the assumption that
contact (saṃyoga) and inherence (samavāya), which are relational concepts of the Vaiśeṣika
school, are a relation, Bhartṛhari says:
[K] VP 2.439: sambandhidharmā saṃyogaḥ svaśabdenābhidhīyate /
sambandhaḥ samavāyaś ca sambandhitvena gamyate //
Contact, even if it is a relation, is conveyed as something whose property
(dharman) is like that of a relatum by its own word [i.e., the word saṃyoga].
Inherence, even if it is a relation, is also understood as a relatum [from its own
word samavāya].29
In his Vṛtti Bhartṛhari comments as follows:
[L] Vṛtti on VP 2.439 (312.17–22): (a) aṅgulyoḥ saṃyoga iti sambandho ’pi
saṃyogaḥ saṃvijñānapadena svatantro ’bhidhīyamānaḥ puruṣādivat sambandhidharmābhidhīyate /
28
29
The following kārikā is useful in considering what it means to say that an entity has a property which is
like that of another entity. VP 3.3.6: taddharmaṇos tu tācchabdyaṃ saṃyogasamavāyayoḥ / tayor apy
upakārārthā niyatās tadupādhayaḥ // (“But, contact (saṃyoga) and inherence (samavāya) are termed
sambandha because they have the property which is like the property of that [i.e. the relation]. Even those
two have specific factors which delimit that [i.e., the relation] and which serve to render service to the
relation”); Iyer 1971: 81: “Conjunction and inherence are called relations because they have the attribute
thereof. They have definite conditions the purpose of which is to render service;” Houben 1995: 176;
341: “As regards saṃyoga (connection) and samavāya (inherence), they (are tacchabdāḥ) are called by
that word (sc. ‘relation’), because they have (as it were) that property (sc. dependence). Even these two
have restricted functions of rendering service, with this [relation] as limiting factor.” A relation that is
delimited by a specific factor, such as the property of residing only in a substance (dravyaikaniyatatva) or
its being related to being as a quality (guṇatvena sattāsambandhaḥ), is termed saṃyoga; a relation that is
delimited by a specific factor, such as the property of its relata being inseparable (ayutasiddhi), is termed
samavāya. Prakāśa on VP 3.3.7 (130.5–6): tathā ca dravyaikaniyatatvaṃ guṇatvena sattāsambandhaḥ
saṃyogasyopādhir iti tatropakāraḥ sambandhena tasyāvacchidyate /; Prakāśa on VP 3.3.8–11 (131.11):
anena cāyutasiddhiḥ samavāyasyopādhir uktaḥ / According to Bhartṛhari, contact and inherence, which
the Vaiśeṣikas posit as relations, do not fit the definition of a relation given by him. Suppose that the
defining feature of a relation is to be dependent; contact and inherence have this feature with respect
to substances and qualities, so that the term sambandha is used for them. The point is that contact
and inherence are dependent entities but not absolutely dependent entities. They are metaphorically
spoken of as relations. An important thing to note is Bhartṛhari’s use of the phrase taddharmaṇoḥ
“those two whose property is like the property of that [i.e., the relation],” which is an uṣṭramukha
“camel-faced” type of bahuvrīhi. MBh on A 1.1.70 (I.180.18–19): uṣṭramukham iva mukham asya so
’yam uṣṭramukhaḥ / The bahuvrīhi uṣṭramukha signifies someone who has a face (mukha) like (iva)
that of a camel (uṣṭra).
Iyer 1977: 187: “The relation called contact is expressed as that which is related (sambandhin) by its
own word (namely, saṃyoga); similarly, inherence is also expressed as that which has inherence.”
Hideyo Ogawa
323
(b) sambandhisambandhas tu ṣaṣṭhyā nimittatvāya kalpate /
(c) yadi śāstrāntaradarśanam avaśyam abhyupagantavyam etad āśrīyate
saṃyogasamavāyaṣaṣṭhīviṣaya (read: saṃyogasamavāyaḥ ṣaṣṭhīviṣaya) iti /
(d) samavāye ’pi ca saṃyogasaṃyoginoḥ samavāya iti sati vyapadeśe sambandhitvena samavāye ’bhidhīyamāne bhavitavyaṃ sambandhāntareṇa /
(e) yadi tu śāstravyapadeśa evāyaṃ lokavyavahārānupātī tatra pratipādanārtham upacāreṇa saṃvyavahāraḥ kriyate //
(a) In the utterance aṅgulyoḥ saṃyogaḥ “There is contact related to two fingers,”
contact (saṃyoga), even if it is a relation, is conveyed as something independent
(svatantra) by a saṃvijñānapada “a conveying word” like saṃyoga,30 just like
the servant in the utterance rājñaḥ puruṣaḥ “the king’s servant.” The contact
is conveyed here as that which has the property that is like the property of a
relatum (sambandhidharman).
(b) In the same utterance, on the other hand, the relation with the contact as a
relatum becomes the cause for the occurrence of a sixth-triplet ending.
(c) If another philosophical doctrine is necessarily to be accepted, this is
admitted: [in the utterance in question] inherence-relation with the contact
enters the domain of [A 2.3.50 ṣaṣṭhī śeṣe that provides for] the occurrence of
a sixth-triplet ending.31
(d) And in the case in which the expression saṃyogasaṃyoginoḥ samavāyaḥ
“There is inherence related to the contact and its bearer” is used of the inherence,
too, there must be another relation since the inherence is conveyed as a relatum.
(e) If, however, this very expression used in a scientific field follows everyday
speech, the inherence is metaphorically spoken of as a relatum in order to
afford an understanding of it.32
In [L](a) it has been shown that contact, if it is denoted by the term saṃyoga, is conveyed as
something independent (svatantra). The property of a relatum (sambandhidharma) is that of
being independent. In addition, in [L](e) it is stated that in the utterance saṃyogasaṃyoginoḥ
samavāyaḥ the inherence denoted by the term samavāya is metaphorically (upacāreṇa)
spoken of as a relatum.33
30
31
32
33
On the concept of saṃvijñānapada, see Ogawa 2010.
On A 2.3.50, see PWT 251.
See n. 28. [L](e) suggests that there is a twofold relatum: primary and metaphorical.
Puṇyarāja says that in the utterance saṃyogo dravyayoḥ “There is contact between two substances”
contact is understood from the term saṃyoga as something that has been brought to accomplishment
(pariniṣṭhitasvabhāva), that is, as a substance. Ṭīkā on VP 2.439 (173.5–10): saṃyogaśabdāt saṃyogo
dravyayor iti yathāvat pariniṣṭhasvarūpaḥ [read: pariniṣṭhitasvarūpaḥ] tatra sambandhisvabhāvaḥ
pratīyate, na ca tasya tadrūpam / evaṃ samavāyaśabdāt samavāyaḥ pratīyata ity āha – saṃbandhaḥ
samavāya ityādi / sambandhitvena sambandhasvarūpaviraheṇa samavāya ity asmāt pratīyate / athavā
yadā saṃyogaśabdaś cāsau sambandhaḥ svatantro ’bhidhīyate tadā tatra samavāyaḥ sambandhatvena
gamyata iti yadā sambandhatvena iti pāṭhas tadaivaṃ yojanīyam // Puṇyarāja tells us that there is a
variant reading of pāda d: sambandhatvena gamyate. Given this reading, the kārikā is interpreted as
arguing that in the utterance saṃyogo dravyayoḥ the inherence-relation is understood as a relation to be
denoted by the sixth-triplet endings.
324
Dignāga on the View of a Generic Term as Denoting a Relation
Now we are in a good position to give an interpretation of statement A: sambandhidharmeṇa sambandho vācyaḥ and statement B: svadharmeṇa nāsti sambandhasya vācakaḥ.
Recall that the third-triplet ending both in sambandhidharmeṇa and in svadharmeṇa can
be taken as itthambhūtalakṣaṇa-tṛtīyā (A 2.3.21).34 Thus, A and B respectively mean:
A: “A relation as characterized by the property of a relatum is to be denoted
[by a generic term];”
B: “There is no nominal which denotes a relation as characterized by its own
property.”
2.3.3 In PSV on PS V.3 (c) ([I]) as it is given by Pind, we find the expression idaṃ tat
prati, which can be deleted from the text.35 We must note the following comments by
Jinendrabuddhi:
[M] PSṬ Ms B 195a5–195b1 (Pind 2015: I 4, n. 15): [(a)] [sambandhanaṃ hi
sambandha iti … rāgavat … tasmāt sambandhidharmeṇa sambandho vācya
iti kṛtvāśaṅkitam iti … svadharmeṇa tv iti. svarūpeṇa nāsti sambandhasya
vācakaḥ śabda iti.]
[(b)] idaṃ tad iti pratyavamarśāyogyarūpatvenāsattvabhūtatvāt svarūpābhidhānaṃ praty āśaṅkaiva nāsti. ato naivāsya jātiśabdena vācyatvam upapadyata iti.
((a) omitted.) (b) A relation is not a substance (sattva) since it is incapable of
being reflexively grasped as “this” or “that;” therefore, there never arises a
suspicion that [a generic term] denotes a relation per se.
The terms dravya, sattva, and vastu are synonymous with one another in the context
of grammar.36 According to Helārāja, a substance (dravya) has the following properties
(dravyadharma):
1. the property of being capable of being reflexively grasped as “this” or “that” (idaṃ
tad iti pratyavamarśayogyatvam);
2. the property of having been brought to completion (pariniṣpannatā);
3. the property of being independent (svātantrya);
4. the property of taking on a fixed gender and number (liṅgasaṃkhyāyoga).37
34
35
36
37
See n. 26.
Pind 2015: I 5, n. 15: “(3)KV do not reproduce the expression idaṃ tad prati that is to be construed
with āśaṅkitam.”
Paddhati on Vṛtti to VP 1.13 (45.26–46.7): vastu iti / nāmapadavācyam / tyadādīnām iti sarvanāmopalakṣaṇam / sattvalakṣaṇānām iti / sattvaṃ dravyaṃ tal lakṣyata ebhir iti / vastūpalakṣaṇaṃ yatra
sarvanāma iti dravyalakṣaṇam /
Prakāśa on VP 3.1.13 (26.14–27.1): tatra dravyadharmā idaṃ tad iti pratyavamarśayogyatvam, pariniṣpannatā, svātantryam, liṅgasaṅkhyāyogaś cetyevamādayaḥ / From the point of view of those who hold
that a word denotes a substance (dravyapadārthapakṣa) Bhartṛhari says: VP 3.1.13: dravyadharmā
padārthe tu dravye sarvo ’rtha ucyate / dravyadharmāśrayād dravyam ataḥ sarvo ’rtha iṣyate // (“In
the view that a substance is a word meaning, it is stated that any word meaning has the property of
Hideyo Ogawa
325
The first property is what Bhartṛhari explains in the Bhūyodravyasamuddeśa.
[N] VP 3.4.3: vastūpalakṣaṇaṃ yatra sarvanāma prayujyate /
dravyam ity ucyate so ’rtho bhedyatvena vivakṣitaḥ //
That object, with reference to which a pronominal that refers to an entity (vastu)
is used, is called dravya when it is intended to be conveyed as something to
be differentiated (bhedyatvena vivakṣitaḥ).38
The point is that whatever the pronominals idam or tad, grammatically assigned the name
sarvanāman,39 can refer to is called a substance (dravya).40
3
Now let us turn to PSV on PS V.3. I will propose the following reading of the text.
[O] PSV on PS V.3: (a) sambandhanaṃ hi sambandhaḥ / 〈rāgādivat /
(b) so ’nyena yujyate〉 /
(c) tasmāt sambandhidharmeṇa sambandho vācya iti kṛtvāśaṅkitam /
(d) svadharmeṇa tu nāsti sambandhasya vācakaḥ śabda iti nāsti /
(e) ato naivāsya jātiśabdena vācyatvam upapadyate //
(a) Indeed, the word sambandha is a word derived in the sense of an action in
the abstract, such as the word rāga “coloring” and the like, meaning “relating,
relation.”
(b) The [relation as denoted by the word sambandha, forming one relatum], is
connected with the other [relatum].
38
39
40
a substance, by resorting to the property of a substance. Hence it is accepted that any word meaning
is a substance”); Iyer 1971: 16: “In the view that substance is the meaning of words, the meaning
of all words can have the characteristic of substance. As the meanings of words are susceptible of
assuming the properties of substance, the latter is said to be the meaning of all words.” According
to Helārāja, in the present view a substance is twofold: primary and metaphorical. Prakāśa on VP
3.1.13 (26.12–14; 27.2): yathā tava dravyavādinaḥ kecid eva śabdā mukhyadravyābhidhāyinas tadanye
tūpacaritadravyābhidhāyinaḥ tathā mamāpi jātivādino mukhyāṃ jātim abhidadhati kecit, upacaritām
anya iti matadvaye ’pi sāmyam /… teṣāṃ śuklādiguṇeṣv apy adhyāropāt te ’pi dravyasyeva dharmā
yeṣām iti dravyadharmāṇaḥ /
Iyer 1971: 123: “That in reference to which a pronoun can be used is substance, presented as something
to be differentiated.” Bhartṛhari introduces the concept of a speaker’s intention (vivakṣā) here. It depends
on the speaker’s intention what dravya is in a given situation. The status of being guṇa and that of being
dravya are not ontologically fixed; they depend on how things are spoken of.
A 1.1.27 sarvādīni sarvanāmāni // See PWT 53.
This concept of a substance (dravya) forms a pair with that of a quality (guṇa), which Bhartṛhari defines
as follows: VP 3.5.1: saṃsargi bhedakaṃ yad yat savyāpāraṃ pratīyate / guṇatvaṃ paratantratvāt
tasya śāstra udāhṛtam // (“Whatever is related [to a certain thing] and differentiates the thing [from
others] is regarded as guṇa when it activates the function [of differentiating], because it is something
dependent. This is what has been illustrated in grammar”); Iyer 1971: 126: “Whatever rests on something
else (saṃsargi) differentiates it (bhedaka) and is understood in that function (savyāpāra), is, being
dependent, called ‘quality’ in the śāstra.” A quality is defined as something dependent (paratantra),
from which it naturally follows that a substance related to the quality must be something independent.
326
Dignāga on the View of a Generic Term as Denoting a Relation
(c) Therefore, assuming that a relation as characterized by the property of
a relatum can be denoted [by the word sambandha], the proponent has a
suspicion that a generic term denotes a relation.
(d) However, considering there is no nominal which denotes a relation as
characterized by its own property, such a suspicion does not arise.
(e) Accordingly, it is absolutely improper to say that the relation is to be
denoted by a generic term.
On the basis of this text of the Vṛtti and its interpretation, I will also propose the following
reading of the third kārikā.
[P] PS V.3: sambandhaś cātra sambandhidharmeṇa vācya ucyate /
tathā hi bhāva ity ukto bhāvaś cānyena yujyate //41
The question arises: what does the word atra “here” in pāda a refer to? Consider the
immediately preceding passage.
[Q] PSV on PS V.2 (Pind 2015: I 3–4): (a) tathā hi 〈sad dravyam, san guṇaḥ,
sat karmeti bhedārthair dravyādiśabdaiḥ〉 sāmānādhikaraṇyaṃ na syāt. tac
ca dṛṣṭam.
(b) na hi sattā 〈tadyogo vā〉 dravyaṃ guṇo vā bhavati, kiṃ tarhi, dravyasya
guṇasya vā.
(c) āha ca:
vibhaktibhedo niyamād guṇaguṇyabhidhāyinoḥ
sāmānādhikaraṇyasya prasiddhir dravyaśabdayoḥ.
(a) To explain. [If a generic term denoted a relation or a generic property,]
the generic term sat could not be co-referential with words like dravya “substance” which signify individuals in the utterances sad dravyam “A substance
is existent,” san guṇaḥ “A quality is existent,” and sat karma “An action is
existent;” but, in reality, this is observed.
(b) Indeed, the generic property being or the relation to it is not a substance or
a quality. [Question] Then what? [Answer] They are something related to a
substance or a quality.42
41
42
I have emended bhāvaḥ kṛtvoktaḥ to bhāva ity ukto. Jambūvijayajī reconstructs this portion as bhāvīkṛtyocyate. NC 607: sambandho ’py atra sambandhidharmavācyo ’bhidhīyate / tathā bhāvīkṛtyocyate
bhāvo ’py anyena yujyate // Pind reports: Ms 195a5: bhāvaḥ kṛtvokta iti (Pind 2015: I 4, n. 14); Ms B
195a5: bhāva ity abhiprāyeṇoktaḥ (Pind 2015: II 19, n. 42). It is hard to justify bhāvaḥ kṛtvoktaḥ grammatically. To be sure, in view of the pratīka given by Jinendrabuddhi, we can assume bhāvīkṛtyoktaḥ
“[the relation is] said to be [a significand of the generic term] after treating it as bhāva.” But, in this
case, we have a vipulā “irregular”-anuṣṭubh. With bhāva ity ukto, on the other hand, we have a pathyā
“regular”-anuṣṭubh (5th , 6th , 7th syllables of pāda c: short, long, long). I gratefully acknowledge helpful
discussions with Dr. Yūto Kawamura on this point.
Pind 2015: II 14–16: “That is (tathā hi), there would be no co-reference (sāmānādhikaraṇyaṃ na syāt)
with words like ‘substance’ (*dravyādiśabdaiḥ) whose referents are particular [general properties]
Hideyo Ogawa
327
(c) Thus he [i.e., Bhartṛhari] says:
“The items that denote respectively something principal (guṇin) and something
subsidiary (guṇa) take different nominal vibhaktis43 because of a restriction
(niyama) [such that a sixth-triplet ending occurs only after an item whose
significand serves as something subsidiary].44 Two words, each of which
denotes a substance, are established to be co-referential with each other[, so
that the principal-subsidiary relation is not known from an utterance consisting
of the two words].”45
Note the statement “They are something related to a substance or a quality” (dravyasya
guṇasya vā) [Q(b)]. What is meant by this statement is that when the term sat denotes
the generic property being or a relation, it becomes a sambandhiśabda because a generic
property implies its bearer (jātimat) and because a relation implies its relatum. This is how
one has to have the utterance *sad dravyasya which is paraphrased as sattā dravyasya or
sambandhaḥ (yogaḥ) dravyasya.
It is important to note that Dignāga quotes VP 3.14.8. In rājñaḥ puruṣaḥ “the king’s
servant,” the significand of the nominal base rājan “king” is a qualifier (viśeṣaṇa) with
respect to the significand of the nominal base puruṣa: what is a qualifier is subsidiary to
what is a qualificand (viśeṣya). In the utterance sambandhaḥ dravyasya “the relation of the
substance” the relation behaves like a substance and serves as a qualificand with respect to
the substance, just as the quality white does in the utterance paṭasya śuklaḥ “the white of
the cloth.”
43
44
45
(*bhedārthaiḥ), like, for instance, ‘existent substance’ (*sad dravyam), ‘existent quality’ (*san guṇaḥ),
and ‘existent action’ (*sat karma); but this is observed (tac ca dṛṣṭam). For existence (sattā) or its
inherence relation (tadyogaḥ) is neither a substance (dravyam) nor a quality (guṇaḥ), but is rather (kiṃ
tarhi) of a substance (dravyasya) or of a quality (guṇasya).”
Here the term vibhakti signifies a nominal ending (sup: A 1.4.104 vibhaktiś ca).
Of this restriction Bhartṛhari says the following. VP 3.7.157: dviṣṭho ’py asau parārthatvād guṇeṣu
vyatiricyate / tatrābhidhīyamānaḥ san pradhāne ’py *upayujyate // (“Even if the relation resides in
two entities (dviṣṭha), it becomes something additional in entities that are subsidiary because of being
intended for others. The relation which resides in the subsidiary entity, when it is denoted [by a
sixth-triplet ending], is, as something residing in the principal entity also, of use to the latter.”) Iyer:
upabhujyate. I have followed Rau’s reading. Iyer 1971: 237: “Even though it (the śeṣa relation) rests on
both it brings about a distinction in what are secondary because of their being subordinate to something
else. Being expressed there (that is, in what are secondary), it touches what is primary also.”
VP 3.14.8. Iyer 1974: 125: “The two words expressive of the secondary and the primary have necessarily
different case-endings. Where they have the same case-endings, both express substance;” Pind 2015:
II 16–17: “It is, moreover, explained that (āha ca) [a word] denoting a quality and one denoting the
bearer of that quality (guṇaguṇyabhidhāyinoḥ) have different case affixes (vibhaktibhedaḥ) because of
a restrictive rule. However, for two words that denote a substance (dravyaśabdayoḥ) co-reference is
acknowledged (sāmānādhikaraṇyasya prasiddhiḥ).” Iyer takes the terms guṇa and guṇin respectively
as meaning “the secondary” and “the primary,” which I think is suited to the given context. Prakāśa
on VP 3.14.8 (154.1–8): paṭasya śukla iti dravyaguṇābhidhāyipadaprayoge śābdo guṇapradhānabhāvaḥ / … vīraḥ puruṣa ityādau tu samānādhikaraṇe viṣaye dvāv api dravyaśabdau svaniṣṭhaṃ svārtham
ācakṣāte / tathā ca prathamaiva / sāmarthyanibandhanas tu guṇapradhānabhāva ukto viśeṣyaṃ syād
anirjñātam ityādinā / evaṃ ca saty api guṇapradhānabhāve śabdāśabdatvakṛto viśeṣaḥ sāmānādhikaraṇyavaiyadhikaraṇyayor … // On the relation of the subsidiary and the principal between word
meanings, see Ogawa 2017.
328
Dignāga on the View of a Generic Term as Denoting a Relation
Thus in the given context the word atra refers to the utterance *sad dravyasya that is
equivalent to sambandhaḥ dravyasya. On this assumption, therefore, we can interpret PS
V.3 as follows:
In addition (ca), it is said (ucyate) that, in this utterance [*sad dravyasya as
paraphrased by sambandhaḥ (yogaḥ) dravyasya] (atra), a relation as characterized by the property of a relatum (sambandhidharmeṇa sambandhaḥ) is to be
denoted (vācya) [by the generic term]. That is to say (tathā hi), a relation, on
the assumption that it is bhāva [i.e., what is denoted by an action noun such as
sambandha],
is said [to be something to be denoted by a generic term] (bhāva ity uktaḥ);
but (ca), [a relation as] bhāva [i.e., a relation as denoted by such an action
noun, forming one relatum,] is connected (yujyate) with the other [relatum]
(anyena).
What is crucial for understanding the present kārikā is that a generic term, insofar as it is a
nominal (nāman), cannot denote a relation qua relation and that a relation which is denoted
by a nominal cannot claim to be a relation per se.
4 Conclusion
A relation is posited as a word meaning through invention in accordance with some tradition.
For Dignāga, who accepts the apoddhāra theory, a relation, which is a conceptual construct,
is posited as the meaning of a nominal such as sambandha on the one hand and as a
meaning of a sixth-triplet ending on the other. As is clear from [O], he accepts that a
relation is a meaning of the sixth-triplet ending. In the everyday world one not only uses the
sixth-triplet ending to convey a relation but also speaks of a relation by using the nominal.
He is aware that one has expressions such as aṅgulyoḥ saṃyogaḥ and saṃyogasaṃyoginoḥ
samavāyaḥ as well as those such as rājñaḥ puruṣaḥ. According to Dignāga, however, the
same observation of everyday usage reveals that the nominal cannot denote a relation qua
relation. Thus a generic term, being a nominal, has no possibility of denoting the relation
per se. We must pay deep attention to the fact that the basis for his arguments about a
relation is everyday speech, just as it is the basis for his arguments for positing anyāpoha
as the word meaning. It is no accident that he enters into the arguments about relations
by grammatically analyzing the term sambandha, which is commonly used in everyday
speech. Dignāga knows that grammar is grounded on everyday speech.
References and abbreviations
A Pāṇini’s Aṣṭādhyāyī. Appendix III (Aṣṭādhyāyīsūtrapāṭha). In: Cardona 1997.
Cardona 1997 G. Cardona, Pāṇini, His Work and Its Traditions. Volume I: Background
and Introduction. Delhi 2 1997.
Cardona 1999 G. Cardona, Approaching the Vākyapadīya. Journal of the American Oriental Society 119.1 (1999) 88–125.
Hideyo Ogawa
329
Houben 1995 J. E. M. Houben, The Saṃbandha-samuddeśa (Chapter on Relation) and
Bhartṛhari’s Philosophy of Language: A Study of Bhartṛhari Saṁbandha-samuddeśa
in the context of the Vākyapadīya with a translation of Helārāja’s commentary
Prakīrṇa-prakāśa. Groningen 1995.
Iyer 1971 K. A. Subramania Iyer, The Vākyapadīya of Bhartṛhari: Chapter III. pt. i;
English Translation. Poona 1971.
Iyer 1974 K. A. Subramania Iyer, The Vākyapadīya of Bhartṛhari: Chapter III. pt. ii;
English Translation with Exegetical Notes. Delhi 1974.
Iyer 1977 K. A. Subramania Iyer, The Vākyapadīya of Bhartṛhari: Kāṇḍa II; English
Translation with Exegetical Notes. Delhi 1977.
MBh Patañjali’s Vyākaraṇamahābhāṣya: (a) Śrībhagavatpatañjaliviracitaṃ Vyākaraṇamahābhāṣyam (Śrīkaiyaṭakṛtapradīpena nāgojībhaṭṭa-kṛtenabhāṣyapradīpoddyotena
ca vibhūṣitam), ed. Vedavrata. 5 vols. Gurukul Jhajjar (Rohatak) 1962–63. (b) The
Vyākaraṇa-mahābhāṣya of Patañjali, edited by F. Kielhorn, third edition, revised
and furnished with additional readings, references and select critical notes by K. V.
Abhyankar. 3 vols. Poona 1962–72. (1: 1962; 2: 1965; 3: 1972.) [References of the
text of the Mahābhāṣya are to volumes, pages, and lines of MBh (b).]
NC Siṃhasūri’s Nyāyāgamānusāriṇī nayacakravālavṛtti: Dvādaśāraṃ Nayacakraṃ of
Ācāryā Śrī Māllavādi Kṣamāśramaṇ; with the Commentary Nyāyāgamānusāriṇī of
Śrī Siṃhasūri Gaṇi Vādi Kṣamāśramaṇa, Part II (5–8 Aras), ed. Muni Jambūvijayajī.
Bhāvnagar 1976.
Ogawa 2009 H. Ogawa, On Bhartṛhari’s notion of ‘power’ (śakti). In: Bhartṛhari: Language, Thought and Reality (Proceedings of the International Seminar, Delhi, December 12–14, 2003), ed. M. Chaturvedi. Delhi 2009, 225–252.
Ogawa 2010 H. Ogawa, Bhartṛhari on unnamable things. In: Logic and Belief in Indian
Philosophy, ed. P. Balcerowicz. Delhi 2010, 403–418.
Ogawa 2017 H. Ogawa, The qualifier-qualificand relation and coreferentiality in Dignāga’s
apoha theory. In: Reading Bhaṭṭa Jayanta on Buddhist Nominalism, ed. P. McAllister.
Vienna 2017, 83–152.
Ogawa forthcoming H. Ogawa, On a bias for doxographical accounts in later commentaries on the Vākyapadīya of Bhartṛhari: With special reference to pratibhā. (Forthcoming.)
Paddhati Vṛṣabhadeva’s Paddhati: See VP (c).
Pind 2009 O. H. Pind, Dignāga’s philosophy of language: Dignāga on anyāpoha; Pramāṇasamuccaya V, texts, translation, and annotation. PhD diss., Universität Wien.
Wien 2009. http://othes.univie.ac.at/8283/, last visited 16-05-2016.
Pind 2015 O. H. Pind, Dignāga’s Philosophy of Language. Dignāga. Pramāṇasamuccayavṛtti V on anyāpoha. Part I: Text. Part II: Translation and Annotation, ed. E.
Steinkellner. Vienna 2015.
Pradīpa Kaiyaṭa’s Pradīpa: See MBh (a).
Prakāśa Helārāja’s Prakāśa: See VP (b) and VP (d).
PS Dignāga’s Pramāṇasamuccaya: See Pind 2015.
PSṬ Jinendrabuddhi’s Pramāṇasamuccayaṭīkā: See Pind 2009.
PSV Dignāga’s Pramāṇasamuccayavṛtti: See Pind 2009.
PWT See Cardona 1997.
330
Dignāga on the View of a Generic Term as Denoting a Relation
SK Bhaṭṭojī Dikṣita’s Vaiyākaraṇasiddhāntakaumudī: Śrī-bhaṭṭojī-dīkṣita-viracitā vaiyākaraṇa-siddhānta-kaumudī śrīmadvāsudeva-dīkṣita-praṇītayā bālamanoramākhyavyākhyayā śrīmaj-jñānendra-sarasvatī-viracitayā tattva-bodhiny-ākhya-vyākhyayā
ca sanāthitā, ed. Giridhara Śarmā Caturveda and Parameśvarānanda Śarmā Bhāskara.
4 vols. Varanasi 1958–61.
Ṭīkā Punyarāja’s Ṭīkā. See VP (e).[References are to pages and lines.]
VP Bhartṛhari’s Vākyapadīya: (a) Bhartṛharis Vākyapadīya: Die Mūlakārikās nach den
Handschriften herausgegeben und mit einem Pāda-Index versehen, ed. W. Rau. Wiesbaden 1977. [References to kārikās of the Vākyapadīya are made according to Rau
1977, so that kārikā numbers are given according to his edition.] (b) Vākyapadīya
of Bhartṛhari with the Commentary of Helārāja, Kāṇḍa III, Part 1, ed. K. A. Subramania Iyer. Poona 1963. (c) Vākyapadīya of Bhartṛhari with the Commentaries
Vṛtti and Paddhati of Vṛṣabhadeva, ed. K. A. Subramania Iyer. Poona 1966. (d)
Vākyapadīya of Bhartṛhari with the Commentary of Helārāja, Kāṇḍa III, Part II, ed.
K. A. Subramania Iyer. Poona 1973. (e) The Vākyapadīya of Bhartṛhari (An Ancient
Treatise on the Philosophy of Sanskrit Grammar), Containing the Ṭīkā of Puṇyarāja
and the Ancient Vṛtti, Kāṇḍa II, with a Foreword by Ashok Aklujkar. Delhi 1983.
Vṛtti Vākyapadīya Vṛtti. See VP (c) and VP (e). [References to kārikās, pages, and lines.]
Communication between a Speaker and a Listener as the
Seeing of a Double Moon — In Light of the Apoha Theory of
Śāntarakṣita and Kamalaśīla
by
Kensho Okada
Introduction
We experience the practical use of words in our everyday activities 1) when a person understands/grasps an object through a word, and 2) when one person tells another something.
The latter case is a person-to-person interaction in which a speaker intends to make a listener
understand something by speaking word(s), while the listener infers the speaker’s intention
by hearing and interpreting them. Communication is said to be established between the
speaker and listener only when they are both present. People are, however, quite distinct
from one another. Hence, even if the speaker and listener are present in the same time
and space, they perceive altogether different things/images respectively. How can they
communicate with each other under such circumstances? Also, what is the way in which
they engage in conversation with one another?
Śāntarakṣita (ca. 725–788) and Kamalaśīla (ca. 740–795) present an explanation of this
issue in the Śabdārthaparīkṣā chapter of their Tattvasaṃgraha (TS) and its pañjikā (TSP).
This paper aims to examine their view to clarify how they answer the question at hand.
As is well known, Śāntarakṣita offers a threefold classification of the theory of the
“exclusion of others” (anyāpoha):
1. Nominally bound negation/Implicative negation (paryudāsa)
a) Exclusion having cognition as its essence (buddhyātmāpoha)
b) Exclusion having a referent as its essence (arthātmāpoha)
2. Verbally bound negation/Non-affirmative negation (prasajyapratiṣedha)1
Of these three exclusions, two of the paryudāsa variety and one prasajyapratiṣedha, Śāntarakṣita asserts that the exclusion having cognition as its essence (= reflection, pratibimba)
is the principal referent denoted by a word, and that the other two are understood by implication. In this way, when reflection is admitted to be the principal referent, the aforementioned
question regarding communication arises, for reflections manifest entirely differently in the
cognition of individuals. How do Śāntarakṣita and Kamalaśīla resolve this issue?
1
Editor’s note: This contribution contains changes that the author has left to the editor’s discretion.
Cf. TS 1003, 1010, and 1014cd.
Birgit Kellner et al., eds., Reverberations of Dharmakīrti’s Philosophy: Proceedings of the Fifth International Dharmakīrti
Conference Heidelberg, August 26 to 30, 2014. Vienna 2020, pp. 331–342.
332
Communication between a Speaker and a Listener
1. Impossibility of verbal convention-setting
First, let us examine Śāntarakṣita’s statement, which is the origin of the question we seek
to answer. In TS 870, he asserts that the 5 categories of particulars, genus, connection to
genus, particulars qualified by genus, and the form of cognition are, ultimately, not suitable
as the referent denoted by a word.2 To begin with, arguing over particulars Śāntarakṣita
and Kamalaśīla state the following:
tatra svalakṣaṇaṃ tāvan na śabdaiḥ pratipādyate /
saṃketavyavahārāptakālavyāptiviyogataḥ // TS 871 //
Of these, first of all, particulars cannot be understood by words. Because
[particulars are] devoid of the pervasion of the time extending from verbal
convention to verbal activity, [convention cannot be set with regard to particulars].3 (871)4
TSP 341, 14–17 ad TS 871:
etad uktaṃ bhavati – samayo hi vyavahārārthaṃ kriyate, na vyasanitayā /
tena yasyaiva saṃketavyavahārāptakālavyāpakatvam asti, tatraiva samayo
vyavahartṝṇāṃ yuktaḥ, nānyatra / na ca svalakṣaṇasya saṃketavyavahārāptakālavyāpakatvam asti / tasmān na tatra samaya iti /
[By this,] the [following] is said: Indeed, verbal convention is made for [use at
the time of] verbal activity,5 not through [mere] desire for [it]. Therefore, it is
suitable to [set] verbal convention only with regard to that which pervades the
time pervaded by verbal convention and verbal activity for [people] living daily
[with use of language], not on anything else. And particulars do not pervade
the time extending from verbal convention to verbal activity. Therefore, verbal
convention [can]not be [set] with regard to them [i.e., particulars].
Here, from the viewpoint of verbal convention (saṃketa) and verbal activity (vyavahāra),
Śāntarakṣita and Kamalaśīla examine whether particulars are suitable to be the referent
denoted by a word and indicate that particulars cannot be the target of verbal convention.
In the following verses, Śāntarakṣita asserts that (1) convention-setting with regard to
particulars is useless for verbal activity6 and that (2) it is impossible to make verbal
2
3
4
5
6
Cf. TS 870: yataḥ svalakṣaṇaṃ jātis tadyogo jātimāṃs tathā / buddhyākāro na śabdārthe ghaṭām
añcati tattvataḥ //; McAllister 2011: 181–182.
Cf. TSP 341,13–14 ad TS 871: na tatra svalakṣaṇe samaya iti śeṣaḥ /
Cf. PV I 92: śabdāḥ saṃketitaṃ prāhur vyavahārāya sa smṛtaḥ / tadā svalakṣaṇaṃ nāsti saṃketas tena
tatra na // “Words express something upon which verbal convention has been agreed and it [i.e., verbal
convention] is recalled for the purpose of verbal activity. At the time [of verbal activity] the particular
[with regard to which verbal convention is set] no longer exists. Therefore, verbal convention is not set
with regard to the particular.”
According to Śākyabuddhi’s commentary, Dharmakīrti asserts that the purposes of verbal convention are
1) the understanding of form of cognition and 2) the apprehension of particulars. Cf. PVṬ D127a7–b5/
P151a3–b2 ad PV I 110d and PVSV 57,15–19.
Cf. TS 873: tasmāt saṃketadṛṣṭo ’rtho vyavahāre na dṛśyate / na cāgṛhītasaṃketo gamyate ’nya
(gamyate ’nya G; (bodhyetā)nya B) iva dhvaneḥ // “Therefore, the thing perceived [at the time of]
Kensho Okada
333
convention in regard to particulars.7 In both cases, on the grounds that particulars cannot
be something on which verbal convention is established, Śāntarakṣita criticizes the view
that particulars are the referent denoted by a word. In the same way, genus, connection to
genus, particulars qualified by genus,8 and the form of cognition are not accepted as the
word-referent due to the impossibility of convention-setting.9
2. Opponent’s criticism
In the final part of the Śabdārthaparīkṣā chapter of TS/TSP, the impossibility of setting
verbal convention and the uselessness of it are discussed once again.10
As previously mentioned, Śāntarakṣita and Kamalaśīla admit that the principal referent
denoted by a word is the reflection manifesting in cognition. Therefore, for them, the very
reflection is the referent which sets with regard to verbal convention. Criticizing their view,
an opponent indicates that the impossibility of setting verbal convention and the uselessness
7
8
9
10
verbal convention is never perceived at [the time of] verbal activity. And, that which is not grasped by a
verbal convention cannot be understood by a word. Just as with the fact that another thing [e.g. horse]
cannot be [understood by the word ‘cow’].”
Cf. TSP 344,1–3 ad TS 875–876: evaṃ tāvat svalakṣaṇe vyavahārānupapatteḥ samayavaiyarthyaprasaṅgān na samayaḥ sambhavatīti pratipāditam / sāmpratam aśakyakriyatvād eva na sambhavatīti
pratipādayann āha – aśakyam ityādi / “In this way, because verbal action toward particulars is impossible, it would follow that verbal convention is useless, [Therefore, the setting of] verbal convention with
regard to particulars is impossible. This is, firstly, understood. Next, in order to explain that, because it is
impossible to make [verbal convention], [the setting of verbal convention] is impossible, [Śāntarakṣita]
states ‘aśakyam’ and so on.”
Cf. TS 882: jātisambandhayoḥ pūrvaṃ vyāsataḥ pratiṣedhanāt / nānantarāḥ prakalpyante śabdārthās
trividhāḥ pare // “Because genus and connection have been already rejected in detail before, the other
three [which are enumerated] immediately after [particulars] are not supposed to be the referent of a
word.” TSP 348,7–8 ad TS 882: evaṃ tāvat svalakṣaṇe samayāsambhavaṃ pratipādya jātyādiṣu triṣu
samayāsambhavaṃ pratipādayann āha – jātisambandhayor ityādi / “In this way, to begin with, the
impossibility of [setting] verbal convention with regard to particulars is explained. After this, in order
to explain the impossibility of [setting] verbal convention with regard to three things – that is, genus
and so on [Śāntarakṣita] states ‘jātisambandhayor’ and so on.”
Cf. TS 884: buddhyākāraś ca buddhistho nārthabuddhyantarānugaḥ / nābhipretārthakārī ca so ’pi
vācyo na tattvataḥ // “Moreover, existing [only] in some cognition [as identical in nature], the form
of cognition follows neither [external] things nor other cognitions, and does not accomplish desired
purposes. Ultimately, that [i.e., the form of cognition] is also not the referent denoted [by a word].” TSP
349,2–4 ad TS 884: buddhyākāro hi tādātmyena buddhāv evāvasthita iti nāsau tadbuddhisvarūpavat
pratipādyam arthaṃ buddhyantaraṃ vānugacchati / tataś ca saṃketavyavahārāptakālāvyāpakatvāt
svalakṣaṇavan na tatrāpi samayaḥ sambhavati / “… Therefore, [the setting of] verbal convention with
regard to it [i.e., form of cognition] is also impossible, because it does not pervade the time extending
from verbal convention to verbal activity, just as particulars.”
Cf. TS 1206ab: saṃketāsambhavo hy atra bhedādau sādhitaḥ purā / “Indeed, the impossibility of [the
setting of] verbal convention with regard to particulars and so on has already been proved before.” TSP
446,8–10 ad TS 1206ab: aśakyasamayatvād ananyabhāktvāc ceti pūrvaṃ svalakṣaṇādau saṃketāsambhavasya saṃketavaiphalyasya ca prasādhitatvāt / “It is because the impossibility of [setting] verbal
convention and the uselessness of it with regard to particulars and so on have already been proved
in such ways as ‘because [the setting of] verbal convention is impossible’ and ‘because of having no
connection with the others’.”
334
Communication between a Speaker and a Listener
of it are also applied to the reflection that they accept as the object of convention. The
opponent says the following:
nanu cāpohapakṣe ’pi kathaṃ saṃketasaṃbhavaḥ /
sāphalyaṃ ca kathaṃ tasya na dvayoḥ sa hi siddhyati // TS 1207 //
vaktṛśrotror na hi jñānaṃ vedyate tat parasparam /
saṃkete na ca tad dṛṣṭaṃ vyavahāre samīkṣyate // TS 1208 //
[Objection:] Even in the apoha theory, (1) how can convention[-setting] be
possible? (2) How can it be useful, either? (1) For it [i.e., the apoha as reflection]
cannot be established between two [persons, i.e.,] a speaker and listener [in
the same way]. It is because [they can]not mutually cognize [one another’s]
cognition. (2) And, that [reflection] which was perceived at the time of [setting
the] verbal convention is not seen [any longer] at the time of verbal activity.
Here, adopting the same method as Śāntarakṣita, the opponent indicates that it can be
neither possible nor useful to set verbal convention with regard to a reflection.
As far as the impossibility of the setting of verbal convention with regard to a reflection is
concerned, according to the opponent, the speaker never cognizes the reflection manifesting
in the listener’s cognition, nor does the listener cognize the reflection manifesting in the
speaker’s cognition. Therefore, it cannot be established that their reflections are one and
the same. This is the reason for the impossibility of the setting of verbal convention with
regard to such reflections.
Regarding the uselessness of verbal convention-setting, the opponent asserts that,
because the reflection perceived at the time of verbal convention and the one perceived
at the time of verbal activity are totally different, it follows that setting verbal convention
with regard to a reflection is not useful for ordinary verbal usage.11
With regard to the former question, Kamalaśīla explains as follows:
TSP 447,1–7 ad TS 1208’ab:
kasmād ity āha – na hi jñānam ityādi / pratyātmasaṃvedanīyam evārvāgdarśanānāṃ jñānam / na hy anyadīyajñānam aparo ’paradarśanaḥ saṃvedayate / jñānād avyatiriktaś ca paramārthataḥ pratibimbātmakalakṣaṇāpohaḥ /
11
Commenting on the opponent’s view, Kamalaśīla explains the uselessness of the setting convention with
regard to reflections as follows: Cf. TSP 447,7–11 ad TS 1208cd: ānarthakyaṃ ca pratipādayann āha –
saṃkete na cetyādi / yat saṃketakāle pratibimbakam anubhūtaṃ śrotrā vaktrā vā, na tad vyavahārakāle
’nubhūyate, tasya kṣaṇakṣayitvena ciraniruddhatvāt / yac ca vyavahārakāle ’nubhūyate na tat saṃketakāle dṛṣṭam, anyasyaiva tadānīm anubhūyamānatvāt / na cānyatra saṃketād anyatra vyavahāro yuktaḥ,
atiprasaṅgād iti / “In addition, in order to explain the uselessness [of setting convention with regard
to reflection, the opponent] states ‘saṃkete na ca’ and so on. Some reflection has been experienced
by a listener or by a speaker at the time of the verbal convention. [However, ] that [reflection] cannot
be experienced at the time of verbal activity. This is because that [reflection], being momentary, has
ceased to existence long before [the time of the verbal activity]. Moreover, that which is experienced at
the time of the verbal activity cannot be that which has been perceived at the time of verbal convention.
This is because a very different [reflection] is experienced at that time (= at the time of verbal activity).
And, it is untenable that verbal activity regarding Y is [established] on the basis of verbal convention
[set with regard to] X. This is because [otherwise] an absurd consequence would follow.”
Kensho Okada
335
tataś ca vaktṛśrotror dvayor api kasyacid ekasya saṃketaviṣayasyārthasyāsiddheḥ kutra saṃketaḥ kriyate gṛhyate vā / na hy asiddhe vastuni vaktā
saṃketaṃ kartum īśaḥ, nāpi12 śrotā gṛhītum, atiprasaṅgāt / tathā hi – śrotrā13
yat pratipadyate svavijñānārūḍham arthapratibimbakaṃ na tad vaktrā saṃvedyate / yac ca vaktrā saṃvedyate na tac chrotrā, svasya svasyaivāvabhāsasya
vedanāt /
[Reply:] Why? [Objection:] “[They can]not [cognize one another’s] cognition” and so on. Ordinary persons can only cognize something to be cognized
by themselves, for, being an ordinary person, no one knows the cognition
[represented in] another’s [mind]. Moreover, the exclusion, having a reflection
as its own nature, is ultimately not different from the cognition. Therefore, [the
following question arises:] With regard to what [object] can the convention
be set [by a speaker]? In what [object] can the convention be comprehended
[by a listener]? It is because the same object is not established as the target
of verbal convention at all between the two persons–that is, the speaker and
listener. Indeed, if nothing real is established, the speaker cannot make any
convention; nor can the listener comprehend it. This is because [if such a
thing were possible,] an absurd consequence would follow. Namely, the listener would comprehend the reflection of an object manifesting in his own
cognition; but the speaker would be unable to cognize it. The listener cannot
comprehend what the speaker cognizes. This is because [each] cognizes only
his own manifestation.
Here, the argument encounters difficulty when a reflection is regarded as the object with
regard to which verbal convention is set. Reflections are different for each individual. An
ordinary person can only cognize the reflection manifesting in his own cognition. It is
impossible for a reflection to move from one person’s cognition to another’s. Therefore,
in reality, one can neither tell the content of his cognition to another, nor perceive that
of another’s cognition. As a result, it may follow that there is not any object of verbal
convention that a speaker and listener share at the time of verbal convention, and thus, the
verbal convention itself cannot be established.
3. Śāntarakṣita and Kamalaśīla’s reply
Śāntarakṣita and Kamalaśīla sought to address this issue. How do they explain the manner
of establishing verbal convention? Śāntarakṣita says the following:
svasya svasyābhāsasya vedane ’pi sa vartate /
bāhyārthādhyavasāyo14 yo15 dvayor api samo yataḥ // 1209 //
timiropahatākṣo hi yathā prāha śaśidvayam /
12
13
14
15
īśaḥ, nāpi B; īśāno ’pi G.
śrotrā em. (cf. nyan pa pos D, P); śrotā G, B.
°ādhyavasāyo em. (cf. zhen pa D, P; bāhyārthādhyavasāyas tulya eva TSP 477,18); °ādhyavasāye G, B.
yo em.; yad G, B; yad om. D, P.
336
Communication between a Speaker and a Listener
svasamāya tathā sarvā śābdī vyavahṛtir matā // 1210 //
vyāpakatvaṃ ca tasyedam iṣṭam ādhyavasāyikam /
mithyāvabhāsino hy ete pratyayāḥ śabdanirmitāḥ // 1211 //
It is because, although [a speaker and listener] cognize their respective manifestations, they both equally determine [their respective manifestations] to
be an external object. Just as one [person] whose eyes are affected by eye
disease says to [another person] like himself “there is a double moon,” all
verbal activities are thought to be the same. That [i.e., reflection] is accepted as
being such a pervader, [i.e., being that which pervades the time extending from
verbal convention to verbal activity,]16 on the basis of [persons] determining
[so]. In actuality, these notions brought about by word(s) are [nothing more
than] something manifesting erroneously.
Commenting on TS 1209, Kamalaśīla states the following:
TSP 447,12–448,3 ad TS 1209:
na hi paramārthato jñānākāro ’pi śabdānāṃ vācyatayā ’bhīṣṭaḥ, yena tatra
saṃketāsambhavaś codyate / yataḥ sarva evāyaṃ śābdo vyavahāraḥ svapratibhāsānurodhena taimirikadvayadvicandradarśanavad bhrānta iṣyate / kevalam arthaśūnyābhijalpavāsanāprabodhāc chabdebhyo ’rthādhyavasāyivikalpamātrotpādāt 17 tat pratibimbakaṃ śabdānāṃ vācyam ity abhidhīyate
jananāt, na tv abhidheyatayā / tatra yady api svasya svasyaivāvabhāsasya
vaktṛśrotṛbhyāṃ paramārthataḥ saṃvedanam, tathāpi taimirikadvayasyeva
bhrāntibījasya tulyatvād dvayor api vaktṛśrotror bāhyārthādhyavasāyas tulya
eva / tathāpi vaktur ayam abhimāno vartate – yam evāham arthaṃ pratipādye
tam evāyaṃ pratipadyata iti / evaṃ śrotur api yojyam /
ekārthādhyavasāyitvaṃ katham anayor vaktṛśrotroḥ parasparaṃ viditam iti
cet, yadi nāma paramārthato na viditam, tathāpi bhrāntibījasya tulyatvād
asty eva paramārthataḥ svapratibhāsānurodhena taimirikadvayavad bhrānta
evāyaṃ vyavahāra iti niveditam etat / tenaikārthādhyavasāyavaśāt saṃketakaraṇam upapadyata eva /
Indeed, ultimately, the form of cognition is also not admitted to be the referent
denoted by words, so that the impossibility of [setting] convention with regard
to it [i.e., the form of cognition] would be criticized. It is because all verbal
activities are admitted to be something erroneous, just as in the case of two
persons affected by eye disease [both] seeing double moons in accordance with
their respective manifestations. Merely by awakening of the latent disposition
which [is deposited] through objectless discourse, only conceptual [cognitions],
by which [a manifestation] is determined as being an object, are produced
by words. [Hence,] on the grounds that [conceptual cognition is] produced
16
17
Cf. TSP 448,6–8 ad TS 1211: saṃketavyavahārāptakālavyāpakatvaṃ ca vaktṛśrotṛbhyām adhyavasitārthapratibimbakasyāvasāyavaśād eveṣṭaṃ, na paramārthataḥ /
°ādhyavasāyi° B; °āvasāyi° G.
Kensho Okada
337
[by a word], that reflection [of conceptual cognition] is said to be the referent
denoted by words; not on the grounds that [the reflection is, in reality, the
referent denoted by a word]. That being the case, although a speaker and
listener ultimately only cognize their respective manifestations, they still both
equally determine [their respective manifestations] to be external objects,
because [they] share the cause of the error, just as in the case of two persons
affected by eye disease [both seeing a double moon]. Even so, the speaker
thinks, “He [i.e., the listener] also understands the object that I understand.”
So does the listener.18
[Question:] How can the speaker and the listener know that they are determining the same object?
[Reply:] Although they ultimately cannot know [it], such a verbal activity,
which is nothing but erroneous, [is still established between them] in accordance with their respective manifestations, because [they] share the cause of
the error, just as the case of the two persons affected by eye disease. It has
already been explained. Therefore, it is quite possible that verbal convention
is made by the force of [people] determining [their respective manifestations]
to be the same.
Śāntarakṣita and Kamalaśīla accept that, in reality, a speaker and listener cognize only
their own respective manifestations. According to them, however, a speaker and listener
determine their own manifestations as external things in the same way. This same determination is the reason for the establishment of verbal convention. That is to say, a convention
is established on the basis of the thinking that “we perceive the same thing.” Based on this
erroneous determination, two persons can communicate.
Moreover, there would appear to be no difference between the verbal activity of persons
affected by eye disease and that of ordinary persons, given that all people cognize only
their own respective manifestations. Śāntarakṣita and Kamalaśīla therefore assert that all
verbal activity is erroneous.
4. Causal relation between word and intention
It is found that Kamalaśīla uses this example of “two persons affected by eye disease seeing
a double moon in the same way” also in the Śābdavicāra section of the Pramāṇāntaraparīkṣā
chapter and Śrutiparīkṣā chapter of the TSP. There, Śāntarakṣita denies that there is an
invariable connection between a word and an external thing, and suggests that a causal
relation is established between a word and speaker’s intention.19 While commenting on
18
19
Kamalaśīla has already stated the same thing in the following portion. Cf. TSP 358,9–12 ad TS 908: etad
uktaṃ bhavati – paramārthataḥ svapratibhāsānubhave ’pi vaktur evam adhyavasāyo bhavati – mayā
’smai bāhya evārthaḥ pratipādyate / śrotur apy evaṃ bhavati – mamāyaṃ bāhyam eva pratipādayatīti /
atas taimirikadvayadvicandradarśanavad ayaṃ sarvaḥ śābdo vyavahāra iti (cf. sgra las byung ba’i tha
snyad ’di thams cad de lta bu yin no D, P) /
Cf. TS 1512 (Śābda): vacasāṃ pratibandho vā ko bāhyeṣv api vastuṣu / pratipādayatāṃ tāni yenaiṣāṃ
syāt pramāṇatā // “How can be there any invariable connection between words and external things? If
338
Communication between a Speaker and a Listener
such statements by Śāntarakṣita, Kamalaśīla gives explanations with use of the example of
“two persons affected by eye disease.”20
Here, let us consider the views of Śāntarakṣita and Kamalaśīla as presented in the
Śābdavicāra section of the TS/TSP. There, as mentioned above, rejecting an invariable
connection between a word and an external thing, Śāntarakṣita explains the manner in
which verbal cognition is meant to be included in the inference.
vacobhyo nikhilebhyo ’pi vivakṣaiṣānumīyate /
pratyakṣānupalambhābhyāṃ taddhetuḥ sā hi niścitā // TS 1514 //
From all words it follows that [speaker’s] intention is inferred. For it [i.e.,
intention] is ascertained to be the cause of a word through perception and
non-perception.
Here, Śāntarakṣita states that an invariable connection having causality as its character is
established between a word and an intention, and suggests that a word is a valid cognition
with regard to intention. In some cases, however, an intention and an uttered word are
known to be different. That is to say, it is just as in the case of some speaker uttering the
word “Y” erroneously while intending to speak about “X.” In this case, it is assumed that
the fallacy lies not in the word but in the erroneous speaker.21 Furthermore, with respect to
a word’s deviation from a particular intention, Kamalaśīla gives explanatory statements as
follows:
20
21
[invariable connection were to be admitted], words expressing them [i.e., external things,] would be
valid cognition.” TS 2618 (Śruti): sākṣāc chabdā na bāhyārthapratibandhavivekataḥ / gamayantīti ca
proktaṃ vivakṣāsūcakās tv amī // “It has already explained that words do not make [external things]
known directly, because there is no invariable connection between external things [and words]. And,
these [i.e., words,] are [nothing but] something indicating [the presence of the speaker’s] intention.” TSP
854,1–3 ad TS 2618: vivakṣām api na vācyatayā pratipādayanti, kiṃ tarhi, liṅgatayā sūcayanti / ata eva
sūcakā ity uktam / tathā hi – śabdād uccaritād arthādhyavasāyī vikalpo jāyate, na vivakṣādhyavasāyī /
“[Words] do not express [the speaker’s] intention as something denoted [by words]. How then? [Words
merely] indicate [the speaker’s intention] as a logical mark. Therefore, [words are] said to be something
indicating. That is to say, [when] a word is uttered, there arises a conceptual cognition which judges an
[external] thing, not [a conceptual cognition which] judges [the speaker’s] intention.”
In his TS 2620, Śāntarakṣita asserts that only when a person has already cognized a causal relation, can
he know a speaker’s intention through words. Against this, the opponent objects that verbal convention
cannot be a means for knowing a particular intention of a speaker. In response, Kamalaśīla criticizes
the opponent’s view with use of the example of “two persons affected by eye disease seeing a double
moon.” Cf. TSP 854,12–22 ad TS 2620.
Cf. TS 1515–1516: bhrāntasyānyavivakṣāyāṃ vākyaṃ ced anyad īkṣyate / yathāvivakṣam (yathāvivakṣam B; tathā vivakṣam G) apy etat tasmān (etat tasmān B; etatta(kṣāsāmānye ta?)smān G [sic])
naiva pravartate // bhrāntābhrāntaprayuktānāṃ vailakṣaṇyaṃ parisphuṭam / vidagdhāḥ prakṛtādibhyo
niścinvanti girām alam // vailakṣaṇyena hetūnāṃ viśeṣaṃ tāsu ye na tu / avagacchanti doṣo ’yaṃ
teṣāṃ liṅgasya nāsti tu // “[Objection:] It is found that as for an erroneous person, his statement and
intention are different. Therefore, it [i.e., his statement] cannot be made in accordance with intention.
[Reply:] There is clearly a difference between words used by an erroneous person and those used by
a non-erroneous person. [However,] clever men [definitely] discern [this difference] through context
and so son. The fallacy lies not in the logical mark, but in the persons who do not notice the difference
among them [i.e., these words] due to difference of their causes.”
Kensho Okada
339
TSP 541, 5–11 ad TS 1517–1518:
api ca, yadā sarva evāyaṃ śābdo vyavahāras taimirikadvayadvicandradarśanavat svapratibhāsānurodhena bhrānta eveṣyate, tadā kathaṃ vivakṣāviśeṣe
vyabhicārodbhāvanād aprāmāṇyaprasañjanaṃ syāt, tatra vivakṣāviśeṣe pāramārthikasya prāmāṇyasyānabhyupagamāt / tathā cāha – saṃketāpekṣayā
tasya hṛdi kṛtvā prakāśanam / anumānatvam uddiṣṭaṃ na tu tattvavyapekṣayā // iti /
Furthermore, when all verbal activity is accepted to be erroneous due to
dependence on [respective] manifestations of one’s own, just as two persons
affected by eye disease see a double moon [in the same way], how can [a word/
verbal cognition] be invalid on the basis of pointing out the deviation from
a particular intention [of speaker]. For it is not accepted that [a word/verbal
cognition has] ultimate validity with regard to a particular intention of speaker.
The same thing is stated as follows: A word (/verbal cognition) is said to be an
inference just when it is taken into consideration that a word makes manifest
[an object] with dependence on verbal convention, not when with dependence
on real truth.22
Here, from the standpoint that all verbal activity is erroneous, Kamalaśīla implicitly indicates that the validity of words/verbal knowledge is not ultimate but conventional. Thus, as
long as the erroneousness of verbal knowledge is considered, it follows that the validity of
a word/verbal knowledge depends only on verbal convention, and has nothing to do with
real truth.
For example, when one taimirika (person affected by eye disease) says to another
taimirika “there is a double moon in the sky” and the latter agrees with the former, it
follows that based on this agreement, verbal convention is established between them. In this
case, in reality, the so-called “double moon” does not exist, and the word “double moon”
would be nonsense. However, to the two taimirika the word “double moon” is thought to
be useful.
Conclusion
To conclude, when reflection is admitted to be the referent denoted by a word, a speaker and
listener cannot share the same real object, because they only cognize different reflections
manifesting in their respective cognitions. They, however, think/believe that they perceive
the same thing by the force of determination. Based on this same determination, the setting
of convention is established. Therefore, according to Śāntarakṣita and Kamalaśīla, all verbal
activity is erroneous, as it is the same as when two persons affected by eye disease both see
a double moon.
22
The source for this citation is uncertain.
340
Communication between a Speaker and a Listener
Supplemental remarks
As discussed above, bearing in mind that a person cognizes only his own representation,
Śāntarakṣita and Kamalaśīla suggest that all verbal activity is established in the same
ways as two persons with eye disease equally see a double moon. Interestingly enough,
similar views have been presented by Dharmakīrti in his Santānāntarasiddhi (SS) and by
Vinītadeva in his Santānāntarasiddhiṭīkā (SSṬ).23 Dharmakīrti and Vinītadeva state the
following:
23
For your information, similar descriptions can also be found in Śubhagupta’s Anyāpohavicārakārikā
(AVK) and Karṇakagomin’s Pramāṇavārttikasvavṛttiṭīkā (PVSVṬ), as far as treatises on apoha theory
are concerned. Cf. Mikogami 1993. Cf. AVK D198a1–2: rang gi blo la snang ba ni // rig pa yin yang
phyi rol dang // gcig par ’dzin pa’i nga rgyal skye // rab rib can du mtshungs blo bzhin // “Although
[persons] cognize their own [respective] manifestations, they [erroneously] think ‘We comprehend the
same external thing,’ just as the cognitions of persons suffering from same eye disease.” This verse
is quoted in Anekāntajayapatākā (AJP 338,6–7 svabuddhipratibhāsasya saṃvittāv api jāyate / bahirarthagrahe mānas tulyataimirabuddhivat //); PVSV 60,26–61,1: … pratipattā pratipattim anusṛtya
ete vṛkṣā iti svaparavikalpeṣv ekapratibhāsān ādarśya vikalpavijñāne vyavasthitas tadvijñānahetūn
bhedena pratipadyetety uktim ataddhetubhyo bhede niyuṅkte. “… [One who makes the setting of a
verbal convention] applies a word to what is excluded from those which are not its causes with the
following hope: The cognizer [for whom the verbal convention is made such as ‘tree’] would be able to
understand that these are trees [when he hears the word ‘tree’] in accordance with his understanding [of
the verbal convention], having referred to [the object] which equally appear in the cognitions of himself
and others, although being dependent [solely on his own conceptual cognition], by differentiating the
causes of this [cognition from others];” PVSVṬ 241,10–21 ad PVSV 60,27: nanu vyāvṛttasya svalakṣaṇasya vyavahārakāle ’nugamo nāsti. nāpi vikalpapratibhāsinaḥ sāmānyākārasya svajñānābhinnatvād
vikalpāntare ’nvayo ’sti. nāpi vaktṛsaṃbandhinas tasya śrotuḥ śrotṛsaṃbandhino vā vaktuḥ pratītiḥ,
anyacetodharmatvenātīndriyatvāt. na cāpratipanne samaṃ pratipādyapratipādakābhyāṃ saṃketaḥ
saṃbhavatīty ata āha — svaparetyādi. … etad uktaṃ bhavati – yathaikas taimiriko dvicandraṃ dṛṣṭvānyataimirikāyopadiśan svadṛṣṭam evopadiśati, na paradṛṣṭam, apratyakṣatvāt / atha ca tasyaivaṃ
bhavati – ayam eva mayā parasmai pratipādita iti / paro ’pi ca svasaṃtānabhāvinam eva dvicandrākāraṃ pratiyan ya eva pratipādakena mama pratipāditas sa eva mayā pratipanna iti manyate / tadvat
pratipādyapratipādakayor buddhyākārasyādhyavasita-bāhyarūpasya bhede ’py ekatvādhyavasāyāt
saṃketakaraṇam / vyavahārakāle ca tasyaiva pratītiḥ, ekatvādhyavasāyāt / “[Objection:] An excluded
particular does not continue to exist at the time of verbal activity. Also, since the form of the universal
manifesting in one’s conceptual cognition is not different from this one’s own cognition, it does not
occur in the other’s conceptual cognition. Neither one who listens to the speaker nor one who speaks to
the listener cognize [the form of the universal manifesting in the others’ conceptual cognition], because
it is not cognizable by sense organs. It is because it is the quality belonging to the other’s mind. When
[the form of the universal appearing in the others’ conceptual cognition] is not cognized, the speaker
(pratipādaka) and listener (pratipādya) cannot equally make verbal convention. [Reply:] [Dharmakīrti]
states ‘svapara’ and so on. … [By this,] the following is said: When some person affected by eye disease
sees a double moon and tells this to another person affected by eye disease, he describes only that which
is perceived by himself, not that which is perceived by anyone else. This is because [he] cannot perceive
[what anyone else perceives]. And he thinks ‘I explain it to him.’ Also, another person, understanding
the form of the double moon that belongs to his own mind, thinks ‘I understand what he has explained
to me.’ In the same way, although the forms of cognition, which are determined to be an external thing,
are different between someone explaining [i.e., speaker] and someone to be explained [i.e., listener],
conventions are made [on the forms of cognition] by the force of the determination of [these forms]
being the same. And, it [i.e., the form of cognition] is cognized at the time of verbal activity, because
[the form of cognition at the time of verbal activity is] determined to be the same [as that at the time of
verbal convention].” (There is no corresponding description in Śākyabuddhi’s Pramāṇavārttikaṭīkā.)
Kensho Okada
SS 65 48,5–10:
bdag dang gzhan la rang rang gi snang ba nyams su myong mod kyi / rab rib
can gnyis kyis zla ba gnyis mthong ba bzhin du rnam par shes pa de lta bu’i
rgyu’i bag chags skye ba’i rang bzhin gyi khyad par thog ma med pa’i dus las
don gcig ’dzin par nges pa’i phyir ro //
[A representation is called a manifest action (vijñapti).] It is because although
each person [i.e., a speaker and listener,]24 experiences his own representations,
just as two persons with eye-disease perceive a double moon [in the same
way], it is determined that the same thing is grasped by a special nature, which
has as its nature the arising of the latent disposition [produced by] the cause
of such cognition, from beginningless time.
SSṬ 49,17–50,18 ad SS 65:
’di ltar smra ba po yang ’di snyam du bdag gis smras pa gang yin pa ’di nyan
pa pos rtogs par ’gyur ro snyam du sems pa la / nyan pa po yang ’di snyam
du ’dis smras pa gang yin pa de ni bdag gis rtogs so snyam du sems par ’gyur
te / … … ’di skad du smra ba po dang nyan pa po dag rang rang gi snang
ba nyams su myong mod kyi / ’on kyang de gnyis kyi bya ba dang tshig gi
rnam pa can gyi shes pa rgyu gcig las ’byung ba gang dag yin pa de dag don
gcig tu shes par nges pa’i phyir rnam par rig byed du tha snyad ’dogs par
byed do // don gcig tu ’dzin par nges pa ni rnam par shes pa de lta bu’i rgyu’i
bag chags skye ba’i rang bzhin gyi khyad par thog ma med pa las byung ba’i
phyir ro // ’dir dpe ni rab rib can gnyis kyis zla ba gnyis su mthong ba bzhin
te / rab rib can gcig gis rab rib can gnyis la ltos zhes zla ba gzhan zhig ston
pa las des mthong ngo zhes smras pa na / de la ni ston pa po yang ’di snyam
du bdag gis ’di la bstan to snyam du nga rgyal byed par ’gyur la nyan pa po
yang ’di snyam du ’dis bdag la bstan to snyam du sems mod kyi / ’on kyang de
gnyis ni rang rang gi snang ba so sor nyams su myong bar zad do //
That is to say, the speaker thinks “the listener will understand what I say.” The
listener, in his turn, thinks “I have understood what he said.” … Although
the speaker and the listener both experience their respective representations,
cognitions having the form of action and speech appear in both from the same
cause, and these [cognitions] determinately grasp the same thing. Therefore,
it is called manifest action (vijñapti) metaphorically. Because there exists a
special nature, which has as its nature the arising of the latent disposition
[produced by] the cause of such cognition, from beginningless time, the same
thing is determinately grasped. “Just as in the case that two persons see a
double moon” is the example [demonstrating] this. If a person affected by eye
disease (A) indicates another moon and says to another person affected by eye
disease (B) “Look!” and he (B) says “I see [it],” then, the speaker thinks “I
have indicated it to him,” and the listener thinks “he indicated it to me.” Even
so, they both only cognize their respective representations.
24
Cf. SSṬ 48,13–14: bdag dang gzhan zhes bya ba ni smra ba po dang nyan pa po’o //
341
342
Communication between a Speaker and a Listener
References and abbreviations
AJP Anekāntajayapatākā (Haribhadra): Anekāntajayapatākā by Haribhadra Sūri: with
his own commentary and Municandra Sūri’s supercommentary, ed. H. R. Kāpadīā.
Baroda 1940.
AVK Anyāpohavicārakārikā (Śubhagupta): D4246, P5744.
McAllister 2011 P. McAllister, Ratnakīrti’s Apohasiddhi: A Critical Edition, Annotated
Translation and Study. Dissertation, Universität Wien. Wien 2011.
Mikogami 1993 E. Mikogami, Śubhagupta no Komyunikēshon Ron. In: Genshi Daijō
Bukkyō to Daijō Bukkyō, ed. E. Maeda. Kyōto 1993 (in Japanese).
PV I Pramāṇavārttika, chapter I (Dharmakīrti): See PVSV.
PVSV Pramāṇavārttikasvavṛtti (Dharmakīrti): The Pramāṇavārttikam of Dharmakīrti:
The First Chapter with The Autocommentary, ed. R. Gnoli. Roma 1960.
PVSVṬ Pramāṇavārttikavṛttiṭīkā (Karṇakagomin): Karṇakagomin’s Commentary on the
Pramāṇavārttikavṛtti of Dharmakīrti, ed. R. Sāṅkṛtyāyana. Repr. Kyoto 1982.
PVṬ Pramāṇavārttikaṭīkā (Śākyabuddhi): D4220, P5718.
SS Saṃtānāntarasiddhi (Dharmakīrti): Santānāntarasiddhi, ed. T. Stcherbatsky. Petrograd
1916.
SSṬ Saṃtānāntarasiddhiṭīkā (Vinītadeva): See SS.
TS Tattvasaṃgraha (Śāntarakṣita): See TSP (G) and TSP (B). The verse-numbers follow
TSP (B).
TS (D) Tattvasaṃgraha (Śāntarakṣita). Tibetan: D4266.
TS (P) Tattvasaṃgraha (Śāntarakṣita). Tibetan: P5764.
TSP Tattvasaṃgrahapañjikā (Kamalaśīla): See TSP (G) and TSP (B). Page- and linenumbers are given in accordance with TSP (B).
TSP (B) Tattvasaṃgrahapañjikā (Kamalaśīla): Tattvasaṅgraha of Ācārya Shāntarakṣita
with the Commentary “Pañjikā” of Shrī Kamalaśīla, ed. S. D. Shastri. Varanasi 1968.
TSP (G) Tattvasaṃgrahapañjikā (Kamalaśīla): Tattvasaṅgraha of Śāntarakṣita: With the
Commentary of Kamalaśīla, ed. E. Krishnamacharya. Baroda 1926.
TSP (D) Tattvasaṃgrahapañjikā (Kamalaśīla). Tibetan: D4267.
TSP (P) Tattvasaṃgrahapañjikā (Kamalaśīla). Tibetan: P5765.
On pramāṇabhūta — The Change of Its Meaning from
Dignāga to Prajñākaragupta
by
Motoi Ono
Introduction
In the maṅgalaśloka of the Pramāṇasamuccaya (=PS), Dignāga (ca. 480–540) describes
the Buddha-Bhagavat by five epithets. Among these epithets, the term pramāṇabhūta
is remarkable because this compound word includes the word pramāṇa (i.e., means of
cognition; valid cognition) in its former part. The term is a key concept to explain the
relationship between pramāṇa as knowledge and the Bhagavat as pramāṇa in the Buddhist
Pramāṇa-school.1
Tilmann Vetter showed that the term in Dignāga should be translated as “who is a
pramāṇa,” by indicating that the translation “who has become a pramāṇa” based on the
interpretation by Dharmakīrti (ca. 600–660) cannot be justified in Dignāga.2 However, even
if pramāṇabhūta in Dignāga means “who is a pramāṇa,” the following question remains
open: How can the Buddha as a person be called a pramāṇa, which is supposed to mean
valid cognition?3
In response to this question, D. Seyfort Ruegg, by investigating exhaustively the use of
the term pramāṇabhūta and the compound words x-bhūta in Sanskrit literature, proposed
to translate pramāṇabhūta in Dignāga as “who is like (i.e., similar to) a pramāṇa.”4 His
1
2
3
4
This paper is a revised English version of the Japanese one presented at the 63th conference of the
Japanese association of Indian and Buddhist Studies, Tsurumi University, June 30–July 1, 2012. The
proceedings of this conference have been already published (cf. Ono 2013). I would like here to express
my gratitude to Prof. Hideyo Ogawa for many kind comments and suggestions on the previous Japanese
paper. I am also grateful to Dr. Patrick McAllister, who kindly checked my draft and corrected my
English.
Cf. pramāṇabhūtāya jagaddhitaiṣine praṇamya śāstre sugatāya tāyine / pramāṇasiddhyai svamatāt
samuccayaḥ kariṣyate viprasṛtād ihaikataḥ // (PS maṅgalaśloka) Masaaki Hattori’s translation in
Hattori 1968: 23 is as follows: “Saluting Him, who is the personification of the means of cognition,
who seeks the benefit of [all] living beings, who is the teacher, sugata, the protector, I shall, for the
purpose of establishing the means of valid cognition, compose the [Pramāṇa-]samuccaya, uniting here
under one head my theories scattered [in many treatises].”
Cf. Vetter 1984: 14; note 6. Vetter has previously translated the word in the same way (cf. Vetter 1964:
32,17). Ernst Steinkellner agreed with Vetter’s view as follows: “Tilmann Vetter gives the reason for his
translation of the term pramāṇabhūta from the maṅgala of Dignāga’s Pramāṇasamuccaya as ‘who is a
means of valid cognition’ (‘der Erkenntnismittel ist.’): the interpretation that the Buddha has become a
means of valid cognition (as if the text read pramāṇībhūta) which does not figure at all in Dignāga’s
own explanation of the verse and can, therefore, hardly be based on the term pramāṇabhūta, was added
by Dharmakīrti, … Since the compound pramāṇabhūta does really not have a cvi-formation, … Vetter
is certainly right in asking for an interpretation of bhūta as the equivalent of the copula serving the
simple adjectivization of the first member” (cf. Steinkellner 1989: 180).
Cf. Vetter 1984:14,16ff.
Cf. Ruegg 1994a; 1994b; 1995.
Birgit Kellner et al., eds., Reverberations of Dharmakīrti’s Philosophy: Proceedings of the Fifth International Dharmakīrti
Conference Heidelberg, August 26 to 30, 2014. Vienna 2020, pp. 343–361.
344
On pramāṇabhūta — The Change of Its Meaning from Dignāga to Prajñākaragupta
view was thereafter criticized by Eli Franco and Helmut Krasser.5 In my opinion, however,
Ruegg’s interpretation is worth reconsidering. This paper aims to provide new materials
for supporting Ruegg’s view and, furthermore, to show that his view is meaningful from
the historical viewpoint of the Buddhist Pramāṇa-school by elucidating the change of the
meaning of the word pramāṇabhūta from Dignāga, through Dharmakīrti, to Prajñākaragupta
(ca. 750–810).
1. The meaning of pramāṇabhūta in Dignāga
1.1. Ruegg’s view
Ruegg pointed out that Patañjali (ca. 150 BCE) called “the teacher (ācārya)” pramāṇabhūta in his Mahābhāṣya (=MBh), and considered this the apparently earliest use of the
word pramāṇabhūta in Sanskrit literature.6 Kaiyaṭa (11th c.), a later commentator on the
MBh, analyzed this compound word as prāmāṇyaṃ prāptaḥ.7 According to Nārāyaṇa,
a subcommentator on the MBh, this explanation serves to avert both interpretations of
pramāṇabhūta as “pramāṇa-like” and as “has become a pramāṇa.”8 Thus, the established
theory by commentators on the MBh seems to be that the word pramāṇabhūta means
“being authority.”9
On the other hand, grammarians explain in some cases that bhūta in the compound
x-bhūta has a meaning of “like (i.e., similar).” According to Ruegg, Yāska (4th c. BCE)
5
6
7
8
9
Cf. Franco 1997: 16, note 3; Krasser 2001: 173–180. Around the time of Krasser’s paper, Noriaki
Hakamaya published two papers dealing with the same issue (cf. Hakamaya 2000; 2001). Jonathan A.
Silk also discussed this issue in terms of Chinese materials (cf. Silk 2002). Based on these articles,
Steinkellner mentioned the issue again (cf. Steinkellner 2003: 337, note 15).
Cf. MBh I 39,10f. ad Vārttika 1,1,1,7: pramāṇabhūta ācāryo darbhapavitrapāṇiḥ śucāv avakāśe
prāṅmukha upaviśya mahatā prayatnena sūtrāṇi praṇayati sma (cf. Ruegg 1994a: 309; 1995: 820f.).
Cf. MBhPr 160r,9ff.: prāmāṇyaṃ prāpta ity arthaḥ. bhū prāptāv ity asyādhṛṣād veti ṇijabhāvapakṣe
rūpam. vṛttiviṣaye ca pramāṇaśabdaḥ prāmāṇye vartate (cf. Ruegg 1994a: note 30; 1995: 820).
Nāgeśa (17th c.), a subcommentator to the MBh, tries to justify Kaiyaṭa’s interpretation of the word
pramāṇabhūta as meaning prāmāṇyaṃ prāpta as follows: “If (the teacher) were supposed to have
become what he was not previously (abhūtatadbhāva) since the verb bhū means ‘become,’ there should
be the form pramāṇībhūta with cvi-suffix. On the other hand, …” (cf. MBhPrU 126,7ff.: nanu bhavater
janmārthatvenābhūtatadbhāvapratītyā cvau sati pramāṇībhūta iti syāt, tadavivakṣāyāṃ tu pramāṇam
ācāryaḥ prakārāntareṇa bhūta ity arthaḥ syād ata āha – prāmāṇyam iti; Ruegg 1994a: 310,7ff.; note
31). Nārāyaṇa also tries to justify Kaiyaṭa’s interpretation as follows: “If the word bhūta means similarity
(upamānārthatve) just as in the case of the word pitṛbhūta etc., it would follow that the trustworthiness
of the teacher is not stated as primary. (On the other hand) if (the word pramāṇabhūta) means that the
teacher has become what he was not previously (abhūtaprādurbhāva), it would follow that there should
be the cvi-suffix just like (in the expression) aṅkurībhūta etc. …” (cf. MBhPrN 232,26ff.: pitṛbhūta
ityādivad bhūtaśabdasyopamānārthatve mukhyaṃ prāmāṇyam ācāryasya noktaṃ syāt, aṅkurībhūta
ityādivad abhūtaprādurbhāvavācitve tadvad eva cviprasaṅgaḥ, …, ato vyācaṣṭe – prāmāṇyam iti;
Ruegg 1994a: 310,23ff.).
Cf. Ruegg 1995: 820,27f. Ruegg also pointed out the uses of the word pramāṇabhūta in the Mahāyānasūtrālaṃkārabhāṣya (cf. Ruegg 1994a: 306f.; 1995: 821f.). Also in this case, the word pramāṇabhūta
can be understood as “being authority” by interpreting bhūta as the equivalent of the copula. Further,
Ruegg pointed out the use of the word in the Lalitavistara. Regarding this use, however, Hakamaya and
Silk indicated that the word pramāṇabhūta extant in the Sanskrit text may have been added in a later
period, based on a comparison to the Chinese translation (cf. Hakamaya 2000: (14); Silk 2002: 113f.).
Motoi Ono
345
gives such an explanation in his Nirukta, and Bhartṛhari (5th c.) explains in his commentary
on the MBh that bhūta in the expression sāmānyabhūta has the meaning of “like.”10 Thus,
Ruegg paid attention to the fact that the use of the expression x-bhūta in the meaning “like”
was already established before Dignāga.11 He further asserted that “the use of °bhūta in
the meaning ‘like’ … is found even in some of the post-Dharmakīrti commentaries.”12
By interpreting bhūta in pramāṇabhūta as meaning “like,” Ruegg suggested that the
whole compound pramāṇabhūta in Dignāga may be translated as “who is like a pramāṇa.”
He summed up the issue as follows: “If understood in this way, there will be no conflict
with Dignāga’s own doctrine that only pratyakṣa and anumāna are pramāṇas.”13
1.2. Krasser’s criticism
Krasser criticized this view of Ruegg. At first, he classified scholars’ interpretation of
the word pramāṇabhūta in Dignāga into three types: “1) pramāṇa should be understood
either 1A) in an extended sense, as authority, or 1B) metaphorically, or 2) bhūta should be
understood in the sense of a comparison.”14 Among these, the interpretation 2) is attributed
to Ruegg, and Krasser’s own position seems to be identical with 1B).
By examining new Sanskrit material of Jinendrabuddhi’s (8th c.) Pramāṇasamuccayaṭīkā (=PSṬ), Krasser indicated that the PSṬ, in contrast to Ruegg’s assertion, does not
explain that bhūta in pramāṇabhūta means “like,” but rather that the former part, i.e.,
pramāṇa, implies the meaning of “like.”15 Namely, Jinendrabuddhi’s explanation bhagavān pramāṇam iva pramāṇam (the Bhagavat is valid cognition [inasmuch as he is] like a
valid cognition) shows the interpretation that the mere expression bhagavān pramāṇam
implies the meaning bhagavān pramāṇam iva.16 Krasser concluded: According to Jinendrabuddhi’s explanation, “the word pramāṇa is applied to the Buddha not in an extended
sense but metaphorically and it does not require any qualification in order to be understood
as metaphor …, and it is also clear that his explanation does not support the interpretation
of °bhūta as ‘like’.”17
10
11
12
13
14
15
16
17
Cf. Ruegg 1994a: 311; note 28; 1995: 820f.
Cf. Ruegg 1995: 821,5f.
Cf. Ruegg 1994a: 311f.; 1995: 820f. This argument was thereafter criticized by Krasser. See below.
Cf. Ruegg 1994a: 315; 1995: 822,9ff.
Cf. Krasser 2001: 173,17ff. Here, the subdivisions of the type 1) in Krasser’s paper are called 1A) and
1B) respectively, for convenience of explanation.
Cf. Krasser 2001: 175f.; PSṬ 1,12ff.: yathā – tvaṃ merus tvaṃ samudro ’si nātha tvaṃ kalpapādapaḥ /
tvaṃ suvaidyaḥ pradīpas tvaṃ tvam eva paramaḥ plavaḥ // ity atrevaśabdaprayogam antareṇāpi
tadartho gamyate, tathehāpīti bhagavān pramāṇam iva pramāṇam. […] bhūto jātaḥ, utpanna ity
arthaḥ. bhūtavacanam abhūtasya nityasyeśvarādeḥ pramāṇasya paraparikalpitasya pratiṣedhārtham.
pramāṇaṃ cāsau bhūtaś ceti pramāṇabhūtaḥ, tasmai pramāṇabhūtāya (cf. Kataoka 2007: 14f.).
Cf. Krasser 2001: 179; PVṬ(P) 86b6; (D)71b7: tshad ma’i sgra dpe nang du ’dus pa can yin no zhes
bya ba’i don to //
Cf. Krasser 2001: 176,25–177,5. Further, Krasser pointed out that the Tibetan expression tshad ma
dang ’dra bas in the PVP by Devendrabuddhi (ca. 630–690) is the Tibetan eqivalent to the Sanskrit
pramāṇam iva, and that Jinendrabuddhi’s explanation bhagavān pramāṇam iva pramāṇam is derived
from Devendrabuddhi (cf. Krasser 2001: 177,6ff.; PVP(P)2a1f.; (D)1b2f. ad PV II 1a: tshad mar gyur
pa zhes bya ba ni tshad mar ’khrungs pa’o // tshad ma dang ’dra bas na tshad ma ste bcom ldan ’das
so //).
346
On pramāṇabhūta — The Change of Its Meaning from Dignāga to Prajñākaragupta
Thus, Krasser elucidated that Ruegg’s assertion can be substantiated neither in the
PSṬ nor the Pramāṇavārttikapañjikā (=PVP). Nevertheless, on the basis of this fact, one
cannot necessarily conclude that the expression pramāṇabhūta in Dignāga does not mean
“pramāṇa-like.” Krasser does not seem to intend to deny entirely Ruegg’s interpretation of
the word pramāṇabhūta in Dignāga as meaning “pramāṇa-like.”18
1.3. New materials supporting the interpretation of pramāṇabhūta as “pramāṇa-like”
In my opinion, it is not so relevant which part of the compound word pramāṇabhūta has
the meaning of “like.” What is important for interpreting Dignāga’s thought is, rather,
the difference between the view 1A) of Krasser’s classification that the word should be
translated as “being authority” by understanding the word pramāṇa as having a extended
sense and bhūta as the equivalent of the copula, and the view 1B) or 2) that the compound
pramāṇabhūta as a whole means “pramāṇa-like,” independently of which part has the
meaning of “like.” Both materials which Ruegg and Krasser showed seem to support the
latter view.
1.3.1. The interpretation mentioned by Jayanta and Yamāri
The interpretation of the compound pramāṇabhūta as “pramāṇa-like” is indeed attested
in the literature of the later Buddhist Pramāṇa-school.19 Namely, there are at least two
such passages in the commentaries by Jayanta (10th c.) and Yamāri (ca. 1000–1060) on
Prajñākaragupta’s Pramāṇavārttikālaṃkāra (=PVA).
First, Yamāri, in his remark on PVA’s first maṅgalaśloka, presents two different interpretations of pramāṇabhūta. The following interpretation is the second one of them:
Yamāri [Phe] (D)190b2f.; (P)228a5f.: yang na tshad ma dang ’dra bas (D:
tshad ma dang ’dra bas lacking in P) tshad mar gyur pa ste / ci ltar mngon
sum la sogs pa ma rtogs pa’i don bsal bar byed pa nyid kyis tshad ma yin pa
de bzhin du bcom ldan ’das kyang ’phyags pa’i bden pa bzhi la sogs pa ma
rtogs pa’i don *gsal bar byed pa’i (D: *gsal ba’i P) phyir tshad ma yin no //;
Ms 11b2 (cf. Li, Chu and Franco 2017: 82)20
18
19
20
Krasser mentioned that a remark of Vibhūticandra’s suggests that there are those who interpret bhūta
as “like.” For, Vibhūticandra, after having explained the interpretation of the word bhūta according
to Dharmakīrti by paraphrasing Jinendrabuddhi’s statement, states the following: “(Dharmakīrti) will
indicate that the (word bhūta) does not have the meaning (‘like’) because the meaning ‘like’ is understood
by implication” (cf. Vibhūti 519,3: ivārthas tu sāmarthyagata iti na tadartham etad iti vakṣyate; Krasser
2001: note 16).
Whether such interpretation is grammatical or not is another question (cf. Franco 1997: 16f., note 3:
“On the other hand, even if -bhūta did not mean ‘similar,’ once authoritative treatises state that it does,
it acquires this meaning for the readers who are acquainted with these treatises”).
The Sanskrit manuscript of the first chapter of Yamāri’s commentary has been discovered recently, and
a series of diplomatic editions of its beginning portion (folios 3a1-20a5) has been published by Xuezhu
Li, Junjie Chu and Eli Franco (cf. Li and Chu 2016; Li,Chu and Franco 2017; Li, Chu and Franco 2018)
after this paper was written. Since all three portions of Yamāri’s commentary that this paper discusses
are included in the mentioned diplomatic editions, I indicate the location in the respective edition after
quoting the text of the Tibetan translation. Although my assumptions about the Sanskrit corresponding
Motoi Ono
347
[= Or, (the Bhagavat is) pramāṇabhūta, inasmuch as he is like pramāṇa (*atha
vā pramāṇam iva pramāṇabhūtaḥ).21 Just as direct perception is a pramāṇa
because it reveals an unknown object, also the Bhagavat is a pramāṇa because
he reveals an unknown object like the four noble truths.]
This is really the interpretation that the compound pramāṇabhūta as a whole means “pramāṇa-like” without specifying the part of the compound which means “like.”
The next case, that of Jayanta, is more interesting. Jayanta comments on Prajñākaragupta’s view that the Bhagavat as pramāṇabhūta is necessary, although every purpose
of human beings can be established on the basis of conventional means of cognition
(sāṃvyavahārikapramāṇa),22 as follows:
Jayanta[De] (D)39b7f.; (P)46a5ff.: der yang mngon sum la sogs pa nyid tshad
ma’i gtso bo yin te / rtogs pa’i *bdag nyid can (D: *gtso bo dag P) yin pa’i
phyir ro // bcom ldan ’das ni tshad ma dang ’dra bas tshad mar gyur pa yin
no zhes bya ba’i bsam pas dogs pa la / *khyab par byed pa (D: *khyab par
byed par byed pa P) rtogs pa’i bdag nyid yin pa’i phyir / bcom ldan ’das kho
na tshad ma’i gtso bo yin la / gzhan ni de dang mthun pa nyid kyi gtso bo ma
yin no zhes bstan to //
[= Supposing the view: “In that case too, the actual direct perception etc. are
primary pramāṇas, because they are of the nature of cognition. The Bhagavat is,
on the contrary, pramāṇabhūta, inasmuch as he is like a pramāṇa (*bhagavān
pramāṇam iva pramāṇabhūtaḥ),” (Prajñākaragupta) indicated that the very
Bhagavat is a primary (*mukhya) pramāṇa, because (he) is of the nature of
pervading cognition; other (pramāṇas such as cognition) are (however) similar
to the (Bhagavat) and not primary.]
According to Jayanta, Prajñākaragupta indicates here, by stating that “the pervading concomitance cannot be grasped by those who are not omniscient (na ca saṃbandho vyāpy
asarvavidā grahītuṃ śakyaḥ),” that the Bhagavat is the primary (*mukhya) pramāṇa, and
this statement is aimed against the fallacious view that pramāṇa as knowledge is primary
and the word pramāṇabhūta means “pramāṇa-like.”
Thus, Jayanta also presupposes the possibility of the interpretation that the compound
word pramāṇabhūta means “pramāṇa-like.” It should be noted that this interpretation is
21
22
to the Tibetan translation, or even my interpretations based on the latter, sometimes seem to deviate
from the diplomatic edition of Sanskrit manuscript, I have not changed my expositions according to the
diplomatic edition; this is because, in my opinion, it is an open question whether the present diplomatic
edition does not need to be emended. A more detailed investigation remains as a task for the future.
In this case too, tshad ma dang ’dra bas is probably the rendering of pramāṇa iva (cf. note 17 in this
paper).
For direct perception does not function in terms of transcendent objects, and inference ultimately does
not function without the cognition of the pervading concomitance (sambando vyāpī ) by the omniscient
being (cf. PVAO 12,12ff.: nanu pramāṇabhūtena bhagavatā ko ’rthaḥ, sāṃvyavahārikapramāṇād
eva sarvapuruṣārthasiddheḥ. naitad asti. […] na tāvat pratyakṣaṃ paralokādau pravartate, tasya
svarūpamātragrahaṇād iti pratipādayiṣyate. anumānaṃ tu saṃbandhagrahaṇam antareṇa nāsti, na ca
saṃbandho vyāpy asarvavidā grahītuṃ śakyaḥ).
348
On pramāṇabhūta — The Change of Its Meaning from Dignāga to Prajñākaragupta
equated by him with the view that the Bhagavat is not a primary pramāṇa. This reminds us
of Nārāyaṇa’s above-mentioned interpretation.23
1.3.2. The expression x-bhūta in the introduction of the Abhidharmakośabhāṣya
Ruegg considered that Bhartṛhari’s use of x-bhūta to express likeness could have been familiar to Dignāga, and regarded it as supporting evidence for his interpretation of pramāṇabhūta
in Dignāga.24 Also concerning Vasubandhu’s use of x-bhūta in his Abhidharmakośabhāṣya
(=AKBh), we may point out the same possibility.
The beginning of the AKBh could have been familiar to Dignāga, because the PS’s
maṅgalaśloka and its explanation in the Pramāṇasamuccayavṛtti (=PSV) seem to be
influenced by the AKBh. Previous studies have already indicated that three of the Bhagavat’s
four epithets besides pramāṇabhūta in the PS’s maṅgalaśloka correspond to the three
characteristics describing the trustworthiness of an authoritative person (āpta) in the
Nyāyabhāṣya (=NBh).25 However, we must recognize that the PS’s description of the
Bhagavat is probably influenced not directly by the NBh, but via the AKBh or, rather,
directly by the AKBh. For, the maṅgalaśloka of the AKBh is as follows:
yaḥ sarvathāsarvahatāndhakāraḥ saṃsārapaṅkāj jagad ujjahāra /
tasmai namaskṛtya yathārthaśāstre śāstraṃ pravakṣyāmy abhidharmakośam //
[= Saluting Him who has averted the darkness regarding everything by every manner and means, who emancipates living beings from a morass of
reincarnation, who teaches reality as it is, I will teach the treatise named
Abhidharmakośa.]26
Further, it is to be noted that the compound word saṃsārapaṅka in this maṅgalaśloka is
explained by Vasubandhu himself as follows:
AKBh 1,17: saṃsāro hi jagadāsaṅgasthānatvāt duruttaratvāc ca paṅkabhūtaḥ.
[= Namely, reincarnation is like a morass, because it is the place to which
living beings cling, and from which (living beings) hardly escape.]27
23
24
25
26
27
Cf. note 8 in this paper.
Cf. Ruegg 1994a: 309,20ff.; 1995: 821. Ruegg seems to consider that the attestation of the word x-bhūta
meaning “x-like” in Bhartṛhari is relevant because his writing had great influence on Dignāga.
Cf. Vetter 1984: note 6; Franco 1997: 28ff. Namely, Pakṣilasvāmin (5th c.) enumerated three characteristics establishing the trustworthiness of the āpta, i.e., 1) having direct knowledge of reality
(sākṣātkṛtadharmatā), 2) having compassion towards living beings (bhūtadayā), and 3) having the
desire to teach reality as it is (yathābhūtārthacikhyāpayiṣā) (cf. NBh 565,6ff.). These correspond
respectively to the sugata, the tāyin or the jagaddhitaiṣin, and the śāstṛ in the PS.
sarvathāsarvahatāndhakāraḥ corresponds to sugata, saṃsārapaṅkāj jagad ujjahāra to tāyin or jagaddhitaiṣin, and yathārthaśāstre to śāstṛ. The Bhāṣya explains the phrases sarvathāsarvahatāndhakāraḥ,
saṃsārapaṅkāj jagad ujjahāra. and yathārthaśāstā by the words ātmahitapratipattisaṃpad, parahitapratipattisaṃpad and parahitapratipattyupāya respectively (cf. AKBh 1.8ff.).
Cf. AKBhVy 5,30f.: saṃsāraḥ paṅka iva saṃsārapaṅkaḥ paṅkasādharmyāt. ata āha – saṃsāro
hi jagadāsaṅgasthānatvād duruttaratvāc ca paṅkabhūta iti. [= (Vasubandhu said:) “a morass of
reincarnation,” because the reincarnation is like a morass, since it is similar to a morass. Therefore, he
said: “Reincarnation is like a morass, because it is the place, to which living beings cling to, and
from which (living beings) hardly escape.”]
Motoi Ono
349
Dignāga who depended on the AKBh in writing the maṅgalaśloka and the introduction of
the PSV could have understood the expression paṅkabhūta as meaning “paṅka-like.”
1.4. The meaning of pramāṇabhūta and Dignāga’s purpose of writing the PS
From the above, it is likely that the word pramāṇabhūta in Dignāga means “who is like
pramāṇa (as knowledge),” as Ruegg suggested. This interpretation not only serves to
avoid the difficulty in considering bhūta as the equivalent of the copula. The crucial point
is, rather, that it corresponds well to Dignāga’s thought about the relationship between
pramāṇa as knowledge and the Bhagavat.
As is well known, Dignāga, in contrast to Dharmakīrti, did not talk much about the
Bhagavat and Buddhist ontology at least in his logical works.28 The explanations in the
PS, except for the maṅgalaśloka, are almost entirely restricted to establishing pramāṇa as
knowledge. In this sense, pramāṇa as knowledge is primary (mukhya), and the Bhagavat
as pramāṇa is secondary in the PS. In the closing section of the PSV, Dignāga says the
following:
Thus,29 this (treatise) has been undertaken only for turning around (vyāvartana) those people who cling to the (opponent’s views), not for introducing
(them) to the Tathāgata’s teaching, because his teaching is not the object of
reasoning. Those who have turned away (from clinging to the fallacious views),
however, can easily comprehend the essence of the teaching which is perfectly
manifested, after having listened to it, because (our logic and the logic of the
opponents) are at different degrees of distance (from the truth).30 (Namely,)
28
29
30
Cf. Vetter 1984: 14.
Cf. Krasser 2004: 134f. I give here my tentative translation according to the following reconstruction of
this part which is a result of the joint research workshop on the PSṬ chapter VI held by Dr. Toshikazu
Watanabe (Kokugakuin University), Dr. Yasutaka Muroya (Austrian Academy of Sciences) and myself
under the support of Prof. Franco, Prof. Chizuko Yoshimizu and Dr. Krasser: iti tadāsaktānāṃ vyāvartanārtho ’yam ārambhaḥ, na tv iyatā tathāgataśāsanānupraveśārthaḥ, taddharmasyātarkagocaratvāt. vyāvṛttās tu paridīpitāṃ dharmatām upaśrutyāyatnenaiva pratipatsyante, viprakṛṣṭāntaratvāt. tarkamātrabalās
tīrthyāḥ so ’py anirdiṣṭalakṣaṇaḥ / svaprayogaviruddhaś ca na ceṣṭārthaprasādhakaḥ // 23 // sudūranaṣṭās
tu munīndraśāsanān nayanti ye tarkapathena dharmatām / tathāpi tāthāgatadharmalakṣaṇaṃ parīkṣyatāṃ yady upayāti vikriyām // 24 // (Italics are used for words retranslated from the Tibetan translations,
whereas normal script is used for the words taken from the PSṬ manuscript.)
Krasser interpreted the word viprakṛṣṭāntara as Dvandva, i.e., “remote” and “near” (cf. Krasser 2004:
133f.). However, his interpretation is not supported by PSV’s Tibetan translations, and the PSṬ also
does not necessarily seem to support it. Regarding the word viprakṛṣṭāntara, Kumārila’s use is worth
considering (cf. the Ślokavārttika, śabdanityatādhikaraṇa, v.168–169ab: tadantarādhikātmā ca bhāti
tenāgrataḥ sthitaḥ / viprakṛṣṭāntarāṇāṃ ca stokadeśe ’pi dṛśyate // siddhabhinnāgradeśānāṃ dūradeśasamānatā / [Ganganath Jha’s translation in Jha 1983: 437: “Even in the case of an object which is
comparatively much nearer to us (than the Sun really is), we find that persons, – residing at places that
are at different degrees of distance from that object, and consequently having their fronts decidedly
different from one another, – mistake that distant object to be at equal distances from themselves”];
I would like to thank Dr. Ryō Sasaki for his informing me of this passage). In my opinion, the word
viprakṛṣṭāntara can be meaningfully understood only by relating it to the following two verses. Jinendrabuddhi’s interpretation of this word seems to have somewhat deviated from Dignāga’s original
intention.
350
On pramāṇabhūta — The Change of Its Meaning from Dignāga to Prajñākaragupta
The opponents depend only on (their) logic. But this very (logic) is one in
which (correct) characteristics are not indicated, is one that contradicts its
own syllogism, and cannot establish the intended object. (k.23) On the other
hand, those who lead (the people) to the essence of the teaching by way of
logic, have (also) deviated very far from the teaching by the king of saints.
Nevertheless, the essence of the Tathāgata’s teaching is to be considered (by
logic) if (it) undergoes a change. (k.24)
Needless to say, it does not mean that Dignāga did not give importance to the Bhagavat. I
agree with Krasser in that Dignāga as a pramāṇavādin is definitely Buddhistic.31
2. Dharmakīrti’s interpretation of pramāṇabhūta
Dharmakīrti paraphrases the word pramāṇabhūta in his PV II as follows:
PV II 7 abc’: tadvat pramāṇaṃ bhagavān abhūtavinivṛttaye / bhūtoktiḥ
[= The Bhagavat is a pramāṇa like that (pramāṇa as knowledge) (tadvat).
The expression bhūta serves to avert those who have not become (abhūta) (a
pramāṇa, like Īśvara and the Veda).]
On the basis of this statement, Śākyabuddhi etc. interpret the word pramāṇabhūta as a
compound meaning “he is a pramāṇa, and he has become.”32 However, it is not necessarily
clear whether such an interpretation matches Dharmakīrti’s own intention. At least it is
obvious that Dharmakīrti did not reinterpret the word pramāṇabhūta in order to use it in his
own philosophy. For, unlike in the case of Prajñākaragupta, the word was never really used
in Dharmakīrti’s treatises.33 Dharmakīrti uses the word pramāṇa instead of pramāṇabhūta
to designate “the Bhagavat as pramāṇa.” He seems to intend to only deconstruct this
compound.
Now, how did he understand the original meaning of pramāṇabhūta in Dignāga, so
that he did not use this word? In order to answer this question, we need to reconsider
Devendrabuddhi’s expression pramāṇam iva pramāṇam (tshad ma dang ’dra bas na tshad
ma ste) which is referred to by Krasser as the evidence for refuting Ruegg’s interpretation
of °bhūta as “like.” This expression first appears in Devendrabuddhi’s introductory remark
on PV II 1, where he explains Dignāga’s usage of the word pramāṇabhūta:
31
32
33
Cf. Krasser 2004.
Cf. PVP(D)1b2; (P)2a1: tshad mar gyur pa zhes bya ba ni tshad mar ’khrungs pa’o; PVṬ(D)71b5f.;
(P)86b4f.: tshad mar ’khrungs pa zhes bya bar (D: ba P) gyur pa’i sgra ni ’khrungs pa’i don to // tshad
ma yang de yin la gyur pa yang de yin pas na tshad mar gyur pa’o (cf. Vibhūti 521,27: pramāṇajāta
[em. in Krasser 2001: 178: pramāṇaṃ jāta] iti bhūtaśabdaḥ prādurbhāvārthaḥ; PSṬ 2,7: pramāṇaṃ
cāsau bhūtaś ceti pramāṇabhūtaḥ; note 15 in this paper). The Tibetan translation tshad mar gyur pa
is also derived from such interpretation (cf. Hakamaya 2000: (10)f.).
As a matter of fact, the translation “who has become a pramāṇa” can be applied only to the word
pramāṇabhūta found in the treatises of Dharmakīrti’s followers. Furthermore, whether such a translation
is appropriate depends on the context. For example, in Dharmottara’s case, where the word appears in
his Prāmāṇyaparīkṣā’s maṅgalaśloka, it may be, contrary to Krasser’s view (cf. Krasser 2001: note
44), appropriately translated as “pramāṇa-like” in Dignāga’s sense, because pramāṇa as knowledge is
primary in this treatise, just as in the case of the PS. Regarding Prajñākaragupta, see below.
Motoi Ono
351
PVP (D)1b2f.; (P)2a1f.: tshad mar gyur pa zhes bya ba ni tshad mar ’khrungs
pa’o // tshad ma dang ’dra bas na tshad ma ste bcom ldan ’das so //
[= The compound pramāṇabhūta (means) “who has become a pramāṇa.”
The Bhagavat is a pramāṇa (inasmuch as he is) similar to a valid cognition
(*bhagavān pramāṇam iva pramāṇam).]34
The statement pramāṇam iva pramāṇam, which means that the expression bhagavān pramāṇam implies the meaning iva, is, according to Śākyabuddhi, the reply to the question
“The pramāṇas are direct perception and inference. Thus, how can the Bhagavat who is not
of their nature be said to be pramāṇa?”35 Note that this statement appears immediately after
the explanation of the compound pramāṇabhūta according to PV II 7bc’. This fact seems
to show that such a statement became necessary owing to the Dharmakīrtian interpretation of bhūta. Namely, it is possible that Devendrabuddhi interpreted bhūta as meaning
“has become” and, as a consequence, had to transfer the meaning which the compound
pramāṇabhūta as a whole originally had in Dignāga to the part pramāṇa. If this is the case,
it means that pramāṇabhūta in Dignāga was understood as meaning “pramāṇa-like” by
Devendrabuddhi.
I think that his understanding derives from Dharmakīrti. Devendrabuddhi’s remark on
PV II 7a shows it:
PVP (D)6b5; (P)7b2f.: de bzhin bcom ldan tshad ma nyid (PV II 7a: tadvat
pramāṇaṃ bhagavān) // ’di tshad ma dang yang ci zhig mtshungs na / gang
gis na tshad ma dang ’dra bas tshad ma nyid yin tshad ma’i mtshan nyid rnam
pa gnyis nyid dang mtshungs pa yin no //
[= The Bhagavat is a pramāṇa like that (pramāṇa as knowledge). In what
way is he like a pramāṇa? He is like (a pramāṇa) in having the twofold defining
characteristic of pramāṇa so that, he is (said to be) pramāṇa, inasmuch as he
is like a pramāṇa.].36
According to Devendrabuddhi, Dharmakīrti shows by the expression tadvat that the Bhagavat’s likeness to pramāṇa as knowledge means that the Bhagavat has the same twofold defining characteristic that pramāṇa as knowledge has (but not that the Bhagavat is secondary to
pramāṇa as knowledge). Further, Dharmakīrti calls the Bhagavat, not pramāṇabhūta, but a
pramāṇa meaning pramāṇam iva, because both the Bhagavat and pramāṇa as knowledge
are said to be pramāṇa in having the twofold defining characteristic.
34
35
36
Cf. Krasser 2001: 177f. I follow Krasser’s translation except for some small modifications.
Cf. PVṬ(P)86b5f.; (D)71b6f.: mngon sum dang rjes su dpag pa ni tshad ma yin pa de bas na / ji ltar
na bcom ldan ’das de’i rang bzhin can ma yin pa la de skad du ce na tshad ma dang ’dra bas na tshad
ma ste zhes bya ba smos te; Inami 1994: note 2; Krasser 2001: 178f. I follow Krasser’s translation.
Cf. Krasser 2001: 182f. I follow Krasser’s translation in terms of Devendrabuddhi’s remark with some
modifications. However, I cannot agree with his view that Devendrabuddhi understood –vat in tadvat in
a possessive sense and the Tibetan translation de bzhin to tadvat was a misunderstanding (cf. Krasser
2001: 183). In my opinion, tadvat must carry the meaning “like that (pramāṇa as knowledge)” in
this context, because tadvat is nothing but the paraphrase of the word pramāṇabhūta. Nevertheless,
it is possible that tadvat was understood as an intentionally ambiguous expression (a kind of śleṣa in
Sanskrit rhetoric) by Devendrabuddhi.
352
On pramāṇabhūta — The Change of Its Meaning from Dignāga to Prajñākaragupta
I think Devendrabuddhi is right. What Dharmakīrti is doing here is interpreting the
word pramāṇabhūta; and the Bhagavat’s likeness to pramāṇa as knowledge is nothing but
what is meant by the word pramāṇabhūta in Dignāga.37 Thus, the word pramāṇabhūta in
Dignāga is probably understood as meaning “pramāṇa-like” by Dharmakīrti himself.
3. Prajñākaragupta’s interpretation of pramāṇabhūta
3.1. pramāṇabhūta as a purpose of the PVA
Prajñākaragupta calls the Bhagavat pramāṇabhūta in the first maṅgalaśloka of his PVA,
following Dignāga, and adopts this term as a key concept of his philosophy.38 However, the
meaning of this term in Prajñākaragupta is different from that in Dignāga. The opening
paragraph of the PVA, immediately after the maṅgalaślokas, clearly shows this difference.
In this paragraph, Prajñākaragupta follows the opening section of the PSV with some
modifications.39 Among these modifications, the following two are relevant for our issue:40
(1) PSV I 1,3f.: atra bhagavato hetuphalasaṃpattyā pramāṇabhūtatvena stotrābhidhānaṃ prakaraṇādau gauravotpādanārtham.41
37
38
39
40
41
I would like to thank Prof. Inami for the following suggestion expressed by E-mail on the 27. Sept. 2013:
“It is possible that Dharmakīrti also understands in the meantime bhūta in pramāṇabhūta as meaning
‘like’ by stating tadvat in PV II 7a. From PV II 7a onwards, Dharmakīrti apparently begins to explain
the word pramāṇabhūta, even if commentators would not support it.” (English translation is mine.)
The first maṅgalaśloka, whose first half is identical to that of the PS, states the purpose of writing the
PVA (cf. pramāṇabhūtāya jagaddhitaiṣiṇe praṇamya śāstre sugatāya tāyine / kutarkasaṃbhrāntajanānukampayā pramāṇasiddhir vidhivad vidhīyate //).
Cf. PSV I 1,3–13: atra bhagavato hetuphalasaṃpattyā pramāṇabhūtatvena stotrābhidhānaṃ prakaraṇādau gauravotpādanārtham. tatra hetur āśayaprayogasaṃpat. āśayo jagaddhitaiṣitā. prayogo
jagacchāsanāc chāstṛtvam. phalaṃ svaparārthasaṃpat. svārthasaṃpat sugatatvena trividham artham
upādāya, praśastatvārthaṃ surūpavat, apunarāvṛttyarthaṃ sunaṣṭajvaravat, niḥśeṣārthaṃ supūrṇaghaṭavat. arthatrayaṃ caitad bāhyavītarāgaśaikṣāśaikṣebhyaḥ svārthasaṃpadviśeṣaṇārtham. parārthasaṃpat tāraṇārthena tāyitvam. evaṃguṇaṃ śāstāraṃ praṇamya pramāṇasiddhyai svaprakaraṇebhyo
nyāyamukhādibhya iha samāhṛtya pramāṇasamuccayaḥ kariṣyate parapramāṇapratiṣedhāya svapramāṇaguṇodbhāvanāya ca, yasmāt pramāṇāyattā prameyapratipattir bahavaś cātra vipratipannāḥ;
PVAO 1,10–2,7: atra bhagavato hetuphalasaṃpattyā pramāṇabhūtatvena stotrābhidhānaṃ śāstrādau,
śāstrārthatvāt. bhagavān eva hi pramāṇabhūto ’smin prasādhyate. tatra hetur āśayaprayogasaṃpat sāṃvyavahārikapramāṇāpekṣayā. āśayo jagaddhitaiṣitā. prayogo jagacchāsanāc chāstṛtvam.
phalaṃ svaparārthasaṃpat. svārthasaṃpat sugatatvena trividham artham upādāya, praśastatvaṃ
surūpavat, apunarāvṛttyarthaṃ sunaṣṭajvaravat, niḥśeṣārthaṃ supūrṇaghaṭavat. parārthasaṃpaj jagattāraṇāt tāyitvam, saṃtānārthaṃ cāparinirvāṇadharmatvāt. evaṃbhūtaṃ bhagavantaṃ praṇamya pramāṇasiddhir vidhīyate. pramāṇādhīno hi prameyādhigamaḥ, bhagavān eva ca pramāṇam,
pramāṇalakṣaṇasadbhāvāt. pramīyate ’neneti pramāṇam. (Emphasized parts show modifications
by Prajñākaragupta.)
Regarding the importance of the addition sāṃvyavahārikapramāṇāpekṣayā after the sentence tatra
hetur āśayaprayogasaṃpat for Prajñākaragupta’s thought, see Iwata 2001: (48)ff.
Hattori’s translation: “At the beginning of the treatise, here [in this verse], I express praise in honor of
the Worshipful [Buddha] in order to produce in [the hearts of] men faith in Him who, because of His
perfection in cause (hetu) and effect (phala), is to be regarded as the personification of the means of
cognition (pramāṇa-bhūta).”
Motoi Ono
353
PVAO 1,10–12: atra bhagavato hetuphalasaṃpattyā pramāṇabhūtatvena
stotrābhidhānaṃ śāstrādau, śāstrārthatvāt. bhagavān eva hi pramāṇabhūto ’smin prasādhyate.
(2) PSV I 1,10–13: evaṃguṇaṃ śāstāraṃ praṇamya pramāṇasiddhyai svaprakaraṇebhyo nyāyamukhādibhya iha samāhṛtya pramāṇasamuccayaḥ kariṣyate
parapramāṇapratiṣedhāya svapramāṇaguṇodbhāvanāya ca, yasmāt pramāṇāyattā prameyapratipattir bahavaś cātra vipratipannāḥ.42
PVAO 2,4–7: evaṃbhūtaṃ bhagavantaṃ praṇamya pramāṇasiddhir vidhīyate.
pramāṇādhīno hi prameyādhigamaḥ, bhagavān eva ca pramāṇam, pramāṇalakṣaṇasadbhāvāt. pramīyate ’neneti pramāṇam.
In passage (1), Prajñākaragupta replaces the PSV’s sentence “in order to produce in men
faith (gauravotpādanārtham)” with the sentence “because (the Bhagavat is) the purpose
of the treatise. Namely, the very Bhagavat as pramāṇabhūta is to be established here”
(śāstrārthatvāt. bhagavān eva hi pramāṇabhūto ’smin prasādhyate). Prajñākaragupta shows
by this replacement that the Bhagavat is praised right at the beginning of the treatise in order
to make it clear that the purpose of the treatise is to establish the Bhagavat as pramāṇabhūta.
In passage (2), Dignāga shows that the purpose of the PS is to reject opponents’ theories
of pramāṇa as knowledge and to establish correctly the Buddhist theories of it. Prajñākaragupta, on the other hand, while showing that the purpose of the PVA is to establish
the pramāṇa, emphasizes that the pramāṇa which he will establish in the PVA is nothing
but the Bhagavat.43 Thus, in the PVA, not pramāṇa as knowledge, but the Bhagavat as
pramāṇa, i.e., pramāṇabhūta, is regarded as pramāṇa in the primary sense, and is the main
topic of the treatise.
3.2. The identification of the Bhagavat as pramāṇabhūta with pāramārthikapramāṇa
Dharmakīrti uses the word sāṃvyavahārikapramāṇa, meaning direct perception and inference, in the closing paragraph of the Pramāṇaviniścaya’s first chapter. However, the word
pāramārthikapramāṇa in the same paragraph does not necessarily mean the Bhagavat.44
42
43
44
Hattori’s translation: “Saluting the teacher who is endowed with such merits, the author will compose
the Pramāṇasamuccaya or the Collected Writings on the Means of Cognition by gathering [passages]
from the Nyāyamukha and other of his treatises in order to establish the means of valid cognition. The
purpose [of the work] is to reject the theories concerning the means of cognition maintained by others
and to elucidate the virtues in his own theories concerning the means of cognition, since there are
divergent opinions with regard to [the nature, number, object, and result of] the means of cognition, on
which depends the clear understanding of the object to be cognized.”
The phrase pramāṇalakṣaṇasadbhāvāt is related to Prajñākaragupta’s theory of the definition of pramāṇa
that in the ultimate sense only the Bhagavat has the twofold defining characteristic of pramāṇa (cf. Ono
2014).
Cf. PVin I 44,2ff.: sāṃvyavahārikasya caitat pramāṇasya rūpam uktam, atrāpi pare mūḍhā visaṃvādayanti lokam iti. cintāmayīm eva tu prajñām anuśīlayanto vibhramavivekanirmalam anapāyi pāramārthikapramāṇam abhimukhīkurvanti. tad api leśataḥ sūcitam eveti.
354
On pramāṇabhūta — The Change of Its Meaning from Dignāga to Prajñākaragupta
Prajñākaragupta, however, clearly identifies the Bhagavat with pāramārthikaṃ pramāṇam.45
He calls the Bhagavat paraṃ pramāṇam too.46
The word pramāṇabhūta qualifying the Bhagavat in the PVA can be regarded as a
synonym of these terms. The term pāramārthikapramāṇa is used by Prajñākaragupta
in the sense that the Bhagavat is the ‘ultimate’ pramāṇa insofar as the trustworthiness
of his direct perception is ascertained by itself, whereas ordinary direct perception and
inference are conventional pramāṇas.47 According to Prajñākaragupta, the Bhagavat as
pramāṇabhūta is by nature a direct perception (pratyakṣarūpa).48 It is clear from this point
that Prajñākaragupta interprets pramāṇa in pramāṇabhūta literally.49
3.3. Blo ldan shes rab’s rendering of pramāṇabhūta as tshad ma yang dag (pa)
As is well known, sKal ldan rgyal po and Blo ldan shes rab (1059–1109) translate the
word pramāṇabhūta qualifying the Bhagavat in the PVA as tshad ma yang dag (pa), i.e.,
“true pramāṇa,”50 while rendering the same word qualifying cognitions etc. in the same
text as tshad mar (’)gyur pa.51 Although the translation of bhūta as yang dag (pa) is not
impossible in itself,52 it can be regarded as reflecting Prajñākaragupta’s usage of the word.
However, Prajñākaragupta himself does not give any analysis of the compound by
which such a translation can be justified.53 His new interpretation that the sentence abhūtavinivṛttaye bhūtoktiḥ (PV II 7bc’) means “the Bhagavat tells the truth in order to
45
46
47
48
49
50
51
52
53
Cf. PVA 67,12f.; Ms26a8: bhagavān eva ca paramārthataḥ kāryakāraṇabhāve pāramārthikaṃ pramāṇaṃ vyāpyanvayavyatirekagrahaṇād iti pratipāditam. (cf. PVAO 12,12ff.; note 22 in this paper).
Cf. PVAO 83,15ff.: bhagavatas tarhi kathaṃ prāmāṇyam. pratyakṣānumānayor hi vyavahāramātreṇa
prāmāṇyam, na bhagavataḥ. tad dhi paraṃ pramāṇam. atrocyate – tadvat pramāṇaṃ bhagavān (PV
II 7a).
Insofar as the trustworthiness of ordinary direct perception can be ascertained only in a conventional
sense and inference postulates the omniscience of the Bhagavat in order to be universally right (cf. note
22 in this paper; Ono 1994; 2012; 2014).
Cf. PVAO 84,1f.: tathāgato hi bhagavān tadvān iti kṛtvā pratyakṣarūpa eva bhagavān pramāṇam.
It corresponds to the fact that Prajñākaragupta interprets the word tadvat in PV II 7 in the possessive
meaning (cf. Krasser 2001: 181, and note 36 above).
Cf. Hakamaya 2000: (11). The rendering tshad ma yang dag (pa) is attested many times also in the
Tibetan translation of Yamāri’s commentary on the PVA by Sumati and Blo ldan shes rab. Many
examples of this appear in Yamāri’s remark on the maṅgalaśloka and the introductory paragraph of
the PVA. In the Tibetan translation of Jayanta’s commentary on the PVA by Śrī Dīpaṃkararakṣita and
Byang chub shes rab, the rendering tshad ma yang dag (pa) is attested only once, whereas the rendering
tshad mar gyur pa is attested many times.
The word pramāṇabhūta qualifying cognitions or treatises etc. is also attested in the PVA (cf. PVA
385,1f.: pramāṇabhūtapratyakṣam; PVA 494,9f.: śāstrād eva pramāṇabhūtāt; PVA 568,6ff.: pramāṇabhūtāḥ pratyayāḥ; PVA 619,26ff.: pramāṇabhūtapratyayapratipādyaḥ). In those cases, bhūta can be
understood as the equivalent of the copula without problems and the Tibetan equivalent of it is almost
tshad mar (’)gyur pa. In Yamāri’s commentary as well, the rendering tshad mar gyur pa is attested
many times. Most of them qualify Veda, Niyoga, words, cognitions etc. But some of them qualify the
Bhagavat. Among them we can find some interesting examples, as we shall see later.
Cf. Ruegg 1994a: note 44; Hakamaya 2000: 324ff.; Steinkellner 2003: note 15. Interestingly, Jonathan
A. Silk found the Chinese equivalent 眞實稱量 for pramāṇabhūta in the Chinese translation of the
Śikṣāsamuccaya 大乗集菩薩学論 (cf. Silk 2002: note 26); there bhūta is rendered as 眞實, i.e., true
or real, just like in the case of tshad ma yang dag (pa).
As far as I am aware, Yamāri refers to the word bhūta in pramāṇabhūta as follows: “Further, supposing
the question ‘why is only the Bhagavat pramāṇa (in the true sense) and others not?,’ (Prajñākaragupta)
Motoi Ono
355
avert error”54 appears to justify the translation tshad ma yang dag (pa) because the word
bhūta means the contrary concept of “error” (bhrānti), i.e., “truth” or “true” in this context.
What Prajñākaragupta really intends here is, however, not to interpret the word bhūta in
pramāṇabhūta, but to justify his own usage of the word pramāṇabhūta by interpreting
Dharmakīrti’s word bhūtokti as not referring to bhūta in pramāṇabhūta.
3.4. Yamāri’s interpretation of the compound pramāṇabhūta
Then, how should we interpret the compound pramāṇabhūta in Prajñākaragupta? Yamāri’s
commentary provides us with an important clue to Prajñākaragupta’s understanding. Among
Yamāri’s two different interpretations, the second interpretation explaining pramāṇabhūta
as “pramāṇa-like” has been mentioned above. The first one, which can be regarded as
Yamāri’s own interpretation, is as follows:
Yamāri [Phe] (D)188b5f.; (P)226a3f.: ’di rtsom pa yang ci zhig byas nas yin
zhe na / bshad pa phyag ’chal nas te rab tu btud cing lus phul nas so (D: song
P) // su la zhe na / tshad ma yang dag gam tshad mar gyur pa la’o55 // tshad
mar gyur pa ni tshad mar red pa ’am tshad mar thob pa’o //; Ms 10b1–2 (cf.
Li, Chu and Franco 2017: 80)
[= (Question:) Further, what has been done before beginning to write this
(treatise)? (Prajñākaragupta) replies: “saluting,” i.e., bowing his head and
throwing his body on the ground. (Question:) To whom? (Prajñākaragupta
replies:) To pramāṇabhūta. (The compound) pramāṇabhūta means “who has
reached the means of cognition (*pramāṇaṃ gataḥ),” in other words, “who
has attained trustworthiness (*prāmāṇyaṃ prāptaḥ).”]
By comparing this with Śākyabuddhi’s explanation of the compound parabhāvabhūta found
in the Pramāṇavārttikasvavṛtti, one can assume that the Sanskrit equivalent of the sentence
tshad mar gyur pa ni tshad mar red pa56 is probably *pramāṇaṃ gataḥ pramāṇabhūtaḥ,
and it is also clear that this sentence shows the interpretation of bhūta as the equivalent
of the copula.57 Further, tshad mar thob pa is most likely the rendering of prāmāṇyaṃ
54
55
56
57
explains the reason implied by the word bhūta (in pramāṇabhūta by the statement:) ‘by the perfection
of cause and effect (hetuphalasaṃpattyā)’” (cf. Y(D)197b3; (P)235b8f.: yang ci ltar bcom ldan ’das
*nyid (D: nyid kyi P) tshad ma yin gyi / gzhan ma yin snyam du dogs na yang dag pa’i sgras bsdus pa’i
gtan tshigs smras pa rgyu dang ’bras bu phun sum tshogs pas zhes bya ba’o //). This explanation, that
the word bhūta in pramāṇabhūta implies the Bhagavat’s perfection of cause and effect, seems to justify
the rendering of bhūta as yang dag pa.
Cf. PVAO 84,4–7: abhūtanivṛttaye bhrāntinivṛttyartham yatas tasya bhagavato bhūtoktis tataḥ sa eva
sarvajñaḥ, nāparaḥ (cf. Ruegg 1994a: 306,2f.; Franco 1997: 56, note 25). Regarding the significance
of this interpretation in Prajñākaragupta’s thought, cf. Iwata 2000; 2001: (50)ff.
I assume that gam tshad mar gyur pa is a kind of gloss. The rendering should be merely tshad ma yang
dag, insofar as the word pramāṇabhūta qualifies the Bhagavat in this case, but the general translation
tshad mar gyur pa is added here to the semantic translation, probably because the interpretation of the
compound itself is talked about (cf. Ms 10b2; Li, Chu and Franco 2017: 80).
The Tibetan tshad mar red pa literally means “being pramāṇa.”
Commenting on the compound parabhāvabhūta in the PVSV, which cannot but be understood as a
pleonastic expression, Śākyabuddhi explains it as gzhan gyi ngo bor red pa, the Sanskrit equivalent
356
On pramāṇabhūta — The Change of Its Meaning from Dignāga to Prajñākaragupta
prāptaḥ,58 which is the same as Kaiyaṭa’s explanation of Patañjali’s pramāṇabhūta and
therefore means “being pramāṇa” as well. Thus, Yamāri shows here the interpretation of
pramāṇabhūta as “being pramāṇa.”59
Further, in commenting on Prajñākaragupta’s above-mentioned statement “because
(the Bhagavat is) the purpose of the treatise” (śāstrārthatvāt), Yamāri says as follows:
Yamāri [Phe](D)198b5f.; (P)237a6ff.: gal te ’o na bcom ldan ’das mngon sum
gyi khyad par gyi rang bzhin yin mod / gang gi bstan bcos kyi don yin pas
bstan bcos blang byar ’gyur ba de tsam gyis (D: gyi P) bdag cag la nye bar
mkho ba ni ma yin no // tshad ma yang dag pas zhes bya ba la sogs pa smos
so // tshad mar gyur pa ni tshad mar red pa zhes bya ba’i don to // ma byung
ba las byung ba ni yin yang brjod par mi ’dod pa’i phyir rtsi ba’i rkyen med
do // dper na “au pa las yang man (D: u tpa las yan P) ni rang gi rgyu las
so” zhes bya ba ’dir / Pa tan dzi lis bshad pa “de la phan pa ’jug pa ni / sna
tshogs pa’i phyir ro // des na (D: na lacking in P) de la phan pa ma byas so”
zhes bya ba de bzhin du ’dir yang lta’o (D: blta P) //; Ms 16a4–5 (cf. Li, Chu
and Franco 2018: 44)
[= (Objection:) Then, even if the Bhagavat is by nature a special direct perception, it is so far not helpful for us in accepting the treatise because (the Bhagavat) is the purpose of the treatise. (Prajñākaragupta) replies: “(the very
Bhagavat) as pramāṇabhūta” etc. (The compound) pramāṇabhūta means
“who has reached the means of cognition (*pramāṇaṃ gataḥ).” Namely, the
suffix meaning “become” (*cvipratyaya) does not exist (in the expression
pramāṇabhūta) because, even though (the Bhagavat) has become what he was
not previously, it is not intended to mention (it). Also in this case, one should
consider (it) according to Patañjali who explains: “The operations of taddhita
are manifold and hence the taddhita-suffix is not found (vicitrās taddhitavṛttayo nātas taddhita utpadyate)” regarding (Pāṇini 1.3.56:) “After the verb
yam- preceded by upa-, when used in the sense of ‘espousing,’ (ātmanepada
is employed) (upād yamaḥ svakaraṇe).”]
Although this paragraph is not easy to understand, it is at least clear that Yamāri presents
here again the interpretation of pramāṇabhūta as “being pramāṇa” and mentions that
the expression pramāṇabhūta does not mean “has become what he was not previously”
(*abhūtatadbhāva)60 since the cvi-suffix does not exist in it. At the end of this paragraph,
58
59
60
of which is parabhāvaṃ gataḥ, as found in Karṇakagomin’s commentary (cf. PVṬ(D)37b3; (P)44b5:
gzhan gyi (D: gyis P) ngo bor gyur pa ni gzhan gyi ngo bor red pa ste; PVSVṬ 72,16f.: parabhāvaṃ
gataḥ parabhāvabhūtaḥ).
This assumption can be ascertained by the Sanskrit manuscript (cf. Ms 10b2; Li, Chu and Franco 2017:
80). The explanation prāmāṇyaṃ prāptaḥ is referred to also by Vibhūticandra (cf. note 62 in this paper).
In Yamāri’s case, however, “being pramāṇa” does not mean “being authority,” as in the case of
grammarians, but “being valid cognition.” See above, section 1.1 in this paper.
The Sanskrit equivalent of ma byung ba las byung ba is normally abhūtvā bhāva (Mahāvyutpatti 2182).
But, abhūtatadbhāva or abhūtaprādurbhāva is also possible (cf. note 8 in this paper). The meaning is
not different in any case. Cf. Ms 16a5; Li, Chu and Franco 2018: 44.
Motoi Ono
357
Yamāri tries to reinforce his argument by referring to Patañjali’s explanation to Pāṇini-sūtra
1.3.56.61
Thus, we can find some parallels between Yamāri’s statements including his reference
to the interpretation prāmāṇyaṃ prāptaḥ and the explanations of pramāṇabhūta by MBhcommentators.62 It is noteworthy that the time of Yamāri’s activity (ca. 1000–1060)63 seems
to be close to that of Kaiyaṭa.64 The parallel between the grammarians and the Buddhist
Pramāṇa-school which Ruegg suggested may be a result of actual influence of one on the
other.65
In any case, it is clear from the above that Yamāri interprets the compound pramāṇabhūta as meaning “being pramāṇa,” while presupposing the existence of other interpretations like “pramāṇa-like” and “who has become a pramāṇa.” In my view, Yamāri’s
interpretation is not contradictory to either Prajñākaragupta’s identification of the word
with pāramārthikapramāṇa, or to Blo ldan shes rab’s translation tshad ma yang dag (pa).
Rather, among various interpretations of the word, only the interpretation “being pramāṇa”
can be in harmony with Prajñākaragupta’s usage of the word. For, the expression “who is
A” can carry the meaning “who is the true A” or “who is A in the true sense.”66
Taking the above into consideration, I would like to propose the translation “who is a
pramāṇa in the true sense” for the term pramāṇabhūta when qualifying the Bhagavat in
Prajñākaragupta.
4. Conclusion
The results of the above consideration are summarized in Table 1.
61
62
63
64
65
66
Commenting on Pāṇini 1.3.56 which prescribes that ātmanepada should be employed after the verb
upa-yam- in the sense of “espousing,” Patañjali states the following: “(Question:) How does (the active
voice) not appear here, like in the sentence ‘he holds (upayacchati) the end of his cloth’? (Answer:)
(the active voice) should appear where one makes one’s own what was not previously one’s own (like in
the case that one has illicit intercourse with another’s wife). (However) if so, it follows that the word
in the Sūtra should be svīkaraṇe (instead of svakaraṇe according to Pāṇini 5.4.50: [abhūtatadbhāve]
kṛbhvastiyoge saṃpadyakartari cviḥ). The operations of taddhita are manifold (i.e., the cvi-suffix does
not appear if one does not wish to say that one makes one’s own what was not previously one’s own) and
hence the taddhita-suffix is not found (cf. MBh I,284,10–12: iha kasmān na bhavati – svaṃ śāṭakāntam
upayacchatīti. asvaṃ yadā svaṃ karoti, tadā bhavitavyam. yady evaṃ svīkaraṇa iti prāpnoti. vicitrās
taddhitavṛttayaḥ, nātas taddhita utpadyate).” Concerning the identification of Yamāri’s citation, I would
like to thank Prof. Vincent Eltschinger, Prof. Hiroshi Nemoto, Dr. Junjie Chu and Dr. Yasutaka Muroya
for their valuable suggestions.
Cf. notes 7 and 8 in this paper. In addition, we can find the following notable sentences in Vibhūticandra’s
remark (cf. Vibhūti 521,29ff.: pramāṇaśabdo jñāne mukhya itaratra tu kena sādharmyenopamānopameyatvam ity āha – avisaṃvādīti. prāptivācī tu bhavatiḥ spaṣṭārtha ity asau na vivṛtaḥ prāmāṇyaṃ
prāptaḥ pramāṇabhūta iti). Their meaning is unfortunately not entirely clear for me.
Cf. Ono 2000: xxiii–xxiv.
Cf. Cardona 1976: 347, note 344.
Cf. Ruegg 1995: 822,17ff.
Cf. Steinkellner 2003: note 15.
358
What is the Bhagavat
called?
Meaning of pramāṇabhūta
Function of bhūta
Commentator’s
explanation of the compound
pramāṇabhūta
Relationship between pramāṇa and the Bhagavat
† pramāṇa
Dignāga
Dharmakīrti
Prajñākaragupta
pramāṇabhūta
pramāṇa (=pramāṇam iva)
pramāṇabhūta
“pramāṇa-like”
“x-bhūta” means “x-like”
*pramāṇam iva pramāṇabhūtaḥ.
(Jayanta, Yamāri)
[“who has become a pramāṇa”]
abhūtavinivṛttaye bhūtoktiḥ
[pramāṇaṃ cāsau bhūtaś ceti
pramāṇabhūtaḥ.] (Śākyabuddhi,
Jinendrabuddhi)
pramāṇa = Bhagavat (in having
the twofold defining characteristic of pramāṇa)
“pramāṇa in the true sense”
copula implying “in true sense”
*pramāṇaṃ gataḥ pramāṇabhūtaḥ; *prāmāṇyaṃ prāptaḥ
pramāṇabhūtaḥ. (Yamāri)
pramāṇa < Bhagavat (Bhagavat
is mukhya)
pramāṇa > Bhagavat (pramāṇa is
mukhya)
means “valid cognition” in all cases
On pramāṇabhūta — The Change of Its Meaning from Dignāga to Prajñākaragupta
Table 1: The interpretations of pramāṇabhūta
Motoi Ono
359
References and abbreviations
Primary literature
AKBh Abhidharmakośabhāṣya (Vasubandhu): Abhidharmakoṣabhāṣyam of Vasubandhu,
ed. Pralhad Pradhan. Patna 1975.
AKBhVy Abhidharmakośavyākhyā (Yaśomitra): Sphuṭārthā Abhidharmakośavyākhyā by
Yaśomitra. Part I, ed. U. Wogihara. Tokyo 1932.
Jayanta Pramāṇavārttikālaṃkāraṭīkā (Jayanta): P 5720, vol. 133, Tshad ma, De 1b1–434a8;
Ne 1b1–375a8; D 4222, vol. 7–8, Tshad ma, De 1b1–365a7; Ne 1b1–312a7.
Mahāvyutpatti Bonzōkannwa Shiyakutaikō Honyaku Meigi Taishū. Tokyō 1917.
MBh I Mahābhāṣya (Patañjali), ed. F. Kielhorn. Revised by K. V. Abhyankar. Vol. I. Poona
1985.
MBhPr Mahābhāṣyapradīpa (Kaiyaṭa): The Vyākaraṇamahābhāṣya of Patañjali with
the commentary Bhāṣyapradīpa of Kaiyaṭa Upādhyāya and the supercommentary
Bhāṣyapradīpoddyota of Nāgeśa Bhaṭṭa. Vol.1 navāhnika, ed. Bh. S. Joshi. Delhi
1987.
MBhPrN Nārāyaṇīya (Nārāyaṇa): Mahābhāṣyapradīpavyākhyānāni. Adhyāya 1 Pāda 1
Āhnika 1–4, ed. Narasimhacharya. Présentation par P.-S. Filliozat. Pondichéry 1973.
MBhPrU Mahābhāṣyapradīpoddyota (Nāgeśa): Mahābhāṣyapradīpoddyota by Nāgeśa
Bhaṭṭa. Vol. I, ed. B Śastri. Calcutta 1899.
NBh Nyāyabhāṣya (Pakṣilasvāmin): Nyāyadarśanam Bhāṣya-Vārttika-Tātparyaṭīkā-sahitam. Vol. 1, ed. Taranatha Nyayatarkatirtha and Amarendramohan Tarkatirtha. Calcutta 1936. Vol. 2, ed. Amarendramohan Tarkatirtha and Hemanta Kumar Tarkatirtha.
Calcutta 1944.
PS Pramāṇasamuccaya (Dignāga).
PSṬ Pramāṇasamuccayaṭīkā (Jinendrabuddhi): Jinendrabuddhi’s Pramāṇasamuccayaṭīkā,
Chapter 1 Part 1: Critical Edition. Part 2: Diplomatic Edition, ed. Ernst Steinkellner,
Helmut Krasser, and Horst Lasic. Beijing/Vienna 2005.
PSV Pramāṇasamuccayavṛtti (Dignāga).
PSV I Pramāṇasamuccayavṛtti I (Dignāga): Dignāga’s Pramāṇasamuccaya, Chapter 1: A
Hypothetical Reconstruction of the Sanskrit Text with the Help of the Two Tibetan
Translations on the Basis of the Hitherto Known Sanskrit Fragments and the Linguistic Materials Gained from Jinendrabuddhi’s Ṭīkā, ed. E. Steinkellner. Wien 2005.
https://www.oeaw.ac.at/fileadmin/Institute/IKGA/PDF/forschung/buddhismuskund
e/dignaga_PS_1.pdf, last visited 10-12-2019.
PV Pramāṇavārttika, II, III, IV (Dharmakīrti): Pramāṇavārttikakārikā (Sanskrit and Tibetan), ed. Y. Miyasaka. Acta Indologica 2 (1971/72) 1–206.
PVA Pramāṇavārttikālaṃkāra (Prajñākaragupta): Pramāṇavārtikabhāshyam or Vārtikālaṅkāraḥ of Prajñākaragupta (Being a commentary on Dharmakīrti’s Pramāṇavārtikam), ed. Rahula Sankṛtyayana. Patna 1953.
PVAO See Ono 2000.
PVin I Dharmakīrti’s Pramāṇaviniścaya. Chapters 1 and 2, ed. E. Steinkellner. Beijing/
Vienna 2007.
360
On pramāṇabhūta — The Change of Its Meaning from Dignāga to Prajñākaragupta
PVP Pramāṇavārttikapañjikā (Devendrabuddhi): D 4217, Vol. 2, Tshad ma, Che
1b1–326b4; P 5717, Vol. 130–131, Tshad ma, Che 1b1–390a8.
PVSV Pramāṇavārttikasvavṛtti (Dharmakīrti): The Pramāṇavārttikam of Dharmakīrti, the
first chapter with the autocommentary, ed. R. Gnoli. Roma 1960.
PVSVṬ Pramāṇavārttikasvavṛttiṭīkā (Karṇakagomin): Ācārya-Dharmakīrteḥ Pramāṇavārttikam (svārthānumānaparicchedaḥ) svopajñavṛttyā Karṇakagomi-viracitayā
taṭṭīkayā ca sahitam, ed. R Sankṛtyayana. Allahabad 1943.
PVṬ Pramāṇavārttikaṭīkā (Śākyabuddhi): D 4220, vol. 3, Tshad ma, Je 1b1–328a7; P
5718, vol. 131, Tshad ma, Je 1b1–402a8.
Vibhūti Vibhūticandra’s notes in Pramāṇavārttikavṛtti (Manorathanandin): Ācārya-Dharmakīrteḥ Pramāṇavārtikam ācārya-Manorathanandi-kṛtayā vṛtyā saṃvalitam, ed.
R. Sankṛtyayana. Patna 1938–1940.
Yamāri Pramāṇavārttikālaṃkāraṭīkā Supariśuddhā (Yamāri): P 5723, vol. 134–136, Tshad
ma, Phe 208a7–345a8; Be 1b1–290a7; Me 1b1–436a8; Tse 1b1–321a5; D 4226, vol.
10–13, Tshad ma, Phe 174b1–287a7; Be 1b1–261a7; Me 1b1–328a7; Tse 1b1–251a7.
Secondary literature
Cardona 1976 George R. Cardona, Pāṇini: A Survey of Research. The Hague 1976.
Franco 1997 Eli Franco, Dharmakīrti on Compassion and Rebirth. Wien 1997.
Hakamaya 2000 Noriaki Hakamaya, Pramāṇa-bhūta to kumāra-bhūta no gogi – bhūta no
yōhō o chūshin ni. Komazawa Tanki Daigaku Bukkyō Ronshū 6 (2000) (1)–(30).
Hakamaya 2001 Noriaki Hakamaya, Pramāṇa-bhūta hoki. Komazawa Tanki Daigaku
Kenkyū Kiyō 29 (2001) 433–448.
Hattori 1968 Masaaki Hattori, Dignāga on perception. Cambridge, Massachusetts 1968.
Inami 1994 Masahiro Inami, “Pramāṇavārttika” Pramāṇasiddhi shō no kenkyū (3). Shimanekenritsu Kokusai Tankidaigaku kiyō 1 (1994) 17–53.
Iwata 2000 Takashi Iwata, Seson wa ikanishite kōjun (pramāṇa) to nattanoka. Komazawa
Tanki Daigaku Bukkyō Ronshū 6 (2000) 1–38.
Iwata 2001 Takashi Iwata, Seson no ryōsei no shōmei no ichi kaishaku – Prajñākaragupta
no shiten kara –. Indo tetsugaku Bukkyōgaku 16 (2001) (44)–(74).
Jha 1983 Ganganath Jha, Ślokavārttika. Translated from the original Sanskrit with extracts
from the commentaries “Kāśikā” of Sucarita Miśra and “Nyāyaratnākara” of Pārtha
Sārathi Miśra. Delhi 2 1983.
Kataoka 2007 Kei Kataoka, Pramāṇasamuccayaṭīkā ad 1.1 wayaku. Minami Ajia Kotengaku 2 (2007) 1–79.
Krasser 2001 Helmut Krasser, On Dharmakīrti’s Understanding of pramāṇabhūta and
His Definition of pramāṇa. Wiener Zeitschrift für die Kunde Südasiens 45 (2001)
173–199.
Krasser 2004 Helmut Krasser, Are Buddhist Pramāṇavādins non-Buddhistic? Dignāga
and Dharmakīrti on the impact of logic and epistemology on emancipation. Hōrin
11 (2004) 129–149.
Li and Chu 2016 Xuezhu Li and Junjie Chu, A Diplomatic Edition of the Introductory
Section of Yamāri’s Pramāṇavārttikālaṅkāraṭīkā Supariśuddhā: Folios 3a1–10a1.
China Tibetology 26 (2016) 3–20.
Motoi Ono
361
Li, Chu and Franco 2017 Xuezhu Li, Junjie Chu, and Eli Franco, A Diplomatic Edition
of the Introductory Section of Yamāri’s Pramāṇavārttikālaṅkāraṭīkā Supariśuddhā:
Folios 10a1–14b2. China Tibetology 28 (2017) 78–87.
Li, Chu, and Franco 2018 Xuezhu Li, Junjie Chu, and Eli Franco, A Diplomatic Edition
of the Introductory Section of Yamāri’s Pramāṇavārttikālaṅkāraṭīkā Supariśuddhā:
Folios 14b2–20a5. China Tibetology 30 (2018) 40–52.
Ono 1994 Motoi Ono, Prajñākaragupta ni yoru Dharmakīrti no pramāṇa no teigi no
kaishaku – Prajñākaragupta no shinriron. Indogaku bukkyōgaku kenkyū 42/2 (1994)
878–885.
Ono 2000 Motoi Ono, Prajñākaraguptas Erklärung der Definition gültiger Erkenntnis
(Pramāṇavārttikālaṃkāra zu Pramāṇavārttika II 1–7). Teil I. Sanskrit-Text und
Materialien. Wien 2000.
Ono 2012 Motoi Ono, Shinriron – pramāṇa towa nanika. In: Ninshikiron to Ronrigaku,
ed. Shōryu Katsura et al. Tokyo 2012, 156–188.
Ono 2013 Motoi Ono, Pramāṇabhūta no imi no hensen. Indogaku bukkyōgaku kenkyū
61/2 (2013) 934–938.
Ono 2014 Motoi Ono, Pramāṇa no teigi – Prajñākaragupta no kaishaku o megutte. Indogaku bukkyōgaku kenkyū 62/2 (2014) 964–970.
Ruegg 1994a D. Seyfort Ruegg, Pramāṇabhūta, *Pramāṇa(bhūta)-puruṣa, Pratyakṣadharman and Sākṣātkṛtadharman as Epithets of the Ŗṣi, Ācārya and Tathāgata in
Grammatical, Epistemological and Madhyamaka Texts. Bulletin of the School of
Oriental and African Studies 57/2 (1994) 303–320.
Ruegg 1994b D. Seyfort Ruegg, La notion du voyant et du «connaisseur suprême» et la
question de l’autorité épistémique. Wiener Zeitschrift für die Kunde Südasiens 38
(1994) 403–419.
Ruegg 1995 D. Seyfort Ruegg, Validity and Authority or Cognitive Rightness and
Pragmatic Efficacy? On the Concepts of Pramāṇa, Pramāṇabhūta and Pramāṇa(bhūta)puruṣa. Asiatische Studien/Études Asiatiques 49 (1995) 817–827.
Silk 2002 Jonathan A. Silk, Possible Indian Sources for the Term tshad ma’i skyes bu as
pramāṇapuruṣa. Journal of Indian Philosophy 30 (2002) 111–160.
Steinkellner 1989 Ernst Steinkellner, Miszellen zur erkenntnistheoretisch-logischen
Schule des Buddhismus. Wiener Zeitschrift für die Kunde Südasiens 33 (1989)
177–181.
Steinkellner 2003 Ernst Steinkellner, Once more on Circles. Journal of Indian Philosophy
31 (2003) 323–341.
Vetter 1964 Tilman Vetter, Erkenntnisprobleme bei Dharmakīrti. Wien 1964.
Vetter 1984 Tilman Vetter, Der Buddha und seine Lehre in Dharmakīrtis Pramāṇavārttika.
Wien 1984.
Dharmakīrti on the Role of Salvific Initiation and the
Reception of His Critique in the Later Śaiva Tradition1
by
Cristina Pecchia
In memory of Helmut Krasser
Towards the end of the Pramāṇasiddhi chapter of the Pramāṇavārttika (= PV II) Dharmakīrti
examines various non-Buddhist theories about how to attain liberation. In this context, he
discusses the function of the ritual of dīkṣā, “initiation,” in attaining the elimination of the
soul’s sins or impurities and, thus, the liberation from the cycle of transmigrations.
An original point-by-point refutation of the Buddhist opinion represented in Dharmakīrti’s text is expounded – as Attilia Sironi has indicated2 – by Kṣemarāja (approximately
1000–1050 CE)3 in his commentary on Svacchandatantra V.88, where he provides a survey
of different views on initiation. In this way, the target of Dharmakīrti’s critique is confirmed
as corresponding to the followers of tantric4 ideas and practices who advocate a Śaiva type
of salvific initiation. Indeed, as observed by Alexis Sanderson and Raffaele Torella, PV
II.259ab5 refers to an initiate who is lighter than before the performance of the initiation,6
very likely alluding to the tulādīkṣā, a ritual characterized by the use of a balance (tulā).
More generally, dīkṣā is described in Śaiva sources as the action that removes all the bonds
(pāśas) or innate impurity (mala) which causes rebirth and, therefore, is the necessary
step to be made in order to attain liberation.7 The special type of ritual that includes a
balance is widely attested in Śaiva sources, from the early Niśvāsatattvasaṃhitā,8 from
1
2
3
4
5
6
7
8
Research for the present article, which was completed in 2015, was generously funded by the Austrian
Science Fund (FWF) in the framework of the stand-alone project “Indian Buddhist epistemology and
the path to liberation” (P 26120–G15), based at the Austrian Academy of Sciences, Vienna. I would
like to thank Diwakar Acharya and Patrick McAllister for their careful reading of this article and their
very helpful observations.
Attilia Sironi, assisted by Raniero Gnoli, has published an Italian translation of the Uddyota passage
here under consideration and the relevant part of the PV (Sironi 1988).
See Sanderson 1988: 700.
As it is not clear when the term tantra became a standard designation for texts of tantric revelation (see
Niśvāsa, pp. 30f.), it might be anachronistic to speak of ‘tantric’ ideas and practices in Dharmakīrti’s
time.
The numbering of the kārikās accords to that established in Vetter 1964: 116f.
Sanderson 2001: 10f., n. 7. Raffaele Torella’s identification of the ritual as a Śaiva type of initiation is
reported in Franco 2001, n. 24.
See, for example, Sanderson 1988: 662, 664ff. and 691; Sanderson 1992: 285; and Acharya 2014: 16f.,
with nn. 25 and 27.
Niśvāsa, Mūlasūtra 7:15ab: tulayā śodhayet pāpam ātmanasya parasya vā; see also Niśvāsa, p. 324,
Acharya 2014: 16, and Eltschinger 2014: 123, n. 102.
Birgit Kellner et al., eds., Reverberations of Dharmakīrti’s Philosophy: Proceedings of the Fifth International Dharmakīrti
Conference Heidelberg, August 26 to 30, 2014. Vienna 2020, pp. 363–374.
364
Dharmakīrti on the Role of Salvific Initiation
sometime between 450 and 550 CE,9 to later works of eminent authors such as Sadyojyotis
and Abhinavagupta.10
In what follows I will analyse the PV section on initiation and the related response
of Kṣemarāja. The skillful use of rhetorical means seems to shape both texts, in the
former adding a sarcastic nuance to the refutation of the opponents’ view, in the latter
covertly minimizing the long past Buddhist attack. This investigation will contribute to
our understanding of Dharmakīrti’s engagement in discussing contemporary soteriological
ideologies and, on the other hand, to our understanding of how his thought reverberated in
a later and different intellectual environment such as the Śaiva one.11
In this study I take Manorathanandin’s Vṛtti (PVV) as the guiding commentary. The
indicators typical of commentarial phraseology (e.g., nanu, cet, and syād etat) show that, in
his opinion, PV II.257–267 forms a section in which Dharmakīrti addresses the followers
of the Scriptures of the Lord (īśvarāgama) with regard to salvific initiation (hereafter
“section on initiation”).12 The two subsequent kārikas, kk. 268–269, present a discussion
of the existence of the self that can be regarded as continuing the refutation of the previous
soteriological view; however, it does not display any explicit indications with regard to it.13
The section on initiation consists of six short thematic units:14
a) K. 257, where, against those who claim the salvific effects of an initiation, Dharmakīrti argues that they explain such an effect by resorting to the authority of the
Scriptures only – which is not satisfactory.
b) Kk. 258–259, where, showing the undesired consequences of their claim, he argues
that a ritual such as initiation is not sufficient to stop rebirth and that bad deeds are
not something that can be embodied.
c) Kk. 260–261, where Dharmakīrti presents the Buddhist view of how to attain liberation, especially focusing on the role of karman in the perpetuation of rebirths.
d) Kk. 262–264, which concern the nature of karmic impulses (saṃskāras). These
are related to the unseen force (adṛṣṭa) of merits and demerits for the upholders of
salvific initiation, but to cetanā for the Buddhists.
e) Kk. 265–266, which explain the role of the mind and the nature of mental faults with
regard to the cognitive faculties’ agency and rebirth.
f) K. 267, where Dharmakīrti explains that the nature of the mind cannot be permanent.
After stating the unacceptability of resorting to Scriptures to argue for salvific initiation,
Dharmakīrti points out that, when urged to prove the efficacy of initiation, the followers of
9
10
11
12
13
14
Goodall & Isaacson 2007: 6. Some further remarks concerning the Niśvāsatattvasaṃhitā are provided
in Sanderson 2001, n. 2, item 7; nn. 5–6; pp. 22f. and 29–31, with notes therein.
See TAK III, s.v. tulādīkṣā.
See Torella 1992 for some reflexions on the relationships between Pratyabhijñā thought and the Buddhist
logical-epistemological system.
PVV 98.17: nanūktam īśvareṇāgame ’sty ātmā mokṣaś cāsya dīkṣāvidhineti, and 100.15f.:… syād etad
ātmano ’pi garbhagatakaraṇādijanane vyāpāraḥ sa eva dīkṣayā niruddha iti na punarjanmety.
The Tibetan tradition represented in dGe ’dun grub pa’s topical outlines (sa bcad) takes the kk. 264–269
as one thematic unit within the section that begins at k. 257 (dbang phyug pa’i grol lam dgag pa –
“Refuting the path to deliverance of the devotees of īśvara;” see Inami & Tillemans 1986: 133f. and
140, items 166–171). Note that the word dbang denotes special rituals such as initiation.
For the text and translation, see the Appendix.
Cristina Pecchia
365
the Scriptures of the Lord adduce arguments which lead to prasaṅgas. For they compare
the efficacy of initiation with the efficacy of rituals applied to physical objects and consider
the initiate’s weighing less after the performance of the ritual a proof of the initiate’s
elimination of sins and attainment of liberation.
Dharmakīrti explicitly identifies the former alleged proof as implying the undesired
consequence that one would then have to believe in the salvific power of other rituals,
too, such as oil massage or scorching oneself with fire. A human being would be like a
seed: if treated with oil or burnt by fire, it does not generate a sprout (k. 258cd). Further,
Dharmakīrti covertly suggests that the opponents’ proof implies a prasaṅga; for, saying
that even the initiates’ loss of their entire weight would not be a proof of their loss of sins (k.
259c), he alludes to the eventual implication that liberation corresponds to having no weight
at all. To this he adds the inference “sin is not heavy because it is not embodied” (k. 259d),
which refutes the opponents’ implicit inference that is based on the reason “because sin is
embodied” – a reason that may be indicative of another fact such as the loss of sins after
initiation only within a physicalist view of sins. Although Dharmakīrti does not expand on
this, his audience can go back to the refutation of the materialist view on rebirth explained
in a previous part of the Pramāṇasiddhi chapter,15 where, showing that the mental does
not depend on the corporeal, he argued for the non-physicality of faults that prevent living
beings from attaining liberation.
The opening of Dharmakīrti’s critique of salvific initiation seems to be designed
to persuade the audience, before any demonstration, of the evident implausibility of a
soteriological method based on a ritual ceremony. In fact, the examples of an oil massage
and a ritual with a balance easily remind one of magical treatments and freak shows.
Mentioning them, Dharmakīrti seems to be adopting a rhetorical strategy that reinforces
the prasaṅgas with which his refutation begins: he intentionally exposes the simplicity
of the opponents’ soteriological proposal and the unsophisticatedness of their proof, also
evoking the unreliability and negligibility of the latter.
His refutation continues with a more general assertion, at kk. 260–261, concerning
the causes of rebirth from the Buddhist point of view. The content of the text is quite
similar to the idea expounded at PV II.81 and 189, where Dharmakīrti states that rebirth is
rooted in misjudgement and thirst, and it is thirst rather than karma that ultimately effects
the setting in motion of the continuum of the five skandhas.16 Arguing against salvific
initiation, he not only reasserts the primacy of cognitive and emotional experience, but
also emphasizes the subordinate role of karma in the Buddhist discourse on liberation.
Moreover, he elaborates on the nature of ignorance and thirst by saying that they are cetanā
(“intention,” or “volition”) and are connected with karma (te cetane svayaṃ karmety –
k. 261c). His formulation refers to a notion that appears in canonical and Abhidharmic
sources17 – famously in AN III.41518 and in AK 4.1ab, where it is said that the manifold
15
16
17
18
See, in particular, PV II.34–53. For a study and English translation of the text, see Franco 1997, chapter
4 and pp. 159–258, and Taber 2003.
See Pecchia 2015: 26–28. Franco 1997: 71ff. provides some considerations on karma in relation with
rebirth in the Pramāṇasiddhi chapter.
See Heim 2013 for a comprehensive study of cetanā as presented in relevant Pāli sources.
“It is volition (cetanā), monks, that I call karma. Having willed (cetayitvā), one performs an action
(kammaṃ karoti) by body, by speech, by mind” (Harvey 2011: 182).
366
Dharmakīrti on the Role of Salvific Initiation
world is caused by karma, which is characterized by intention (karmajaṃ lokavaicitryaṃ
cetanā tatkṛtaṃ ca tat).
The subsequent text features a more committed Dharmakīrti, who explains why salvific
initiation is considered problematic and takes into serious account his opponents’ argument.
In a Buddhist soteriological perspective, he states, it is the series of seeds continuously
generating mental faults that has to be stopped, but initiation, as a ritual, does not have any
impact on mental faults.
Upholders of salvific initiation can argue against this stance focussing on the nature and
role of karma. Their main argument is concerned with the nature of saṃskāras, “karmic
impulses,” which in their view are linked to adṛṣṭa (“the unseen”), a latent force from which
physical and mental actions ultimately derive and which is eliminated through initiation.
From a Buddhist point of view, however, the saṃskāras are connected to cetanā,19 which
performs its function as long as the sense faculties exist. Therefore, a ritual performance
such as initiation, which does not affect the sense faculties’ capacity of being operative,
cannot hinder their capacity of generating physical and mental activity (kk. 262–264).
The two positions are based on antithetic doctrinal views. Dharmakīrti takes the opportunity to discuss them as regards their implications in view of rebirth, shifting the focus to
the causes of rebirth, i.e., mental faults. He has the upholders of salvific initiation point
out that if actions are linked to the mind only, the absence of mind – which is the case at
the time of death – would also imply the absence of rebirth. The Buddhist reply to this is
that mental impurities are what links the mind to another birth; therefore, one could accept
initiation as a means to liberation only if such a special ritual affected this capacity of the
impurities. But the Buddhists do not believe that a ritual performance can determine the
results of future actions, since it cannot affect the series of mental faults which arise from
their own seeds, and whose capacity and perpetuation depends on the presence of what
nourishes them, i.e., the view of a self (kk. 265–266).
However, the mention of bījas in the present context seems to be parallel to the example
of a bīja in k. 258, where Dharmakīrti refers to the special treatment of a seed that hinders the
seed’s capacity of generating a sprout. Thus, the discussion of the effects of initiation seems
to be intentionally enclosed between two references to bījas which, given their difference,
amplify the polarization of the debate over the causes of rebirth. On the Buddhist side, the
metaphor of the seed (bīja) in relation to the series of faults is linked to the traditional view
of causation referred to in texts such as the Śālistambasūtra and the AK, where the stock
example is the arising of a sprout from a seed and other causes. It should be noted that
this metaphor is not predictable in Dharmakīrti’s work. In fact, it piles up especially in the
Pramāṇasiddhi chapter and occurs in a few places of the Svārthānumāna and Pratyakṣa
chapters of the Pramāṇavārttika, suggesting that Dharmakīrti mentioned the example of the
seed in order to show how his philosophical discourse linked to the Abhidharmic tradition.
The informed discussion of salvific initiation presented in the PV confirms that materials
relating to groups of Śaiva worshippers were available to Dharmakīrti and his audience
19
As observed by D. Keown, saṃskāras “designate the transformative effect that moral action has upon
the character of the agent” and “Phalas (referred to in Buddhism as karma-vipāka) denote not the
end product of a transpersonal causal chain but the effect of saṃskāric change as experienced by the
actor” (Keown 1996: 336f.). Dharmakīrti’s text can be taken as evidence for Keown’s statement that “a
coherent account of karma can be given purely in terms of saṃskāras” (ibid., p. 337).
Cristina Pecchia
367
as significant parts of their religious and intellectual environment.20 Further, although
the intellectual apparatus of the tantric communities was very likely still quite thin, these
groups were probably well aware of the necessity of having their soteriological programme
legitimized at various levels. The fact that they provide a physical proof of the validity of
their ritual procedure is indeed telling about their programmatic effort of gaining a place
in society for their soteriological proposal. When Dharmakīrti specifically refers to the
tulādīkṣā, he is not only arguing against another way of attaining liberation, but also, or
primarily, against the demonstrative value attached to a ritual performance, as if liberation
from suffering could be equated to a reduction of bodily weight. The tulādīkṣā was a perfect
example to this end, namely casting a bad light on the tantric proposal for liberation.
However, it is very difficult, if not impossible, to specify the identity of Dharmakīrti’s
target, whether he was addressing various groups that shared some basic ideas based on their
belief in īśvara, or practitioners whose ideology was represented in a particular tantric work.
Some key terms in the section on initiation point to ideas and practices that correspond to
those attested in the Niśvāsatattvasaṃhitā, testifying to early forms of concepts that will
develop and appear in later tantric texts in different forms and sometimes under different
names. A case in point is the idea of impurity, which is referred to as pāpa (“bad deed”
or “sin”) in the Niśvāsatattvasaṃhitā and in the PV section on initiation. The term pāpa
mentioned at k. 259 refers to an imperceptible material substance (dravya) and is contrasted
with mala (k. 265), which Dharmakīrti uses to designate mental impurity within a Buddhist
discourse, as can be seen in PV II.208–209 and 212.21 Thus, the PV section on initiation
bears witness to a stage of the development of tantric doctrines earlier than that attested in
the Svāyambhuvasūtrasaṅgraha (not later than the middle of the seventh century),22 where
the term mala, together with pāśas (bonds), denotes three distinct types of impurities that
cover the soul. Indeed, D. Acharya has argued that strong criticism from the Buddhists
urged the Śaivas to reformulate their claims on the removal of sin through initiation. It is
to be noted that the tantric doctrinal development reflected by the use of the term mala
in extant written sources does not correspond to a change in the role of initiation, which
maintains its function of making Śiva intervene in the initiand’s life and allowing the
initiand to eventually attain liberation.23
The concept of impurities features prominently in section V.88 of Kṣemarāja’s Uddyota
on Svacchandatantra (SvaTU), which concludes the chapter concerning tantric initiation.
In the pūrvapakṣa he has the Buddhists (saugatāḥ) dispute tantric purification and its four
possible objects, which are the self, the mind, actions, and impurities such as the view of
a self.24 Also, at the beginning of his response to the Buddhist criticism he explains the
20
21
22
23
24
See Eltschinger 2014: 125ff. for a consideration of the mention of tantric works in the Svārthānumāna
chapter of the PV, and Bisschop 2010: 483–486 for some considerations about the presence of the
Pāśupata movement around the middle of the first millennium.
On these kārikās see Pecchia 2015: 148–153 and the respective sections in Part 3, ”Comments upon
the Kārikās ...”.
See Niśvāsa, pp. 40f., and Acharya 2014: 10f.
Sanderson 1992: 285f.
SvaTU 73.10–13: tad atra dīkṣāyām eva pratyavatiṣṭhante saugatāḥ – iha dīkṣayā kim ātmanaḥ saṃskāraḥ kriyate buddher vā? kim ātmagrahādīnāṃ malānāṃ, kiṃ vā karmaṇām?
368
Dharmakīrti on the Role of Salvific Initiation
Śaiva typology of impurities which cover the self.25 In both cases, he seems to be blissfully
unaware of earlier articulations of the tantric concept of impurity.
The Buddhists mentioned by Kṣemarāja, as Sironi has shown, can be easily identified
with Dharmakīrti and followers of his ideas, since most of the arguments in the pūrvapakṣa
consist in a free paraphrase of formulations found in the PV section on initiation. Also,
the uttarapakṣa includes citations from the same passage. The following issues from the
pūrvapakṣa, in particular, can be linked to the PV:
1. If the mind were not capable of moving the senses after initiation, the senses of an
initiate could not be active (SvaTU 75.1f. → PV II.265cd).
2. Mental faults together with their karmic impressions do not disappear in an initiate
(SvaTU 75.2f. → PV II.266cd).
3. If the mind were eradicated immediately after initiation, the body would dissolve
and there would not be any activity (SvaTU 75.3ff. → PV II.264d–265a).
4. The āgama as a valid means of cognition (SvaTU 75.10f. → PV II.257).
5. If actions could be destroyed only by initiation because they are not embodied, they
would not possess any power even in the case of an oil massage or scorching oneself
with fire; also, if the initiation with scales (dhaṭadīkṣā)26 makes the initiate lighter,
sin should be embodied (SvaTU 75.11–76.1 → PV II.258–259).
In his discussion Kṣemarāja avails himself of the devices elaborated in a centuries-long
reflection within non-dualist Śaiva thought. However, while arguing from a coherent and
mature perspective, he also seems to employ a rhetorical strategy aiming to neutralize the
negative nuance covertly generated by Dharmakīrti’s remarks on initiation. As can be seen
in Table 1, he in fact reverses the PV sequence of arguments. The segment PV II.258–259,
which presents a prasaṅga and Dharmakīrti’s sarcastic remarks against initiation, appears
only at the end of Kṣemarāja’s pūrvapakṣa and is split in two parts in the uttarapakṣa,
where k. 258 is discussed at the very beginning and k. 259 towards the end, followed by a
defense of the validity of the Śaiva Scriptures.
In his response, Kṣemarāja first states that Dharmakīrti’s critique concerning the undesired consequence of ritual initiation in fact reveals the risibility of the Buddhist point of
view.27 He then adduces the argument of mantras, which, given their inconceivable power,
also possess the capacity of eliminating bonds.28
25
26
27
28
SvaTU 76.4–10: ayam ātmā … āṇavena malena … kārmeṇa … māyākhyena malena ca valitaḥ.
Sironi’s Italian translation of dhaṭadīkṣā° (SvaTU 75.17) with “iniziazione del vaso” (1988: 93) seems
to presuppose the reading ghaṭadīkṣā°, because “vaso” (in English “pot”) is a usual translation of
ghaṭa. However, the reading ghaṭa is problematic in the present SvaTU context and might be due to
an oversight. An analogous oversight might have caused Sironi’s translations of dhaṭaśuddhivat and
dhaṭavat (SvaTU 81.4 and 7) with “come quando si pulisce un vaso” and “così come accade di un
vaso,” respectively (Sironi 1988: 97). With regard to the latter passage Sironi explicitly states that she
reads ghaṭa°, and not dhaṭa°, by saying that she emends the edited text ghaṭataḥ to ghaṭavat (Sironi
1988: 97, n. 37).
SvaTU 77.7f.: tat teṣām eva upahāsyatām āviṣkaroti.
SvaTU 77.13f.: acintyaprabhāvatvāt teṣāṃ pāśapraśamane ’pi sāmarthyaṃ kiṃ na sahyate.
369
Cristina Pecchia
Table 1: Arrangement of PV II.257–267 (on ritual initiation) in Kṣemarāja’s
Uddyota on Svacchandatantra V.88
PV kārikās grouped in the- PV kārikās in the SvaTU
matic units according to pūrvapakṣaa
the PVV
PV kārikās in the SvaTU
uttarapakṣa
257
258–259
260–261
262–264
265–266
267
258 (77.5f.)
266cd (77.15f.)
262a–c, 264 (80.14f., 18f.)
259c (81.1)
257 (82.7f.)
[256 (82.10f.)]
a
265cd (75.1f.)
266cd (75.2f.)
264d–265a (75.3ff.)
257 (75.10f.)
258ac–259 (75.11–76.1)
Page and line numbers of the SvaTU edition are given in brackets.
One of his main concerns, however, is the nature of faults and the means for their
final elimination. The view expounded in PV II.266, in particular, engages him in a longer
analysis aiming to demonstrate, first, that the cleansing of impurities is not distinct from
that of the self and, second, that initiation can hinder the arising of faults because it has the
capacity of eventually effecting the purification of the self. In this connection, Kṣemarāja
argues again for the special powers of mantras. Even if they cannot cancel the karman that
supports the present body and its related passions, they are able to hinder any future arising
of bonds (pāśas). Their way of operating is exemplified by the case of an ugly man who by
means of mantras obtains a very beautiful wife, although his ugliness is not cancelled.29
Kṣemarāja’s emphasis on the force of mantras in connection with karman and impurities
points out his commitment in holding onto the tradition and providing evidence for its
tenability. A similar commitment is shown when he explains the loss of weight after
initiation, which he regards as being due to the elimination of the cause of weight, namely
tamas. A quite different attitude was possible, though, as is evident from Abhinavagupta’s
consideration of the tulādīkṣā as a means to give confidence to deluded people.30
It is worth noting that in both the pūrvapakṣa and the uttarapakṣa Kṣemarāja does
not refer to contents of the PV section on initiation which focus on the role of karman
(PV II.260–261) and in the pūrvapakṣa he also omits to mention Dharmakīrti’s remarks
on the nature of karmic impulses (PV II.262–264a). Rather, in his response, Kṣemarāja
elaborates on the view of the self according to a non-dualist Śaiva doctrine and the value
of its purification through initiation:
But in truth liberation is the viewing of the self in that which has the nature of
truth, which is made of consciousness, and is admitted as the ultimate reality
29
30
SvaTU 78.1–5: yathā hi virūpasya vairūpyam anivartyāpi lokottararamaṇīvaśīkaraṇaṃ mantraiḥ kriyate,
tadvad dehārambhikarmāśodhanāt vartamānadehe rāgādyanivṛttav api bhāviprarohapratirodhanaṃ
pāśānāṃ kurvatāṃ mantrāṇāṃ kim āyātam.
TĀ 20.1: atha dīkṣāṃ bruve mūḍhajanāśvāsapradāyinīm. Italian translation in Gnoli 1999: 458.
370
Dharmakīrti on the Role of Salvific Initiation
of any form of action and cognition because it is the ultimate reality of the
I-awareness. From any point of view, … liberation is just the manifestation of
being Śiva. And in this regard the performance of initiation is for the sake of
realizing that one attains that everything is of the same essence after knowing
it.
This forms a stark contrast with the Buddhist approach to the way of attaining liberation,
which Kṣemarāja describes in the concluding part of the pūrvapakṣa:
Therefore, liberation is nothing but the insight of selflessness preceded by the
meditative practice on momentariness and so forth.31
His statement shows how a late representative of the Indian intellectual tradition perceived
Dharmakīrti’s soteriological stance and can be taken as an answer from within the tradition
to the question “Are Buddhist Pramāṇavādins non-Buddhistic?,” which Helmut Krasser
asked and discussed in an article of 2004. The SvaTU presentation of Dharmakīrti’s critique
on initiation as a method to attain liberation suggests that Kṣemarāja regarded such a critique
as being part of a soteriological approach in which typically Buddhist concepts such as
momentariness are considered instrumental to practices that lead to liberation according
to a Buddhist point of view, namely meditative practices that lead to the realization of
selflessness.
Appendix
Text and translation of PV II.257–267
The Sanskrit text and footnotes presented in this appendix are not a critical edition of the
PV kārikās, but only a provisional presentation of them, as they are transmitted in the
available witnesses in Sanskrit, i.e., KH , the manuscript bearing the PV kārikās; ASa , the
manuscript bearing Prajñākaragupta’s commentary on the PV, which includes the kārikās;
and V, which indicates the kārikās as reflected in Manorathanandin’s commentary on the
PV. The text in Sāṅkṛtyāyana’s printed edition of the latter does not deviate from the text in
the manuscript. It is to be noted that the PV kārikās in the PVV result from Sāṅkṛtyāyana’s
reconstruction of the PV.32
(257) āgamasya tathābhāvanibandhanam apaśyatām /
muktim āgamamātreṇa vadan33 na paritoṣakṛt //
One who, based on the Scripture alone, proclaims liberation, does not satisfy
31
32
33
SvaTU 76.1f.: tasmāt kṣaṇikatvādibhāvanāpūrvikā nairātmyadṛṣṭir eva mokṣa iti.
For details about these witnesses, see Pecchia 2015, chapter 9. A critical edition of the PV kārikās here
under examination and Manorathanandin’s commentary thereon is under preparation in the framework
of the FWF project “Indian Buddhist epistemology and the path to liberation” (see first note). For a
German translation of this section of the PV, see Vetter 1990; for some considerations, see Eltschinger
2014: 122ff. My translation of kk. 258–59 and 262–265 only slightly differs from the translation
provided in Acharya 2014: 14ff.
vadan KH V: bruvan ASa .
Cristina Pecchia
(258)
(259)
(260)
(261)
(262)
(263)
34
35
36
37
38
39
371
those who do not see the reason why the Scripture is of that kind [i.e., reliable with
regard to its content].
nālaṃ bījādisaṃsiddho vidhiḥ puṃsām ajanmane /
tailābhyaṅgāgnidāhāder api muktiprasaṅgataḥ //
The rite [of initiation] validated by the example of a seed et cetera is not sufficient
for the absence of rebirth of embodied souls,
because there would be the undesired consequence of liberation even due to an oil
massage, scorching [oneself] with fire, and the like.
prāg guror lāghavāt paścān na pāpaharaṇaṃ kṛtam /
mā bhūd gauravam evāsya na pāpaṃ gurv amūrtitaḥ34 //
That a man who weighed heavier before becomes lighter after [initiation] does not
mean that his sin is removed.
Let it [even] be the case that he has no weight at all; [but] sin is not heavy because it
is not embodied.
mithyājñānatadudbhūtatarṣasaṃcetanāvaśāt /
hīnasthānagatir janma tena35 tacchin na jāyate //
It is due to intentional mental acts associated with false cognition and the craving
arising from it
that there is rebirth, which is going to an inferior place [such as a womb]. Therefore,
one who cuts them is not reborn.
tayor eva hi sāmarthyaṃ jātau tanmātrabhāvataḥ /
te cetane svayaṃ karmety36 akhaṇḍaṃ janmakāraṇam //
Indeed, these two alone are capable of causing rebirth, because the latter occurs only
due to them.
Those two [types of] intentional mental acts are by themselves karma. Therefore, the
cause of rebirth is not fragmentary (i.e., defective, in our description).
gatipratītyoḥ karaṇāny37 āśrayas tāny adṛṣṭataḥ /
adṛṣṭanāśād agatis38 tat saṃskāro na cetanā //
[Ob.:] The sense faculties are the basis of [instances of] cognition and motion; they
are due to the unseen force.
Because the unseen force is destroyed, there is no motion. This [force] is the karmic
impulse, not an intentional mental act.
sāmarthyaṃ karaṇotpatter bhāvābhāvānuvṛttitaḥ /
dṛṣṭaṃ buddher na cānyasya santi tāni na yanti kim39 //
[Re.:] The capacity of arising that belongs to [the activity of] the sense faculties
is observed in connection with the presence or absence
of the mind, and nothing else. When these [faculties] exist, why should they not
continue [to be operative]?
amūrtitaḥ Asa V: amūrttataḥ KH .
tena KH ASa : tatas V.
karmety KH V ASa (post correctionem): karmmāty ASa (ante correctionem).
The reading karaṇāny is attested in KH Asa V, but the printed edition of Prajñākaragupta’s commentary
has kāraṇāny.
agatis KH V: na gatis ASa .
santi tāni na yanti kim KH V: tāni santi na santi kim ASa .
372
Dharmakīrti on the Role of Salvific Initiation
(264) dhāraṇapreraṇakṣobhanirodhāś cetanāvaśāḥ /
na syus teṣām asāmarthye tasya dīkṣādyanantaram //
Concentrating, moving, being agitated or withdrawing are due to intentional mental
acts.
These [actions] would not occur if those [faculties] were not possessed of any capacity
immediately after one’s initiation et cetera.
(265) atha buddhes tadā ’bhāvān na syuḥ sandhīyate malaiḥ /
buddhis40 teṣām asāmarthye jīvato ’pi syur akṣamāḥ //
[Ob.:] If [actions] were due to the mind, they would not occur, because then (i.e.
after initiation) [the mind] would not be there. [Re.:] Because of impurities
the mind is linked [to other births]. If [impurities] were not possessed of any capacity
[to impel action after initiation], even in the case of an [initiated] living being they
would be unable [to produce any effect].
(266) nirhrāsātiśayāt puṣṭau pratipakṣasvapakṣayoḥ /
doṣāḥ svabījasantānā dīkṣite ’py anivāritāḥ //
Since they decrease or increase according to the degree of prosperity of what hinders
or favours them,
mental faults, whose series arise from their own seeds, are unimpeded even in one
who is initiated.
(267) nityasya nirapekṣatvāt kramotpattir virudhyate /
kriyāyām akriyāyāṃ ca kriyayoḥ41 sadṛśātmanaḥ //
The successive arising of something permanent is contradictory because of the
independence of the latter—
[something permanent which is] of the same nature in the case of action and nonaction [of the faculties etc.] in both times (e.g., before death and after death).
References and abbreviations
ASa Manuscript copy of the Alaṅkāra of Prajñākaragupta as reproduced in: Sanskrit
manuscripts of Prajñākaragupta’s Pramāṇavārttikabhāṣyam. Facsimile edition by
S. Watanabe. Patna/Narita 1998.
Acharya 2014 D. Acharya, On the Śaiva concept of innate impurity (mala) and the function
of the rite of initiation. Journal of Indian Philosophy 42 (2014) 9–25.
AK Abhidharm-koshabhāṣya of Vasubandhu, ed. P. Pradhan. Patna 1967.
AN The Aṅguttara-Nikāya. Part III, ed. E. Hardy. London 3 1976 (first published 1897).
Bisschop 2010 P. Bisschop, Śaivism in the Gupta-Vākāṭaka age. Journal of the Royal
Asiatic Society 20 (2010) 477–488.
Eltschinger 2014 V. Eltschinger, Buddhist epistemology as apologetics. Studies on the
history, self-understanding and dogmatic foundations of late Indian Buddhist philosophy. Wien 2014.
Franco 1997 E. Franco, Dharmakīrti on compassion and rebirth. Wien 1997.
40
41
buddhis KH V (see PVV 101.3): buddhes ASa (and Sāṅkṛtyāyana’s reconstruction of the pāda in his
edition of the PVV).
kriyayoḥ KH V: kriyā ca ASa .
Cristina Pecchia
373
Franco 2001 E. Franco, Dharmakīrti’s reductionism in religion and logic. In: Le parole e
i marmi. Studi in onore di Raniero Gnoli, ed. R. Torella. Roma 2001, 285–308.
Gnoli 1999 Abhinavagupta, Luce dei Tantra. Tantrāloka, ed. Raniero Gnoli. Milano 1999.
Goodall & Isaacson 2007 D. Goodall and H. Isaacson, Workshop on the Niśvāsatattvasaṃhitā, Newsletter of the NGMCP 3 (2007) 4–6.
Harvey 2011 P. Harvey, An analysis of factors related to the kusala/akusala quality of
actions in the Pāli tradition. Journal of the International Association of Buddhist
Studies 33.1–2 (2011) 175–209.
Heim 2013 M. Heim, The forerunner of all things. Buddhaghosa on cetanā. New York
2013.
Inami & Tillemans 1986 M. Inami and T.J.F. Tillemans, Another look at the framework
of the Pramāṇasiddhi chapter of Pramāṇavārttika. Wiener Zeitschrift für die Kunde
Südasiens 30 (1986) 123–142.
Keown 1996 D. Keown, Karma, character, and consequentialism. The Journal of Religious
Ethics 24.2 (1996) 329–350.
KH Pramāṇavārttika (kārikās) of Dharmakīrti, facsimile of the manuscript in: B. Kellner
and F. Sferra, A palm-leaf manuscript of Dharmakīrti’s Pramāṇavārttika from the
collection kept by the Nepalese rājaguru Hemarāja Śarman. In: Sanskrit texts from
Giuseppe Tucci’s collection, ed. F. Sferra. Roma 2008, 443–483.
Krasser 2004 H. Krasser, Are Buddhist Pramāṇavādins non-Buddhistic? Dignāga and
Dharmakīrti on the impact of logic and epistemology on emancipation. Hōrin 11
(2004) 129–146.
Niśvāsa The Niśvāsatattvasaṃhitā. The earliest surviving Śaiva Tantra. Vol. 1. A critical
edition & annotated translation of the Mūlasūtra, Uttarasūtra & Nayasūtra, ed.
Dominic Goodall, in collaboration with Alexis Sanderson and Harunaga Isaacson,
with contributions of Nirajan Kafle, Diwakar Acharya and others. Pondicherry 2015.
Pecchia 2015 C. Pecchia, Dharmakīrti on the cessation of suffering. A critical edition
with translation and comments of Manorathanandin’s Vṛtti and Vibhūticandra’s
glosses on Pramāṇavārttika II.190–216. Leiden/Boston 2015.
PV Pramāṇavārttika of Dharmakīrti, in PVV and ASa .
PV II Pramāṇasiddhi chapter of the PV.
PVV Pramāṇavārttikavṛtti of Manorathanandin, in: Dharmakīrti’s Pramāṇavārttika with
a commentary by Manorathanandin, ed. R. Sāṅkṛtyāyana. Appendix to Journal of
the Bihar and Orissa Research Society 24–26 (1938–1940) (for the numbering of
the kārikās, see Vetter 1964: 116f.).
Sanderson 1988 A. Sanderson, Śaivism and the Tantric traditions. In: The world’s religions, ed. S. Sutherland et alii. London 1988, 660–703.
Sanderson 1992 A. Sanderson, The doctrine of the Mālinīvijayottara. In: Ritual and
speculation in early Tantrism. Studies in honor of André Padoux, ed. T. Goudriaan.
Albany 1992, 281–312.
Sanderson 2001 A. Sanderson, History through textual criticism. In the study of Śaivism,
the Pañcarātra and the Buddhist Yoginītantras. In: Le sources et le temps. Sources
and time: A colloquium, Pondicherry, 11–13 January 1997, ed. François Grimal.
Pondicherry 2001, 1–47.
374
Dharmakīrti on the Role of Salvific Initiation
Sironi 1988 A. Sironi, Il commento di Kṣemarāja alla stanza V, 88 dello Svacchandatantra:
natura e scopo della dīkṣā. Rivista degli Studi Orientali 61 ([1987] 1988) 89–113.
SvaTU Kṣemarāja, Svacchandatantra-uddyota, ed. M.S. Kaul. 6 vols. Bombay 1921–1935.
TĀ Abhinava Gupta, The Tantrāloka. With commentary by Jayaratha Rājānaka, ed.
Mukund Rām Shāstri. 6 vols. Bombay 1921.
Taber 2003 J. Taber, Dharmakīrti against physicalism. Journal of Indian Philosophy 31
(2003) 479–502.
TAK III Tāntrikābhidhānakośa III, T–Ph. Dictionnaire des termes techniques de la litterature hindoue tantrique. A dictionary of technical terms from Hindu tantric literature.
Wörterbuch zur Terminologie hinduistischer Tantren. Wien 2013.
Torella 1992 R. Torella, The pratyabhijña and the logical-epistemological school of Buddhism. In: Ritual and speculation in early Tantrism. Studies in honor of André
Padoux, ed. T. Goudriaan. Albany 1992, 327–345.
V Pramāṇavārttika kārikās as reflected in the PVV.
Vetter 1964 T. Vetter, Erkenntnisprobleme bei Dharmakīrti. Wien 1964.
Vetter 1990 T. Vetter, Der Buddha und seine Lehre in Dharmakīrti’s Pramāṇavārttika.
Der Abschnitt über den Buddha und die vier edlen Wahrheiten im PramāṇasiddhiKapitel. Wien 2 1990 (1st ed. 1984).
Dharmottara on the viparyaye bādhakapramāṇa and
trairūpya in Dharmakīrti’s sattvānumāna
by
Masamichi Sakai
Introduction
It goes without saying that Dharmakīrti’s new approach of inferring momentariness (kṣanikatvānumāna) – i.e., the sattvānumāna, the inference of momentariness based on the
inferential reason property “existence” (sattva) – had a decisive impact on the later development of this kind of Buddhist inference. At the same time, it is also true that the
sattvānumāna generated a number of interpretative tasks for Dharmakīrti’s successors.
The problem is that the sattvānumāna seems to contain many aspects that threaten
to shake the foundation of the traditional Buddhist logic in place since Dignāga.1 Prof.
Katsumi Mimaki gives a clear synopsis of the problems confronting later Buddhist logicians,
basing himself mainly on the works of Jñānaśrīmitra, Ratnakīrti, Ratnākaraśānti, and
Mokṣākaragupta, who constitute the last phase of Buddhist philosophers.2 I agree with
Prof. Mimaki’s contention that the problems they dealt with did not arise suddenly at that
time, but rather had been developing gradually over the course of history.3 I have shown
for example that one of the most crucial problems of the sattvānumāna – namely that the
“example” (henceforth: dṛṣṭānta) is of no use, which seemingly forces the Buddhist to
discard the second trairūpya condition (sapakṣa eva sattvam; henceforth T2) – was argued
by Arcaṭa with a keen awareness.4
Arcaṭa’s pupil, Dharmottara, shares his teacher’s awareness of the problem. In line with
Arcaṭa’s argument, he also asks whether or not the dṛṣṭānta in the sattvānumāna is necessary.
However, Dharmottara goes even further, contesting the necessity of the inferential reason
property “existence” itself, thereby deepening the problem and developing the argument.
1
2
3
4
I am very grateful to Mr. Tyler Neill (PhD candidate, Leipzig University) for correcting and improving
my English. Through the conversation with him, I could also improve this essay as a whole. With my
deepest gratitude, I dedicate this article to my beloved Teacher Dr. Helmut Krasser, who privileged me
to be a member of his project of editing the codex unicus of Dharmottara’s Pramāṇaviniścayaṭīkā.
It is interesting that Dharmakīrti himself presents the sattvānumāna as if there were no conflict between
the logic underlying the sattvānumāna and the traditional Buddhist logic in place since Dignāga. For
now I cannot make any judgments about whether Dharmakīrti himself thinks that the logic of the
sattvānumāna does not deviate from the traditional Dignāgean logic, but his followers’ activities make
me think this is unlikely.
Cf. Mimaki 1976 and Mimaki 1984.
Cf. Mimaki 1984: 238.
I read a paper on this topic at the XVIIth Congress of the International Association of Buddhist Studies,
Vienna 2014; see Sakai 2015. In Arcaṭa’s argument, we can see a direct link with the antarvyāpti theory
advocated by Ratnākaraśānti and reported by Mokṣākaragupta in comparison with the bahirvyāpti
theory (cf. TBh 47,1–6).
Birgit Kellner et al., eds., Reverberations of Dharmakīrti’s Philosophy: Proceedings of the Fifth International Dharmakīrti
Conference Heidelberg, August 26 to 30, 2014. Vienna 2020, pp. 375–389.
376
Dharmottara on viparyaye bādhakapramāṇa and trairūpya
The aim of this article is to introduce and clarify Dharmottara’s arguments for solving
the problematic nature of the sattvānumāna,5 by shedding light on what the exact problem
is that Dharmottara inherits from his teacher, and how he addresses and develops it. In doing
so, I attempt to properly locate Dharmottara’s contribution within the larger interpretative
history of the sattvānumāna.
1 Arcaṭa and Dharmottara: sharing the problem of the dṛṣṭānta and
the viparyaye bādhakapramāṇa
1.1 Arcaṭa’s argument: the dṛṣṭānta is useless and redundant
Arcaṭa’s greatest concern, as well as his pupil’s, is the relation between the threefold
characteristic of a good reason property, i.e., trairūpya, and the so-called viparyaye bādhakapramāṇa in the sattvānumāna.
In Dharmakīrtian logic, the viparyaye bādhakapramāṇa in the sattvānumāna plays
the critical role of demonstrating the essential connection (svabhāvapratibandha) or the
pervasion (vyāpti) between the inferential reason property (henceforth: hetu) “existence”
and the target property (henceforth: sādhyadharma) “momentariness.”6
In the Hetubinduṭīkā (HBṬ), Arcaṭa argues7 that, in the case of kṣaṇikatvānumāna, since
its sādhyadharma “momentariness” is imperceptible by nature, it is therefore impossible
to find and show a dṛṣṭānta via perception (pratyakṣa). Generally speaking, in Dignāgean
inference, in order to say that a hetu satisfies T2, one must exhibit at least one thing that
possesses the hetu as well as the sādhyadharma and that is ontologically different from the
site of inference (sādhyadharmin; henceforth: pakṣa). However, in the kṣaṇikatvānumāna,
it is in fact impossible to find any momentary thing at all via perception. How then can
one find a momentary thing to serve as dṛṣṭānta? Concerning this problem, Arcaṭa is of
5
6
7
Perceiving the importance of Dharmottara’s arguments, Prof. Tadashi Tani was early in exploring them
as they appear both in Dharmottara’s Kṣaṇabhaṅgasiddhi (KBhS) and Pramāṇaviniścayaṭīkā, second
chapter (PVinṬ 2). See Tani 1997 and Tani 2000. At that time, however, the Sanskrit manuscript of
the latter was unavailable. But now, the situation has changed, and I am luckily in a position to consult
the codex unicus of the PVinṬ 2 (for the project of editing the codex unicus of the PVinṬ, see https:
//www.oeaw.ac.at/ikga/forschung/buddhismuskunde-sinologie/schwerpunkte/pramanaviniscayatika/),
which covers the entirety of the important portion in which Dharmottara discusses the problem of
the sattvānumāna most intensively. The Sanskrit original surely enables us to grasp Dharmottara’s
arguments with more clarity.
Dharmakīrti’s first use of this defeating source of knowledge (bādhakapramāṇa) for the sattvānumāna
is in the Pramāṇaviniścaya, second chapter (cf. PVin 2 80,1–8), and also in his later works, the
Hetubindu and the Vādanyāya (cf. HB 4,9–12, VN 2,1–4, respectively). He puts forward the viparyaye
bādhakapramāṇa as the method of proving the essential connection for the sattvānumāna. Its basic and
simple form can be demonstrated as follows: “Given that there is non-momentariness, since there is
the inconsistency with purpose fulfillmenta , the being-real which is characterized as that [i.e., purpose
fulfillment,] is abandoned” (HB 4,11-12: akṣaṇikatve ’rthakriyāvirodhāt tallakṣaṇaṃ vastutvaṃ hīyata
iti.). For a German translation, cf. Steinkellner 1967: 37. a My understanding of the word arthakriyā
is based on Dharmottara’s elucidation in his PVinṬ and the Nyāyabinduṭīkā. Cf. PVinṬ 2 Ms 89a2–3
(PVinṬS 17,11): arthaḥ prayojanam, tasya kriyā niṣpattiḥ; NBṬ 76,3–4: arthasya prayojanasya kriyā
niṣpattiḥ.
This paragraph is a summary of Sakai 2015: 284–289 (sections 3.1.–3.2.).
Masamichi Sakai
377
the opinion that, since the viparyaye bādhakapramāṇa can prove the essential connection,
it can also prove that whatever possesses the hetu has the sādhyadharma. Thus, if one
applies the viparyaye bādhakapramāṇa to something possessing the hetu and ontologically
different from the pakṣa, one can show that the hetu satisfies T2. In the final analysis,
however, Arcaṭa sees this activity of finding a dṛṣṭānta as nonsensical, given the natural
objection in favor of directly applying the viparyaye bādhakapramāṇa to the pakṣa. That
is, the logical method used to find something momentary as a dṛṣṭānta can also be used to
prove that the pakṣa is momentary. With this in mind, Arcaṭa views the dṛṣṭānta as being
redundant and of no use in the case of the sattvānumāna.8 He asserts that T2 in the case of
the sattvānumāna should be understood to be fictitious (kālpanika).
1.2 Dharmottara’s treatment of the dṛṣṭānta
Dharmottara takes the same position as Arcaṭa, namely that what proves the momentariness
of a dṛṣṭānta is the viparyaye bādhakapramāṇa and not any other source of knowledge,
and that the same viparyaye bādhakapramāṇa can also be used to prove that the pakṣa
is momentary.9 Given this position, the dṛṣṭānta seems to be of no use, according to
Arcaṭa. However, in the Pramāṇaviniścayaṭīkā (PVinṬ), Dharmottara elucidates a positive
significance of, and role for, the dṛṣṭānta in the case of the sattvānumāna.10
The following series of arguments in the PVinṬ is strongly influenced by Arcaṭa’s
position, and it seems here that Dharmottara, in an attempt to preserve some kind of raison
d’être for the trairūpya, tries to overturn Arcaṭa’s conclusion.
8
9
10
In other words, the dṛṣṭānta never contributes in any way to a proof that the pakṣa has the sādhyadharma.
This is so in the sense that the dṛṣṭānta cannot play even the Dharmakīrtian role of conveying the
essential connection (svabhāvapratibandha) to those who do not remember or do not know it, since
without the functioning of the viparyaye bādhakapramāna, one can never know that the dṛṣṭānta
possesses the sādhyadharma “momentariness.” Given that momentariness itself can never be perceived,
there is no difference between the dṛṣṭānta and the pakṣa. For the difference between the Dharmakīrtian
and the Dignāgean role of dṛṣṭānta, see Steinkellner 2004.
In the KBhS, Dharmottara asserts that even if there can be dṛṣṭāntas, namely things whose momentariness is accepted, it cannot thereby be proved that the pakṣa too is of that nature, and this is because a
mere similarity with such dṛṣṭāntas cannot conclusively prove that the other thing is also momentary.
Thus, Dharmottara requires, in order to prove the momentariness of the pakṣa, the viparyaye bādhakapramāṇa to be applied to the pakṣa. Cf. KBhSF 223,15–17, 21–23: ’di la dpe ni yod kyang chos mthun
pa tsam ’ba’ zhig gis (’ba’ zhig gis KBhSF : ’ba’ zhig gi KBhSP ) dngos po gzhan dag de’i rang bzhin
du ’gyur ba rigs pa dang ldan pa ma yin no // … gal te yang dpe la skad cig mar ’jig pa yod na (’jig pa
yod na KBhSF : ’jig pa KBhSP ) / de lta na yang khyab pa rab tu sgrub pa’i (rab tu sgrub pa’i KBhSF :
rab tu bsgrubs pa’i KBhSP ) tshad mar rab tu bstan pa ñid las skad cig mar ’jig par grub bo // “In this
case [i.e., when momentariness is to be proved], even if there are dṛṣṭāntas, it is not tenable that other
things are [also] of that nature [i.e., = of the nature of momentariness] due to a mere similarity [with
dṛṣṭāntas]. …even if there can be perishing within a moment in dṛṣṭāntas, nevertheless it is only on the
basis of showing the source of knowledge which in fact demonstrates pervasion (i.e., the viparyaye
bādhakapramāṇa) that perishing within a moment is proved.” For Frauwallner’s German translation, cf.
Frauwallner 1935: 241.
In both the KBhS and the PVinṬ, Dharmottara discusses various logical problems with the sattvānumāna,
and although many of these discussions are common to both works, it is only in the PVinṬ that
Dharmottara cares about the positive aspect of the dṛṣṭānta.
378
Dharmottara on viparyaye bādhakapramāṇa and trairūpya
1.2.1 dṛṣṭānta’s role and significance
In the PVinṬ, Dharmottara states:
If impermanence [i.e., momentariness,] is to be proved, what is a dṛṣṭānta,
where the relation between that which is to be pervaded and that which pervades
[it] should be demonstrated? And he [i.e., Dharmakīrti] will declare [later, in
PVin 3 128,1–131,5] that, “One should not take up a hetu, relying on scriptural
doctrine (samaya/gzhung PVinṬt ).” And, when an opponent who does not
adhere to [any] settled doctrine (siddhānta) objects to momentariness, in that
case, he should be forced to provisionally accept the momentariness of a
certain thing even by receiving bribes (utkoca). This is because, otherwise,
owing to the lack of a dṛṣṭānta, there would be no hetu. Therefore, there is no
[dṛṣṭānta] at all whose sādhyadharma has already been well established [for
such an opponent].11
First, in asking himself what to make of the dṛṣṭānta in the inference of momentariness,
Dharmottara refers to Dharmakīrti’s discourse on the antinomic reasons (viruddhāvyabhicārihetu) in the third chapter of the Pramāṇaviniścaya. There Dharmakīrti argues that
his three kinds of hetu – essential feature (svabhāva), effect (kārya), and non-perception
(anupalabdhi) – can never be antinomic. This is because they are all based on real things
(vastu). According to him, any hetus based on scriptures (āgama) are antinomic, since
they do not issue from the force of seeing real things (avastudarśanabalapravṛtta).12 With
respect to this position of Dharmakīrti, Dharmottara sees a role for a dṛṣṭānta. Namely,
11
12
PVinṬ 2 Ms 99a3–6: athānityatve sādhye ko dṛṣṭāntaḥ, yatra vyāpyavyāpakabhāvaḥ sidhyeta. a na
ca samayam āśritya hetur upādeyaa iti vakṣyati. anāśritasiddhāntakaś ca paro yadā kṣaṇikatvaṃ
(kṣaṇikatvaṃ emended [cf. skad cig ma nyid PVinṬt ]: kṣaṇikaṃ Ms) prati pratyavatiṣṭhate, tadāsāv
utkocādānenāpi kasyacit padārthasya kṣaṇikatvābhyupagamaṃ (padārthasya kṣaṇikatvā° emended:
padārthakṣaṇikatvā° Ms) kārayitavyaḥ. b itarathā hi dṛṣṭāntābhāvād dhetur na syāt. tasmān na prasiddhasādhyadharmaḥ kaścit b . For Tani’s Japanese translation from the Tibetan version, see Tani 2000:
167. a–a To my understanding, the bold part is Dharmottara’s short summary of PVin 3 128,1–131,5,
where Dharmakīrti details antinomic reasons (viruddhāvyabhicāra). I have not yet been able to find an
exact passage in the PVin. For the Tibetan translation corresponding to PVin 3 128,1–131,5, which
has been translated into Japanese with a clear analysis, see Ono 2010: 136–139. b–b For this part,
the Tibetan version reads as follows: PVinṬt P305b3;D253a3: de lta ma yin na dpe med pas gtan
tshigs ma yin par ’gyur ro // de’i phyir bsgrub par bya ba’i chos ma grub pa la ni dpe ’ga’ yang
med do // “Otherwise, because there is no dṛṣṭānta, [a relevant reason property] would not indeed
be a reason property. Therefore, with regard to [a reason property] whose sādhyadharma is not well
established (*aprasiddhasādhyadharme/*asiddhasādhyadharme), there is no dṛṣṭānta at all (*na kaścid
dṛṣṭāntaḥ).” For me, the content of this Tibetan translation is difficult.
Cf. PVin 3 128,2–5: na hi sambhavo ’sti kāryasvabhāvayor uktalakṣaṇayor anupalambhasya vā viruddhāvyabhicāritāyām. na cānyo ’vyabhicārī . tasmād avastudarśanabalapravṛttam āgamāśrayam anumānam āśritya tadarthavicāreṣu viruddhāvyabhicārī sādhanadoṣa uktaḥ, śāstrakārāṇām artheṣu
bhrāntyā viparītasvabhāvopasaṃhārasambhavāt. na hy asti sambhavo yathāvasthitavastusthitiṣv ātmakāryānupalambheṣu.
Masamichi Sakai
379
Dharmottara regards a dṛṣṭānta as a place where it is guaranteed that a relevant hetu is
based on a real thing and not on scripture.13
Then, Dharmottara imagines a certain situation in which a dṛṣṭānta is needed: Suppose
there is an opponent who is open-minded, i.e., free from any dogmas, but who never accepts
momentariness. In that case, an advocate of momentariness must force that opponent into
provisionally accepting (abhi-upa-√gam) the momentariness of a certain thing – this
‘certain thing’ is a dṛṣṭānta. Otherwise, not only could he not present the sādhyadharma
to him, but he could not present even the hetu. That is, for such an opponent a certain
thing serving as dṛṣṭānta would be the first place where both the sādhyadharma and the
hetu coexist. Therefore, with this dṛṣṭānta the proponent can first establish both for that
opponent. So, the proponent must employ the following procedure: first, force the opponent
into provisionally accepting the momentariness of a certain thing, and second, make it
known to him that the hetu is also there. After that, he can eventually move on to a proof of
the momentariness of the pakṣa. However, at this point, the sādhyadharma of that dṛṣṭānta
has not yet been proved.
1.2.2 Definition of the dṛṣṭānta in the case of the sattvānumāna
In this line of argumentation, Dharmottara defines the dṛṣṭānta in the inference of momentariness as follows:
Moreover, with regard to this “grasping/holding of a dṛṣṭānta [in a proof],”
mentioned here and there, the following is the meaning: It is by depending on
the fact that the hetu exists in a thing that is different from the pakṣa that one
can make the defeating source of knowledge function, not in a different manner.
Therefore, the dṛṣṭānta is an object 1) which is a sphere (viṣaya) where hetu is
established and 2) which is a place where the defeating source of knowledge
is shown. For an opponent in turn, however, [dṛṣṭānta] is not [yet] proved as
being caused to be bound to the sādhyadharma, [i.e., momentariness], [until
the defeating source of knowledge is made to function there].14
One aspect of this is that Dharmottara confirms that the hetu should be established on
the basis of reality and that it is in the dṛṣṭānta that this takes place. Another aspect is
that the dṛṣṭānta is a place where the viparyaye bādhakapramāṇa should be shown, since
13
14
In the sattvānumāna, the hetu as “existence” refers to “purpose fulfillment,” and this is of course
established on the basis of seeing real things – for example, in the case of a pot as the dṛṣṭānta, its
purpose fulfillment, holding water etc., is established by observation.
PVinṬ 2 Ms 99a6–7: yaḥ punas tatra tatra dṛṣṭāntaparigraha uktaḥ, tatrāyam abhiprāyaḥ (abhiprāyaḥ
emended [cf. dgongs pa PVinṬt ]: a{pra}y{ā}ayaḥ Ms) – sādhyadharmivyatirikte vastuni hetoḥ sattvam
āśritya bādhakaṃ pramāṇaṃ vyāpārayituṃ śakyam, nānyathā. tasmād dhetor vidhiviṣayo bādhakapramāṇapradarśanagocaro (°pradarśana° emended [cf. rab tu ston pa’i PVinṬt ]: °darśana°) ’rtho
dṛṣṭāntaḥ. na tu prativādinaḥ sādhyadharmasambandhitaḥ siddhaḥ. For Tani’s Japanese translation
from the Tibetan version, see Tani 2000: 168.
380
Dharmottara on viparyaye bādhakapramāṇa and trairūpya
otherwise, the proponent cannot prove the dṛṣtānta’s momentariness, which has been only
forcibly and provisionally accepted by the opponent.15
2 The crucial problem of the hetu sattva: No need for the trairūpya condition as a whole
However, the position that the viparyaye bādhakapramāṇa establishes as proved the provisionally accepted momentariness of a dṛṣṭānta evokes the very crucial problem, which, to
the best of my knowledge, was first propounded by Arcaṭa, and which equally confronted
later Buddhist logicians.
Dharmottara introduces the following, seemingly rhetorical pūrvapakṣa:
[Objection 1:] If the defeating source of knowledge does not require [a dṛṣṭānta] of which the sādhyadharma is already admitted, then, in this manner,
when the defeating source of knowledge shows the pervasion of [a property]
“being produced” by [a property] “being impermanent” in a property possessor as dṛṣṭānta, this very same defeating source of knowledge will prove the
impermanence in the pakṣa too. Thus, it is in every case the defeating source
of knowledge that is capable of proving the sādhya[dharma]. For this reason,
there is the undesirable consequence that the hetu, which is of the essential
feature type, is not a real hetu (ahetutvaprasaṅga). [Objection 2:] Moreover,
even if the following is the case: “Defeating (bādhaka) itself does not occur
without depending upon [the hetus] ‘being existent’ and ‘being produced,’
therefore, [the hetu] ‘being existent’ should be needed,” it is nevertheless only
on the basis of pakṣadharmatva [i.e., the fact that the hetu is a property of
the pakṣa], which is accompanied by the defeating source of knowledge, that
the sādhyadharma is proved. Thus, there is the undesirable consequence that
there is not a threefold condition [as a whole, namely because only the first
condition is needed].16
The points of this objection are:
15
16
It should be noted here that in the KBhS Dharmottara does not mention the role and significance of
dṛṣṭānta in the case of the sattvānumāna that he discusses in the PVinṬ (cf. the text portions of the
PVinṬ 2 cited in nn. 11 and 14 above).
PVinṬ 2 Ms 99a7–99b3: a yadi badhakaṃ pramāṇaṃ na siddhasādhyadharmāpekṣam, evaṃ tarhi
yenaiva bādhakena pramāṇena dṛṣṭāntadharmiṇi kṛtakatvasyānityatvena vyāptir upadarśyate, tenaiva
pramāṇena sādhyadharmiṇy apy anityatvaṃ setsyatīti sarvatra bādhakam eva pramāṇaṃ sādhyasādhanasamartham. ataḥ svabhāvahetor ahetutvaprasaṅgaḥ.a athāpi sattvaṃ kṛtakatvaṃ cānāśritya
bādhakam eva na pravartata iti sattvam āśrayaṇīyaṃ, evam api pakṣadharmatvād eva bādhakapramāṇasahāyāt sādhyasiddhir iti trairūpyābhāvaprasaṅgaḥ. For Tani’s Japanese translation from the
Tibetan version, see Tani 2000: 168–169. a–a For this part we have a parallel argument in the KBhS. Cf.
KBhSF 223,23–25: gal te de ltar na bsgrub par bya ba’i chos can nyid la khyab pa (khyab pa KBhSF :
khyab pa’i KBhSP ) sgrub pa’i tshad mas skad cig ma nyid du sgrub na yod pa zhes bya ba’i gtan
tshigs nye bar blangs pa don med par ’gyur ro zhe na / “[Objection:] If, in this way [i.e., as it is in the
dṛṣṭānta], the source of knowledge demonstrating pervasion proves momentariness just in the pakṣa
itself, it should be useless to take/employ the hetu called sattva [in an inference].” For Frauwallner’s
German translation, cf. Frauwallner 1935: 241–242.
Masamichi Sakai
381
1. Given that the viparyaye bādhakapramāṇa can in fact prove the dṛṣṭānta’s momentariness, it can prove the pakṣa’s momentariness too. It follows then that the hetu
sattva in the sattvānumāna is not a real hetu that can in fact prove sādhyadharma –
the real hetu, i.e., reasoning, is the very viparyaye bādhakapramāṇa instead.
2. It may be the case that defeating does not occur without this hetu – since what is
defeated by the viparyaye bādhakapramāṇa is this sattva in the case where nonmomentariness is supposed17 – and so, in this sense, the hetu seems to be needed.18
Nevertheless, it should actually be the case that only the pakṣadharmatva – the fact
that the hetu is a property of the pakṣa – is needed. This is because the viparyaye
bādhakapramāṇa can function in the pakṣa if and only if that pakṣa possesses sattva,
so that its possession of the sādhyadharma can be proved by it. In this case, what is
needed for a sound inference is only the first trairūpya condition (henceforth: T1).
Thus, the trairūpya condition considered as a whole would be useless.
It is this same second point of this objection on the basis of which Arcaṭa discards the
necessity of T2.
2.1 Counterexample and the second trairūpya condition
To respond to these undesired consequences on behalf of the trairūpya theory, Dharmottara
appeals to the concept of a ‘counterexample’ in a possible debate with opponents:
[Answer:] [To this objection,] we say (ucyate): What has been said, namely
that, “[It is sufficient that] there is pervasion only in the pakṣa, and so what
can be gained by grasping other property possessors [other than the pakṣa]?”
is not tenable. This is because it is not possible to show the defeating source of
knowledge in a single locus [i.e., in the pakṣa], after setting aside visible objects
that are different from the pakṣa. Suppose the defeating source of knowledge
should be shown in the following manner: “If there were not the momentariness
of a sound, there would not be even [its] existence.” If an opponent were to
say, “Just as a visible thing, such as a pot etc., though not momentary, achieves
purpose fulfillment, this sound too must be so,” [then] those visible [things],
such as pots, should also in the same manner [i.e., like in the case of a sound]
be brought into the scope of the defeating source of knowledge. And, with
regards to those [visible things such as a pot and the like] that have [already]
been brought into the scope [of the defeating source of knowledge], he [i.e., the
opponent] may respond: “A certain invisible thing, though it achieves purpose
[fulfillment], should be (bhaviṣyati) non-momentary, and these visible [things,
the pot and so on] are like this.” Therefore, [this] invisible thing that has in
turn been put forth as a [counter]example by the opponent should be brought
into the scope of the defeating source of knowledge. Therefore, why is it the
17
18
Cf. n. 6 above.
In other words, this position refers to the view that the hetu sattva and the property “being existent”
defeated by the viparyaye bādhakapramāṇa are the same.
382
Dharmottara on viparyaye bādhakapramāṇa and trairūpya
case that the defeating source of knowledge is shown only in the specific and
single pakṣa?19
Dharmottara’s point is this: Even though the proponent applies the viparyaye bādhakapramāṇa directly to the pakṣa and thereby completes the proof of its momentariness, his
19
PVinṬ 2 Ms 99b3–7: ucyate – yad uktam – sādhyadharmiṇy eva vyāptiḥ, tat kiṃ dharmyantaraparigraheṇeti, tad ayuktam, yasmād dṛṣṭān sādhyadharmivyatiriktān bhāvān parityajya naikasmin dharmiṇi
bādhakaṃ pramāṇaṃ śakyaṃ darśayitum. yadi kṣanikatvaṃ na syāc chabdasya, sattvam eva na syād
ity evaṃ bādhake pramāṇe darśayitavye yadi paro brūyāt – yathā ghaṭādir dṛṣṭo bhāvo ’kṣaṇikatve ’py
arthakriyākārī , tadvad ayaṃ śabdo ’pi syāt iti, evaṃ te ’pi dṛṣṭā ghaṭādayo bādhakapramāṇaviṣayīkartavyāḥ. teṣu ca viṣayīkṛteṣu punarbrūyāt (punarbrūyāt emended: punabrūyāt Ms) – adṛṣṭo bhāvaḥ
kaścid arthakāry apy akṣaṇiko bhaviṣyati, tadvac cāmī dṛṣṭā iti punaḥ pareṇa dṛṣṭāntīkṛto ’dṛṣto bhāvo
bādhakapramāṇaviṣayīkartavya iti (°kartavya iti emended [bya dgos pa’i phyir PVinṬt ]: °kartta iti Ms)
katham ekasminn eva sādhyadharmiṇi bādhakapramāṇapradarśanam. For Tani’s Japanese translation
from the Tibetan version, see Tani 2000: 169–170. For this part we have a parallel argument in the
KBhS (the exact literal parallel is the part a–a). Cf. KBhSF 223,25–224,17: a chos can nyid la khyab pa
sgrub par byed pa’i (khyab pa sgrub par byed pa’i KBhSF : khyab par sgrub par byed pa’i KBhSP )
tshad mar rab tu bstan par ni nus pa ma yin te / ’di ltar khyab pa sgrub pa’i tshad ma ni ’di ltar bstan
par bya ste / gal te sgra rtag par gyur na rim dang cig car dag gis don bya ba la nye bar sbyor bar
mi ’gyur ro zhes so // ’di la ni pha rol pos kyang ji ltar bum pa skad cig ma ma yin yang rim dang
cig car dag gis mngon par ’dod pa’i ’bras bu bsgrub par nus pa de bzhin du sgra yang ’gyur ro zhes
brjod par nus so // ’di la yang lan / bum pa yang ngas gnod par byed pa’i tshad ma’i yul du byas pa
nyid do zhes brjod dgos so // des na rnam grangs ’dis pha rol gyi nye bar bkod pa mthong ba’i dngos
po mtha’ dag rgol bas (rgol bas KBhSF : gol bas KBhSP ) gnod par byed pa’i tshad ma’i yul du bya
ba yin no // gal te pha rol po dngos po ’ga’ zhig skad cig ma ma yin par rim dang cig car gyis (rim
dang cig car gyis KBhSF : rim dang cig car gyi KBhSP ) don byed par nus pa de bzhin sgra yang yin no
zhes ma mthong ba yang dogs par byed na (dogs par byed na KBhSF : dags par byed na KBhSP ) de la
yang gal te ’ga’ zhig der gyur na skad cig ma ma yin pa des kyang rim dang cig car dag gis don byed
par mi nus so zhes spyir brjod par bya’o // des na srid par byas pa’i ma mthong ba de yang gnod par
byed pa’i tshad ma’i yul nyid du khas blang bar bya ba nyid do //a de lta yin pa dang mthong ba dang
ma mthong ba’i dngos po mtha’ dag la (dngos po mtha’ dag la KBhSF : dngos po mtha’ dag KBhSP )
khyab pa sgrub pa’i tshad ma bstan par rnam par gnas pa na (rnam par gnas pa na KBhSF : rnam par
gnas pa dang KBhSP ) / gang dag bsgrub par bya ba’i chos can las (bsgrub par bya ba’i chos can las
KBhSF : bsgrub sgrub par bya ba’i chos can las KBhSP ) tha dad pa’i dngos po de dag dpe’i dngos la
brten pa’i phyir bsgrub par bya ba’i chos can las tha dad pa gang dag la (gang dag la KBhSF : gang la
KBhSP ) gtan tshigs yod pa de dag ni dpe yin te / bsgrub par bya ba’i dngos po nges pa ni (nges pa
ni KBhSF : nges pa na KBhSP ) khyab pa sgrub pa’i tshad ma la rag las pa’i phyir ro // “a [Answer:]
It is not possible to show the source of knowledge proving pervasion only in the pakṣa alone. This is
because the source of knowledge demonstrating pervasion should be shown in the following manner
– ‘If a sound [i.e., the pakṣa] were permanent, it would be employed in fulfilling a purpose neither
in a gradual nor in a simultaneous manner.’ [A proponent says] so. Against this, [his] opponent in
turn can retort – ‘Just as a pot, though being non-momentary, can bring about an expected effect in a
gradual or simultaneous manner, so too must a sound be.’ Against this too, the reply should be stated
[as follows] – ‘A pot too is by all means brought into the scope of the defeating source of knowledge by
me [i.e., the proponent].’ Thus, every visible thing proposed (*upanyasta) by the opponent through
such an enumeration is brought into the scope of the defeating source of knowledge by the disputant
[i.e., the proponent]. If the opponent doubts even an invisible [thing], saying that, ‘A certain thing,
being non-momentary, can fulfill a purpose in a gradual or simultaneous manner, and a sound is so too,’
[then], against this too, it should be generally stated – ‘If there were to exist a certain [thing, and if it
were] not momentary, it also would be incapable of fulfilling a purpose, whether in a gradual or in a
simultaneous manner.’ Therefore, this invisible [thing] too, which has been hypothetically assumed [by
the opponent], is necessarily provisionally accepted as the object of the defeating source of knowledge.a
And when in this manner it is established that the source of knowledge demonstrating pervasion is
Masamichi Sakai
383
opponent might bring other existing things as ‘counterexamples’ which are thought to be
non-momentary but capable of purpose fulfillment. Theoretically speaking, for any existent
things including even invisible things (adṛśya), it is not until the viparyaye bādhakapramāṇa functions that their momentariness is proved, since only it is capable of proving
momentariness. In this sense, the opponent has the right to present counterexamples to
which the viparyaye bādhakapramāṇa has not yet been applied, so as to invalidate the
pakṣa’s momentariness. Dharmottara insists that, in order to rebut such counterexamples,
the proponent must show the viparyaye bādhakapramāṇa in counterexamples. This activity
amounts to the fact that the opponent needs to hold T2, in that such counterexamples will
be ontologically different things from the pakṣa. In this regard, it is not the case that the
trairūpya condition in its totality is useless. Rather, it is an indispensable condition for a
sound inference of the pakṣa’s momentariness. In this way, Dharmottara, although basically
inheriting Arcaṭa’s view, asserts the significance of T2.
2.1.1 Omni-applicability of the viparyaye bādhakapramāṇa
In theory, the opponent can pose an infinite number of counterexamples including even
invisible things. Therefore, the proponent must apply the viparyaye bādhakapramāṇa
to each and every case according to the situation. In other words, the viparyaye bādhakapramāṇa must be such a source of knowledge as should and in fact can be applied
to all counterexamples brought forth by the opponent. In this regard, Dharmottara calls
Dharmakīrti’s viparyaye bādhakapramāṇa “that whose scope is a collection of all things”
(sarvapadārthopasaṃhāraviṣayam).
And for this very reason, for the Teacher [i.e., Dharmakīrti], that which demonstrates the relation between that which is to be pervaded and that which pervades [it, i.e., the viparyaye bādhakapramāṇa,] is intended to have the scope of
a collection of all things, as [he] has said [in the PVin 2 76,3–4]: “whatsoever
is produced is impermanent.” And he does not accept the pervasion which is
subject to distinctions among the pakṣa and the like. Therefore, the logical
mark, whose scope is in fact all things, is ascertained to be pervaded by its
own target [property], which is proved by the defeating source of knowledge.20
As the reason for the viparyaye bādhakapramāṇa’s omni-applicability, Dharmottara refers
to an essential feature of the hetu “existence” (sattva), which is a part of the viparyaye
20
shown in [both] all visible and invisible things – since [in this case] those things that are different from
the pakṣa belong to a subset of example objects – [all] those things which are different from the pakṣa
and in which the hetu exists are [all] examples. This is because the ascertainment of [whether] the target
[property] exists (*bhāva) [or does not exist] depends on the source of knowledge that demonstrates
pervasion.” For Frauwallner’s German translation, cf. Frauwallner 1935: 242.
PVinṬ 2 Ms 99b1–100a1: ata eva ca sarvapadārthopasaṃhāraviṣayaṃ vyāpyavyāpakabhāvasādhakam
iṣṭam ācāryasya, yad āha – yat kiñcit kṛtakam, tat sarvam anityam iti (cf. PVin 2 76,3–4: yat kiñcit
kṛtakaṃ tat sarvam anityam…). pakṣādipravibhāgāpekṣāṃ ca vyāptiṃ necchati. tasmāt sarvapadārthaviṣayam eva liṅgaṃ bādhakena pramāṇena sādhitasvasādhyavyāptikam avasātavyam. For Tani’s
Japanese translation from the Tibetan version, see Tani 2000: 172.
384
Dharmottara on viparyaye bādhakapramāṇa and trairūpya
bādhakapramāṇa: namely, that it pertains to all things.21 Thus, the pervasion ascertained
by the viparyaye bādhakapramāṇa functioning with this special property is not subject
to distinctions between pakṣa and non-pakṣa. In short, the viparyaye bādhakapramāṇa
proves that all things are momentary.
3 Problem of the first trairūpya condition
However, the viparyaye bādhakapramāṇa’s omni-applicability prompts a further crucial
problem for the advocate of trairūpya, namely that T1 (= pakṣadharmatā) is useless. Given
that the viparyaye bādhakapramāṇa proves that all things are momentary, then, since the
pakṣa is already included in the sphere of everything, its momentariness is proved solely by
the viparyaye bādhakapramāṇa. Thus, it is useless to particularly point out, in an inference,
the pakṣa’s possession of the hetu “existence.” Arcaṭa does not discuss any problem of this
kind in his HBṬ.22
The relevant objection in the PVinṬ reads:
[Objection:] But, given that the pervasion is shown [by the defeating source
of knowledge] as encompassing all things, a sound [as the pakṣa] too, which
is included in all [things], is proved to be momentary. Thus, there is no benefit
to including a property of the pakṣa [in the arguments].23
21
22
23
In the PVinṬ, Dharmottara differentiates the logical scope of the inferential reason property “being
existent” (sattva) from that of the inferential reason property “being produced” (kṛtakatva). For the
Buddhists, their ontological scopes are the same, but it might be the case that there are some opponents
for whom this is not the case. According to Dharmottara’s explanation, Dharmakīrti is taking such a
possibility into account when he introduces this hetu. The scope of the former includes something that
has a cause, i.e., is produced, but which is beyond the category of the five aggregates (pañcaskandha),
i.e., invisible, incognizable. Cf. Sakai 2010: 143–144.
Cf. Sakai 2015: 291–292 (section 4.2.). Durvekamiśra elucidates where Arcaṭa sees the significance
of T1 in the sattvānumāna. He reports (HBṬĀ 261,23–262,2) that Arcaṭa’s way of recognizing the
necessity of T1 is rejected by Dharmottara and the like who have a different view on the same issue.
According to Durvekamiśra, Arcaṭa requires T1, so that one can indicate the object to which the
viparyaye bādhakapramāṇa is applied in a relevant inference. On the other hand, Dharmottara’s reason
for needing T1 is so that one can ascertain the pakṣa’s being real (tattva). For this interesting controversy,
cf. Sakai 2019.
PVinṬ 2 Ms 100a1–2: nanu ca sarvapadārthopasaṃhāravati vyāptipradarśane śabdo ’pi sarvāntarbhūtaḥ kṣaṇikaḥ siddha iti na pakṣadharmopasaṃhārasya phalam. For this part we have a parallel
argument in the KBhS. Cf. KBhSF 224,18–22: gal te dngos po mtha’ dag gi yul la (yul la KBhSF : yul
can la KBhSP ) khyab pa sgrub pa’i tshad ma srid na / ’o na ni des na ji ltar dpe dag la bsgrub bya nges
pa khyab pa sgrub pa’i tshad ma dang ’brel pa de bzhin du bsgrub bya’i chos can la yang rjes su ’gro
ldog rab tu grub pa dag las mngon par ’dod pa grub pas phyogs kyi chos nye bar bstan pas ci bya zhe
na / “[Objection:] If the source of knowledge demonstrating pervasion is applicable to the sphere of all
things, then, on the basis of this [fact], with regard to examples, ascertainment of the target [property] is
subject to the source of knowledge demonstrating pervasion. In the same manner, with regard to the site
too, the desired [property, i.e., momentariness] is proved due to the positive and negative concomitances
that are in fact established [by the source of knowledge that demonstrates pervasion]. Thus, what can be
gained by pointing out a property of the site?” For Frauwallner’s German translation, cf. Frauwallner
1935: 242–243.
Masamichi Sakai
385
Dharmottara rebuts this by strictly distinguishing the ascertainment of pervasion from that
of whether or not the relevant pakṣa actually exists, i.e., in reality. That is, the ascertainment
of pervasion by the viparyaye bādhakapramāṇa that everything, including the pakṣa, is
momentary belongs to a theoretical level; on the other hand, the ascertainment of whether
the pakṣa actually exists belongs to an actual, epistemological level. He says:
[Dharmottara:] Well then, is it the case that a hare’s horn is known to be
existent to a person who has already ascertained that “Whatever is existent is
momentary?” [Objector:] Since a hare’s horn does not exist, it is not known
[to him]. But a sound does exist. [Dharmottara:] Then, this [i.e., the existence
of a sound] should be known by means of another source of knowledge,
[since], if this is known, it can be understood that a sound is not like a hare’s
horn. Therefore, though a sound is generally [i.e., without its individuality,]
included in [all things] when pervasion is [demonstrated by the defeating
source of knowledge], its [i.e., a sound’s] existence should be individually
known by means of another source of knowledge in order to ascertain [its
own] momentariness.24
The ascertainment of pervasion and that of the pakṣa’s actual existence are totally different
and independent issues. Knowledge of the former does not imply the latter. Otherwise, once
pervasion is demonstrated, it would be proved that not only a sound, but also a hare’s horn is
momentary. But this is not the case, since a hare’s horn does not exist in reality. According
to Dharmottara, the pakṣa’s actual existence must be ascertained by another source of
24
PVinṬ 2 Ms 100a2–4: tat kim idānīṃ yat sat, tat sarvaṃ kṣaṇikam iti niścitavato ’pi puṃsaḥ śaśaviṣāṇaṃ
sad iti jñātam. śaśaviṣāṇasyāsattvād ajñātam. śabdasya tu sattvam. tat tarhi pramāṇāntarāj jñātavyam,
yasmiṃ jñāte na śaśaviṣāṇavac chabda ity avagamyeta. tasmāt sāmānyenāntarbhūtasyāpi śabdasya
vyāptikāle kṣaṇikatvaniścayārthaṃ viśeṣeṇa sattvaṃ pramāṇāntareṇa jñātavyam. For this part too we
have a parallel argument in the KBhS. Cf. KBhSF 224,22–225,2: bden mod kyi ’on kyang dngos po’i
tshogs ni mtha’ yas la khyab pa bsgrub pa (khyab pa bsgrub pa KBhSF : khyab pa sgrub pa KBhSP )
yang dngos po mtha’ dag gi yul can yin pa des na sgra med kyang bsgrub bya sgrub byed dag gi (sgrub
byed dag gi KBhSF : sgrub byed dag gis KBhSP ) khyab pa ni dngos po gzhan nyid las rab tu grub pa
kho na’o (rab tu grub pa kho na’o KBhSF : rab tu grub tu grub pa kho na’o KBhSP ) // des na ji ltar ri
bong gi rwa med par khyab pa grub pa de bzhin du sgra med par rab tu ’grub pa nyid do // de las khyab
pa mi ’grub pa’i phyir na sgra’i yod pa ni ji ltar yang phan ’dogs par byed pa nyid ma yin no // de lta
yin dang khyab pa ’grub pa la khas ma blangs pa’i byed pa can sgra’i yod pa nyid ni tshad ma gzhan
kho na las rtogs par bya ba yin no // des na nges pa tshad ma gzhan la rag las pa sgra’i yod pa nyid ni
sgra mi rtag par bsgrub par nus pa’i phyir phyogs kyi chos nye bar bsdu bar bya ba (nye bar bsdu bar
bya ba KBhSF : nye bar bsu bar bya ba KBhSP ) kho na’o // “[Answer:] This is right. But, a collection
of things is infinite, and [the source of knowledge] demonstrating pervasion can be applied to all things
(*sakalapadārthaviṣayam). Therefore, even if a sound were not to exist, pervasion between the target
property and the reason property would surely be realized on the basis of other things [other than a
sound]. Thus, just as pervasion is established without the existence of a hare’s horn, so [pervasion] is
surely established without the existence of a sound. Since pervasion is not established on the basis of
that [i.e., a sound], the existence of a sound in no way brings benefits [to the establishment of pervasion].
And it is on the basis of another source of knowledge that the existence of a sound, whose functioning
is not accepted in this way when pervasion is proved, has to be understood. Thus, since the existence
of a sound, whose ascertainment requires another source of knowledge, is in a position to prove the
impermanence of that sound, a property of the site must by all means be included [in the arguments of
a proof].” For Frauwallner’s German translation, cf. Frauwallner 1935: 243.
386
Dharmottara on viparyaye bādhakapramāṇa and trairūpya
knowledge other than the viparyaye bādhakapramāṇa. Showing T1 in an inference amounts
to the fact that the actual existence of the pakṣa is epistemologically ascertained by another
source of the knowledge.25 In this way, Dharmottara insists on the indispensability of T1.
4 Concluding remarks
In short, while absolutely relying on the logical universality of the viparyaye bādhakapramāṇa, Dharmottara strives not for a formal, but rather for a substantial raison d’être of
trairūpya. Furthermore, this reliance is well founded, given that he is basing himself on
Dharmakīrti’s position as expressed in his Hetubindu and Vādanyāya.26 For the discussions
of the kṣaṇikatvānumāna in these works, Dharmakīrti depends exclusively on the viparyaye
bādhakapramāṇa. At the same time, however, his formulation includes T1 and T2.27 It
appears that under this logically ambiguous circumstance, Arcaṭa steers his course of
interpretation toward stressing the viparyaye bādhakapramāṇa, judging Dharmakīrti’s
retention of the trairūpya to be a mere formality, and Dharmottara, in turn, makes efforts
toward harmonizing the viparyaye bādhakapramāṇa with trairūpya.
To the best of my knowledge, in the larger interpretative history of the sattvānumāna,
Dharmottara’s argument examined above is the first systematic defense of the significance of
trairūpya taken as a whole, which well represents the Buddhist logical standpoint of being
neither antarvyāpti nor bahirvyāpti.28 What’s more, this seems to remain the only way
for traditional Buddhists to maintain their argumentative identity until the innovation by
Jñānaśrīmitra and Ratnakīrti of proving the momentariness of a dṛṣṭānta via the combination
of prasaṅga and prasaṅgaviparyaya,29 itself a strategy that avoids exclusive reliance on
the viparyaye bādhakapramāṇa.30
25
26
27
28
29
30
Namely, perception (pratyakṣa). In the previous part of the PVinṬ (= PVinṬ 2 Ms 98b1–99a3) Dharmottara already discussed how and by which source of knowledge the inferential reason property “existence”
is established (siddha). There is also a parallel argument in the KBhS (= KBhSF 221,7–16 [pūrvapakṣa],
231,7–232,13 [uttarapakṣa]). For a detail of the arguments, see Sakai 2013, where I have translated
these texts and analyzed their contents.
For now I am uncertain whether Dharmakīrti, in PVin 2, absolutely relies on the viparyaye bādhakapramāṇa for establishing the pervasion of the reason property “being produced” (kṛtakatva) by the
target property “being impermanent” (anityatva), though in this work he introduces the viparyaye
bādhakapramāṇa in its complete form. This is because Dharmakīrti also introduces other arguments
for establishing pervasion other than the viparyaye bādhakapramāṇa. Cf. Sakai 2014.
Cf. HB 6,9-10: yat sat, tat sarvaṃ kṣaṇikam, yathā ghaṭādayaḥ, saṃś ca śabda iti…, VN 1,13–14: yat
sat kṛtakaṃ vā, tat sarvaṃ anityam, yathā ghaṭādiḥ, san kṛtako vā śabda iti.
Cf. Ono 2004: 481–488.
Cf. Mimaki 1984: 239.
See, for example, Vācaspatimiśra, one of the most renowned representatives of the Brahmanical side,
who probably flourished before the time of Jñānaśrīmitra (for the date of Vācaspatimiśra, cf. Acharya
2006: xviii–xxii. Dr. Yasutaka Muroya recently suggested a new chronological relationship between
Vācaspatimiśra and Jñānaśrīmitra in Muroya 2011). In his Nyāyakaṇikā and Nyāyavārttikatātparyaṭīkā,
he summarizes and then introduces as pūrvapakṣa Dharmottara’s argument defending the Buddhist
trairūpya theory considered as a whole in the case of the sattvānumāna. Cf. NKaṇ 509,7–511,9 (=NVTṬ
515,3–13): a nākṣaṇikasyārthakriyāvirodhād (nākṣaṇika° conjecture; na kṣaṇika° NKaṇ; na ca kṣaṇika° NVTṬ) asādhāraṇatā hetoḥ. na ca sādhyadharmiṇi dṛśyamāne śabdādau vyāptiprasādhanād
(°prasādhanād NKaṇ: °sādhanād NVTṬ) eva sādhyasiddher asādhanāṅgaṃ hetuvacanam. na khalu
Masamichi Sakai
387
References and abbreviations
Acharya 2006 D. Acharya, Vācaspatimiśra’s Tattvasamīkṣā. The Earliest Commentary
on Maṇḍanamiśra’s Brahmasiddhi. Stuttgart 2006.
Frauwallner 1935 E. Frauwallner, Dharmottaras Kṣaṇabhaṅgasiddhiḥ: Text und Übersetzung. Wiener Zeitschrift für die Kunde des Morgenlandes 42 (1935) 217–258.
HB Hetubindu (Dharmakīrti), ed. E. Steinkellner. Beijing/Wien 2016.
HBṬ Hetubinduṭīkā (Arcaṭa), ed. S. Sanghavi and Muni Śrī Jinavijayaji. Baroda 1949.
HBṬĀ Hetubinduṭīkāloka (Durvekamiśra): see HBṬ.
KBhSF Kṣaṇabhaṅgasiddhi (Dharmottara) (Tibetan), ed. E. Frauwallner: In: Frauwallner
1935.
KBhSP Kṣaṇabhaṅgasiddhi (Dharmottara) (Tibetan): Peking No. 5751.
sarvopasaṃhāravatī vyāptir dṛśyamātraviṣayā bhavitum arhati. śakyaṃ hi śaṅkituṃ pareṇādṛśyamānānāṃ sattvam akṣaṇikān na vyāvartitaṃ tvayeti sattvam anaikāntikaṃ kṣaṇikatvasādhana iti.
tasmād yad dṛśyam adṛśyaṃ vā, tat sarvaṃ kṣaṇikam iti darśanīyā vyāptiḥ.a b nanv evam api śabdāder
vivādāspadībhūtasya vyāptidarśanabalād eva siddhā kṣaṇikateti tadavastham evāsādhanāṅgatvaṃ
hetuvacanasya. maivam. asaty api śaśaviṣāṇādau yat sad dṛśyam adṛśyaṃ vā, tat sarvaṃ kṣaṇikam iti,
yathā sarvopasaṃhāravatī vyāptiḥ sidhyati (vyāptiḥ sidhyati NKaṇ: vyāptiḥ NVTṬ). na ca śaśaviṣāṇādayo ’pi bhavanti kṣaṇikāḥ. evaṃ saty api vivādāspadīkṛte (°kṛte NKaṇ: °bhūte NVTṬ) śabdādau
vyāptisiddhāv api na sidhyati kṣaṇikatety avaśyaṃ darśayitavyam eva teṣāṃ (darśayitavyam eva teṣāṃ
NKaṇ: darśayitavyam eteṣāṃ NVTṬ) kṣaṇikatvasādhanāya sattvam iti nāsādhanāṅgatā hetuvacanasya
(°vacanasya NKaṇ: °vacanasyeti NVTṬ).b “a It is not the case that, since there is the inconsistency with
purpose fulfilment for a non-momentary [thing], the inferential reason property [of being existent] is
unique/specific. Nor is it the case that, since the target [property, i.e., momentariness] is proved solely
on the basis of the fact that the pervasion is well established with regard to the pakṣa, i.e., a sound and
the like, which is seen, the statement of the inferential reason property [i.e., the statement of T1] is not a
constituent of the proof. It is indeed impossible for the pervasion that encompasses all [things] to have
as its object seen [things] only. This is because the opponent has the right to doubt [this] in the following
manner: ‘By you [i.e., the proponent] the property of being existent belonging to unseen [things] is not
excluded from a non-momentary [thing]. Thus, the [inferential reason] property of being existent is
indeterminate in the proof of momentariness.’ Therefore, [in order to rebut such doubt,] the pervasion
should be shown in the following way: ‘Whatever a thing is, i.e., whether seen or unseen, it is exclusively
momentary.’ a b [If one says:] even if this is so, the statement of the inferential reason property [i.e., the
statement of T1] is not a constituent of the proof, [something that] is the case when a sound and the like
about which there is a dispute is proved to be momentary only by virtue of showing the pervasion, [then,
we answer:] it should not be like that [for the logical reason sattva]. The pervasion that encompasses all
[things] is established in such a way that, even if a hare’s horn and the like does not exist, that which
exists, whether seen or unseen, is all/exclusively momentary. And it is not the case that a hare’s horn and
the like too are momentary. In this way, even if a sound and the like which becomes the subject of dispute
exists, [its] momentariness is not proved, even when the pervasion is established. Therefore, in order to
prove their momentariness [i.e., that of a sound and the like], their existence must necessarily be shown
in full. Thus, it is not the case that the statement of inferential reason property [i.e., the statement of
T1] is not a constituent of the proof.b ” The part a–a, in terms of content, corresponds to Dharmottara’s
defense of T2 examined above (cf. section 2.1.), the part b–b to the discussion examined in section
3. The part b–b has a parallel in the Tattvasamīkṣā by the same author. Cf. TSam 36,29–35: na ca
sarvopasaṃhāravadvyāptisādhanād eva pramāṇāt sarvamadhyapātināṃ vivādāspadānāṃ śabdādīnām
api kṣaṇikatvasiddhe(r asādhanāṅgaṃ hetuvacanam iti) sāṃpratam. yathaiva hi gagananalinādīnām
asatām asaty api kṣaṇikatve na sarvopasaṃhāravatīṃ vyāptiṃ prasādhyataḥ pramāṇasya vi[ro](dhaḥ,
tathaiva vivādāspadeṣu śabdādiṣu kas tasya virodhaḥ, sattvasiddhau tu syād iti tatsādhanāya hetur
upanetavyo vacanena, saṃś ca vivādāspadībhūtaḥ śabdādi)[r i](ti svabhāvahetuḥ).
388
Dharmottara on viparyaye bādhakapramāṇa and trairūpya
Mimaki 1976 K. Mimaki, La réfutation bouddhique de la permanence des choses (Sthirasiddhidūṣana) et la preuve de la momentanéité des choses (Kṣaṇabhaṅgasiddhi).
Paris 1976.
Mimaki 1984 K. Mimaki, Proof of Momentariness. Kōza Daijōbukkyō 9 (1984) 219–254
(in Japanese).
Muroya 2011 Y. Muroya, Bhāsarvajña’s interpretation of bhāva eva nāśaḥ and a related
chronological problem. In: Religion and Logic in Buddhist Philosophical Analysis.
Proceedings of the Fourth International Dharmakīrti Conference. Vienna, August
23–27, 2005, ed. H. Krasser, H. Lasic, E. Franco, and B. Kellner. Vienna 2011,
341–365.
NBṬ Nyāyabinduṭīkā (Dharmottara), ed. D. Malvania. Patna 1955.
NKaṇ Nyāyakaṇikā (Vācaspatimiśra): E. Stern, “Vidhivivekaḥ” of Maṇḍanamiśra with
Commentary, “Nyāyakaṇikā,” of Vācaspatimiśra, and Supercommentaries, “Juṣadhvaṅkaraṇī ” and “Svaditaṅkaraṇī ,” of Parameśvaraḥ, Critical and Annotated
Edition: the Pūrvapakṣaḥ. Dissertation, University of Pennsylvania. Pennsylvania
1988.
NVTṬ Nyāyavārttikatātparyaṭīkā (Vācaspatimiśra), ed. A. Thakur. New Delhi 1996.
Ono 2004 M. Ono, “Antarvyāpti” and “bahirvyāpti” in the Buddhist logic. Indotetsugaku
Bukkyōshisō Ronshū (2004) 457–492 (in Japanese).
Ono 2010 M. Ono, On Viruddhavyabhicārin. Indo Ronrigaku Kenkyū 1 (2010) 125–143
(in Japanese).
PVin 2 Pramāṇaviniścaya (Dharmakīrti) Chapter 2, ed. E. Steinkellner. Beijing/Vienna
2007.
PVin 3 Pramāṇaviniścaya (Dharmakīrti) Chapter 3, ed. P. Hugon and T. Tomabechi. Beijing/Vienna 2011.
PVinṬ 2 Ms Sanskrit Manuscript of the second chapter of the Pramāṇaviniścayaṭīkā of
Dharmottara: see PVin 2: Introduction xxx–xxxi.
PVinṬS Pramāṇaviniścayaṭīkā (Dharmottara), ed. M. Sakai: In: Sakai 2010.
PVinṬt Pramāṇaviniścayaṭīkā (Dharmottara) (Tibetan): Peking No. 5727, Derge No. 4227.
Sakai 2010 M. Sakai, Dharmottaras Erklärung von Dharmakīrtis kṣaṇikatvānumāna:
Pramāṇaviniścayaṭīkā zu Pramāṇaviniścaya 2 vv. 53–55 mit Prosa. Dissertation,
University of Vienna. Vienna 2010. http://othes.univie.ac.at/9623/, last visited
27-09-2020.
Sakai 2013 M. Sakai, A Study of Dharmottara’s Theory of Momentariness: The Problem
of the Inferential Reason – Existence (sattva) – in the Sattvānumāna. Indotetsugaku
Bukkyōgaku Kenkyū 20 (2013) 77–93 (in Japanese).
Sakai 2014 M. Sakai, Dharmakīrti’s Proof that “All That Is Produced Is Impermanent:”
Dharmakīrti’s Second Answer in the Pramāṇaviniścaya. Indogaku Bukkyōgaku
Kenkyū 62(2) (2014) 978–971 (in Japanese).
Sakai 2015 M. Sakai, Arcaṭa on dṛṣṭānta, trairūpya, and viparyaye bādhakapramāṇa in
Dharmakīrti’s sattvānumāna. South Asian Classical Studies 10 (2015) 281–296.
Sakai 2019 M. Sakai, Antarvyāpti and Antarvyāptivāditva: Arcaṭa on the Buddhist Inference of Momentariness and Durvekamiśra’s Comparison of him with Dharmottara.
Journal of Indological Studies 30 & 31 (2019) 99–128.
Masamichi Sakai
389
Steinkellner 1967 E. Steinkellner, Dharmakīrti’s Hetubinduḥ: Teil II. Übersetzung und
Anmerkungen. Wien 1967.
Steinkellner 2004 E. Steinkellner, The Early Dharmakīrti on the Purpose of Examples.
In: The Role of the Example (dṛṣṭānta) in Classical Indian Logic, ed. Sh. Katsura
and E. Steinkellner. Wien 2004, 225–250.
Tani 1997 T. Tani, Problems of Interpretation on Dharmottara’s Kṣaṇabhaṅgasiddhi
(1)–(3). Kōchikōgyōkōtōsenmongakkō Gakujyutsukiyō 41 (1997) 19–37(1); 39–57(2);
59–77(3) (in Japanese).
Tani 2000 T. Tani, Study of Momentariness. Tokyo 2000 (in Japanese).
TBh Tarkabhāṣā (Mokṣākaragupta), ed. H.R. Rangaswami Iyengar. Mysore 1944.
TSam Tattvasamīkṣā (Vācaspatimiśra), ed. D. Acharya. Stuttgart 2006.
VN Vādanyāya (Dharmakīrti), ed. M. T. Much. Wien 1991.
Dharmakīrti’s Vādanyāya and the History of Conceptions of
Debate in Indian Logic
by
Ryo Sasaki
1. Introduction
Throughout the history of Indian thought in the classical period, the method of debate
has been a subject of investigation side by side with philosophical ideas. A conception of
“debate” can already be found at an early stage, notably in the medical treatise Carakasaṃhitā (CaS). After passing through the Fangbianxinlun (*Upāyahṛdaya, UH, 方便心
論),1 an early Buddhist debate treatise, the conception of debate was more systematically
approached in the Nyāyasūtras (NS), and further expanded and elaborated in Vātsyāyana’s
Nyāyabhāṣya (NBh) and Uddyotakara’s Nyāyavārttika (NV).
In the Vādanyāya (VN),2 Dharmakīrti redefined a traditional Nyāya concept related
to debate, namely the idea of a “condition of defeat” (nigrahasthāna),3 a criterion that
determines victory or defeat in a debate. In the Vādanyāya’s redefinition of nigrahasthāna,
Dharmakīrti presented a new conception of debate, i.e., the “debate of well-educated
people” (satāṃ vādaḥ). By doing so, he appears to have reacted to traditional ideas found in
the Nyāyasūtras and Nyāyabhāṣya. His ideas in turn influenced the presentation of debate
in later Nyāya works, as will be discussed below with a focus on the Nyāyabhūṣaṇa (NBhū)
of Bhāsarvajña and the Nyāyavārttikatātparyaṭīkā (NVTṬ) of Vācaspati Miśra.
Prets 2000, Kang 2003, Preisendanz 2009 and others have analyzed the ancient Indian
exposition of debate in the Carakasaṃhitā. Pertinent ideas in the *Upāyahṛdaya were
investigated by several Japanese researchers (Ui 1925, Kajiyama 1984, Ishitobi 2006). The
more systematic exposition of debate in the Nyāyasūtras and Nyāyabhāṣya was studied
in detail especially by Matilal 1998, Preisendanz 2000, and Nicholson 2010. Steinkellner
1988 pointed out Dharmakīrti’s new conception of debate. Much’s translation and critical
1
2
3
G. Tucci suggests Upāyahṛdaya as the original Sanskrit title of Fangbianxinlun, while E. Frauwallner
suggested Prayogasāra. At the 17th Congress of the International Association of Buddhist Studies
(IABS) in 2014, Prof. Shōryū Katsura proposed a third possibility: Prayogahṛdaya.
See VN 1,4–5: asādhanāṅgavacanam adoṣodbhāvanaṃ dvayoḥ / nigrahasthānam, anyat tu na yuktam
iti neṣyate //1// “Asādhanāṅgavacana and adoṣodbhāvana are the conditions of defeat for the two
(debaters, i.e. a proponent and an opponent, respectively). However, other [conditions of defeat that the
Nyāya school and the like explain] are not correct, hence [they are] not accepted.” Here Dharmakīrti
presents his original idea of dividing “the condition of defeat” into asādhanāṅgavacana (the condition
of defeat for a proponent) and adoṣodbhāvana (the condition of defeat for an opponent). This compels
us then to conclude that a proponent is judged to be defeated when his behavior corresponds to
asādhanāṅgavacana and an opponent is judged to be defeated when his behavior corresponds to
adoṣodbhāvana, according to the terms of debate set up in the VN. However, based on the descriptions
supplied in the VN, this conclusion must in fact be wrong. See section 5 for details.
As for nigrahasthāna as presented in the Nyāyasūtras, see Vidyabhusana 1921: 84–90. With regard to
nigrahasthāna as defined in the Vādanyāya, see Much 1986 and 1991; Chinchore 1988; Gokhale 1993;
Sasaki 2012a, 2012b, 2013a, 2013b, 2014a, and 2014b.
Birgit Kellner et al., eds., Reverberations of Dharmakīrti’s Philosophy: Proceedings of the Fifth International Dharmakīrti
Conference Heidelberg, August 26 to 30, 2014. Vienna 2020, pp. 391–410.
392
Dharmakīrti’s Vādanyāya and the History of Conceptions of Debate
edition of the Vādanyāya in 1991, together with his comprehensive presentation of results
of the Vādanyāya research up until that point, mark a significant contribution to research
on this work. Ono 2004 and 2006 examined Udayana’s views on debate, and Chinchore
1990 studied reactions to the Vādanyāya by Naiyāyikas active after Udayana.
While several studies have thus been conducted on Indian debate, little attention has
been given to how the Vādanyāya marks a historical transition in the conception of debate.
This paper therefore aims to place the Vādanyāya in the context of expositions of debate in
the above-mentioned texts, and to thereby clarify its significance.
2. Jalpa in the Carakasaṃhitā
As the first step in our analysis, we will examine the idea of debate in the Carakasaṃhitā
because it is one of the earliest works in which the concept of debate is introduced in detail,
even though it was not exactly explained in a systematic manner. In the Carakasaṃhitā,
the parts of the eighth chapter of the Vimānasthāna dealing with debate can be divided
into two sections: the section on colloquy (saṃbhāṣā) in CaS 8.8.15–26 and the section on
disputation (vāda) in CaS 8.8.27–66.4 In the latter section, vāda is defined as follows:
CaS 8.8.28: tatra vādo nāma sa yat pareṇa saha śāstrapūrvakaṃ vigṛhya kathayati. sa ca dvividhaḥ saṃgraheṇa – jalpaḥ, vitaṇḍā ca. tatra pakṣāśritayor
vacanaṃ jalpaḥ, jalpaviparyayo vitaṇḍā. yathā – ekasya pakṣaḥ punarbhavo
’stīti, nāstīty aparasya; tau ca svasvapakṣahetubhiḥ svasvapakṣaṃ sthāpayataḥ, parapakṣam udbhāvayataḥ, eṣa jalpaḥ. jalpaviparyayo vitaṇḍā. vitaṇḍā
nāma parapakṣe doṣavacanamātram eva.
Of these [44 technical terms], disputation (vāda) is [debate] in which one
discusses with another in a hostile manner (vigṛhya), based on scriptures. In
brief, this [disputation] is of two kinds: wrangle (jalpa) and cavil (vitaṇḍā).
Of these, wrangle is the statement of two [disputants] who depend on [their
respective] positions. Cavil is the opposite of wrangle. For instance, one
holds the position that rebirth exists [while] the other holds [the position] that
[rebirth] does not exist. In addition, both [disputants] establish their positions
on the basis of logical reasons for their positions [and] point out [the fault in]
the other’s position. This is wrangle. Cavil is the opposite of wrangle. Cavil is
merely “indicating the fault in the other’s position.”
Disputation (vāda) is here defined as a “[debate] in which one discusses with another in a
hostile manner (vigṛhya).” There are two types of vāda: wrangle (jalpa) and cavil (vitaṇḍā).
As discussed below in section 4, this classification differs from the one in the Nyāyasūtras.
Let us note that jalpa here involves hostility to the other disputant, and that it consists of
individual assertions of a proponent and opponent, and mutual ripostes between them.
While the vāda section of the Carakasaṃhitā provides a brief account of vāda in which
jalpa occurs as a subcategory of vāda, the concept of jalpa, the action of √jalp, as well as
4
See Preisendanz 2009: 266–268.
Ryo Sasaki
393
the idea of a hostile (vigṛhya) verbal confrontation are already introduced in the saṃbhāṣā
section.5
CaS 8.8.15: saṃbhāṣāvidhim ata ūrdhvaṃ vyākhyāsyāmaḥ – bhiṣag bhiṣajā
saha saṃbhāṣeta. tadvidyasaṃbhāṣā hi jñānābhiyogasaṃharṣakarī bhavati,
vaiśāradyam api cābhinirvartayati, vacanaśaktim api cādhatte, yaśaś cābhidīpayati, … yac cācāryaḥ śiṣyāya śuśrūṣave prasannaḥ krameṇopadiśati
guhyābhimatam arthajātaṃ tat paraspareṇa saha jalpan piṇḍena vijigīṣur
āha saṃharṣāt, tasmāt tadvidyasaṃbhāṣām abhipraśaṃsanti kuśalāḥ.
CaS 8.8.16: dvividhā tu khalu tadvidyasaṃbhāṣā bhavati – saṃdhāyasaṃbhāṣā, vigṛhyasaṃbhāṣā ca.
Hereafter, we shall describe the method of colloquy. A physician should discuss
with a physician. Colloquy with experts (tadvidyasaṃbhāṣā) increases the
pleasure of the application of knowledge, provides dexterity [in debate], gives
skill of speech, illuminates fame (yaśas)…. Besides, the teacher who is pleased
with the disciple desirous of hearing [teachings] teaches things intended to
be kept secret in an orderly manner. [The same teacher] who disputes with
another [disputant] in wrangle (√jalp) excitedly states [the secret] in one breath
in order to gain victory (vijigīṣu). Therefore, the wise highly praise colloquy
with experts.
One should know (khalu) that colloquy with experts (tadvidyasaṃbhāṣā) takes
two forms: friendly colloquy and hostile colloquy.
The saṃbhāṣā section describes the method and purpose of colloquy with experts (tadvidyasaṃbhāṣā) more concretely and vividly than the brief definitions of vāda, or jalpa or
vitaṇḍā in the vāda section. Although the relationship between saṃbhāṣā and jalpa is
problematic, it is likely that vigṛhyasaṃbhāṣā and jalpa are the same or at least very similar
concepts, considering such expressions as vigṛhyasaṃbhāṣāyāṃ jalpet (CaS 8.8.18).6
Here “the person who disputes in wrangle” (jalpat) is considered to be desirous of
victory (vijigīṣu). Hence, we can say that one purpose of wrangle (jalpa) is victory. This
purpose deserves careful attention for two reasons: (i) the Nyāyabhāṣya adopted the same
idea and (ii) the Vādanyāya criticizes the purpose of victory. Both points will be examined
in more detail later.
Fame is another of the purposes listed in connection with saṃbhāṣā. It must be noted
that fame or something akin to fame is not considered to be the purpose worth accomplishing
in the context of debate in the *Upāyahṛdaya, the Nyāyasūtras, several commentaries of
the Nyāyasūtras, and the Vādanyāya. This point will be examined later again.
Regarding the means of debate, acts by a disputant such as ridiculing the opponent are
also allowed in a saṃbhāṣā:
5
6
For jalpa or √jalp, cf. CaS 8.8.15, CaS 8.8.18, CaS 8.8.20, CaS 8.8.21. As Preisendanz 2009: 268
pointed out, the word vāda is also already introduced in the saṃbhāṣā section.
Matilal 1998: 38–41 also points out that Caraka divides the “hostile debate” (vigṛhyasaṃbhāṣā) into
two main types, jalpa and vitaṇḍā. Matilal calls the first the “j-type” hostile debate and the second the
“v-type” hostile debate.
394
Dharmakīrti’s Vādanyāya and the History of Conceptions of Debate
CaS 8.8.20: pariṣat tu khalu dvividhā – jñānavatī , mūḍhapariṣac ca. saiva
dvividhā satī trividhā punar anena kāraṇavibhāgena – suhṛtpariṣat, udāsīnapariṣat, pratiniviṣṭapariṣac ceti. tatra pratiniviṣṭāyāṃ pariṣadi jñānavijñānavacanaprativacanaśaktisaṃpannāyāṃ mūḍhāyāṃ vā na kathaṃcit kenacit
saha jalpo vidhīyate; mūḍhāyāṃ tu suhṛtpariṣady udāsīnāyāṃ vā jñānavijñānavacanaprativacanaśaktīr antareṇāpy adīptayaśasā mahājanavidviṣṭenāpi
saha jalpo vidhīyate. tadvidhena ca saha kathayatāviddhadīrghasūtrasaṃkulair vākyadaṇḍakaiḥ kathayitavyam, atihṛṣṭaṃ muhur muhur upahasatā
paraṃ nirūpayatā ca parṣadam ākāraiḥ, bruvataś cāsya vākyāvakāśo na deyaḥ; kaṣṭaśabdaṃ ca bruvatā vaktavyo nocyate, athavā punar hīnā te pratijñā,
iti. punaś cāhvayamānaḥ7 prativaktavyaḥ – parisaṃvatsaro bhavāñ chikṣasva
tāvat; na tvayā gurur upāsito nūnam, athavā paryāptam etāvat te; sakṛd api
hi parikṣepikaṃ nihataṃ nihatam āhur iti nāsya yogaḥ kartavyaḥ kathaṃcit.
apy evaṃ śreyasā saha vigṛhya vaktavyam ity āhur eke; na tv evaṃ jyāyasā
saha vigrahaṃ praśaṃsanti kuśalāḥ.
There are two types of congregations: the intellectual [congregation] and the
ignorant congregation. Even though [the congregation] takes two forms, the
very same [congregation can be divided into] three kinds through the following classification based on [different] grounds – the friendly congregation,
the neutral congregation and the prejudiced congregation. Among these, the
prejudiced congregation consists of persons endowed with the ability to learn,
understand, speak and reply, and of ignorant ones; those who [should] in no
way be involved in wrangle (jalpa) with anyone else. However, in the friendly
congregation or the neutral [congregation], which consists of ignorant persons,
one [should] work on wrangle with [another who is] devoid of the ability
to learn, understand, speak and reply, does not illumine fame (yaśas) and is
hated by great men. Furthermore, while disputing with such a person, one
should dispute by means of long sentences mingled with distorted and long
holy scriptures and should not give the opportunity to the [opposing] speaker
by ridiculing another (i.e., the opposing speaker) in high spirits again and
again and making gestures in the presence of the congregation. Furthermore,
[one should say,] “while [the opponent] makes a mischievous remark, he does
not say what he should say,” or indeed “Your (i.e., the opponent’s) thesis has
been abandoned.” Again, when [the opponent] challenges [the disputant], [the
disputant] should reply, “You should learn for another year,” “You have not
yet honored [your] preceptor,” or “That’s enough of your [talking]!” If [the
opponent] is condemned to be defeated even once, [people will] say that he
is defeated and, therefore, will lack the ability to concentrate on what he is
saying. Besides, some say that one should talk in a hostile manner (vigṛhya)
with a superior [opponent] in the same way, but [to begin with] the wise does
not recommend the discord with a more excellent [opponent] in this manner.
7
For cāhvayamānaḥ, CaS prints cāhū(hva)yamānaḥ.
Ryo Sasaki
395
Here we should note that a disputant is recommended to use incomprehensible sentences and
ridicule an opponent when engaging in wrangle with a friendly but ignorant congregation,
a neutral but ignorant congregation, or sometimes a superior congregation. Such a hostile
manner of debate is also allowed in wrangle (jalpa) and cavil (vitaṇḍā) defined in the
Nyāyasūtras and in the Nyāyabhāṣya; on the other hand, in the Vādanyāya, acts such as
embarrassing others (parapaṃsana), which are regarded as bad deeds (asadvyavahāra),
are not considered to be an admissible method of debate.
3. The discussion of the composition of a treatise (造論) in the *Upāyahṛdaya
Having discussed the conception of debate in the Carakasaṃhitā, we will now consider
debate in the *Upāyahṛdaya.8 There is no Sanskrit text of the *Upāyahṛdaya, but the extant
Chinese translations permit to conclude that the *Upāyahṛdaya contains remarkable ideas
that can be connected with the Nyāyasūtras.
UH (T1632) 23b14–24: 答曰不然。今造此論不爲勝負利養名聞。但欲顯
示善惡諸相故造此論。世若無論迷惑者衆、則爲世間邪智巧辯所共誑
惑、起不善業、輪迴惡趣、失眞實利。若達論者則自分別善惡空相、
衆魔外道邪見之人、無能惱壞作障礙也。故我爲欲利益衆生造此正
論。又欲令正法流布於世。如爲修治菴婆羅果、而外廣植荊棘之林爲
防果故、今我造論亦復如是、欲護正法不求名聞故。汝前説長諍論者
是事不然。爲護法故、故應造論。
I will answer that [this is] not so. Now, I have not composed this treatise
(造此論) for the purpose of victory, profit, or reputation (勝負利養名聞).
I compose this treatise because I only wish to reveal diverse good and bad
features [of debate]. If the world had no treatise [of debate], there would be
many confused people. Then, [the confused] people would be deceived by the
world’s perverse ideas, and wily rhetoric would give rise to bad deeds, which
would be reborn in an evil world and would lose real benefits. If debate (論) is
understood, [its] good, bad, and useless features are distinguished as a matter
of course. [Then], evildoers, non-Buddhists, and adherents of perverse views
would not be able to harm [people] and obstruct [their nirvāṇa]. Therefore, to
benefit people (利益衆生), I compose this correct treatise.9 Furthermore, I
wish to disseminate the true teaching [of Buddha] (正法) to the world. Just as
in order to cultivate the fruits of mango trees one plants a thicket of thorns
(荊棘之林) widely around them, so now I will compose [this] treatise in the
same way because I wish to protect the true teaching [of the Buddha] and I do
not seek [to enhance my] reputation. You explained earlier that [I am] good
8
9
Kajiyama 1984 assumes that the author of the *Upāyahṛdaya is Nāgārjuna.
The same basic point is argued by Candrakīrti in the Madhyamakāvatāra as follows, MA 6.118 (p. 231):
bstan bcos las dpyad rtsod la chags pa’i phyir // ma mdzad rnam grol phyir ni de nyid bstan // gal te de
nyid rnam par bshad pa na // gzhan gzhung ’jig par ’gyur na nyes pa med // Cf. Uryuzu and Nakazawa
2012: 214. I am indebted to Dr. Shenghai Li for having provided this useful information.
396
Dharmakīrti’s Vādanyāya and the History of Conceptions of Debate
at debate (諍論), [but] this is not true. In order to protect the teaching [of the
Buddha], I must compose [this] treatise.10
Compared with claims made in the Carakasaṃhitā about jalpa or saṃbhāṣā, the way in
which the *Upāyahṛdaya explains the purposes and methods of composing a treatise can
be summed up as follows:
1.
2.
3.
4.
5.
The purpose is not “victory;”
The purpose is not “(self-)profit” or “reputation;”
The purpose is the “benefit of people;”
The purpose is “protection of the true teaching [of the Buddha];”
The method is expressed using the metaphor of “thorns.”
According to the author’s intention, the *Upāyahṛdaya was composed in order to protect
the true teaching of the Buddha. This means of protection is metaphorically explained as
planting a thicket of thorns around mango trees. As will be seen in the following section, a
similar expression appears in the Nyāyasūtras.
4. Jalpa and vitaṇḍā in the Nyāyasūtras and Nyāyabhāṣya
With these considerations in mind, we will now examine the concept of debate in the
Nyāyasūtras and Nyāyabhāṣya in order to trace conceptual changes in the idea of “debate.”
In NS 1.2.1–3, the concept of debate is classified into three categories, namely, vāda, jalpa
and vitaṇḍā, while jalpa and vitaṇḍā are presented as subcategories of vāda in CaS 8.8.28.
On the other hand, NS 4.2.47–51 proposes three types of debate: saṃvāda, jalpa, and
vitaṇḍā. Although this discrepancy in terminology and classification is problematic,11 we
will not discuss this and rather concentrate on jalpa and vitaṇḍā, which are explained with
the help of metaphors as follows:
NS 4.2.50: tattvādhyavasāyasaṃrakṣaṇārthaṃ jalpavitaṇḍe bījaprarohasaṃrakṣaṇārthaṃ kaṇṭakaśākhāvaraṇavat //
Just as thorny branches cover [seeds] for the purpose of protecting seed germination, so wrangle (jalpa) and cavil (vitaṇḍā) [are undertaken] for the purpose
of protecting the ascertainment of truth.12
As pointed out in Preisendanz 2000: 236, Kang 2003: 36–37 and Ishitobi 2003, this sentence
parallels the very beginning of the *Upāyahṛdaya discussed above. It can be assumed that
the Nyāyasūtras adopted the metaphor of “thorns” from the *Upāyahṛdaya, or that both
the Nyāyasūtras and *Upāyahṛdaya drew on another text which is their origin. In the same
context of explaining jalpa and vitaṇḍā, Vātsyāyana commented on the sūtra as follows:
10
11
12
Ishitobi 2006: 42–44 and Eltschinger 2012: 471–472 translate this passage and analyze the motives and
aims of treatise or debate presented in the *Upāyahṛdaya.
Preisendanz 2000 considered this problem in detail.
Cf. NBh 1099,4: anutpannatattvajñānānām aprahīṇadoṣāṇāṃ tadarthaṃ ghaṭamānānām etad iti; NV
1099,8: anutpannatattvajñānenaitat kartavyam iti sūtrārthaḥ.
Ryo Sasaki
397
NS 4.2.51: tābhyāṃ vigṛhyakathanam //
NBh 2000,2–3 (ad NS 4.2.51): vigṛhyeti vijigīṣayā, na tattvabubhutsayeti, tad
etad vidyāpālanārtham, na lābhapūjākhyātyartham iti.
By means of these two, [i.e., the wrangle (jalpa) and cavil (vitaṇḍā),] hostile
(vigṛhya) dispute [is undertaken].
“Hostile” (vigṛhya) is [equivalent to] “by a desire to win” (vijigīṣayā); [however] is [this] not [equivalent to] “by a desire to know truth.” This [hostile
dispute is undertaken] for the purpose of defending expertise (vidyā), [but] is
not [undertaken] for the purpose of profit, honor, or reputation.13
As to the purpose of, respectively, jalpa/vitaṇḍā and the composition of a treatise (造
論), both NBh and UH exclude (self-)profit and endorse the protection of their own doctrine. On the other hand, the presentation in the Nyāyabhāṣya is different from that in the
*Upāyahṛdaya in terms of whether the desire for victory is approved of as a legitimate
purpose.
The main points regarding purpose and method of jalpa and vitaṇḍā according to the
Nyāyabhāṣya can be summed up as follows:
1.
2.
3.
4.
5.
The purpose is “victory;”
The purpose is not “(self-)profit,” “honor” or “reputation;”
The idea of “benefit of people” is not mentioned;
The purpose is “defense of expertise;”
The method is expressed using the metaphor of “thorns.”
5. The concept of debate in the Vādanyāya
As Ruegg 2000: 137–138, n. 41 acutely pointed out, Nāgārjuna, Bhāviveka, Dharmakīrti,
and probably also Vasubandhu, reject jalpa or vitaṇḍā. Here we limit the discussion to Dharmakīrti’s criticism of jalpa and vitaṇḍā. Dharmakīrti does not mention the thorn metaphor
used in the Nyāyasūtras. However, Śāntarakṣita quotes NS 4.2.50 in his commentary on
the Vādanyāya, the Vādanyāyaṭīkā Vipañcitārthā (VA):
VA 70,25–26: yathoktaṃ tattvādhyavasāyasaṃrakṣaṇārthaṃ jalpavitaṇḍe
bījaprarohasaṃrakṣaṇārthaṃ kaṇṭakaśākhāvaraṇavad iti.
This quotation occurs in a context where Dharmakīrti criticized the traditional concept of
debate – particularly wrangle (jalpa) and cavil (vitaṇḍā) – and instead promoted his own
original conception.
13
Cf. NV 1099,10–2000,4: tābhyāṃ jalpavitaṇḍābhyāṃ vigṛhyeti vijigīṣayā, na tattvabubhutsayeti. vidyāpālanārthaṃ caitat, na lābhapūjākhyātyartham iti.
398
Dharmakīrti’s Vādanyāya and the History of Conceptions of Debate
5.1 The “debate of people with a desire to win” (vijigīṣūṇāṃ vādaḥ) in the Vādanyāya
First, Dharmakīrti criticizes jalpa and vitaṇḍā by considering these two concepts to mean
“debate of people with a desire to win” (vijigīṣūṇāṃ vādaḥ).
VN 22,8–16: chalavyavahāre ’pi vijigīṣūṇāṃ vāda iti cet, na, durjanavipratipattyadhikāre satāṃ śāstrāpravṛtteḥ. na hi parānugrahapravṛttā mithyāpralāpārambhātmotkarṣaparapaṃsanādīn asadvyavahārān upadiśanti. na ca
paravipaṃsanena lābhasatkāraślokopārjanaṃ satām ācāraḥ. nāpi tathāpravṛttebhyaḥ svahastadānena prāṇinām upatāpanaṃ satsaṃmatānāṃ śāstrakārasabhāsadāṃ yuktam. na ca nyāyaśāstrāṇi sadbhir lābhādyupārjanāya
praṇīyante. tasmān na yogavihitaḥ kaścid vijigīṣuvādo nāma.
(Objection:) Even if distortion (chala) is used, the “debate of people with a
desire to win” (vijigīṣūṇāṃ vādaḥ) [is undertaken]. (Answer:) [This is] not
[so] because well-educated people are not engaged in the doctrinal system
in consideration of bad people’s evil deeds (vipratipatti).14 That is to say,
[well-educated] people who are engaged in the benefit of others do not teach
bad deeds (asadvyavahāra), such as boasting and embarrassing others through
beginning incoherent speech. Besides, winning profit, reverence, or praise by
disgracing others is not [suitable] behavior for well-educated people. Furthermore, troubling [other] people by giving a [helping] hand to those who are
engaged in this manner [i.e., disgracing others] is not suitable for those who
are considered to be well-educated people, i.e., those who are participants in
the meeting of learned men (śāstrakārasabhāsad). Additionally, well-educated
people don’t compose methodically written works for the purpose of gaining
profit and the rest. Because of these [evil deeds, such as distortion (chala)], a socalled “debate of people with a desire to win” is not reasonable (yogavihita)15
at all.
On the surface, it would seem that Dharmakīrti does not directly explain the characteristics
of “debate of people with a desire to win,” but merely states evil deeds that are not
appropriate for well-educated people. However, Dharmakīrti here comes to the conclusion
that “debate of people with a desire to win” is not reasonable. From this viewpoint, one
may say that the “evil deeds” should be considered to be precisely the behavior exhibited
in a debate of people with a desire to win.
Dharmakīrti does not use the terms jalpa or vitaṇḍā.16 However, according to the
explanation in the Nyāyasūtras,17 “distortion” (chala) is the hostile method used in jalpa
and vitaṇḍā. Additionally, with the passage of time, the terms vijigīṣukathā18 or vijigīṣuvāda
14
15
16
17
18
See VA 70,19–21: durjanānāṃ vipratipattir aśobhano vyavahāraḥ tasmān na yogavihito nyāyyaḥ
kaścid vijigīṣuvādo(1) nāma, yac chalādibhiḥ kriyata ity adhyāhāraḥ. ((1) vijigīṣu° em. : vijagīṣu°.)
McClintock 2010: 70, n. 169 has pointed out that Much uses the inappropriate translation “regellos” for
the yogavihita in Much 1991: 51,2.
Vitaṇḍā is criticized in VN 61,1–2: etenaiva vitaṇḍā pratyuktābhyupagamābhāve vivādābhāvāt.
See NS 1.2.2: yathoktopapannaś chalajātinigrahasthānasādhanopālambho jalpaḥ //
Cf. ṢDS 30: vijigīṣukathā(1) yā tu chalajātyādidūṣaṇā / sa jalpaḥ, sā vitaṇḍā tu yā pratipakṣavarjitā //
((1) vijigīṣu° em. : vijagīṣu°.)
Ryo Sasaki
399
came to be considered alternative expressions for jalpa and vitaṇḍā. Therefore, it seems
reasonable to conclude that Dharmakīrti criticizes the characteristics of jalpa and vitaṇḍā
in these phrases.
Let me summarize the purposes and methods of “debate of people with a desire to win”
presented in the Vādanyāya as follows:
1.
2.
3.
4.
5.
The purpose is “victory;”
The purpose is “(self-)profit,” “reverence” or “praise;”
The purpose is not “benefit of others;”
The idea of “defense of expertise” and the like is not mentioned;
The method is “evil deeds,” “distortion,” “embarrassing others” and so forth.
5.2 The “debate of well-educated people” (satāṃ vādaḥ) in the Vādanyāya
After criticizing the debate of people with a desire to win, i.e., wrangle and cavil, Dharmakīrti presents his original conception of debate, namely, “debate of well-educated people”
(satāṃ vādaḥ).
VN 22,16–21: parānugrahapravṛttās tu santo vipratipannaṃ pratipādayanto
nyāyam anusareyuḥ satsādhanābhidhānena bhūtadoṣodbhāvanena vā, sākṣipratyakṣaṃ tasyaivānuprabodhāya. tad eva nyāyānusaraṇaṃ satāṃ vādaḥ,
ukte nyāye tattvārthī cet pratipadyeta, tadapratipattāv apy anyo na vipratipadyeteti.19
On the other hand, when [well-educated people] persuade one who has fallacious notions, well-educated people who aim at the benefit of others (parānugraha) would abide by the reasonable rule (nyāya) by stating a correct
piece of proof or pointing out a real fault for making him (i.e., the one who
has fallacious notions) notice [his fault] in the presence of witnesses. The
debate of well-educated people (satāṃ vādaḥ) is precisely that, i.e., abiding by
reasonable rules if [an opponent] seeking the truth were to properly understand
the logical argument (nyāya) stated [by the proponent,20 and] even if [he] does
not understand it, others [in proximity] were not to misunderstand it.21
Elsewhere, Dharmakīrti claims that the proper means for a proponent is “stating a piece of
correct proof” (satsādhanābhidhāna), while the means for an opponent is “pointing out a
real fault” (bhūtadoṣodbhāvana). The means of “debate of people with a desire to win” are
described as “evil deeds” (vipratipatti/asadvyavahāra). Dharmakīrti compared the former
to the latter in the following passage:
VN 22,22–23,6: tattvarakṣaṇārthaṃ sadbhir upahartavyam eva chalādi vijigīṣubhir iti cet, na, nakhacapeṭaśastraprahārādīpanādibhir apīti vaktavyam.
19
20
21
vipratipadyeta° em. [D337a6/P377b8: mi rtogs pa; VA 70,24: vipratipadyeta] : pratipadyeta° VN.
See VA 70,21–22: ukte sati nyāye tattvārthī cet prativādī pratipadyeta tam arthaṃ nyāyopetam.
See VA 70,22–24: atha svapakṣarāgasya balīyastvād ukte ’pi nyāye na pratipadyeta. tadā tena prativādinā tasya nyāyasyārthasyāpratipattāv anyaḥ(1) samīpavartyātmajño janakāyo na vipratipadyeteti
kṛtvā nyāyānusāraṇam eva satāṃ vāda iti vartate. ((1) anyaḥ VAMS em. : anya° VA.)
400
Dharmakīrti’s Vādanyāya and the History of Conceptions of Debate
tasmān na jyāyān ayaṃ tattvarakṣaṇopāyaḥ. sādhanaprakhyāpanaṃ satāṃ
tattvarakṣaṇopāyaḥ sādhanābhāsadūṣaṇaṃ ca, tadabhāve mithyāpralāpād
atra paropatāpavidhāne ’pi tattvāpratiṣṭhāpanāt. anyathāpi nyāyopavarṇane
vidvatpratiṣṭhānāt. tasmāt parānugrahāya tattvakhyāpanaṃ vādino vijayaḥ,
bhūtadoṣadarśanena mithyāpratipattinivartanaṃ prativādinaḥ.
(Objection:)22 Well-educated people with a desire to win have to offer a distortion (chala) and so forth23 in order to protect truth (tattvarakṣaṇa). (Answer:)
No, [this is not true]. [If that were to be so,] it would have to be said that [welleducated people with a desire to win have to protect truth] even by attacking
[an opponent] with fingernails, an open palm, or weapons, or by setting [the
opponent] on fire. Therefore, this means of protecting truth [used by people
with a desire to win] is not superior [to the means of protecting truth used by
well-educated people]. The means of protecting truth [used by] well-educated
people are the explanation of proof (sādhanaprakhyāpana) and the refutation
of pseudo-proof (sādhanābhāsadūṣaṇa) because there is no establishment
of truth without them (i.e., these two means)24 even if [the well-educated
people] trouble others with incoherent speech in this case [and] because there
is firm ground for the learned men in case that they tell a logical argument
(nyāya) even if [they] do not [begin incoherent speech].25 Hence, the victory
of a proponent is the explanation of truth (tattvakhyāpana)26 for the purpose
of benefitting others (parānugraha); [on the other hand, the victory] of an
opponent is the removal of misapprehension (mithyāpratipattinivartana) by
showing the real fault [of the proponent’s proof].
It is important to consider the following features of “debate of well-educated people” when
comparing it with the definition of “debate of people with a desire to win:”27
1. Dharmakīrti does not consider the concept of “victory” in the usual sense of the
word to be the purpose of debate of well-educated people, and instead reinterprets
this concept. In a debate of well-educated people, a proponent is victorious when
22
23
24
25
26
27
The Tibetan translation of the Vādanyāya gives the lines a different reading. Cf. D337a6–7/
P377b8–378a1: gal te de kho na nyid bsrung ba’i don du dam pa rnams kyang rgyal bar ’dod pa
sgyu(1) la sogs pa dag gis tshar gcad (D; bcad P) par bya ba yin no zhe na / ma yin te khu tshur
dang / thal lcag dang / mtshon chas bsnun (D; bsnan P) pa dang / me la sogs pa dag gis kyang zhes
brjod par bya na / ((1) sgyu em. : rgyu D337a7/P377b8.)
The word ādi seems to imply jāti or nigrahasthāna. Cf. NS 1.2.2: yathoktopapannaś chalajātinigrahasthānasādhanopālambho jalpaḥ //
See VA 70,27–28: tadabhāva iti sādhanaprakhyāpanasādhanābhāsadūṣaṇayor abhāve.
See VA 70,28: anyathāpīti mithyāpralāpādyabhāve ’pi.
Steinkellner 1988: 1441–1442 pointed out that [satāṃ-]vādaḥ has to serve the investigation of truth
(tattvacintā VN 21,22) and the explanation of truth (tattvakhyāpana VN 23,5). In the discussion at the
conference, Prof. Steinkellner made the important remark that tattva in the case of tattvacintā should
mean “true reality.”
As for the difference between Dharmakīrti’s satāṃ vādaḥ and the notions of vāda, jalpa or vitaṇḍā
as defined in the Nyāyasūtras, see Much 1991: Einleitung 2.2 and Gokhale 1993: Introduction, pp.
xv–xvii.
Ryo Sasaki
2.
3.
4.
5.
401
he “explains truth [to others]” and an opponent is victorious when he “removes
misapprehension [of others].”
The purpose is not “(self-)profit,” “reverence” or “praise.”
Dharmakīrti considers the concept of “benefitting others” (parānugraha) to be the
purpose of debate of well-educated people. The Nyāyasūtras, Nyāyabhāṣya, and
Nyāyavārttika do not propose this concept to be the purpose of jalpa or vitaṇḍā, at
least not in the context of debate.
The purpose is “protection of truth.”
The proponent’s method is “the explanation of proof” or “stating a piece of correct
proof” and the opponent’s method is “the refutation of pseudo-proof” or “pointing
out a real fault.”
Regarding the first point, we must draw attention to the condition of defeat (nigrahasthāna)
and victory as constructed systematically by Dharmakīrti in the Vādanyāya. In Dharmakīrti’s theory, neither a proponent nor an opponent are to be defeated, even if they act solely
within the guidelines of the nigrahasthāna. On the basis of Figure 1, which shows the flow
chart of the conditions of victory or defeat defined in the Vādanyāya, the following situation
serves as an example: in the beginning, a proponent intends to prove his statement using
correct proof, i.e., the proof-action is not the case of nigrahasthāna. An opponent then tries
to refute the proof of the proponent, but what the opponent points out is a pseudo-fault,
not a real fault. If the proponent does not refute the pseudo-fault, neither proponent nor
opponent will win or lose, in spite of the fact that the proponent’s proof is correct. The
reason why Dharmakīrti regards the winner and the loser as undecided in this situation is
that the victory of the proponent is defined to be the explanation of truth, but the proponent
does not achieve this, since he fails to correct the opponent’s mistake. This reinterpreted
idea of victory is consistently maintained throughout the entire text of the Vādanyāya.
6. Responses to Dharmakīrti’s ideas in the Nyāya school
It remains to be studied how different Naiyāyikas accepted or refuted the ideas on debate
from the Vādanyāya. In this paper, we shall concentrate on Vācaspati Miśra’s Nyāyavārttikatātparyaṭīkā (NVTṬ) and Bhāsarvajña’s Nyāyabhūṣaṇa (NBhū).
6.1 Jalpa and vitaṇḍā in the Nyāyavārttikatātparyaṭīkā
Vācaspati Miśra explained wrangle and cavil as follows:
NVTṬ 1099,15–2000,11 (ad NS 4.2.51): na kevalaṃ tadarthaṃ ghaṭamānānāṃ jalpavitaṇḍe, api tu vidyānirvedādibhiś ca pareṇāvajñāyamānasya;
tābhyāṃ vigṛhyakathanam iti sūtram. yas tu svadarśanavilasitamithyājñānāvalepadurvidagdhatayā sadvidyāvairāgyād vā lābhapūjākhyātyarthitayā
kuhetubhir īśvarāṇāṃ janādhārāṇāṃ purato vedabrāhmaṇaparalokādidūṣaṇapravṛttaḥ, taṃ prati vādī samīcīnadūṣaṇam apratibhayāpaśyan jalpavitaṇḍe
402
Dharmakīrti’s Vādanyāya and the History of Conceptions of Debate
Figure 1: The conditions of victory or defeat defined in the Vādanyāya. (This figure has been presented in Sasaki 2013a)
Ryo Sasaki
403
avatārya vigṛhya28 jalpavitaṇḍābhyāṃ tattvakathanaṃ karoti vidyāparipālanāya…. idam api prayojanaṃ jalpavitaṇḍayoḥ. na tu lābhakhyātyādi dṛṣṭam.
na hi parahitapravṛttaḥ paramakāruṇiko munir dṛṣṭārthaṃ paravañcanopāyam upadiśatīti.
Wrangle (jalpa) and cavil (vitaṇḍā) [take place] not only between those who
strive for it (i.e., protecting the ascertainment of truth)29 but also between those
who are derogatorily considered to disregard expertise30 by others. [According
to the Nyāya-]sūtra (i.e., NS 4.2.51),31 “by means of these two, [i.e., wrangle
and cavil,] the hostile dispute [is undertaken].” However, the debater who
doesn’t find an appropriate objection against the [following] person introduces
(avatārya), namely, divides (vigṛhya) wrangle and cavil and then makes statements about truth through wrangle and cavil for the purpose of defending
expertise. The [above-mentioned] person is engaged in objecting to Veda,
Brahman, the future world and the rest in the presence of the Gods held firm
by people through fallacious logical reasons because of a false conception,
haughtiness, or unsophisticatedness that appears in his own view or [because
of] a desire for profit, honor, or reputation on account of aversion to the true
expertise…. This (i.e., defending expertise) is also the purpose of the wrangling and the cavil. However, profit, reputation and the like are considered to
be neither [the purpose of the wrangling nor the cavil] because the extremely
compassionate sage who is engaged in the welfare of others (parahita) does
not teach obvious methods for deceiving others.
Vācaspati seems to reinterpret the traditional definition of jalpa and vitaṇḍā by introducing
a new concept in the Nyāyavārttikatātparyaṭīkā. The main points of his reinterpretation
can be summarized as follows:
1. Vācaspati reinterpreted vigṛhya in NS 4.2.51 as avatārya, while Vātsyāyana interpreted vigṛhya as vijigīṣayā. It is assumed that Vācaspati intended to change the
traditional idea that debaters seek victory in jalpa and vitaṇḍā. This probably indicates his acceptance of Dharmakīrti’s criticism of the “debate of people with a desire
to win” (vijigīṣūṇāṃ vādaḥ). However, according to Thakur’s edition, this interpretation is problematic because he reads avatārya vijigīṣayā taṃ vigṛhya. Therefore, I
would like to avoid drawing a definitive conclusion here.
2. The purpose is not “(self-)profit,” “honor” or “reputation.”
3. Vācaspati introduced the idea of “welfare of others” (parahita). He does not clearly
describe this concept as the purpose or motivation of jalpa or vitaṇḍā. However,
it is likely that Vācaspati integrated the Vādanyāya’s idea of “benefit of others”
(parānugraha) into his own approach.
28
29
30
31
avatārya vigṛhya NVTṬ(CSS), NVTṬ(KSS); avatārya vijigīṣayā taṃ vigṛhya NVTṬ(Thakur).
See NS 4.2.50: tattvādhyavasāyasaṃrakṣaṇārthaṃ jalpavitaṇḍe… // “The wrangling (jalpa) and the
cavil (vitaṇḍā) [are undertaken] for the purpose of protecting the ascertainment of truth.” See section 4
for details.
See NVTP 558,2–3: vidyānirveda ity asya vivaraṇaṃ sadvidyāvairāgyād iti. ādigrahaṇavivaraṇaṃ
lābhapūjeti.
See section 4 for details.
404
Dharmakīrti’s Vādanyāya and the History of Conceptions of Debate
4. The purpose is “defense of expertise.”
5. The method is expressed using the metaphor of “thorns.”32
Although there remain some uncertainties as to how Vācaspati Miśra responded to the
ideas on debate presented in the Vādanyāya, it seems clear that he adopted some of these
ideas to fit his own concept of debate.
6.2 Jalpa and vitaṇḍā in the Nyāyabhūṣaṇa
In the case of the Nyāyabhūṣaṇa, an influence from the Vādanyāya is more evident than in
the case of the Nyāyavārttikatātparyaṭīkā.
NBhū 332,11–23: yatra tu vijigīṣuṇā saha lābhapūjākhyātikāmo jayārthaṃ
pravartate, sā vijigīṣukathā. nanu ca mokṣamārgaviruddhatvāl lābhādiprāptes
tatprayojanā kathā na yukteti, satyam; neyaṃ mumukṣuṇā kartavyā, kiṃ tu
jñātvā varjanīyeti nirūpitā. yadā tu vītarāgo vijigīṣuṇākṣiptaḥ kathāṃ parihartuṃ paroparodhān na śaknoti, tadāsau vītarāgas tena vijigīṣuṇā saha
parānugrahārthaṃ jñānāṅkurarakṣaṇārthaṃ ca tām eva caturaṅgāṃ kathāṃ
kuryāt…. svātmani śiṣyādyātmani cotpannas tattvādhyavasāyāṅkuraḥ śākyādimṛgair bhakṣyetāpi yadi jalpavitaṇḍābhyāṃ kaṇṭakaśākhābhyām āvaraṇaṃ
na kriyeta. ye cānutpannatattvajñānāḥ śiṣyādayaḥ, teṣāṃ prativādinaś cānugrahārthaṃ vītarāgeṇāpi jalpādau pravartitavyam ity uktam. anugrahaś ca
mokṣaśāstrācāryādiṣu śraddhotpattidvāreṇa bhavati.
However, when those who long for profit, honor, or reputation undertake a
certain [debate] for the purpose of victory with those [disputants who are]
desirous of victory, the [debate] is the “dispute of people with a desire to win”
(vijigīṣukathā). (Objection:) Because the acquisition of profit and the rest is
contrary to the path to liberation, the dispute that has it (i.e., victory) as its
purpose is not appropriate. (Answer:) Yes, [you are correct]. It is determined
that those who desire liberation ought not to do this (i.e., the disputation
whose purpose is victory), but rather ought to avoid [such a disputation] after
becoming aware of [such a disputation]. However, unless a person without
passion who is provoked by a person with the desire to win can repel the
dispute by troubling others, this passionless person will undertake the very
dispute, which consists of four component parts33 with this person with a
32
33
Cf. NVTṬ 1099,13–14 (ad NS 4.2.50): tattvādhyavasāyasaṃrakṣaṇārthaṃ jalpavitaṇḍe bījaprarohasaṃrakṣaṇārthaṃ kaṇṭakaśākhāvaraṇavat iti sūtram. tad vyācaṣṭe – anutpannatattvajñānānām
iti.
Basically, in the Nyāya school vītarāgakathā corresponds to vāda, and vijigīṣukathā corresponds to jalpa
or vitaṇḍā. Bhāsarvajña, however, shows another interpretative possibility – that vītarāgakathā and vijigīṣukathā are divided into four sub-types: (i) sapratipakṣā vītarāgakathā, (ii) apratipakṣā vītarāgakathā,
(iii) sapratipakṣā vijigīṣukathā, and (iv) apratipakṣā vijigīṣukathā. See NBhū 332.7–11: vītarāgakathāpi
dvividhā bhavati – sapratipakṣā vāpratipakṣā(1) ca…. evaṃ cāvāntarabhedavivakṣāyāṃ catasraḥ kathā
bhavanti, na tisra iti. vyavahāras tu viśeṣasaṃjñātrayeṇaivāstīti tisra eva viśeṣasaṃjñā uddiṣṭāḥ. atha
vā tisra eva bhavantu, vādo ’pi hi pratipakṣahīno vitaṇḍaiveti vakṣyāmaḥ. ((1) vāpratipakṣā em. : vā
pratipakṣā.) “The dispute of people without passion (vītarāgakathā) also has two types: [the dispute
Ryo Sasaki
405
desire to win for the purpose of benefitting others (parānugrahārtham) and
for protecting the sprout of knowledge…. The sprout of the ascertainment of
truth that surfaces in [a debater] himself and in his disciples would be eaten by
savage animals, such as Buddhists, if [the sprout] were not covered by thorny
branches, i.e., the wrangling and the cavil. Furthermore, it is said that even
the person without passion has to undertake the wrangling and so on for the
purpose of benefitting (anugrahārtham) disciples and the rest who have no
knowledge of truth and for [benefitting] an opponent. Additionally, the benefit
[of others] occurs through having faith in the masters [who teach] the doctrine
of liberation and the like.
As in Vācaspati’s text cited above, Bhāsarvajña also proposes a new interpretation of jalpa
and vitaṇḍā. The main points are summarized as follows:
1. Bhāsarvajña considers that a person without passion (vītarāga) has to participate
in the “dispute of people with a desire to win” (vijigīṣukathā) when he is provoked
by a person with the desire to win (vijigīṣu). Even in that case, vītarāga, as its
name suggests, is free from passion, such as the desire to win. Therefore, in the
case of vijigīṣukathā carried out by vītarāga, it seems that Bhāsarvajña intends, like
Dharmakīrti, to remove the idea of victory as the purpose of jalpa and vitaṇḍā. In
other words, both vītarāga in the Nyāyabhūṣaṇa and sat in the Vādanyāya have no
desire to win (vijigīṣā). It seems quite probable that the Nyāyabhūṣaṇa’s view is
influenced by the Vādanyāya.
2. The purpose is not “(self-)profit,” “honor,” or “reputation” for those who desire
liberation (mumukṣu).
3. Bhāsarvajña introduced the idea of parānugrahārtham “for the purpose of benefitting
others” as the motivation behind vītarāga in the context of jalpa and vitaṇḍā. Being
engaged in parānugraha “benefitting others” is common among vītarāga in the
Nyāyabhūṣaṇa and sat in the Vādanyāya. One may say that Bhāsarvajña covertly
imported Dharmakīrti’s idea of parānugrahāya.
4. The purpose is the “protection of the sprout of knowledge [of the ascertainment of
truth].”
5. The method is expressed using the metaphor of “thorns.”
Judging from the above, we can be fairly certain that Bhāsarvajña adopted the viewpoint
and concept described in the Vādanyāya to his own notion of debate, presented in the
Nyāyabhūṣaṇa.
that] has an opposing view and [the dispute that] has no opposing views…. Furthermore, when one
wishes to express the division [of the disputes] in this way (i.e., based on the existence or nonexistence
of opposing views) respectively, there are four types of disputes, not three. On the other hand, it is
[already] taught that there are only three particular names (i.e., vāda, jalpa, and vitaṇḍā) because there
is conventional usage only by the three particular names. We, however, will state as follows: if only the
three [particular names] exist, the debate (vāda) that has no opposing views is also a cavil (vitaṇḍā).”
Dharmakīrti’s Vādanyāya and the History of Conceptions of Debate
Figure 2: The comparison of conceptions of debate in Indian Logic
406
Ryo Sasaki
407
7. Concluding remarks
The comparison of “debate” concepts in the sources examined in this paper is presented in
Figure 2. In addition, this comparison yields the following main points:
1. In the Carakasaṃhitā the term jalpa is employed to refer to a type of hostile debate
conducted for the purpose of victory.
2. In the *Upāyahṛdaya the composition of a treatise (造論) is explained using the
metaphor of thorns (荊棘). The same metaphor is used in the Nyāyasūtras. Either it was adapted from the *Upāyahṛdaya or from another text upon which both
the Nyāyasūtras and the *Upāyahṛdaya draw. In the Nyāyasūtras the metaphor is
used to explain jalpa and vitaṇḍā, technical terms that are also employed in the
Carakasaṃhitā.
3. In the Nyāyabhāṣya both jalpa and vitaṇḍā are interpreted as forms of debate for
the purpose of victory. As for jalpa, this interpretation is similar to that found in the
Carakasaṃhitā.
4. In the Vādanyāya Dharmakīrti criticized the purpose of victory attributed to jalpa and
vitaṇḍā in the Nyāyabhāṣya, presenting instead the idea of parānugraha “benefitting
others,” as the purpose of debate. Dharmakīrti’s negation of the purpose of victory
and the affirmation of the purpose of “benefit of others” is similar to the position
found in the *Upāyahṛdaya.
5. From among the later Naiyāyikas, at least Bhāsarvajña, in the Nyāyabhūṣaṇa, accepted implicitly Dharmakīrti’s construction of the debate concept in the Vādanyāya
by adopting the idea of “benefitting others” and rejecting the idea of “victory.” It is
possible that the same might be said of Vācaspati’s Nyāyavārttikatātparyaṭīkā as
well, although the text is not explicit.
We may reasonably conclude that the Vādanyāya marked a turning point in the historical
transition of the concept of debate. Dharmakīrti criticized the earlier concept of debate in
the Nyāya school as “debate of people with a desire to win” (vijigīṣūṇāṃ vādaḥ) and created
a new one, i.e., “debate of well-educated people” (satāṃ vādaḥ). His new understanding
of the nature of the debate affected the later Nyāya school and led some Naiyāyikas to
modify their approach. However, the extent to which the Vādanyāya influenced later works
in the Nyāya school remains a matter for further research. An extended examination of
the Vādanyāya’s commentaries, such as Śāntarakṣita’s Vipañcitārthā, as well as more
comprehensive studies on later Nyāya literature, are needed to more fully understand these
relationships.
References and abbreviations
Primary sources
CaS Carakasaṃhitā, ed. Vaidya Jadavaji Trikamji Acharya. Varanasi 2008.
D sDe dge edition of the Tripiṭaka in Tibetan.
MA Madhyamakāvatāra, ed. L. de la Vallée Poussin. St. Petersburg 1912.
408
Dharmakīrti’s Vādanyāya and the History of Conceptions of Debate
NBh Nyāyabhāṣya. See NV.
NBhū Nyāyabhūṣaṇa, ed. Svāmin Yogīndrānanda.Varanasi 1968.
NS Nyāyasūtras. See NV.
NV Nyāyavārttika. See NVTṬ.
NVTP Nyāyavārttikatātparyapariśuddhi, ed. A. Thakur. New Delhi 1996.
NVTṬ Nyāyavārttikatātparyaṭīkā, ed. Tāranātha Nyāya-Tarkatīrtha and Amarendramohan
Tarkatīrtha. Calcutta 1936–1944. (Repr. Kyoto 1982.)
NVTṬ(KSS) Nyāyavārttikatātparyaṭīkā, ed. Rajeshwara Sastri Sravida. Benares 1925.
NVTṬ(Thakur) Nyāyavārttikatātparyaṭīkā, ed. A. Thakur. New Delhi 1996.
P Peking edition of the Tripiṭaka in Tibetan.
ṢDS Ṣaḍdarśanasamuccaya, ed. Mahendra Kumar Jain. Calcutta 1969. (Repr. Calcutta
2
1981.)
T Taishō shinshū daizōkyō. (Taishō edition of the Tripiṭaka in Chinese.)
UH *Upāyahṛdaya. See T [Vol. 32, No. 1632].
VA Vipañcitārthā, ed. R. Sāṅkṛtyāyana. Patna 1935–1936. (Tib. D No. 4239, P1 No. 5725,
P2 No. 5738.)
VAD Vipañcitārthā, ed. Svāmi Dvārikādās Śāstrī. Varanasi 1972.
VAMS Vipañcitārthā: Photostat copy of the Sanskrit manuscript in the library of GeorgAugust-Universität Göttingen. (Color photostat copies of the same manuscript: 2a,
3a, 4a, 5a, 6a are contained in 尼瑪旦増編,『羅布林卡珍蔵文物輯選』,中国
蔵学出版社,2011.)
VN Vādanyāya, ed. M. T. Much. Wien 1991. (Tib. D No. 4218, P No. 5715.)
Secondary sources
Chinchore 1988 M. R. Chinchore, Vādanyāya: A Glimpse of Nyāya-Buddhist Controversy.
Delhi 1988.
Chinchore 1990 M. R. Chinchore, Post-Udayana Nyāya Reactions to Dharmakīrti’s Vādanyāya: An Evaluation. Indian Philosophical Quarterly 17.1 (1990) 1–31.
Eltschinger 2012 V. Eltschinger, Debate, Salvation and Apologetics: On the Institutionalization of Dialectics in the Buddhist Monastic Environment. In: Devadattīyam. Johannes Bronkhorst Felicitation Volume, ed. F. Voegeli et al. Bern etc. 2012, 429–489.
Gokhale 1993 P. P. Gokhale, Vādanyāya of Dharmakīrti: The Logic of Debate. Delhi
1993.
Ishitobi 2003 M. Ishitobi, Ryūju to indoronrigaku no tanjō. Kokugakuin zasshi 104.6
(2003) 34–45.
Ishitobi 2006 M. Ishitobi, Ryūju zō hōbenshinron no kenkyū. Tokyo 2006.
Kajiyama 1984 Y. Kajiyama, Bukkyō chishikiron no keisei. In: Ninshikiron to ronrigaku,
ed. A. Hirakawa et al. Tokyo 1984, 1–101.
Kang 2003 S. Y. Kang, Die Debatte im alten Indien: Untersuchungen zum Sambhāṣāvidhi
und verwandten Themen in der Carakasaṃhitā Vimānasthāna 8.15–28. Reinbek
2003.
Matilal 1998 B. K. Matilal, The Character of Logic in India, ed. J. Ganeri et al. Albany
1998.
Ryo Sasaki
409
McClintock 2010 S. L. McClintock, Omniscience and the Rhetoric of Reason: Śāntarakṣita and Kamalaśīla on Rationality, Argumentation, & Religious Authority. Boston
2010.
Much 1986 M. T. Much, Dharmakīrti’s Definition of “Points of Defeat” (Nigrahasthāna).
In: Buddhist Logic and Epistemology: Studies in the Buddhist Analysis of Inference
and Language, ed. B. K. Matilal et al. Dordrecht/Boston/Lancaster/Tokyo 1986,
133–142.
Much 1991 M. T. Much, Dharmakīrtis Vādanyāyaḥ, Teil II: Übersetzung und Anmerkungen. Vienna 1991.
Nicholson 2010 H. Nicholson, The Shift from Agonistic to Non-Agonistic Debate in Early
Nyāya. Journal of Indian Philosophy 38 (2010) 75–95.
Ono 2004 T. Ono, Nyāyapariśiṣṭa ni tsuite: Indo koten tōronjutsu no dentō [Nyāyapariśiṣṭa
or Bodhasiddhi by Udayana]. Sōtō shū kenkyūin kenkyū kiyō 34 (2004) 73–94.
Ono 2006 T. Ono, Nyāyapariśiṣṭa ni tsuite (3): Indo koten tōronjutsu no dentō [On Nyāyapariśiṣṭa (3): some Rules of Philosophical Argumentation]. Sōtō shū kenkyūin kenkyū
kiyō 36 (2006) 101–121.
Preisendanz 2000 K. Preisendanz, Debate and Independent Reasoning vs. Tradition: On
the Precarious Position of Early Nyāya. In: Harānandalaharī : Volume in Honour of
Professor Minoru Hara on his Seventieth Birthday, ed. R. Tsuchida et al. Reinbek
2000, 221–251.
Preisendanz 2009 K. Preisendanz, Logic, Debate and Epistemology in Ancient Indian
Medical Science: An Investigation into the History and Historiography of Indian
Philosophy. Indian Journal of History of Science 44.2 (2009) 261–312.
Prets 2000 E. Prets, Theories of Debate, Proof and Counter-Proof in the Early Indian
Dialectical Tradition. In: On the Understanding of Other Cultures: Proceedings
of the International Conference on Sanskrit and Related Studies to Commemorate
the Centenary of the Birth of Stanislaw Schayer (1899–1941), Warsaw University,
Poland, October 7–10, 1999, ed. P. Balcerowicz et al. Warsaw 2000, 369–382.
Ruegg 2000 D. S. Ruegg, Three Studies in the History of Indian and Tibetan Madhyamaka
Philosophy: Studies in Indian and Tibetan Madhyamaka Thought Part 1. Vienna
2000.
Sasaki 2012a R. Sasaki, Vādanyāya ni okeru han-shoshō-kyoseki-ninshiki-shudan: Dharmakīrti ni yoru sonzaisēni motozuku Setsunametsu-ronshō [Sādhyaviparyaye bādhakapramāṇam in the Vādanyāya: Dharmakīrti’s Proof of Momentariness from
Existence]. Tōyō no shisō to shūkyō 29 (2012) 1–22.
Sasaki 2012b R. Sasaki, Dharmakīrti no nigrahasthāna kaishaku (1): asādhanāṅgavacana
ni tsuite [Dharmakīrti’s interpretation of nigrahasthāna (1): On asādhanāṅgavacana].
Kuwon: kenkyū ronbun shū 3 (2012) 69–90.
Sasaki 2013a R. Sasaki, Dharmakīrti no nigrahasthāna kaishaku (2): adoṣodbhāvana ni
tsuite [Dharmakīrti’s interpretation of nigrahasthāna (2): On adoṣodbhāvana]. Kuwon:
kenkyū ronbun shū 4 (2013) 55–75.
Sasaki 2013b R. Sasaki, Nigrahasthāna in the Vādanyāya: Controversy between Dharmakīrti and the Nyāya School. Journal of Indian and Buddhist Studies 61.3 (2013)
178–182.
410
Dharmakīrti’s Vādanyāya and the History of Conceptions of Debate
Sasaki 2014a R. Sasaki, Dharmakīrti ni yoru tōron-shisō no taikeika: Vādanyāya ni okeru
Nyāya-gakuha no hihan to juyō [Systematization of the Thought of Debate by Dharmakīrti: Criticism and Acceptance of the Nyāya School’s Theory in the Vādanyāya].
Minami Ajia Kotengaku 9 (2014) 319–370.
Sasaki 2014b R. Sasaki, Acceptance and interpretation of Dharmakīrti’s theory of nigrahasthāna in the Nyāyamañjarī. Kuwon: kenkyū ronbun shū 5 (2014) 40–62.
Steinkellner 1988 E. Steinkellner, Remarks on Niścitagrahaṇa. In: Orientalia Iosephi
Tucci Memoriae Dicata, ed. G. Gnoli. Roma 1988, 1427–1444.
Ui 1925 H. Ui, Hōbenshinron no chūshakuteki-kenkyū. In: Indo tetsugaku kenkyū dai ni.
Tokyo 1925, 473–585.
Uryuzu and Nakazawa 2012 R. Uryuzu and M. Nakazawa, Zenyaku Candrakīrti nyūchūron. Chiba 2012.
Vidyabhusana 1921 S. C. Vidyabhusana, A History of Indian Logic: Ancient, Mediaeval
and Modern Schools. Calcutta 1921. (Repr. Delhi 2006.)
The Meaning of bāhyārtha in Dignāga’s and
Jinendrabuddhi’s Theories of Inference
by
Kiyokuni Shiga
1. Introductory remarks
In an earlier paper, “Remarks on the Origin of All-Inclusive Pervasion” (henceforth Shiga
2011b), I proposed the hypothesis that the theory of all-inclusive pervasion (sarvopasaṃhāravyāpti), considered to have been created by Dharmakīrti, in fact originated in Dignāga’s
theory of inference. The statement of Dignāga’s in question, which appears in the third
chapter of the Pramāṇasamuccayavṛtti (henceforth PSV), is the following:
To be more precise, the co-existence [of a logical reason] with such an [object to be proved] is understood by means of two [types of] exemplification
based on similarity or dissimilarity, [in] which [the property to be proved] is
associated1 with external thing[s2 that respectively have the proving property]
(bāhyārthopasaṃhṛtena).3
1
2
3
This research was supported in part by JSPS KAKENHI Grant Number 24720029 and 25284014. I
would like to express my gratitude to Prof. Kei Kataoka for having given me valuable comments after
the conference.
As we will see below, Jinendrabuddhi understands the word upasaṃhṛta or upasaṃhāra as forming
part of the compound bāhyārthopasaṃhṛta or bāhyārthānupasaṃhāra as upadarśita (PSṬ D185b3;
P211a7; Skt. B146a2), upanayana or prakāśana (PSṬ D230a3; P260b5; Skt. PSṬ B185a2). Moreover,
he interprets the word upasaṃhāra as “associating/bringing the [property] to be proved with/near all
the proving [properties]” elsewhere in the same chapter (PSṬ D227b1; P257b7f.: nye bar bsdu ba
ni med na mi ’byung ba nyid de[ste P] / thams cad du sgrub par byed pa la bsgrub par bya ba nye
bar bsdu ba ste / nye bar ’dren pa’o // Skt. B182b6f.: upasaṃhāro ’vinābhāvitvam, sarvatra sādhane
sādhyasyopasaṃharaṇam upanayaḥ). Therefore, it would be more appropriate in this case to translate
upasaṃhṛta as “associated,” “applied” or “brought near” than as “summed up” (Shiga 2011b: 525 with
n. 14) or “included,” as in the case of Dharmakīrti’s term sarvopasaṃhāra (HB 5*, 21), which could be
translated as “including” or “summing up [the property to be proved in] all the [property-bearers that
respectively have the proving property]” (HBṬ 62, 18–20: sarvasmin sādhanadharmavati dharmiṇi, na
dṛṣṭāntadharmiṇy eva, sādhyadharmasyopasaṃharaṇam upasaṃhāro ḍhaukanaṃ …). Meanwhile, it
should be noted here that Arcaṭa glosses upasaṃhāra as ḍhaukana (“bringing [the property to be proved]
near [all the property-bearers that respectively have the proving property]”). Taking this interpretation
into consideration, it is probable that the word upasaṃhāra, even in the compound sarvopasaṃhāra,
could mean “associating” or “applying” rather than “summing up” or “including.”
According to Jinendrabuddhi, bāhyārtha is to be understood not as a single and particular individual
(PSṬ D215a4f.; P244b2; Skt. B171a2f.: ... ekatraiva vyaktiviśeṣe pradarśanasya prādhānyanirāsaparaṃ
draṣṭavyam), but as a whole assembly of things (PSṬ D215a3; P244a8f.; Skt. B171b7: sāmānyena sarvo
yathoktadṛṣṭāntalakṣaṇo ’rtharāśir dṛṣṭāntaḥ, na tu ghaṭa eva) or the generalized property-bearer as
opposed to the particular property-bearer to be proved (PSṬ D185b3; P211a7; Skt. B146a2: dharmiṇaḥ
pakṣīkṛtād viśeṣād anyatra sāmānye ...).
PSV(K) P138a6f.: de ’dra ba’i lhan cig rgyu ba nyid ni phyi rol gyi don nye bar sdud par byed
pa chos mthun pa dang mi mthun pa’i dpe gnyis po dag gis rtogs pa yin te / PSV(V) D51a7–b1;
Birgit Kellner et al., eds., Reverberations of Dharmakīrti’s Philosophy: Proceedings of the Fifth International Dharmakīrti
Conference Heidelberg, August 26 to 30, 2014. Vienna 2020, pp. 411–427.
412 The Meaning of bāhyārtha in Dignāga’s and Jinendrabuddhi’s Theories of Inference
This statement coincides with Dharmakīrti’s formulation of all-inclusive pervasion in his
Hetubindu4 (henceforth HB). Considering various factors, I concluded that Dignāga and
Dharmakīrti shared the idea that pervasion is established or confirmed by taking individual
cases5 into account.6
As I mentioned in Shiga 2011b, the term bāhyārtha also appears several times in the
fourth chapter of the Pramāṇasamuccaya (henceforth PS) and the PSV in the expressions
bāhyārthānupasaṃhāra, bāhyārthapradarśana, bāhyārthāpekṣā, as well as bāhyārthopasaṃhṛta. These examples of its use need to be examined in a new light to clarify the
relationship between bāhyārtha and dṛṣṭānta. This paper aims to further investigate various
issues that were not settled in Shiga 2011b, including the following: (1) What do Dignāga
and Jinendrabuddhi mean by the term bāhyārtha used in logical contexts, mainly in the PS
and PSV 4? (2) Does it literally denote “an external object,” that is, “a thing in the external
world as an object of cognition,” or does it denote “something external to the subject
or the property-bearer to be proved,” as Jinendrabuddhi understands it? (3) In this case,
does Jinendrabuddhi deviate from Dignāga’s original intention by adopting Dharmakīrti’s
theories when he interprets Dignāga, or are his comments on the term bāhyārtha in keeping
with Dignāga’s intention? (4) What is the relationship between the terms bāhyārtha and
bahirvyāpti?
2. Usage examples for the term bāhyārtha in the PS/PSV 4
2.1. Regarding the role of dṛṣṭānta7
First, we will observe an example of Dignāga’s use of the term bāhyārtha in PS/PSV 4,
where he defines the role of dṛṣṭānta.8
4
5
6
7
8
P54b6f.: de dang mthun pa dang lhan cig tu rgyu ba nyid dang / chos mthun pa’am chos mi mthun
pa nyid kyis[nyid can gyis D] phyi rol gyi don nye bar bsdus nas dpe gnyis kyis go bar byed do //
(Skt. reconstruction: *tādṛksāhabhavyaṃ hi sādharmyeṇa vaidharmyeṇa vā bāhyārthopasaṃhṛtena
dṛṣṭāntadvayena gamyate.) See also Shiga 2011b: 524–526 with n. 11–18.
HB 5*, 18–22: tasya dvividhā prayogaḥ; sādharmyeṇa vaidharmyeṇa ca, yathā yat sat tat sarvaṃ
kṣaṇikaṃ yathā ghaṭādayaḥ, saṃś ca śabda iti, tathā kṣaṇikatvābhāve sattvābhāvaḥ, saṃś ca śabda ity
anvayavyatirekābhyāṃ sarvopasaṃhāreṇa vyāptipradarśanalakṣaṇau sādharmyavaidharmyaprayogau.
While Dignāga states “external case[s]” (bāhyārtha), Dharmakīrti refers to “all cases” (sarva).
See Shiga 2011b: 532f.
In Dignāga’s system of logic, the word dṛṣṭānta has two different meanings: an actual example or a
thing (artha, abhidheya) and an exemplification or a statement (vacana, abhidhāna). (See Katsura 2004:
141f.)
See Shiga 2011b: 527. In PS 4.2 dṛṣṭānta is defined differently as follows: sādhyenānugamo hetoḥ
sādhyābhāve ca nāstitā / khyāpyate yatra dṛṣṭāntaḥ sa sādharmyetaro dvidhā // (See Jambuvijaya
1966:133. This verse is quoted in Daśavaikālikasūtrahāribhadrīvṛtti 34b. Cf. NMukh 11.) “The example
is a [thing] in which it is conveyed that a logical reason is accompanied by what is to be proved and that
[the logical reason] is absent when what is to be proved is absent. This is in two forms: [the example on
the basis of] similarity and the other (= the example on the basis of dissimilarity).” (See also Katsura
2004: 141 with n. 11.)
Kiyokuni Shiga
413
Reconstructed text9 of PSV on PS 4.3 (= Appendix [2]):10
evaṃ tarhi ghaṭo ’nudāharaṇam. tatra hi yathā hetuḥ sādhyānugataḥ, tathā
sādhyam api hetvanugatam iti cet. na, avivakṣitatvāt [yathā]11 kṛtakatvaviśeṣasya hetutvena, tathā ghaṭe sādhyānugamasya. bāhyārthapradarśanaṃ hi
nidarśye pradhānam.12
[Question:] If so (= if the example on the basis of similarity is defined as a
thing in which a logical reason is accompanied by what is to be proved),13 a
pot would not [constitute a valid] exemplification, because in the [pot], just as
a logical reason is accompanied by what is to be proved, what is to be proved
is [also] accompanied by the logical reason.
[Answer:] This is not [right], because just as being produced as a particular
property is not intended to be the logical reason, what is to be proved is
[not intended to be] accompanied [by the logical reason] in a pot[, which is
a particular thing, even though that is the case],14 for the primary [role] of
exemplification is to indicate external thing[s].15
Jinendrabuddhi comments on the term bāhyārtha as follows:
9
10
11
12
13
14
15
Words in roman type are found in the PSṬ or in the fragments, whereas those in italics are reconstructed.
The Sanskrit reconstructions of the Tibetan translations of PS/PSV 3 and 4 cited in this paper, other
than a few passages, are part of the achievements of the PSṬ seminar organized and led by Prof. Shōryū
Katsura at Ryūkoku University. I would like to specially thank all the participants of this seminar,
especially Dr. Yasuhiro Okazaki, who made Sanskrit reconstructions and editions of PS/PSV/PSṬ 4; Dr.
Toshikazu Watanabe, who checked and corrected Dr. Okazaki’s draft; Prof. Diwakar Acharya for his
valuable comments and suggestions on the Sanskrit reconstructions; and Prof. Katsura for permission
to use these results and other relevant materials. Any errors that remain are my own.
See also Shiga 2011b: 527 with n. 21–24.
No equivalent in PSV(K).
PSV(K): P148b4–6: ’on te de ltar na bum pa dper mi bya ste / de la ni ci ltar gtan tshigs bsgrub bya’i
rjes su ’gro ba de bzhin du / bsgrub bya yang gtan tshigs kyi rjes su ’gro ba yin no zhe na / ma yin
te / byas pa’i khyad par gtan tshigs su brjod par ’dod pa ma yin pa bzhin du bum pa la bsgrub bya rjes
su ’gro ba brjod pa ma yin pa’i phyir ro // phyi rol gyi don la bstan pa ni dpe la gtso bo yin no // PSV(V)
D60b1–3; P64a6f.: gal te ’di ltar bum pa dper brjod pa de lta na ni ji ltar gtan tshigs bsgrub bya dang
ldan pa de ltar bsgrub bya gtan tshigs dang ldan par ’gyur ro zhe na ma yin te / brjod par mi bya
ba’i phyir te / bum pa ni gang bsgrub bya’i rjes su ’gro ba can gyis[gyi P] byas pa nyid kyi khyad par
du byas pa brjod par bya ba’i phyir ro // ngag gi don bstan pa ni nges par bstan pa’i don gtso bo yin
no // (See also Katsura 2004: 155 with n.28.)
PSṬ D214b5f.; P244a2f.: ’o na de lta na zhes pa / gal te gang du gtan tshigs bsgrub byas rjes su ’gro
ba yin pa de chos mthun pa nyid kyi dpe’o zhes pa mtshan nyid yin na / (Skt. B171b3: evaṃ tarhīti
yadi hetoḥ sādhyānvayo yatra sa sādharmyadṛṣṭānta iti lakṣaṇam.)
PSṬ D214b7; P244a4f.: ma yin te brjod par mi ’dod pa nyid kyi phyir zhes pa bsgrub bya’i gtan tshigs
kyi rjes su ’gro ba yod kyang / brjod par ’dod pa yod pa ma yin te / (Skt. B171b4: nāvivakṣitatvād iti.
sann api sādhyasya hetunānugamo na vivakṣyate.)
Cf. Katsura 2004: 143: “Dignāga clearly states that the main purpose of an example statement is to
indicate an external object (bāhyārtha) as an example. This seems to suggest that as long as he is
discussing logic and epistemology, he is assuming external reality.” Also cf. Katsura 2004: 155: “He (=
Dignāga) further states that the main purpose in referring to a particular object like a pot is to indicate
some positive support in external reality.”
414 The Meaning of bāhyārtha in Dignāga’s and Jinendrabuddhi’s Theories of Inference
External thing[s] mean [those things] in general that are [similar] in kind
to what is to be proved, which are [external] to a particular [thing, that is, a
particular property-bearer] such as a pot. [The expression bāhyārthapradarśanam, i.e.] “to manifest external thing[s]” means to generally manifest the
pervasion of a logical reason by what is to be proved in the [external things in
the following way:] “Whatever is produced is all necessarily impermanent.”16
Here, Jinendrabuddhi interprets bāhyārtha differently from how he commented on the
phrase bāhyārtha (-upasaṃhṛta) in the third chapter, where he interpreted bāhyārtha as
“elsewhere than in a particular [property-bearer] that is made to be the subject, i.e., the
[property-bearer] in general” (dharmiṇaḥ pakṣīkṛtād viśeṣād anyatra sāmānye).17 This
gloss shows the contrast between the inside and the outside of the subject (pakṣa), and the
contrast between particular and generalized property-bearers.
The above quotation, on the other hand, suggests that the term bāhyārtha does not
mean a particular thing such as a pot, which is one of similar examples, but the generality
of things that are similar in kind to what is to be proved (sādhyajātīyasāmānya). In this
case, particularity (= an internal thing) is contrasted to generality (= external things).
Jinendrabuddhi focuses on the division between an example as an individual thing and a
generalized example, but not on the division between a thing inside the subject and things
outside the subject.
Next, let us look into the compound bāhyārthapradarśana. According to the Pramāṇasamuccayaṭīkā, the first half bāhyārtha is to be interpreted in the locative sense, i.e. as
“in external thing[s],” and the word vyāpti is complemented as an object of the last half
-pradarśana. Thus, the compound as a whole means “indicating the pervasion in external
thing[s].” Elsewhere in the same chapter, Jinendrabuddhi presents another interpretation
with regard to bāhyārtha. He understands dṛṣṭānta as meaning an assembly of things (artharāśi), but not as a particular individual thing (vyaktiviśeṣa). And that particular individual
16
17
PSṬ D215a1f.; P244a6f.: bum pa la sogs pa’i khyad par las / bsgrub bya’i rigs can gyi spyi ni phyi rol
gyi don no[to D] // de la spyir gang cung zad byas pa zhes bya ba de[de’i P] thams cad mi rtag pa kho
na’o zhes bsgrub byas gtan tshigs la khyab pa gsal bar byed pa ni phyi rol gyi don gsal bar byed pa’o //
(Skt. B171b5f.: ghaṭāder viśeṣāt sādhyajātīyasāmānyaṃ bāhyo ’rthaḥ, tatra sāmānyena yan nāma
kiṃcit kṛtakaṃ tat sarvam anityam eveti sādhyena hetor vyāptiprakāśanaṃ bāhyārthaprakāśanam.)
PSṬ D185b2–4; P211a7f. (= Appendix [1]): phyi rol gyi don nye bar bsdus pas zhes pa / phyogs su
byas pa’i chos can gyi khyad par las gzhan du spyi la nye bar bstan pas zhes pa’i don to // spyi yang
khyad par yongs su mi spong pa’i phyir / bsgrub par bya ba’i chos can yang der nang du ’dus pa kho na
ste / bsgrub bya’i chos can kho na la med na mi ’byung ba nyid ston pa bsal ba lhur len pa ni phyi rol
gyi don smos pa’o // (Skt. B146a1–3: bāhyārthopasaṃhṛteneti. dharmiṇaḥ pakṣīkṛtād viśeṣād anyatra
sāmānya upadarśitenety arthaḥ. sāmānyasya ca viśeṣāparityāgāt, sādhyadharmy api tatrāntargata
eva. sādhyadharmiṇy evāvinābhāvitvapradarśananirāsaparaṃ tu bāhyārthagrahaṇam.) “[The phrase]
bāhyārthopasaṃhṛtena means ‘by [means of two types of examplification based on the similarity or
dissimilarity, in which the inseparable relation] is indicated elsewhere (anyatra) than in the particular
[property-bearer] that is made to be the subject, i.e. in the [property-bearer] in general.’ And because
the general does not abandon a particular, the property-bearer to be proved is also included in it (= the
property-bearer in general). The word bāhyārtha, on the other hand, is mentioned for the purpose of
negating the indication of the inseparable relation only in the property-bearer to be proved.” See also
Shiga 2011b: 525f. with n. 18. It is to be noted that Jinendrabuddhi paraphrases the word bāhyārtha to
anyatra here.
Kiyokuni Shiga
415
thing is described as being included in the assembly of things.18 For Jinendrabuddhi, the
term bāhyārtha means “something external to a single and particular individual thing,” or
“an assembly of similar things.”
It is worth mentioning in passing that an opponent paraphrases this role of dṛṣṭānta
as: “This exemplification depends on external thing[s]” (*bāhyārthāpekṣam idaṃ nidarśanam).19 Jinendrabuddhi paraphrases this as “pervasion that depends on external thing[s].”20
3. Examination of the instances in PS/PSV 1
As Katsura (2004) points out, there is still a possibility that the term bāhyārtha in bāhyārthapradarśana means “an object in the external world” and implies “some positive support
in external reality.”21 The expression bāhyārtha is used at least nine times in PS/PSV 1. An
examination of these instances (PSV 1 4, 8; 18, 13; 18, 25; 19, 2; 19, 6; 19, 11; 19, 15; 19,
16; 19, 18) reveals that all of them mean “an external object” or “an object in the external
world.”22 It is easily understood that the counterpart of bāhyārtha is “an internal thing,”
such as cognition, knowledge or the mind.
However, if the term bāhyārtha, which is used in logical contexts in PS/PSV 3 and
4, means “an external object,” we encounter certain difficulties. First, Dignāga uses the
example buddhi in his formulation of this inference: “Sound is permanent, because it is
intangible, like cognition.”23 It is obvious that this buddhi is not a so-called ‘external object.’
18
19
20
21
22
23
PSṬ D215a3–5; P244a8–3: de’i phyir spyis[spyi yis P] ji skad bshad pa’i dpe’i mtshan nyid can gyi
don gyi phung po thams cad dpe ste / bum pa kho na[na ni D] ma yin no // bum pa ni de’i nang du ’dus
pas te de[om. P] kho na nye bar mtshan pa’i don du dper rjod[brjod P] pa’o // ... phyi rol gyi don ston
pa ni nges par bstan par[par om. P] bya la gtso bo’o zhes pa’i tshig ni gsal ba’i khyad par gcig kho
na la rab tu ston pa gtso bo nyid yin pa sel ba lhur byed par blta bar bya’o // (Skt. B171b6–172a2:
tasmāt sāmānyena sarvo yathoktadṛṣṭāntalakṣaṇo ’rtharāśir dṛṣṭāntaḥ, na tu ghaṭa eva. ghaṭas tu
tadantargatas tasyaivopalakṣaṇārtham udāhriyate. ... bāhyārthapradarśanaṃ hi nidarśye pradhānam
iti vacanam ekatraiva vyaktiviśeṣe pradarśanasya prādhānyanirāsaparaṃ draṣṭavyam.) “Therefore the
example is a whole assembly of things in general that have the characteristic of the example as already
stated, but not just a pot. A pot, on the other hand, being included within it (= the assembly of things),
is given as an [actual] example for the purpose of implying the very [assembly of things]. ... It should
be understood that [Dignāga’s] statement: ‘For the primary [role] of exemplification is to indicate
external thing[s]’ is [made] for the purpose of negating [the view that] to indicate [pervasion] only
[in] a single particular individual is its primary [role].” See also Shiga 2011b: 527, n. 23.
PSV on PS 4.4 (K) 149b1–3; (V) D60b5f.; P64b2f.
PSṬ D217a6; P246b7f.: ’dir phyi rol gyi don la ltos pas ci zhig / khyab pa kho nas zhes pa gang du bsgrub
byas rjes su ’gro ba zhes pa’i mtshan nyid las so // (Skt. B173b7–174a1: kim atra bāhyārthāpekṣayā
vyāptyaiveti, sādhyenānugamo yatreti lakṣaṇāt.)
See Katsura 2004: 143; 155.
The expression bāhyārthopasaṃhāra is found in PSṬ 1 160, 7–9: yadi tāvad viṣaye pravartamānaṃ
mana indriyavṛttikṛtam anugrahaṃ nāpekṣate, evaṃ satīndriyāṇāṃ sarvathaivānarthakyaṃ syāt, manasaiva bāhyārthopasaṃhārāt puruṣasyopabhogasiddheḥ. In this case, bāhyārthopasaṃhāra means
“to comprehend/cover objects in the external world.”
PSV(K) P125b3f. (Kitagawa 1965: 474, 2–5): dper na sgra ni rtag[em. : mi rtag P/Kitagawa] ste / reg
bya ma yin pa’i phyir blo bzhin no // de bzhin du mig gi gzung bar bya ba yin pa’i phyir mi rtag zhes
bya ba ’di yang bsgrub byar bstan pa’i phyir dam bca’ ba thob po // PSV(V) D41a5f.; P43b8–44a1
(Kitagawa 1965: 474, 2–5): dper na sgra rtag ste / reg par bya ba yin pa’i phyir blo bzhin no zhes bya
ba dang / de bzhin du mig gis gzung bya yin pa’i phyir mi rtag ces bya ba ’di yang bsgrub bya bstan
416 The Meaning of bāhyārtha in Dignāga’s and Jinendrabuddhi’s Theories of Inference
Although the formulation that contains the example buddhi is presented as fallacious, this
is not because buddhi is not “an external object,” but because buddhi is a pseudo-example
that does not have the property to be proved (nityatva).24
Furthermore, if we take a close look at the expression bāhyārtha, the word artha itself
can mean “an actual thing” with some factual basis.25 So it is likely that the word bāhya is
used in the sense of “outside” or “external [to the subject]” and artha is used in the sense
of “actual example” or a thing that can be verbally expressed or named, i.e., in the sense of
padārtha.26
4. The expression bāhyārtha-upasaṃhāra or -anupasaṃhāra
4.1. A usage example found in the criticism of the Naiyāyika definition of exemplification (udāharaṇa)
Next, we shall examine cases where the word upasaṃhāra or anupasaṃhāra comes after
bāhyārtha.
[Even] if [an opponent states that] the example is qualified by the first half [of
the definition of udāharaṇa in NS 1.1.36],27 the similarity or dissimilarity to
the [property-bearer] to be proved does not need to be mentioned [as a part of
the definition of example], because:
[If] so, it (= the example) is not seen as what conveys the knowledge [of what
is to be proved]. (PS 4.19b)
To wit, when [the example] does not associate [the property to be proved] with
external thing[s that respectively have the proving property] (bāhyārthānupasaṃhāre), the example does not convey the knowledge of the property to be
proved. Alternatively, for that reason, [the necessity of mentioning similarity
or dissimilarity] is not established, [because similarity and dissimilarity are
24
25
26
27
pa’i phyir dam bca’ bar thal bar ’gyur ro // (Skt. reconstruction: *tad yathā nityaḥ śabdaḥ, asparśatvāt,
buddhivat, evam anityaḥ śabdaḥ, cākṣuśatvāc cety etad api sādhyanirdeśāt pratijñā prasajyate. Cf. NV
274, 4f.) Cf. NP 8, 9: sādhyadharmāsiddho yathā nityaḥ śabdo ’mūrtatvād buddhivat.
NP 8, 10f.: yad amūrtaṃ vastu tan nityaṃ dṛṣṭaṃ yathā buddhir iti. buddhau hi sādhanadharmo
’mūrtatvam asti, sādhyadharmo nityatvaṃ nāsti, anityatvād buddher.
Also cf. PV 1.26: tasmād vaidharmyadṛṣṭānte neṣṭo ’vaśyam ihāśrayaḥ / tadabhāve ca tan neti vacanād
api tadgateḥ // This is Dignāga’s view on the substratum (āśraya) of vaidharmyadṛṣṭānta, which is
indirectly quoted by Dharmakīrti. They both think that it is not always necessary for a dissimilar example
to have a substratum.
Cf. Hetumukha (?) sarva evāyam anumānānumeyavyavahāro buddhyārūḍhena dharmadharmibhedena,
[na bahiḥ sadasattvam apekṣate NVTṬ]. (Quoted in PVSV 2, 22–3, 1; NVTṬ 51, 11f.; 162, 28f.) See
also Frauwallner 1959: 164.
The entire Naiyāyika definition of udāharaṇa is as follows. NS 1.1.36–37: sādhyasādharmyāt taddharmabhāvī dṛṣṭānta udāharaṇam, tadviparyayād vā viparītam. “The exemplification is an [actual] example
that [is supposed to] have [another] property (i.e., the property to be proved) of it (= the property-bearer
to be proved) due to the [example’s] similarity (i.e., the proving property) to the [property-bearer] to be
proved. The counter[-example] is [an actual example that is not supposed to have another property (i.e.,
the property to be proved) of the property-bearer to be proved] due to its opposite (= the similarity to
what is to be proved) (i.e., the dissimilarity to what is to be proved).”
Kiyokuni Shiga
417
implied] just by [the qualifier:] “having the property of the [property-bearer to
be proved]” (taddharmabhāvitvena).28
This is found in the criticism of the Naiyāyika definition of the exemplification (udāharaṇa) in the paramata-section of PS/PSV 4. Dignāga asserts that if dṛṣṭānta “does not
associate [the property to be proved] with external thing[s that respectively have the proving
property],”29 such dṛṣṭānta cannot prove what is to be proved. This suggests that bāhyārthopasaṃhāra is an indispensable condition for dṛṣṭānta to prove what is to be proved. Just
before this assertion, he also states that if dṛṣṭānta is treated as separate from the content
of the logical reason, a fallacy would occur: a certain thing (e.g. space) could become
both a similar and dissimilar example because of its similarity (e.g. being existent) and
dissimilarity (e.g. being inaudible) to the property-bearer to be proved (e.g. sound).30
Jinendrabuddhi comments on the use of the words bāhyārtha and upasaṃhāra in the
above quotation as follows:
[The word] bāhyārtha in [the phrase] na hi bāhyārthānupasaṃhāre means
another thing that is different from mere similarity or dissimilarity. Furthermore, it (= bāhyārtha) [actually] means a pervasion or inseparable relation
(avinābhāvitva). The association (upasaṃhāra) of it (= bāhyārtha = avinābhāvitva) means applying (upanayana), i.e. manifesting (prakāśana) that
[bāhyārtha] is an object to be explained. When it (= upasaṃhāra) is not present,
the example is not what conveys the knowledge [of what is to be proved]. ...31
28
29
30
31
PSV(K) P153b6–8: gal te dpe snga mas khyad par du bya ba yin na bsgrub bya dang chos mthun pa
dang mi mthun pa smos par mi bya ste / gang gi phyir, de ltar go byed de ma mthong / (PS 4.19c) phyi
rol gyi don rjes su ma bsdus pa la dpes bsgrub bya’i chos rtogs pa ni yod pa ma yin pa’i phyir de’i
chos rtogs pa nyid du ma grub pa kho na yin no // PSV(V) D64a7–b1; P68a6–8: gzhan yang gal te
dpes[dpe yis P] sngon du khyad par du byed na ni bsgrub par bya ba’i chos dang mthun pa dang / chos
dang mi mthun par ’dzin pa yin[ma yin P] te / go byed du de mthong ma yin // (PS 4.19c) phyi rol
gyi don gyi rjes thogs su ma smos pa’i dpes ni bsgrub par bya ba’i chos la go bar bya ba yin no zhes
chos de rtogs pa nyid du yang mi ’grub po // (Skt. reconstruction: *yadi pūrveṇa dṛṣṭānto viśeṣyate, na
sādhyasādharmyavaidharmyagrahaṇaṃ kartavyam. yasmāt, naivaṃ sa gamako dṛṣṭaḥ (PS 4.19b) na
hi bāhyārthānupasaṃhāre dṛṣṭāntasya sādhyadharmagamaka[tvam] astīti taddharmabhāvitvenaiva
vāsiddham.)
Or “does not associate [the pervasion or inseparable relation of the proving property with the property
to be proved] with external thing[s],” according to Jinendrabuddhi.
PSV(K) 153b5f.: ’di la yang gal te gtan tshigs kyi don las dpe tha dad pa yin na dpe gtan tshigs kyi don
la ’brel pa nyid du gdon mi za bar brjod par bya bar mi ’gyur ro // de lta na yang byas pa’i phyir mi
rtag ste nam mkha’ bzhin zhes bya ba yang dper ’gyur te yod pas yod pa’i phyir dang / mnyan bya ba
ma yin pa’i phyir zhes bya ba ’di la bsgrub par bya va dang chos mthun pa dang mi mthun pa yod pa yin
no // PSV(V) D64a6f.; P68a5f.: de yang[de yang om. P] gal te gtan tshigs kyi don las dpe logs shig pa
yin na ni dpe gtan tshigs kyi don dang rjes su ’brel par brjod par mi bya bar ’gyur ro // de bzhin du byas
pa’i phyir na mi rtag pa ste / nam mkha’ bzhin no zhes bya bar yang dper ’gyur ro // de’i phyir bsgrub
par bya ba’i chos dang mi mthun pa dang / chos dang mthun pa ni yod pa dang mnyan bya nyid dag la
sogs pa la yang yod pa[yod pa om. P] yin no // (Skt. reconstruction: *atrāpi yadi hetvarthāt pṛthag
dṛṣṭāntaḥ, na dṛṣṭānto hetvarthānugata evāvaśyaṃ vācyaṃ syāt. tathā ca kṛtakatvād anitya ākāśavad
ity api dṛṣṭāntaḥ syāt. asti hy asya sādhyena sādharmyaṃ vaidharmyaṃ ca sattvāśrāvaṇatvādi.)
PSṬ D230a2f.; P260b4–6 (= Appendix [4]): phyi rol gyi don nye bar ma bsdus pa zhes pa chos mthun
pa nyid dang chos mi mthun pa nyid tsam las gzhan pa’i don ni phyi rol gyi don te / de yang khyab
pa ste / med na mi ’byung ba nyid do zhes pa’i tha tshig go // de’i nye bar bsdus pa ni rtogs par bya
418 The Meaning of bāhyārtha in Dignāga’s and Jinendrabuddhi’s Theories of Inference
Jinendrabuddhi understands bāhyārtha as “another thing that is different from mere similarity or dissimilarity,”32 then regards it as “pervasion” or “inseparable relation.” These
interpretations are unique, in that he takes bāhya in the sense of “different” and then
identifies it as pervasion.33
In any case, regarding the word bāhyārtha, Jinendrabuddhi calls attention to the fact that
dṛṣṭānta does not consist of the combination of an actual example (e.g. a pot) and its mere
similarity (e.g. being produced)34 or dissimilarity (e.g. not being produced) to the subject
(e.g. sound), but instead the combination of dṛṣṭāntadharmin and the inseparable relation
of a logical reason (e.g. being produced) with what is to be proved (e.g. impermanence).
4.2. Usage examples regarding the Naiyāyika definition of the application (upanaya)
Here we will examine the uses of bāhyārtha and bahiḥ found in the criticism35 against the
Naiyāyika definition of the application (upanaya).36
32
33
34
35
36
ba’i yul nyid du nye bar sbyar ba ste / gsal bar byed pa zhes pa’i don to // de med na dpe la go bar
byed pa nyid med do zhes pa ngag[dag P] gi don to // (Skt. B185a1–3: na hi bāhyārthānupasaṃhāra
iti sādharmyavaidharmyamātrād anyo ’rtho bāhyārthaḥ. sa punar vyāptir avinābhāvitvam iti yāvat.
tasyopasaṃhāraḥ pratipādyaviṣayatvopanayanaṃ prakāśanam ity arthaḥ. tasminn asati nāsti dṛṣṭāntasya gamakatvam iti vākyārthaḥ.) Cf. HBṬ 62, 18–20: sarvasmin sādhanadharmavati dharmiṇi na
dṛṣṭāntadharmiṇy eva sādhyadharmasyopasaṃharaṇam upasaṃhāro ḍhaukanaṃ ...
This interpretation seems to reflect Dignāga’s preceding assertion that the part of the Naiyāyika definition
of exemplification sādhyasādharmyāt is useless, because an example defined as such does not convey
the knowledge of the property to be proved. Cf. PSṬ D232b1f.; P263a7: chos mthun nyid kyang mi
rigs ’gyur / zhes[shes P] khyad par med par brjod pa’i phyir / (Skt. sādharmyaṃ ca na yujyata ity
aviśeṣaṇābhidhānāt.)
In other words, Jinendrabuddhi here equates ‘the place’ where the association is made (= bāhyārtha)
with ‘the object’ of the associaion (= vyāpti). And there is another passage in the PSṬ where bāhyārtha
is interpreted as pervasion. PSV(K) P154a6f.: gal te khyod kyis bsgrub bya dang sgrub byed dang
ldan pa dpe’i don yin na gnyi ga’i chos dang ldan du zin kyang phyi rol gyi don nye bar ma bsdus
na bsgrub bya’i don rtog par nus pa ma yin no zhes bshad zin to // PSV(V) D64b6; P68b5f.: gal te
khyod kyis bsgrub par bya ba dang sgrub par byed pa’i chos dang ldan pa’i dpe las don yin na ni
gnyis ka’i chos dang ldan pas kyang ngag gi don mjug bsdus pa las bsgrub par bya ba’i don rtogs par
mi nus so zhes sngar brjod zin to // (Skt. reconstruction: *yadi bhavatāṃ sādhyasādhanadharmavatā
dṛṣṭāntenārthaḥ, na tūbhayadharmavatāpi bāhyārthānupasaṃhāreṇa sādhyo ’rthaḥ śakyo bhāvayitum
ity uktam.) PSṬ D231b1; P262a5: khyab pa gsal bar byed pa ’di tsam med par bsgrub par bya ba’i don
rtogs par nus pa ma yin no zhes sngar bshad zin to // (Skt. B186a4: na tv etāvatā vyāptiprakāśanam
antareṇa sādhyo ’rthaḥ śakyo gamayitum ity uktaṃ prāk.)
For the meaning of sādharmya, see NBṬ 152, 8–10 on NB 3.5: samāno dharmo ’sya so ’yaṃ sadharmā. tasya bhāvaḥ sādharmyam ... dṛṣṭāntadharmiṇā saha sādhyadharmiṇaḥ sādṛśyaṃ hetukṛtaṃ
sādharmyam ucyate.
The outline of Dignāga’s criticism is as follows. According to the Naiyāyika definition of upanaya,
the property as a logical reason, such as “being produced” (kṛtakatva), is applied to the subject, such
as “sound.” In that case, it follows that the property kṛtakatva could be applied as either a common
property or a particularized property. However, fallacies would occur in both cases. It is not possible to
apply all kinds of common properties of the example to the subject. (PSV(K) 154b5, PSV(V) D65a4f.;
P69a3–5)
NS 1.1.38: udāharaṇāpekṣas tathety upasaṃhāro na tatheti vā sādhyasyopanayaḥ. “The application is
to associate [the property of the example, i.e., the proving property] with what is to be proved in relation
Kiyokuni Shiga
419
On the other hand,37 the application [of the property to be proved and/or the
proving property] to [a dissimilar example such as] space [in the inference:
“Sound is impermanent, because it is produced”] by depending on external
thing[s] is correct, because, unlike space [and so on], there is nothing that is
produced [and] permanent38 outside [the subject, i.e. sound], whereas, like
a pot [and so on], there is something [produced] and impermanent [outside
the subject, i.e. sound]. Therefore [sound that is produced] is impermanent.
Hence, [the application] is [valid only when it is made] with regard to certain
external case[s].39 ...40
As we have already seen, both Dignāga and Jinendrabuddhi recognize “associating [the
property to be proved or inseparable relation] with external thing[s]” (bāhyārthopasaṃhāra) as an essential factor of dṛṣṭānta for a valid inference. And here “external thing[s]”
(bāhyārtha) are restated as being “outside” (bahiḥ), and space and a pot are given as actual
examples. According to Jinendrabuddhi, common absence (vyatireka) is indicated in space,
whereas common occurrence (anvaya) is indicated in a pot. And the word bahiḥ is men-
37
38
39
40
to the examples: [the subject] is like [the similar example] or [the subject] is unlike [the dissimilar
example].” It should be noted in passing that Viśvanātha glosses the word upasaṃhāra as “placing near”
(upanyāsa) in his Vṛtti 313, 26 on NS 1.1.38. Cf. YD 91, 3: sādhyadṛṣṭāntayor ekakriyopasaṃhāra
upanayaḥ. “The association, i.e., the application is to equate an example with what is to be proved.”
The preceding passage reads as follows: PSV(K) 155b3f.: mi rtag pa nyid kyi gtan tshigs lus can nyid
la sogs pa yang med pa’i phyir mi rtag pa nyid ’gog par yang ma gyur cig snyam nas / gang las de
med pas rtag par mi ’gyur ba ma byas pa kho na rtag pa nyid kyi gtan tshigs ni ma yin no // PSV(V)
D66a1; P69b8–70a1.: ma byas pa nyid ni rtag pa nyid kyi gtan tshigs ma yin te / gang de med pa las
rtag par mi ’gyur ro // mi rtag pa nyid kyi gtan tshigs lus can ma yin pa la sogs pa yang med pa’i
phyir mi rtag pa nyid ’gegs pa yang ma yin no // (Skt. reconstruction: *mā bhūd mūrtatvāder apy
anityatvahetor abhāvād anityatvapratiṣedho ’pi. na hy akṛtaka eva nityatvahetuḥ, yataḥ tadabhāvān
na nityaḥ syāt.) “Also regarding [the logical reason] having a fixed form (mūrtatva) and the like, it is
not [possible] to negate impermanence either, because [mūrtatva] is not a [valid] logical reason for
[proving] impermanence, for [the fact of] not being produced [would] never be a [valid] logical reason
for [proving] permanence; otherwise (= if the fact of not being produced were a valid logical reason for
proving permanence) what is permanent would not be present for the reason that it (= the fact of not
being produced) is not present.”
PSṬ D236a3; P267a4f.: de nyid kyi phyir gsungs pa / rtag pa nyid byas pa ni ’ga’ yang yod pa ma
yin no zhes pa ste / mi rtag pa nyid dang bral ba’i byas pa yod pa ma yin no zhes pa’i don to // (Skt.
B190a3f.: na hi kiṃcin nityaṃ kṛtakam astīti na kiṃcid anityatvarahitaṃ kṛtakam asti.)
PSṬ D236a4f.; P267a6f.: rjes ’gro dang ldog pa la ltos[bltos P] nas / de bstan pa’i ched du nam
mkha’ bzhin dang bum pa bzhin zhes phyogs gcig dper brjod pa tsam nye bar sbyar ba’o zhes pas kho
bo’i lugs kyis nam mkha’ sogs nye bar sbyar ba[ba’i P] rigs so // (Skt. B190a4f.: anvayavyatirekaṃ
cāpekṣya tatpradarśanārtham ākāśavad ghaṭavac cety ekadeśa udāharaṇamātram upanīyata iti yukto
’smanmatenākāśādyupanayaḥ.)
PSV(K) 155b4–6: phyi rol gyi don la bltos nas kyang nam mkha’ dper byed pa gang gi phyir phyi rol
na nam mkha’ bzhin du byas pa rtag pa ni med la bum pa bzhin du mi rtag pa yod la de’i phyir mi rtag
go zhes bya ba ’ga’ zhig gi ni phyi rol na yod pa yin no // nye bar sbyor ba yang de ltar mi rigs pa yin
no // PSV(V) D66a1–3; P70a1–3: nam mkha’ ni ngag gi don la ltos nas nye bar sbyar bar rigs te gang
gi phyir ma byas pa las phyi rol du[tu P] gyur pa’i rtag pa ni cung zhig kyang yod pa ma yin te / nam
mkha’ bzhin no // mi rtag pa la yang bum pa bzhin no zhes bya ba yin no // de’i phyir mi rtag pa’o zhes
bya ba nyid kyis nye bar sbyar bar yang rigs pa ma yin no // (Skt. reconstruction: *bāhyārthāpekṣayā tv
ākāśasyopanayo yuktaḥ [ākāśam udāhriyate K], yasmād bahir ākāśavat, na hi kiṃcin nityaṃ kṛtakam
asti, anityaṃ cāsti ghaṭavat. tasmād anityam iti kvacid bahir asti. evam upanayo ’py ayuktaḥ.)
420 The Meaning of bāhyārtha in Dignāga’s and Jinendrabuddhi’s Theories of Inference
tioned “for the purpose of negating the dependence on the mere [fact that the property to
be proved and/or the proving property] is present only in a pot and absent only in space,”41
which implies that bāhyārtha does not refer to a single case, but plural cases. Indeed, on
the basis of this passage alone, we cannot decide whether bāhyārtha means “an object in
the external world” or “external things outside [the subject],” but at least Jinendrabuddhi’s
comments consistently support the meaning of “external thing[s] outside [the subject].”42
5. bāhyārtha and the term and notion of bahirvyāpti
As I observed in Shiga 2011b, the expression anyatra, which means “in another place
[than the location of the property-bearer to be proved],” is found in PS 2.11.43 Dignāga
himself comments on this word, presenting a remarkable idea. He introduces the concept
of “the generalized substratum” (ādhārasāmānya)44 .45 Jinendrabuddhi glosses the word
anyatra as “generally (sāmānyena) in all cases, e.g. a kitchen.”46 He also uses the words
anyatra and sāmānya when he comments on the expression bāhyārthopasaṃhṛta in PSṬ
3.47 From this fact, we can safely say that Jinendrabuddhi, for one, sees the terms bāhyārtha
and anyatra as referring to substantially the same thing: that is, property-bearer[s] of the
example outside the subject. It is worth noting that he also states that a logical reason (e.g.
smoke) is inseparably related not only with the property to be proved (e.g. fire), but also
with the generalized substratum (e.g. any place that has fire), because a property-bearer
depends on its property, and the property to be proved is particularized by the generalized
substratum.48
41
42
43
44
45
46
47
48
PSṬ D236a2f.; P267a4 (= Appendix [8]): phyi rol smos pa ni ’ba’ zhig pa’i bum pa dang / nam mkha’
yod pa dang med pa nyid tsam la ltos[bltos P] pa sems[sem pa P] lhur byed pa’o // (Skt. B190a3:
kevalaghaṭākāśasadasattvamātrāpekṣatvanirāsaparaṃ bahirgrahaṇam.)
Cf. PSṬ D185b2–4; P211a7f. (Skt. B146a1–3).
PS 2.11: liṅgasyāvyabhicāras tu dharmeṇānyatra darśyate / tatra prasiddhaṃ tadyuktaṃ dhamiṇaṃ
gamayiṣyati // See also Shiga 2011b: 528–532.
This is paraphrased as “the generalized property-bearer” (dharmisāmānya) by Jinendrabuddhi.
PSV(K) P112b5 (Kitagawa 1965: 461, 11f.): spyi dang ’di bstan pa nyid yin te gang na du ba yod pa de
na me yod do zhes rab tu bstan phyir ro // PSV(V) D30b1; P31b1 (Kitagawa 1965: 461, 11f.): gzhi
thun mong ba nyid du bstan pa ste gang na dud pa yod pa de na me yod[de na med P] do zhes bstan
pa’i phyir ro // (Skt. reconstruction: *ādhārasāmānyena tu pradarśita eva, yatra dhūmas tatrāgnir iti
pradarśanāt.)
PSṬ 2 45, 6: anyatreti sāmānyena sarvatra mahānasādau. Jinendrabuddhi glosses the word anyatra in
the PSV as “the generalized place that has smoke” (PSṬ 2 45, 10: dhūmavatpradeśasāmānye).
PSṬ D185b2–4; P211a7f. (= Appendix [1]): phyi rol gyi don nye bar bsdus pas zhes pa / phyogs su
byas pa’i chos can gyi khyad par las gzhan du spyi la nye bar bstan pas zhes pa’i don to // spyi yang
khyad par yongs su mi spong pa’i phyir / bsgrub par bya ba’i chos can yang der nang du ’dus pa kho na
ste / bsgrub bya’i chos can kho na la med na mi ’byung ba nyid ston pa bsal ba lhur len pa ni phyi rol
gyi don smos pa’o // (Skt. B146a1–3: bāhyārthopasaṃhṛteneti. dharmiṇaḥ pakṣīkṛtād viśeṣād anyatra
sāmānya upadarśitenety arthaḥ. sāmānyasya ca viśeṣāparityāgāt, sādhyadharmy api tatrāntargata eva.
sādhyadharmiṇy evāvinābhāvitvapradarśananirāsaparaṃ tu bāhyārthagrahaṇam.)
PSṬ 2 46, 4–9: ādheyenāvinābhāvitve darśyamāna ādhāreṇāpi darśitaṃ bhavati. ādheyatantratvād
ādhārasya. atha vā yatra dhūmaḥ, tatrāgnir iti nānena dhūmasyāgnimātreṇāvinābhāvitvaṃ kathyate,
kiṃ tarhi, ādhārasāmānyaviśiṣṭena. tasmād ādhārasāmānyenāpi tad darśitam eva. “When it is indicated that [a logical reason] is inseparably related with what is to be contained, it follows that [the
inseparable relation of the logical reason] with the container (i.e. the generalized substratum) is also
Kiyokuni Shiga
421
If these observations thus far are valid, it follows that bāhya in the term bāhyārtha and
bahir- in the term “external pervasion” (bahirvyāpti) have the same logical value. Dignāga
is sometimes described as a “typical” bahirvyāptivādin by modern scholars, presumably
because he constructed the system of inference, especially the theory of trairūpya, which
requires presenting the similar (sapakṣa) and the dissimilar (vipakṣa) as separate from
the subject (pakṣa).49 Nevertheless, Dignāga does not declare himself a proponent of
bahirvyāpti. In Buddhist treatises on logic and epistemology, the term bahirvyāpti first
appeared in Arcaṭa’s HBṬ 62, 9; 62, 23; 62, 27.50 Śāntarakṣita uses both bahirvyāpti
and antarvyāpti in his VNṬ 5, 30–6, 7. It should be noted here that Arcaṭa and others
understand bahirvyāpti as “pervasion that is indicated ‘only’ outside the property-bearer
to be proved, that is, ‘only’ in a property-bearer of the example,” but this is different
from Dignāga’s view. Dignāga does not state that pervasion is indicated ‘only’ outside the
property-bearer to be proved. Rather, he devised the notion of “the generalized propertybearer” (ādhārasāmānya), which can implicitly include the property-bearer to be proved
(sādhyadharmin, pakṣa).51
Pātrasvāmin (7–8 cent.),52 a Jaina logician who is considered to have been the first to
advocate the theory of the single characteristic (ekalakṣaṇa) of a logical reason, which
is virtually identical to the theory of antarvyāpti,53 states that the inseparable relation
(avinābhāvitva) asserted by Buddhists is acknowledged outside sādhya or in dṛṣṭānta,
whereas “being otherwise impossible” (anyathānupapannatva) is only acknowledged
in the property-bearer to be proved.54 Pātrasvāmin’s view was followed by other Jaina
49
50
51
52
53
54
indicated, because a container depends on what is to be contained. Alternatively, it is not stated by this
[formulation:] ‘Wherever there is smoke, there is fire,’ that smoke is inseparably related [only] with
mere fire, but that [the smoke is inseparably related] with the [fire] that is qualified by the generalized
substratum. Therefore, that (= the inseparable relation) [of the logical reason] with the generalized
substratum is actually indicated.”
We can describe the relationship between trairūpya and bāhyārthopasaṃhāra as follows: to confirm the
three characteristics of a logical reason (hetu), i.e., its being the property of the subject, its being present
in sapakṣa (which is similar to the subject in that it has the property to be proved [sādhyadharma]) and
its not being present in vipakṣa (which is dissimilar to pakṣa because it does not have sādhyadharma),
means to check whether the logical reason is valid, when or after the inference in question is made or
formulated. This could be called “the process of applying the logical reason to those cases other than
the subject” or “the process of inference from the perspective of hetu.” bāhyārthopasaṃhāra, on the
other hand, means to associate/apply the property to be proved (sādhyadharma) with/to external things
(= dṛṣṭāntadharmin), having the proving property (sādhanadharma) in the exemplification (dṛṣṭānta)
as a member of proof, when or after the inference in question is made or formulated. This could be
called “the process of applying the property to be proved to those cases other than the subject” or “the
process of inference from the perspective of sādhyadharma.”
Cf. HBṬ(V) D117a3f.; P145b2f.: ji ltar gzhan dag gis bsgrub par bya ba’i chos can yongs su spangs
nas khyab pa gzhan rab tu ston pa lta bu ni ma yin te / thams cad smos pas[pa P] ni khyab pa gzhan
dgag pa’i phyir ro // (For this passage, see also Funayama 2001.)
See Shiga 2011b: 528–533 and Matilal 1985: 129, n. 1.
For the detail of Pātrasvāmin, see Shiga 2011a: 423–426.
See Shiga 2011a: 425f.
TSP 500, 13f. and PSṬ 2 2, 13f.: vinā sādhyād adṛṣṭasya dṛṣṭānte hetuteṣyate / parair mayā punar
dharmiṇy asambhūṣṇor vināmunā // PSṬ 2 2, 10f.: avinābhāvitvaṃ hi sādhyād bahir iṣyate, anyathānupapannatvaṃ tu dharmiṇy eva sādhya eveti; 2 3, 4: avinābhāvitvaṃ hi paraiḥ sādhyād bahir iṣyata
iti, eṣo ’dhyāropaḥ. Cf. HBṬ 62, 27–63, 2: na hi sa śyāmaḥ, tatputratvāt, paridṛśyamānaputravad iti
422 The Meaning of bāhyārtha in Dignāga’s and Jinendrabuddhi’s Theories of Inference
logicians such as Akalaṅka.55 They regarded themselves as proponents of antarvyāpti and
the Buddhists as proponents of bahirvyāpti.56 Therefore, it seems reasonable to suppose
that Arcaṭa and other successors of Dharmakīrti did not intend to criticize Dignāga’s view
directly, but rather to point out that Dignāga was allegedly credited with creating the theory
of the so-called bahirvyāpti.
6. Conclusion
We can safely state that, as long as we do not disregard Jinendrabuddhi’s annotations,57
when Dignāga uses the term bāhyārtha in connection with upasaṃhāra, pradarśana or
apekṣā in the context of inference, he does not mean “an external object” or “an object in
the external world,”58 but “external thing[s] outside the subject.” It is possible to suppose
that Dignāga was well aware of the division between the inside and outside of the subject
or property-bearer to be proved.
More specifically, the term bāhyārtha found in PS/PSV 3 and 4 refers to (1) “[external]
things, that is, the generalized [property-bearer], which is other than a particular [propertybearer] that is made to be the subject” (PSṬ B146a2, Appendix [1]); (2) “the generalized
[property-bearer] that is similar in kind to what is to be proved, which is external to a
particular [property-bearer] such as a pot” (PSṬ B171b5, Appendix [2]); (3) “an [external]
thing that is different from mere similarity or dissimilarity,” that is, “pervasion or inseparable
relation” (PSṬ B185a2, Appendix [4]); (4) more than simply a single example such as a pot
or space (PSṬ B190a3, Appendix [8]). Apart from (3), Jinendrabuddhi’s interpretations do
not seem to deviate from Dignāga’s original intention.
Due to his trairūpya-theory, Dignāga is often regarded as the proponent of bahirvyāpti.
What he intended to maintain, however, might not be the so-called position of bahirvyāpti
that means, according to Arcaṭa, the pervasion that is to be grasped or observed ‘only’
in a property-bearer of the example, that is, outside the subject. Rather, it is probable
that Dignāga, like Dharmakīrti, assumed some sort of generality or universality regarding
pervasion and its substratum.
55
56
57
58
tatputratvasya śyāmatvena sādhyād bahiḥ paridṛśyamānaputre vyāptipradarśane ’pi sādhyasiddhir
bhavati.
See SVin 5.15cd: antarvyāptāv asiddhāyāṃ bahirvyāptir asādhanam // PSaṃ 32cd: avinābhāvasambandhe ’py antarvyāptyāvatiṣṭhate // PSaṃ 50cd: antarvyāptāv asiddhāyāṃ bahiraṅgam anarthakam //
Cf. NA 20: antarvyāptyaiva sādhyasya siddher bahirudāhṛtiḥ / vyarthā syāt tadasadbhāve ’py evaṃ
nyāyavido viduḥ // See also NAVi 401, 10f. on bahirudāhṛtiḥ: bahir vivakṣitadharmiṇo ’nyatra dṛṣṭāntadharmiṇy udāhṛtir vyāptidarśanarūpā ...
Indeed Jinendrabuddhi’s interpretation may have been influenced by Dharmakīrti’s thoughts, but it
is also problematic to think that Jinendrabuddhi always comments on the PS/PSV by adopting and
following Dharmakīrti’s theories. There may be cases where Jinendrabuddhi’s comments conform to
Dignāga’s original intention. (Cf. Steinkellner 2004: 227f. with n. 5.)
We cannot find passages that positively support the meaning of “an object in the external world” in
Dignāga’s statements on inference either.
Kiyokuni Shiga
423
As for pervasion, on the other hand, Dignāga asserts the logical predominance of
the common absence (vyatireka) over the common occurrence (anvaya),59 and states that
vyatireka can be grasped or understood by mere non-observation (adarśanamātra).60 Indeed,
the pervasion based on Dignāga’s system appears to be hypothetical, because pervasion
would not be valid if even a single counter-example is found, but it is likely that he thought
that pervasion should be universal once it was established.61 We can assume that Dignāga
thought that this universal pervasion should be applied to external cases outside the subject.
In addition, since the subject is supposed to be implicitly included in those external cases,
it is possible to apply the pervasion to the subject and infer what is to be proved.
Table 1: Dignāga’s usage of bāhyārtha in logical contexts, with Jinendrabuddhi’s
comments (PS/PSV/PSṬ 3 and 4)
No.
Dignāga
[1]
PSV on PS 3.36b:62 tādṛksāhabhavyaṃ
hi sādharmyeṇa vaidharmyeṇa vā bāhyārthopasaṃhṛtena dṛṣṭāntadvayena
gamyate. (PSV(K) P138a5–7 ; PSV(V)
D51a6–b1; P54b5–7)
[2]
59
60
61
62
Jinendrabuddhi
PSṬ B146a1–3: darmiṇaḥ pakṣīkṛtād viśeṣād anyatra sāmānya upadarśitenety
arthaḥ. sāmānyasya ca viśeṣāparityāgāt, sādhyadharmy api tatrāntargata
eva. sādhyadharmiṇy evāvinābhāvitvapradarśananirāsaparaṃ tu bāhyārthagrahaṇam.
PSV on PS 4.3: bāhyārthapradarśa- PSṬ B171b5–6: bāhyārthapradarśanaṃ hi nidarśye pradhānam. (PSV(K) naṃ hīti ghaṭāder viśeṣāt sādhyajātīP148b6; PSV(V) D60b2f.; P64a7)
yasāmānyaṃ bāhyo ’rthaḥ, tatra sāmānyena yan nāma kiṃcit kṛtakaṃ tat sarvam anityam eveti sādhyena hetor vyāptiprakāśanaṃ bāhyārthaprakāśanam.
See also PSṬ B171b6–172a2: tasmāt
sāmānyena sarvo yathoktadṛṣṭāntalakṣaṇo ’rtharāśir dṛṣṭāntaḥ, na tu ghaṭa
eva; PSṬ B171b6–172a2: bāhyārthapradarśanaṃ hi nidarśye pradhānam
iti vacanam ekatraiva vyaktiviśeṣe pradarśanasya prādhānyanirāsaparaṃ draṣṭavyam.
PSṬ 2 47, 16–18: sarvatrānagnau na dṛṣṭa ity anena vyatirekasya prādhānyaṃ vipakṣe sarvatrādarśanena khyāpayati. anyatrāpi ca dṛṣṭa iti. apiśabdena kvacin na dṛṣṭo ’pīti dyotayann anvayasyāprādhānyam. (See also Pind 2015: II 232.)
See PS/PSV 5.34 and Pind 2011: 70–73.
For the interpretation of adarśanamātra, see Kataoka 2011: 195 with n. 12f. and Pind 1999: 323–330.
Words in italics are not attested in the PSṬ or other texts.
424 The Meaning of bāhyārtha in Dignāga’s and Jinendrabuddhi’s Theories of Inference
Table continued
No.
Dignāga
Jinendrabuddhi
[3]
PSV on PS 4.4: bāhyārthāpekṣam idaṃ
nidarśanam. (PSV(K) P149b3; PSV(V)
D60b6; P64b3)
[4]
PSV on PS 4.19: na hi bāhyārthānupasaṃhāre dṛṣṭāntena sādhyadharmo
gamyate. (PSV(K) P153b7; PSV(V)
D64b1; P68a7)
PSV on PS 4.20: yadi bhavatāṃ sādhyasādhanadharmavatā dṛṣṭāntenārthaḥ,
na tūbhayadharmavatāpi bāhyārthānupasaṃhāreṇa sādhyo ’rthaḥ śakyo
bhāvayitum ity uktam. (PSV(K) P154a6;
PSV(V) D64b6; P68b5f.)
No direct comment. Cf. PSṬ B173b6:
na hi sadbhāvamātradarśane bāhyārthāpekṣatvaṃ nidarśanasya yujyate;
PSṬ B173b6–174a1: tataś ca ghaṭe ’nityatvaṃ prayatnānantarīyakatvaṃ ca
dṛṣṭam iti kṛte hetoḥ sapakṣe gamyata
eva sadbhāvamātram iti kim atra bāhyārthāpekṣayā vyāptyaiveti ...
PSṬ B185a1–3: na hi bāhyārthānupasaṃhāra iti sādharmyavaidharmyamātrād anyo ’rtho bāhyārthaḥ. sa punar
vyāptir avinābhāvitvam iti yāvat.
No direct comment. Cf. PSṬ B186a3–4:
na tv etāvatā vyāptiprakāśanam antareṇa sādhyo ’rthaḥ śakyo gamayitum
ity uktaṃ prāk; PSṬ B186a4–6: na hi
vastudharmavatāpi dṛṣṭāntena vinā bāhyārthopasaṃhāreṇa sādhyo ’rthaḥ śakyo bhāvayitum.
No direct comment.
[5]
[6]
[7]
[8]
PSV on PS 4.20: na ca dṛṣṭāntasādhyayor ekīkaraṇād evaṃ anityatvaṃ siddham, bāhyārthānupasaṃhārāt. (PSV(K) P154b8; PSV(V) D65a6f.;
P69a6)
PSV on PS 4.20: bāhyārthāpekṣayā
tv ākāśasyopanayo yuktaḥ. (PSV(K)
P155b4f.; PSV(V) D66a1f.; P70a1f.)
PSV on PS 4.20: yasmād ākāśavad bahiḥ, na hi kiṃcin nityaṃ kṛtakam asti,
anityaṃ cāsti ghaṭavat. tasmād anityam
iti kvacid bahir asti. (PSV(K) P155b5f.;
PSV(V) D66a2; D70a2f.)
No direct comment. See the example [8].
PSṬ B190a2–3: kathaṃ bāhyārthāpekṣa ity āha: yasmād ityādi.
kevalaghaṭākāśasadasattvamātrāpekṣatvanirāsaparaṃ bahirgrahaṇam.
See also PSṬ B190a3–5: anvayavyatirekaṃ cāpekṣya tatpradarśanārtham
ākāśavad ghaṭavac cety ekadeśa
udāharaṇamātram upanīyate.
References and abbreviations
Primary Sources
HB Hetubindu (Dharmakīrti): Ernst Steinkellner, Dharmakīrti’s Hetubinduḥ, Teil I: Tibetischer Text und rekonstruierter Sanskrit-Text. Vienna 1967.
Kiyokuni Shiga
425
HBṬ Hetubinduṭīkā (Arcaṭa): Hetubinduṭīkā of Bhaṭṭa Arcaṭa with Sub-Commentary
Entitled Āloka of Durveka Miśra, ed. S. Sanghavi and Jinavijayaji. Baroda 1949.
HBṬ(V) Hetubinduṭīkā (Vinītadeva): (Tib.) D4234, P5733.
NA Nyāyāvatāra (Siddhasena Divākara): Jaina epistemology in historical and comparative
perspective: critical edition and English translation of logical epistemological treatises: Nyāyâvatāra, Nyāyâvatāra-vivṛti and Nyāyâvatāra-ṭippana with introduction
and notes, ed. P. Balcerowicz. 2 vols. Stuttgart 2001.
NAVi Nyāyāvatāravivṛti (Siddharṣi): See NA.
NB Nyāyabindu (Dharmakīrti): See NBṬ
NBṬ Nyāyabinduṭīkā (Dharmottara): Paṇḍita Durvekamiśra’s Dharmottarapradīpa [Being a sub-commentary on Dharmottara’s Nyāyabinduṭīkā, a commentary on Dharmakīrti’s Nyāyabindu], ed. Dalsukhbhai Malvania. Patna 2 1971.
NMukh Nyāyamukha (Dignāga): See Katsura 1981.
NP Nyāyapraveśakasūtra (Śaṅkarasvāmin): Nyāyapraveśakaśāstra of Baudh Ācārya Diṅnāga (The father of the Buddhist Logic). With the commentary of Ācārya Haribhadrasūri and with the subcommentary of Pārśvadevagaṇi, ed. Muni Jambuvijaya.
Delhi/Ahmedabad 2009.
NS Nyāyasūtra (Gautama): Nyāyadarśanam with Vātsyāyana’s Bhāṣya, Uddyotakara’s
Vārttika, Vācaspati Miśra’s Tātparyaṭīkā and Viśvanātha’s Vṛtti, ed. Taranatha
Nyaya-Tarkatirtha and Amarendramohan Tarkatirtha. Calcutta 1936.
NV Nyāyavārttika (Uddyotakara): See NS.
NVTṬ Nyāyavārttikatātparyaṭīkā (Vācaspatimiśra): See NS.
PS Pramāṇasamuccaya (Dignāga): See PSV.
PS/PSV 1 Pramāṇasamuccaya/Pramāṇasamuccayavṛtti (Dignāga), chapter 1: See Steinkellner 2005.
PSaṃ Pramāṇasaṃgraha (Akalaṅka): Akalaṅka Granthatrayam (Svopajñavivṛtisahitam
Laghīyastrayam, Nyāyaviniścayaḥ, Pramāṇasaṅgrahaś ca) of Śrī Bhattākalaṅkadeva,
ed. Mahendra Kumar Shastri. Ahmedabad/Calcutta 1939.
PSṬ Viśālāmalavatī Pramāṇasamuccayaṭīkā (Jinendrabuddhi): (Tib.) D4268; P5766.
PSṬ 1 Pramāṇasamuccayaṭīkā (Jinendrabuddhi), chapter 1: Jinendrabuddhi’s Viśālāmalavatī Pramāṇasamuccayaṭīkā. Chapter 1. Part I: Critical Edition, ed. E. Steinkellner,
H. Krasser, and H. Lasic. Beijing/Vienna 2005.
PSṬ 2 Pramāṇasamuccayaṭīkā (Jinendrabuddhi), chapter 2: Jinendrabuddhi’s Viśālāmalavatī Pramāṇasamuccayaṭīkā. Chapter 2. Part I: Critical Edition, ed. H. Lasic, H.
Krasser, and E. Steinkellner. Beijing/Vienna 2012.
PSV(K) Pramāṇasamuccayavṛtti (Dignāga) translated by Kanakavarman: (Tib.) P5702.
See also Kitagawa 1965: 440–579.
PSV(V) Pramāṇasamuccayavṛtti (Dignāga) translated by Vasudhararakṣita: (Tib.) D4204;
P5701. See also Kitagawa 1965: 440–579.
PV Pramāṇavārttika (Dharmakīrti): See PVSV.
PVSV Pramāṇavārttikasvavṛtti (Dharmakīrti): The Pramāṇavārttikam of Dharmakīrti.
The First Chapter with the Autocommentary, ed. R. Gnoli. Roma 1960.
SVin Siddhiviniścaya (Akalaṅka): Siddhiviniścayaṭīkā of Anantavīryāchārya: The commentary on Siddhiviniścaya and its vṛtti of Bhaṭṭa Akalaṅkadeva, ed. Mahendra
Kumar Jain. 2 vols. Varanasi 1959.
426 The Meaning of bāhyārtha in Dignāga’s and Jinendrabuddhi’s Theories of Inference
TSP Tattvasaṃgrahapañjikā (Kamalaśīla): Tattvasaṅgraha of Ācārya Shāntarakṣita with
the Commentary ‘Pañjikā’ of Shri Kamalashīla, ed. S. D. Shastri. 2 vols. Varanasi
1968.
VNṬ Vādanyāyaṭīkā (Śāntarakṣita): Dharmakīrti’s Vādanyāya: With the Commentary of
Śāntarakṣita, ed R. Sāṅkṛtyāyana. Appendix to Journal of the Bihar and Orissa
Reseach Society: New Series 21–22 (1935–36).
YD Yuktidīpikā: The most significant commentary on the Sāṃkhyakārikā, ed. A. Wezler
and Sh. Motegi. Vol. 1. Stuttgart 1998.
Secondary Sources
Frauwallner 1959 E. Frauwallner, Dignāga, sein Werk und seine Entwicklung. Wiener
Zeitschrift für die Kunde Süd- und Ostasiens 3 (1959) 83–164.
Funayama 2001 T. Funayama, On the date of Vinītadeva. In: Le parole e i marmi, studi in
onore di Raniero Gnoli nel suo 70゜compleanno, ed. Raffaele Torella. Rome 2001,
309–325.
Jambuvijaya 1966 Muni Jambuvijaya, Dvādaśāraṃ Nayacakram of Āchārya Śrī Mallavādi Kṣamāśramaṇa. With the commentary Nyāyāgamānusāriṇī of Śrī Siṃhasūri
Gaṇi Vādi Kṣamāśramaṇa. Part I. Bhavnagar 1966.
Kataoka 2012 K. Kataoka, Gengo-tetsugaku – Apoha-ron [Linguistic Philosophy – Apoha
Theory]. In: Ninshiki-ron to ronri-gaku [Epistemology and Logic], ed. Sh. Katsura,
A. Saito, M. Shimoda, and F. Sueki. Tokyo 2012, 189–226.
Katsura 1981 Sh. Katsura, Inmyō shōrimonron kenkyū (4) [A Study of Nyāyamukha].
Hiroshima Daigaku Bugakubu Kiyō 41 (1981) 62–82.
Katsura 2004 Sh. Katsura, The Role of dṛṣṭānta in Dignāga’s logic. In: The Role of the
Example (dṛṣṭānta) in Classical Indian Logic, ed. E. Steinkellner and Sh. Katsura.
Vienna 2004, 135–174.
Kitagawa 1965 H. Kitagawa, Indo-koten-ronrigaku no kenkyū – Jinna (Dignāga) no taikei
– [A study on classical Indian logic – Dignāga’s system]. Tokyo 1965.
Matilal 1985 B. K. Matilal, Logic, Language and Reality. An introduction to Indian
Philosophical Studies. Delhi 1985.
Pind 1999 O. H. Pind, Dharmakīrti’s Interpretation of Pramāṇasamuccayavṛtti V 36:
śabdo ’rthāntaranivṛttiviśiṣṭān eva bhāvān āha. In: Dharmakīrti’s Thought and Its
Impact on Indian and Tibetan Philosophy, Proceedings of the Third International
Dharmakīrti Conference Hiroshima, November 4–6, 1997, ed. Sh. Katsura. Vienna
1999, 317–332.
Pind 2011 O. H. Pind, Dignāga’s Apoha Theory: Its Presuppositions and Main Theoretical
Implications. In: Apoha. Buddhist Nominalism and Human Cognition, ed. M. Siderits,
T. Tillemans, and A. Chakravarti. New York 2011, 64–83.
Pind 2015 O. H. Pind, Dignāga’s Philosophy of Language. Dignāga. Pramāṇasamuccayavṛtti V on anyāpoha. Part I: Text. Part II: Translation and Annotation, ed. E.
Steinkellner. Vienna 2015.
Shiga 2011a K. Shiga, antarvyāpti and bahirvyāpti re-examined. In: Religion and Logic
in Buddhist Philosophical Analysis, Proceedings of the Fourth International Dhar-
Kiyokuni Shiga
427
makīrti Conference Vienna, August 23–27, 2005, ed. H. Krasser, H. Lasic, E. Franco,
and B. Kellner. Vienna 2011, 423–435.
Shiga 2011b K. Shiga, Remarks on the Origin of All-inclusive Pervasion. Journal of
Indian Philosophy 39 (2011) 521–534.
Steinkellner 2004 E. Steinkellner, The Early Dharmakīrti on the Purpose of Examples.
In: The Role of the Example (dṛṣṭānta) in Classical Indian Logic, ed. E. Steinkellner
and Sh. Katsura. Vienna 2004, 225–251.
Steinkellner 2005 E. Seinkellner, Dignāga’s Pramāṇasamuccaya, Chapter 1: A hypothetical reconstruction of the Sanskrit text with the help of the two Tibetan translations
on the basis of the hitherto known Sanskrit fragments and the linguistic materials
gained from Jinendrabuddhi’s Ṭīkā. 2005. https://www.oeaw.ac.at/fileadmin/Ins
titute/IKGA/PDF/forschung/buddhismuskunde/dignaga_PS_1.pdf, last visited
10-12-2019.
Closing a Gap in the Interpretation of Dharmakīrti’s Logic
by
Ernst Steinkellner
Dedicated to Hans Lenk in gratitude.1
It is the ontic character2 of Dharmakīrti’s logic that allows us to consider his method for
ascertaining the necessary concomitance (vyāpti) of the concept of an “effect” as logical
reason (kāryahetu) with that of a “cause” as its consequent (sādhya) to be a method for
ascertaining a causal relation (kāryakāraṇabhāva) as well.
Scholars have long examined the formulation of this method from various perspectives,
seeing it as two different topics, one causal, the other logical. I will begin by taking a
closer look at the first, the causal. In my opinion, it was separated from the second due
to its focus on causality, and moreover, its proper understanding was also impeded by a
weighty interpretational mistake. Although the second topic, the logical one, fared better, its
systematic purpose in the context of Dharmakīrti’s work has not been clarified sufficiently.
Three papers presented in Vienna at the Second International Dharmakīrti Conference in
1989 are my starting point: one by Brendan Gillon, who criticized Dharmakīrti’s proposed
method for ascertaining a causal relation or, in other words, for solving the induction
problem (Gillon 1991), and the papers by Tadashi Tani (1991) and myself (Steinkellner
1
2
The reason for this dedication to the German philosopher Hans Lenk is my assumption that certain
basic Buddhist epistemic and ontologic conceptions, such as vikalpa, vāsanā, anāditva, can fruitfully
be stimulated by his epistemological thought: Grasping as acting on various levels of interventional
interpretations on the basis of schemata that are bio-genetically determined on all levels. For some
concise introductions to Lenk’s thought, cf. Lenk 1998 and 2003; the collection of articles Global
TechnoScience and Responsibility: Schemes Applied to Human Values, Technology, Creativity, and
Globalisation, 2007; and chapters III and IV in Lenk and Paul 2014: 71–117.
While, as a strict nominalist, Dharmakīrti is not bothered by a correspondence principle for the relationship between concepts or words and their referents in reality, his theory of concepts and their generation
(apohavāda) provides, nevertheless, a link to reality via actual experiences. Actual experience gives
some additional precision to the general profile of the concepts which are inherited from a past without
beginning and continuously transported as residual impregnations in the continuum of cognition. Experience gradually reduces the innate erroneous misrepresentation of what there is in these impregnations.
As for the concepts themselves, Dharmakīrti thinks that they can be more or less closely related to
reality. In order to cope with a world of universal dissatisfaction, what is needed is a way to distinguish
concepts that are closer to reality from those that are less or not at all related to reality. The former can
be considered sound or true because they can be relied on in activities. To determine their truth is the
task of logic, a method for deriving the truth of one concept from its necessary concomitance with
another concept whose truth is already established or generally accepted. With Dharmakīrti, in this
respect going beyond Dignāga’s logical conceptions, the result of distinguishing sound or true concepts
from others is possible because their relationship is grounded in relations that are considered real, such
as factual identity and causality. It is this idea that makes for nothing less than an ontic character of
Dharmakīrti’s logic. Yet, even if Dharmakīrti admits at the same time that factual identity and causality,
too, are only conceptual constructs, he considers them as the only reliable constructs that satisfyingly
function when applied in worldly affairs and with the ambition to transgress the world of suffering.
Birgit Kellner et al., eds., Reverberations of Dharmakīrti’s Philosophy: Proceedings of the Fifth International Dharmakīrti
Conference Heidelberg, August 26 to 30, 2014. Vienna 2020, pp. 429–440.
430
Closing a Gap in the Interpretation of Dharmakīrti’s Logic
1991), which dealt with Dharmakīrti’s proposal of how to ascertain concomitance (vyāpti)
in the case of a svabhāvahetu.
Tadashi Tani, on his part, was able to show that in Dharmakīrti’s first work (that is, the
first chapter of the Pramāṇavārttika with the Vṛtti usually called *Svavṛtti),3 a prasaṅga
assumes this function.4 In my own paper, I examined the meaning of Dharmakīrti’s later
formulation of this device, in the Vādanyāya, as viparyaye bādhakapramāṇa. Dharmakīrti
developed this method on the basis of the sattvānumāna, and by means of an experiment
I was able to show that it is applicable to all svabhāvahetu inferences, such as the one
of the designation “tree” from the designation “Śiṃśapā.” The question left open at the
time I wrote my paper was “… whether the different treatments of the svabhāvahetu and
the kāryahetu in this respect were not also resolved in a certain sense in order to design a
homogeneous logical system, or at least, whether there are no indications to be found in
Dharmakīrti’s work that he was aiming in this direction” (Steinkellner 1991: 323).
Based on PV 1.23’cd and passages from the Hetubindu, Tani in fact already answered
this question affirmatively, summarizing: “The kāryahetu too can be interpreted by the
same model …” and “The necessary relation can be determined by SVB-pramāṇa.” Yet,
he added: “Nevertheless, Dharmakīrti did not explicitly explain so” (Tani 1991: 337).
In fact, Tani only very closely missed the mark. The passage he refers to (PV 1.23’cd–
24ab with PVSV 17,5–7) belongs to a section of the Vṛtti in which Dharmakīrti establishes
the essential relation (svabhāvapratibandha) as the fundament of a reason’s non-deviation
from the property to be proved (PVSV 10,13–20,13). It is, however, only in the section
beginning with PV 1.34ab (from PVSV 21,24 to the end of the work) that Dharmakīrti
actually deals with the question of how, in the case of the two types of acceptable reasons,
such an essential relation can be ascertained. And it is here, in PV 1.34 with PVSV 22,2–4
and 6f, that Dharmakīrti’s answer is clearly spelled out in regard to the kāryahetu. It is this
clarification that links my present paper to both Tani’s explanation and Gillon’s critique.
I will start with Gillon, because only with a more exact view of the object of his
critique are we in a position to see that Dharmakīrti did, in fact, already in this first
work of his, provide everything needed for designing a comprehensive and consistent
logical system, even if he did not use the later and subsequently traditional term viparyaye
bādhakapramāṇa.
In his 1991 paper, Gillon demonstrated that “Dharmakīrti’s solution” to the problem in
his formulating a method for ascertaining a causal relation “can be seen not to work” (Gillon
1991: 57). As far as I can see, Gillon’s critique seems invalid with regard to Dharmakīrti’s
statement as it has been interpreted until now by all who have set their minds to it, including
myself.5
3
4
5
For the historical position of this work, cf. Frauwallner 1954: 144–148 and 152, where Frauwallner
tentatively names it *Hetuprakaraṇa; for its character, cf. Steinkellner 2013: I xxviiif.
Tani 1991: 332–338. Tani refers to prasaṅga as “hypothetical negative reasoning.”
Gillon’s observations were later refined in two papers by Horst Lasic, who closely analysed the cognitional process implied in the steps of this method (1999) and interpreted it as serving another aim,
namely, that of justifying a vyāpti already ascertained (2003: 194). But the existence of an induction
problem remained unquestioned; this was finally also affirmed by John Dunne (2004: 191–192) and
Tom Tillemans (2004: note 25; cf. below).
Ernst Steinkellner
431
I must admit that I could not approve when, in 1989, I heard Dharmakīrti being charged
with such a blunder, or with not having seen the pitfalls implied in his proposed method.
This irritation lingered on: While on one hand, Dharmakīrti’s fallacy seemed evident, on
the other hand, it did not coincide with my appreciation of the acuteness of his thinking
and expression. Yet it was only after my retirement that I had the leisure to look at the
issue again in the hope of being able to disperse my qualms. When a seemingly correct
translation nonetheless doesn’t make sense, the first step I have always taken is to examine
the original again. So that is what I did. The fault may lie, after all, not in the interpreted
object but in the interpreter’s perception. The results of my re-examination have already
been published (Steinkellner 2013), and are reflected in three sentences of translation and
25 pages of notes. Alas, all in German!6
I will, therefore, first remind you of Gillon’s critique, which, on the basis of Kajiyama’s
translation, is directed against Dharmakīrti’s earliest formulation. Then I will indicate
the crux of all earlier translations and summarize the main points of my new understanding.7 Finally, I will draw support for this interpretation from Dharmakīrti’s last work, the
Vādanyāya.
The formulation at stake consists of three sentences that belong together. So far, everyone
has only dealt with the first sentence (except for John Dunne, who includes the second but
misunderstands it, and Gillon and Richard Hayes, who include the third but jump over the
second). These sentences are:
1. yeṣām upalambhe tallakṣaṇam anupalabdhaṃ yad upalabhyate, tatraikābhāve ’pi nopalabhyate, tat tasya kāryam.
2. tac ca dhūme ’sti.
3. sakṛd api tathādarśanāt kāryaḥ siddhaḥ, akāryatve ’kāraṇāt sakṛd apy
abhāvāt (PVSV 22,2–4 and 6f).8
The salient point of Gillon’s critique9 has been made quite clear by Tom Tillemans:
6
7
8
9
The conclusion of these notes (II.209f) also finds support from Śāntarakṣita’s Tattvasaṅgraha 1695–1697,
where the causality relation and non-existence are similarly seen: in both cases only the respective
verbal usage (vyavahāra) is inferred, not a real entity as such (cf. Steinkellner 2019: 68-69).
In this section, I repeat parts of a paper delivered at the Lumbini conference in 2013 (cf. Steinkellner
2015).
Kajiyama translates the first: “If a thing (E) which, having perceptible characteristics, was not perceived,
is perceived when other things (Cs) have been perceived, and if, when even one thing (C) among these
has disappeared E is not perceived, then E is the effect of C …” (Kajiyama 1963: 2f; E = effect, C =
cause). Other samples of translations are: “If a previously unperceived thing defined as perceptible is
later perceived when other things are perceived, and if that thing is not perceived when one among
those other things is absent, then it is the effect of that thing. That kind of definition of an effect applies
to smoke” (Dunne 2004: 335). “That which, not having been apprehended, is apprehended, when
its conditions have been apprehended, [but] is not apprehended, when even one of them is absent, is
[ascertained] to be their effect. … An effect is established because of being observed even once in that
way, because, if it were not an effect, it would not arise even once from what is not its cause” (Gillon
and Hayes 2008: 353 and 398).
In his paper, Gillon only concludes: “… Dharmakīrti seems to believe that a sequence of five simple
non-relational observations results in relational knowledge. … The problem is that this sequence cannot
discriminate between genuine causes and spurious correlations. … Moreover, further observations will
432
Closing a Gap in the Interpretation of Dharmakīrti’s Logic
The root of the problem of determining causality is that people just have no
way of knowing definitely when they have correctly isolated the true cause
from the myriad of other background things. Using Mill’s method or that of
Dharmakīrti, there can always be the doubt that, in spite of something being
thought to be the cause, the presence of that circumstance was not in fact what
was actually responsible for the effect and its absence was not what brought
about an absence of the effect – some other hidden factor that we didn’t know
about, didn’t think about and perhaps didn’t see at all was first present and
then absent at the key stages of our tests (Tillemans 2004: note 25).
The following is my present translation. The italicized words indicate deviations from my
earlier ones.
1. That (entity) (tat) which (yat) as unperceived (at first), although (on
principle) perceptible, is perceived on the perception of which (other
entities present) (yeṣām), (and again) is not perceived when even a single
(entity) among these (other entities) (tatra) is absent, is the effect of this
(single other entity that is absent).
2. And this [fact of being conditionally10 known through perceptions and
non-perceptions] does occur in the case of smoke.
3. From being observed in this way even once (something, such as smoke,)
is established as that which is to be effected (kāryaḥ) (by something else,
such as fire); for, were it not (that which is) to be effected (by this), it
would not even once occur on account of that which is not (its) cause.
In terms of mere translating, this new version is roughly the same, with the exception of
some added phrases, as all those earlier ones that evoked Gillon’s critique. My interpretation
of the italicized items, however, differs in the following ways:
1. Hitherto almost everyone11 has assumed the pronoun yeṣām to refer to the many
different individual entities as causes that, by coming together, constitute a single
causal complex (hetusāmagrī ). I now consider the plural pronoun, in the light of
Dharmakīrti’s concept of “cause,” which always means a causal complex, to refer to
different causal complexes and not different individual causes that together constitute
a single causal complex.12
10
11
12
never eliminate the possibility of spurious correlation. But this is just the induction problem again”
(Gillon 1991: 58).
“Conditionally,” in this context, is short for “in respect to the fact of the presence or absence of something
else.”
The exception being Horst Lasic who already in 1999: 237 and 238 clearly excluded an interpretation of
yeṣām as referring to the constituents of a single causal complex when he paraphrases with “a perception
of everything present at the place of observation.” In my note (2013: 354) I overlooked this paraphrase.
Mind, however, that in my opinion Lasic’s “things” should also be understood as causal complexes.
For Dharmakīrti’s concept of cause as a causal complex, cf. various passages in HB § b.1221 with
Steinkellner 1967: 44–55 (see also Shah 1967: 45–59), Steinkellner 1971: 184–188, and Dunne 2004:
161–169. Cf. also the collection of excerpts from the Vādanyāyaṭīkā and the references to Arcaṭa’s
Hetubinduṭīkā in the appendix of Steinkellner 2015.
Ernst Steinkellner
433
2. The process of cognition described in the first sentence consists of several perceptions
and non-perceptions under certain conditions, namely, the presence and absence of
another entity, and it refers to a specific case of two different entities. In order to
generalize the cognition derived from a specific case, Dharmakīrti uses a reductio
ad absurdum argument, a prasaṅga, when he says, in the third sentence: “Were
(something, say smoke,) not to be effected (by another thing, say fire,) it (could)
not even once occur on account of that which is not (its) cause.”13 This second
step indicates an unwanted consequence, should the truth of the cognition gained
by observing a specific concrete case not be accepted. It thereby is supposed to
provide the qualities of necessity and general validity for this cognition. In other
words, the prasaṅga “transforms,” so to speak, the formulation into one of a necessary
concomitance (vyāpti),14 and in this function of justifying a specific common presence
and absence as being necessary in general, the prasaṅga is an integral part of the
proposed method.
3. Thus, the three cited sentences belong together. Together they make for a veritable
proof formula (prayoga): The first sentence states a specific cognition of the common
presence and common absence (anvayavyatireka) of the property “being conditionally perceived and non-perceived” as the logical reason (hetu) with the property
to be established, the consequent (sādhya), namely, “being the effect of a certain
other entity.” This cognition of a mere anvaya and vyatireka is transformed into, i.e.
established as, a generally valid vyāpti by the third sentence, the prasaṅga. And the
second sentence states the presence of the reason (“being conditionally perceived
and non-perceived”) in the locus of smoke (pakṣadharmatā).
In this way, I think, Dharmakīrti tried to overcome the problem of an incomplete induction.
I consider his completing the concomitance formulation with a prasaṅga, whose function
is to justify that formulation, to be a veritable stroke of genius. Whether he really succeeded
or only shifted the problem to yet another level is a question beyond the scope of this article.
Dharmakīrti actually modified this early formulation15 in his later works, the Hetubindu
and the Vādanyāya.16 In these, the prasaṅga argument is either not included, as in the
Hetubindu’s first passage (HB § 3.32), which is focussed on the formal aspects of inference
and proof, or it is present in a more elaborate form, as in the Hetubindu’s second passage
(HB § 4.22) and in the Vādanyāya.17
13
14
15
16
17
PVSV 22,6f: akāryatve ’kāraṇāt sakṛd apy abhāvāt.
Dharmakīrti expressly states this to be the function of the prasaṅga in PVin 3. 4,7–9 (cf. Watanabe, this
volume: § 2.3).
It was repeated with only syntactic adaptions in the Pramāṇaviniścaya (PVin 2. 85,6–8).
For the differences of these formulations from the earlier ones, cf. Steinkellner 2013: II 193–198; for a
synoptic survey of the formulations, see Steinkellner 2013: II 197.
Throughout the later tradition, it was particularly the formulation offered in the first passage of the
Hetubindu that was adopted as the authoritative Dharmakīrtian statement on the issue. In my study
of 2013, much to my shame, I completely forgot the prasaṅga-related passages in his later works
and judged this wrongly asserted lack of a prasaṅga argument as a set-back in Dharmakīrti’s theory
(Steinkellner 2013: II 204). I now see that Dharmakīrti did not (!) change his conception of ascertaining
a causal relation, but rather further elaborated on the original version in his later works with the purpose
of strengthening exactly this prasaṅga.
434
Closing a Gap in the Interpretation of Dharmakīrti’s Logic
We can thus jump straight to the Vādanyāya, his last work, where Dharmakīrti’s
awareness of the induction problem is most clearly revealed. Already in the Vṛtti (PVSV
22,10–23,6 with PV 1.35), Dharmakīrti deals with the issue that an effect, say, smoke,
does not deviate from fire as its cause (avyabhicāra) by parrying the idea that it may have
“another cause,”18 such as a termite hill.19 In his answer there, Dharmakīrti repudiates the
opponent’s proposal by clarifying that this deviating item is only something similar to
smoke, but not one of “such type” (tādṛśa) that it is the product of something that is “such,”
namely, being able to produce real smoke.20
In the Vādanyāya, however, where Dharmakīrti demonstrates how “effect” as a means
of proof (sādhanāṅga) is justified (samarth-), the notion of “another cause” is given a more
stringent meaning. The relevant passage in the Vādanyāya (VN 4,2–10) consists of two
parts: One is Dharmakīrti’s last formulation (VN 4,2f) of the content to be ascertained, a
causal relation; the other is his answer to an alternative, in a sentence nicely resembling
Tilleman’s conclusion cited above: “Another (entity) may be capable in regard to (the
production of) this (effect); because of the absence of that (other entity) this (effect) did
not come about” (anyat tatra samartham. tadabhāvāt tan na bhūtam, VN 4,6f).
The formulation in the Vādanyāya is as follows: “This (entity) comes to be, when that
(other entity) is present, (and) even if the (various) causes of this (entity) which are different
from that entity (and) capable (of producing its perception) are present, (this entity) does
not come about in the absence of that (entity).”21 Here, the problem seen by Gillon in the
earlier formulation is still present.
18
19
20
21
PVSV 22,10: anyahetukatvān nāhetukatvam iti cet.
Cf. PVṬt 60a7 = PVSVṬ 100,12. When summarizing his argument, Dharmakīrti refers to the termite
hill (śakramūrdhan) again: “If a termite hill is of fiery nature, it is surely fire. If it is not of fiery
nature, how can smoke come about there?” (agnisvabhāvaḥ śakrasya mūrdhā yady agnir eva saḥ /
athānagnisvabhāvo ’sau dhūmas tatra kathaṃ bhavet // PV 1.36). To understand this example of
the termite hill, I can offer a quite plausible physical explanation that I gratefully received from the
entomologist Dominique Zimmermann of the Natural History Museum Vienna: Temperature and
humidity in termite structures are kept constant by means of ventilation systems (which maintain the
temperature at around 30 degree Celsius). The air that escapes through the central chimney can be up
to 15 degrees warmer and also more humid than the surrounding air. Given the appropriate conditions,
the escaping air, through effects of condensation, can become visible as vapour. Zimmermann could
not find any references to this phenomenon in scientific publications. She sees this as possibly due to
the fact that biologically it is irrelevant whether the air escaping from a termite hill is visible as vapour
or not. But since a termite hill is used as an example by our philosopher, we can safely assume that this
sort of smoke-like vapour from termite hills was commonly known in India.
PVSV 22,14f: tādṛśād dhi bhavan sa tādṛśaḥ syāt. And tādṛśa is “something with a certain nature
when it comes about through nothing but its own cause,” as he states later in the Hetubindu (yo hi yasya
svabhāvaḥ, sa svahetor evotpadyamānas tādṛśo bhavati, HB 11,7f).
idam asmin sati bhavati, satsv api tadanyeṣu samartheṣu taddhetuṣu tadabhāve na bhavatīti (VN 4,2f).
In this formulation, a form was chosen the proximity of which to the Venerable’s description of a
causal relation in the short formula at the beginning of the pratītyasamutpāda is obvious and seemingly
intended. For the ādi-formula of the pratītyasamutpāda: asmin satīdaṃ bhavati, asyotpādād idam
utpadyate (“If that is present, this comes to be; on account of the arising of that this arises”), cf. La
Vallée Poussin 1913: 49–51.
Ernst Steinkellner
435
But then, Dharmakīrti considers the option of “another cause,” not of something similar
to smoke, but of smoke as such.22 And in order to indicate the unwanted consequence
should the above justification not be acknowledged, he adds:
Otherwise, (i.e.) if only (the proposition) “(this) is not present in the absence
of that” (tadabhāve na bhavati) were communicated, the (causal) capacity of
the latter would be in doubt, since also another (entity) is absent there (where
the latter is absent). Another (entity, then) may be capable in regard to (the
production of) this (effect); because of the absence of that (other entity,) this
(effect) did not come about. (In this case) moreover, the absence (of the effect)
in the absence of that (cause) would be (nothing but) a fortuitous conformity
(yādṛcchāsaṃvāda), just like the absence of a date palm that grows in regions
where mother marriage is customary in the absence of mother marriage in
other regions.23
By introducing a prasaṅga into his proof of a causal relation and by this final elaboration of
the prasaṅga’s function, Dharmakīrti seems to have solved, for his own specific purposes24
and historical context, the problem of induction found in the need to know, with indisputable
certainty, the common absence of reason and consequence as a necessary feature of a good
reason.
Now, induction in a modern, post-Baconian sense is a method for acquiring knowledge.
This was certainly true for Dharmakīrti as well. There is, however, possibly a limitation to
this comparison due to the fact that Dharmakīrti deals with it only in connection with the
need in Dignāga’s system of logic to ascertain the knowledge of the common absence of
two properties (vyatireka). This problem was already realized by Dharmakīrti’s teacher
Īśvarasena, who felt motivated to increase the number of characteristics of a good logical
reason to six, instead of the three as defined by Dignāga.25 It is for this reason that I believe
the “taste” of the induction problem in the Indian context to be different. The important
remaining questions are not only whether indeed different concepts of induction and its
problem are at stake, but also whether Dharmakīrti actually succeeded in working out a
solution to the induction problem as such, and if not, where exactly he failed under the new
interpretation presented. Yet, with much curiosity I will have to leave it to better equipped
and interested philosophers to find answers to questions such as these.26
We can now shift to the same statement as a logical topic:
The three sentences we dealt with so far from the perspective of being a method for
ascertaining causality must also be seen, in terms of the central purport of Dharmakīrti’s
22
23
24
25
26
VN 4,6f: anyat tatra samartham, tadabhāvāt tan na bhūtam.
VN 4,5–10: anyathā kevalaṃ tadabhāve na bhavatīty upadarṣane ’nyasyāpi tatrābhāve sandigdham
asya sāmarthyam. anyat tatra samartham, tadabhāvāt tan na bhūtam. etannivṛttau punar nivṛttir
yadṛcchāsaṃvādaḥ, mātṛvivāhocitadeśajanmanaḥ piṇḍakharjūrasya deśāntareṣu mātṛvivāhābhāve
’bhāvavat.
Such as to avoid the insufficiencies of Īśvarasena’s adarśanamātra-theorem (cf. Steinkellner 1966 and
2013: note 180) or his increasing Dignāga’s three characteristics of a good reason to six (cf. HB § d.,
Steinkellner 1967: 70–78 with notes, and Steinkellner forthcoming).
On Īśvarasena’s ṣaḍlakṣaṇo hetuḥ, cf. HB 40,1–47,7 and Steinkellner 1967: 70–78; 1979: note 475; as
well as Tillemans 2000: note 335.
See the appendix.
436
Closing a Gap in the Interpretation of Dharmakīrti’s Logic
work, from the logical perspective. As said above, the three sentences amount to a proof
that consists of a vyāpti formulation, the indication of a pakṣadharmatā and a prasaṅga to
establish the vyāpti, because the section of PV 1.34–38 (PVSV 21,24–24,7) is dedicated to
the question of how the essential relation that a kāryahetu presupposes can be ascertained.27
The question is then whether we can identify, also in this case, the prasaṅga component
applied for establishing the vyāpti of a kāryahetu, as in the case of the svabhāvahetu (Steinkellner 1991: 319ff.), as structurally constituting or being equivalent to a sādhyaviparyaye
bādhakapramāṇa. I think we can. Again, however, only by way of an experiment because –
as Tani has said – he “did not explicitly explain so.”
According to Dharmakīrti’s later work, the bādhakapramāṇa is the non-perception of
a pervading property (vyāpakadharmānupalabdhi, VN 8,6) through which the absence
of the reason in the contradictory opposite (viparyaya) of the consequent (sādhya) can be
inferred.
In this experiment, the principal uncertainty lies in determining the pervading property
(vyāpaka). If Dharmakīrti had a pre-formation of the later term for this structure in mind, as
I think he did, this property should not be one that is referred to in the proof itself, but one
that should nevertheless be referred to in his prasaṅga. As far as I can see, the only candidate
to offer itself is the property of “being observed in this way at least once” (*sakṛttathādṛṣṭatva).28 Under this hypothesis, we may conclude that the property *sakṛttathādṛṣṭatva,
which pervades the reason “being conditionally perceived and non-perceived,”29 is not
perceived and thus is negated in the locus of the contradictory opposite of the property to
be proven, i.e., the property of “not being the effect of that” (*atatkāryatva).
Moreover, I think we can discover this structure in the words Dharmakīrti uses to
complete the prasaṅga establishing the vyāpti, namely: “For, were it not (that which is)
to be effected (by this), it would not even once occur on account of that which is not (its)
cause,”30 when he continues with the words “If an effect, however, would occur without its
cause, it would not have a cause at all,”31 and to this adds, with the pepper of irony, a proof
formula in corroboration of this last statement: “That, namely, is not a cause of this, which
occurs without that; and smoke occurs without fire; therefore this (smoke) would not have
that (fire) as (its) cause.”32
To conclude: If my experiment can be considered sound, it means that Dharmakīrti
applied the structure of the later (sādhya)viparyaye bādhakapramāṇa already in his first
work, and at the appropriate place in his work, that is, where the ascertainment of a logical
concomitance as based on essential relations is the subject matter. Already in his first work,
then, Dharmakīrti seems to have given a comprehensive and consistent account of all the
basic constituents of his new logic, even if he did not yet use the terminology he coined
later.
27
28
29
30
31
32
This section is followed by PV 1.39 (PVSV 7–15), where the same question is answered for the
svabhāvahetu.
I derive this hypothetically constructed property from sakṛd api tathādaryanāt (PVSV 22,6).
A property derived from bhāvābhāvasādhanapramāṇābhyām (VN 4,1f).
akāryatve ’karaṇāt sakṛd apy abhāvāt (PVSV 22,6f).
kāryasya ca svakāraṇam antareṇa bhāve ’hetumattaiva syāt (PVSV 22,7f).
na hi yasya yam antareṇa bhāvaḥ, sa tasya hetur bhavati; bhavati ca dhūmo ’gnim antareṇa; tan na
taddhetuḥ syāt (PVSV 22,8–10).
Ernst Steinkellner
437
At this point we also have to pay attention to the wider context of Dharmakīrti’s logical
thought and to follow up on some more questions that need answers. For I think there
is yet another way to get to grips with Dharmakīrti’s “failure” to solve the induction
problem beyond being possibly satisfied with the solution presented above. And this line
of interpretation will have to be elaborated, too, although I do not think it will make for a
substantial difference in respect to the induction issue. Here I can only indicate the general
direction of the way.
Right at the beginning, when he introduces the new theorem of the three kinds of logical
reason (trividhā hetu) for the first time, Dharmakīrti cites a statement that is possibly from
Dignāga’s lost Hetumukha (cf. Frauwallner 1959: 103f.):
All this dealing with inference and what can be inferred is entirely based on a
distinction between properties and a bearer of properties, (a distinction) that
is mounted on the mind (buddhyārūḍha).
(sarva evāyam anumānānumeyavyavahāro buddhyārūḍhena dharmadharmibhedena, PVSV 2,22–3,1).
These words express, above all, the trivial acknowledgement that a logician is dealing with
concepts and linguistic items, also when cause and effect and their relation are at stake.
This means that Dharmakīrti is clear about the fact that inference, as resulting knowledge,
belongs to the realm of concepts which are inherited from a beginningless past. As such, it
is also erroneous because it does not truly represent reality as it is (yathābhūta). But the
application of logical rules and his specific theory of concept formation (apohavāda) allow
for a differentiation between concepts that can be considered true, i.e., referring to real
(sat) entities, and false, i.e., referring to unreal (asat) entities. Exactly this differentiation,
then, is the task of inference when it attributes or excludes properties from entities, and its
success is the reason for judging inferential cognition as valid (pramāṇa), meaning reliable
when adopted in any kind of activity, be it worldly or aiming beyond.
Moreover, by purging false concepts, inference is essential for weeding out all those
misapprehensions that are the cause for being bound to continuous suffering existence
(saṃsāra), the cause of which is for Dharmakīrti, above all, the belief in a permanent self
(ātman), tantamount to nescience (avidyā).33
Terms such as “cause,” “effect,” and their “relation” refer to concepts and as such have
no referent in reality. This is clearly spelled out in Dharmakīrti’s Sambandhaparīkṣā, a
short work that I believe was composed in close proximity to his first work on logic. If it
is necessary, as I said at the beginning, to identify causes and what effects they have in
order to act successfully as a worldling or as someone already on the Buddhist path, one
must clarify what a cause and what an effect are. In other words: On what grounds can we
attribute the word or concept of “cause” or “effect” to a certain entity? Or: Why can we
correctly say that something is the “effect” of something else as its “cause”?
The answer to such questions is Dharmakīrti’s yeṣām upalambhe-formula presented
above. It offers no less than a logical model that determines exactly in what sense the
concepts or terms of “cause” and “effect” can be correctly applied, for it provides a reason
33
Cf. Vetter 1990: 22–26, 42f., 112–114; Eltschinger and Ratié 2013: 4–36.
438
Closing a Gap in the Interpretation of Dharmakīrti’s Logic
on account of which something can be conceived and named as being the effect of a specific
cause.
Not being a post-Baconian scholar, causality as a natural phenomenon may not have
been Dharmakīrti’s concern. Yet what he laboured over was also not a mere “language
game.” On the contrary, his examination had the most serious purpose of uprooting the
false conceptions that constitute impediments for the mind in its progress towards liberation
from suffering.34 In the same manner as his elaborations on the inference from essence as
reason (svabhāvahetu), which is based on the common acceptance of linguistic conventions
(tathāprasiddhi, PVSV 16,30f), Dharmakīrti’s thought on causality and on the inference
from effect as reason is grounded in the very core of the Buddhist world view.35
References and abbreviations
Dunne 2004 J. D. Dunne, Foundations of Dharmakīrti’s Philosophy. Boston 2004.
Eltschinger and Ratié 2013 V. Eltschinger and I. Ratié, Self, No-Self, and Salvation. Dharmakīrti’s Critique of the Notions of Self and Person. Wien 2013.
Frauwallner 1954 E. Frauwallner, Die Reihenfolge und Entstehung der Werke Dharmakīrti’s. In: Asiatica. Festschrift Friedrich Weller zum 65. Geburtstag. Leipzig 1954,
142–154.
Frauwallner 1959 E. Frauwallner, Dignāga, sein Werk und seine Entwicklung. Wiener
Zeitschrift für die Kunde Südasiens 3 (1959) 83–164.
Gillon 1991 B. S. Gillon, Dharmakīrti and the Problem of Induction. In: SBET, 53–58.
Gillon and Hayes 2008 B. S. Gillon and R. P. Hayes, Dharmakīrti on the role of causation
in inference as presented in Pramāṇavārttika Svopajñavṛtti 11–38. Journal of Indian
Philosophy 36 (2008) 335–404.
Hamilton 2000 S. Hamilton, Early Buddhism: A New Approach. The I of the Beholder.
Richmond 2000.
HB E. Steinkellner, Dharmakīrti’s Hetubinduḥ. Critically edited. Beijing/Vienna 2016.
Kajiyama 1963 Y. Kajiyama, Trikapañcakacintā. Development of the Buddhist Theory
on the Determination of Causality. Miscellanea Indologica Kiotiensia 4–5 (1963)
1–15.
Kellner 2004 B. Kellner, Why Infer and not just Look? Dharmakīrti on the Psychology
of Inferential Processes. In: The Role of the Example (dṛṣṭānta) in Classical Indian
Logic, ed. Sh. Katsura and E. Steinkellner. Wien 2004, 1–51.
La Vallée Poussin 1913 L. de La Vallée Poussin, Théorie des douze causes. Gand 1913.
Lasic 1999 H. Lasic, Dharmakīrti and His Successors on the Determination of Causality.
In: Dharmakīrti’s Thought and Its Impact on Indian and Tibetan Philosophy. Proceedings of the Third International Dharmakīrti Conference Hiroshima, November
4–6, 1997, ed. Sh. Katsura. Wien 1999, 233–242.
Lasic 2003 H. Lasic, On the Utilisation of Causality as a Basis of Inference. Dharmakīrti’s Statements and Their Interpretation. Journal of Indian Philosophy 31 (2003)
185–197.
34
35
On this purpose in earlier layers of Buddhism, cf. Hamilton 2000 and Ronkin 2005: 193ff.
Cf. Dunne 2004: 193–222; Kellner 2004; Steinkellner 2013: II 204–210.
Ernst Steinkellner
439
Lenk 1998 H. Lenk, Einführung in die Erkenntnistheorie. München 1998.
Lenk 2003 H. Lenk, Grasping Reality: an Interpretation-Realistic Epistemology. Singapore 2003.
Lenk 2007 H. Lenk, Global TechnoScience and Responsibility: Schemes Applied to Human
Values, Technology, Creativity, and Globalisation. Berlin 2007.
Lenk and Paul 2014 H. Lenk and G. Paul, Transkulturelle Logik. Universalität in der
Vielfalt. Bochum/Freiburg 2014.
PV 1 See PVSV.
PVin 2 Dharmakīrti’s Pramāṇaviniścaya. Chapters 1 and 2. Critically ed. by E. Steinkellner. Beijing/Vienna 2007.
PVin 3 Dharmakīrti’s Pramāṇaviniścaya. Chapter 3. Critically ed. by P. Hugon and T.
Tomabechi. Beijing/Vienna 2011.
PVSV R. Gnoli, The Pramāṇavārttikam of Dharmakīrti. The first Chapter with the Autocommentary. Text and Critical Notes. Roma 1960.
PVSVṬ Ācārya-Dharmakīrteḥ Pramāṇavārttikam (svārthānumāna-paricchedaḥ) svopajñavṛttyā, Karṇakagomiviracitayā taṭṭīkayā ca sahitam, ed. R. Sāṅkṛtyāyana.
Ilāhābād 1943.
PVṬt Pramāṇavārttikaṭīkā (Śākyabuddhi), Kapitel 1 – P 5718, Je 1b–348a8, Ñe 1b–85b2.
Ronkin 2005 N. Ronkin, Early Buddhist Metaphysics. The making of a philosophical
tradition. London/New York 2005.
SBET Studies in the Buddhist Epistemological Tradition. Proceedings of the Second
International Dharmakīrti Conference Vienna, June 11–16, 1989, ed. E. Steinkellner.
Wien 1991.
Shah 1967 N. J. Shah, Akalaṅka’s Criticism of Dharmakīrti’s Philosophy. A Study.
Ahmedabad 1967.
Steinkellner 1966 E. Steinkellner, Bemerkungen zu Īśvarasenas Lehre vom Grund. Wiener
Zeitschrift für die Kunde Südasiens 10 (1966) 73–85.
Steinkellner 1967 E. Steinkellner, Dharmakīrti’s Hetubinduḥ. Teil II. Übersetzung und
Anmerkungen. Wien 1967.
Steinkellner 1971 E. Steinkellner, Wirklichkeit und Begriff bei Dharmakīrti. Wiener Zeitschrift für die Kunde Südasiens 15 (1971) 179–211.
Steinkellner 1979 E. Steinkellner, Dharmakīrti’s Pramāṇaviniścayaḥ, zweites Kapitel:
Svārthānumānam. Teil II. Übersetzung und Anmerkungen. Wien 1979.
Steinkellner 1991 E. Steinkellner, The Logic of the svabhāvahetu in Dharmakīrti’s Vādanyāya. In: SBET, 311–324.
Steinkellner 2013 E. Steinkellner, Dharmakīrtis frühe Logik. Annotierte Übersetzung der
logischen Teile von Pramāṇavārttika 1 mit der Vṛtti. 2 vols. Tokyo 2013.
Steinkellner 2015 E. Steinkellner, On Dharmakīrti’s Proposal to Solve the Induction
Problem. In: Logic in Buddhist Scholasticism: From Philosophical, Historical and
Comparative Perspectives, ed. Gregor Paul. Lumbini 2015, 169–185.
Steinkellner 2019 E. Steinkellner, Miszellen zur erkenntnistheoretisch-logischen Schule
des Buddhismus XII. anupalabdhi as pramāṇāntara—Īśvarasena is the Opponent in
Tattvasaṅgraha 1693–1694. With an Edition of Tattvasaṅgraha 1691–1697 and the
Pañjikā. Wiener Zeitschrift für die Kunde Südasiens 56–57 (2019) 67–75.
440
Closing a Gap in the Interpretation of Dharmakīrti’s Logic
Steinkellner forthcoming E. Steinkellner, Dharmakīrti and Īśvarasena. In: Archaeologies
of the Written: Indian, Tibetan, and Buddhist Studies in Honour of Cristina ScherrerSchaub, ed. V. Tournier, V. Eltschinger, and M. Sernesi. Naples Forthcoming.
Tani 1991 T. Tani, Logic and Time-ness in Dharmakīrti’s Philosophy – Hypothetical
Negative Reasoning (prasaṅga) and Momentary Existence (kṣaṇikatva). In: SBET,
325–401.
Tillemans 2000 T. J. F. Tillemans, Dharmakīrti’s Pramāṇavārttika. An annotated translation of the fourth chapter (parārthānumāna). Vol.1 (k. 1–148). Wien 2000.
Tillemans 2004 T. J. F. Tillemans, Inductiveness, Deductiveness and Examples in Buddhist Logic. In: The Role of the Example (dṛṣṭānta) in Classical Indian Logic, ed.
Sh. Katsura and E. Steinkellner. Wien 2004, 251–275.
Vetter 1990 T. Vetter, Der Buddha und seine Lehre in Dharmakīrtis Pramāṇavārttika. Der
Abschnitt über den Buddha und die vier edlen Wahrheiten im Pramāṇasiddhi-Kapitel.
Wien 1990 (second revised edition).
VN M. T. Much, Dharmakīrtis Vādanyāyaḥ. Teil I. Sanskrit-Text. Wien 1991.
Watanabe 2020 T. Watanabe, Dharmakīrti and His Successors on āśrayāsiddha and prasaṅgaviparyaya. In: Reverberations of Dharmakīrti’s Philosophy, ed. B. Kellner, P.
McAllister, H. Lasic, and S. McClintock. Vienna: 2020.
Appendix
Having seen my paper of 2015, Claus Oetke sent me a mail (November 26, 2016) on the
subject of induction with its different aspects that I consider valuable sharing in the hope
to make the present contribution more interesting to philosophers. Claus Oetke says (in my
translation from the German):
There are (at least) four variants of problems related to induction that are
reflected by the following questions:
1. How is it possible, a) to distinguish between causal and merely accompanying states of affairs and b) to find out which (preceding) states of
affairs are causes of (later) appearing facts and which are not?
2. How is it possible to justify that for a certain (type of) fact only a certain
other (type of) fact can be the cause?
3. How is it possible to ascertain that a regularity given within certain framing conditions must also be given within alternative framing conditions?
4. How is it possible to ascertain that in the world of experience there are
causal regularities (at all) and that laws referring to causality must be
valid?
… Regarding Dharmakīrti, he apparently offers a plausible answer to 1), while
it looks as if he was aware of questions 2) and 3) yet did not offer a cogent
solution. Against this the problematic question of 4) seems to be altogether
beyond his sense of problems, while it is, seen philosophically, the perhaps
most significant question.
Philosophical Reflections on the sahopalambhaniyama
Argument
by
John Taber
Introduction
In 1991 the Japanese scholar T. Iwata published a masterful study of the sahopalambhaniyama argument. In his work he carefully traces the development and interpretation of
the argument throughout the entire Buddhist epistemological tradition, from Dignāga to
Jñānaśrīmitra, and also surveys responses to the argument from a variety of Buddhist and
non-Buddhist opponents. He is not, however, primarily concerned with evaluating the
argument philosophically – though questions of logic inevitably arise when interpreting
the argument – and, it would seem, not at all with assessing it as a piece of philosophical
reasoning independently of its historical context. That is what I would like to do in this
essay. In what follows I shall construct a version of the sahopalambhaniyama (henceforth
SUN) argument and attempt to judge its strengths and weaknesses as an argument, in
particular, its formal validity and the defensibility of what I shall represent as its premises.
Although I shall indeed allow myself to be prompted by the statements of Dharmakīrti (as
well as other classical authors), and will even at the beginning touch on the question how
Dharmakīrti might have used the argument – what it was supposed to prove – I make no
claim that the version of the SUN argument I will be considering was Dharmakīrti’s. Aside
from the fact that other scholars are much more qualified to ascertain that than I,1 that is,
tell us how the SUN argument as presented by Dharmakīrti should be understood, I wish
to be relieved of the burden of historical accuracy in this essay in order better to focus on
the logical and philosophical features of my reconstruction of it.
I believe it is worthwhile extracting the SUN argument from its historical and textual
context and considering it in isolation just as an argument for at least two reasons. First, I
believe that the version of the argument I shall devise is an argument for idealism unknown
to Western philosophers. One could call it a ‘new’ argument for idealism if there were any
traffic these days in arguments for idealism!2 Although it is not the case that idealism is
a hotly contested view now and a ‘new’ argument might just push it over the top, so to
speak – that is, bring it within the range of plausible or at least defensible philosophical
positions – any argument for it that is not obviously fallacious is of inherent interest. For
philosophers remain interested in the possibilities of reason. An argument that purports to
prove so unexpected a thesis as that nothing can exist outside of consciousness represents,
in the first instance, a bold claim about what philosophical reasoning can accomplish.
1
2
In addition to Iwata’s study see Arnold 2008, Ratié 2014, and Kellner 2017. Kellner’s study is, to date,
the most penetrating.
I know of only two defenders of idealism among contemporary philosophers: S. Rickless (2013) and J.
Foster (2008).
Birgit Kellner et al., eds., Reverberations of Dharmakīrti’s Philosophy: Proceedings of the Fifth International Dharmakīrti
Conference Heidelberg, August 26 to 30, 2014. Vienna 2020, pp. 441–462.
442
Philosophical Reflections on the sahopalambhaniyama Argument
Second, although I would hesitate to say that I consider the argument a viable one, it
does not, as far as I can tell, have any obvious flaws. It is valid, and its premises appear to
be at least tenable. I believe it is a strong argument – which is not to say, of course, that I
think it is conclusive.
I should say at the outset that I believe that the usual prejudice against idealism, the
opinion that there could not possibly be any good arguments for it, is wholly unphilosophical.
For what is it based on? Merely, as far as I can tell, the unshakeable conviction that we
live in a world populated by mind-independent, physical objects. The very thrust of any
argument for idealism, however, is to call this very conviction into question on the basis
of common observations about the nature of our experience. Typically – we will consider
an exception later – those who would see any argument for idealism as a non-starter, as
already refuted by perception, are simply begging the question.
Perhaps one final remark is in order to reinforce the value of investigating the SUN
philosophically as well as historically. Despite the interest in aspects of Indian philosophical
thought that are contiguous with contemporary analytic philosophy, one cannot deny that
many of the greatest Indian thinkers – such as Nāgārjuna, Dharmakīrti, and Śaṅkara – were
attempting to demonstrate views about the nature of reality that are radically opposed to
those of contemporary philosophers. In particular, they held that the empirical world is
an illusion. As some scholars now seem to recognize, in order to give a balanced picture
of Indian thought we should take Indian arguments for anti-realism just as seriously as
arguments concerning topics in philosophy of logic, epistemology, and philosophy of
language.3
The SUN argument as an argument for idealism
Thus far I have referred to the SUN argument as an argument for idealism. But was it really
meant to prove that? Here, I would like tentatively to explore this admittedly historical
question – though still in a philosophical way – which will set up the examination of the
argument itself. Iwata 1991 (henceforth “Iwata”) I, 18–20 suggests that Dharmakīrti, at
least, used it to prove only dvirūpatā: the thesis that a cognition includes within itself
two aspects, the object-form (arthākāra) and the subject-form (grāhakākāra, i.e., the
cognition itself as the “grasper”), hence that the immediate object of experience is the form
within cognition. Iwata cites as his main evidence the fact that, when presenting the SUN
argument both in his Pramāṇavārttika (PV 3.387–397) and his Pramāṇaviniścaya (PVin
1.54–58), Dharmakīrti either directly states or implies that the conclusion of the argument
– namely, that object or object-form and cognition are not different – holds even if there is
an external object.4 Thus, evidently, he believed that the argument would be acceptable
even to a Sautrāntika, who postulates something outside of cognition as the cause of the
object-form. The SUN argument is thus intended to refute the bāhyārthavādin who is also
3
4
I have in mind J. Westerhoff in particular. See, e.g., Westerhoff 2009 and 2010.
PVin 1 43,4–5: saty api bāhye ’rthe sahopalambhavedanābhyāṃ bhāsamānasya nīlades tatsaṃvidaś
cāvivekaḥ siddhaḥ. In PV 3.387–397 this is implied when Dharmakīrti, after developing the argument
in vv. 387–390c, concedes apparently to the Sautrāntika that one could infer an external object as the
cause of the arising of a perceptual cognition were it not possible to explain it as the effect of the
immediately preceding cognition (390d–391).
John Taber
443
a nirākāravādin, that is, in modern terminology, the ‘naive’ or ‘direct’ realist who holds
that we directly perceive physical objects. It does not rule out the existence of an external
object by itself.
Nevertheless, while some of Dharmakīrti’s successors followed him in viewing the
argument as a proof of dvirūpatā, hence in the first instance as a refutation of the nirākārajñānavāda, others, Buddhists and Brahmins alike, saw it as a proof of vijñaptimātratā,
mere-cognition: the thesis that there are no uncognized objects or that nothing exists outside
of cognition. Iwata notes how the argument might have been seen by Dharmakīrti himself
as practically amounting to a proof of mere-cognition. If what we are directly experiencing
are only forms in cognition, that is to say, if our experience of a world that seemingly
contains physical objects can be accounted for solely in terms of changes taking place
within cognition itself, then it becomes superfluous to postulate physical objects at all.
As Iwata puts it, “Insofar as the external object, in the sense in which it is postulated
in Sautrāntika … is fundamentally imperceptible, that is, it can only be represented by
means of inference, there is no longer any reason for it to be presupposed as the object of
knowledge” (Iwata I, 6). Thus, the external object becomes “epistemologically completely
superfluous” (Iwata I, 6). But perhaps that meant for Dharmakīrti that it does not exist
at all? Although Dharmakīrti has moved beyond the direct refutation of the possibility
of physical objects that Vasubandhu appears to undertake in his Viṃśikā – when arguing
that the Buddha could not have literally meant there are such things when he referred to
the āyatanas5 – he might still have been in sympathy with the overall strategy for proving
vijñaptimātratā in that text, as Kellner/Taber 2014 have now explained it. Namely, there
are no objects outside of cognition because there is no evidence for them. This strategy
tacitly appeals to the epistemological principle, employed in Indian philosophy prior to
Vasubandhu and still widely used for centuries after him (and after Dharmakīrti), that
something exists if there is a pramāṇa for it, and something does not exist if there is not.6
Kellner (2017: 29–31) is quite right to emphasize (contra Ratié 2014) that the constraints
of his logical theory would have prevented Dharmakīrti from devising a formal anumāna
based on an anupalabdhihetu demonstrating that objects outside of cognition do not exist.
Yet he surely must have known that when asking questions like, “If the cognition has the
form of blue, etc., what is the pramāṇa for an external object?”7 – implying that in light of
arguments like the SUN which undermine perception as evidence for external objects and
the lack of conclusive inferential evidence, there is no reason to postulate them – he was
coming as close as he could to asserting that they do not exist, and that is what most of his
readers would probably take away.8
5
6
7
8
See Arnold 2008. Although Manorathanandin, in one passage, recommends that one who demands a
proof that there is no external object even after it has been shown that the object is never experienced
independently of cognition should consult Vasubandhu’s mereological arguments, I don’t think one can
conclude that Dharmakīrti, who nowhere mentions them, endorsed them too. See Ratié 2014: 358–362.
Or, something does not exist anupalabdheḥ, “because it is not apprehended.” See Kellner/Taber 2014:
727–734.
dhiyo nīlādirūpatve bāhyo ’rthaḥ kiṃpramāṇakaḥ / PV 3.432ab. Or PV 3.333: “[Objection:] If an
external object were experienced, what mistake would there be? [Reply:] None at all! Only this [remains
to be asked:] Why would it be said that an external object is perceived?” Kellner 2017: 24.
Thus, Ratié 2014 shows that Abhinavagupta understood the SUN in this way while adapting it to his
own purposes: “…For the Śaivas the epistemic argument is not confined to a demonstration that there is
444
Philosophical Reflections on the sahopalambhaniyama Argument
Iwata generally presupposes that when arguing for the two-fold nature of cognition
(dvirūpatā) Dharmakīrti is in dialogue with the Sautrāntika or Buddhist ‘externalist’ (as
some prefer). There is good evidence for this. The framework of the initial discussion of
dvirūpatā in both PV 3 and PVin 1 is the identification of self-awareness as the result
(phala) of the pramāṇa perception, and Dharmakīrti wants to show that his solution is
acceptable to both the Sautrāntika and the Yogācāra. If the immediate result of the SUN
argument, however, is the demonstration of dvirūpatā, then the logical opponent against
whom it is directed would be, as I’ve already suggested, a bāhyārthavādin who is also
a nirākāravādin, rather than a Sautrāntika.9 Indeed, as is well known, much of the time
in his works Dharmakīrti is arguing against a Mīmāṃsaka, an incorrigible direct realist
who thinks that perception by itself establishes the existence of external objects. Especially
in this case – if the SUN targets a direct realist – it would be tantamount to a proof of
vijñaptimātratā. For the direct realist, unlike the representationalist (the Sautrāntika), does
not think we can infer the existence of physical objects; we just see them. Once it has been
shown to such a person, by proving dvirūpatā, that what he is seeing is really just the form
in cognition, then what he thought were external objects are eliminated and he has nothing
else, no inferred entities, with which to replace them.
Thus, the SUN argument, even when seen as a proof of dvirūpatā, could also be seen
as amounting to a proof of vijñaptimātratā if supplemented by a further step – whether or
not this was Dharmakīrti’s actual intention – namely, the claim that if there is no reason
to believe in objects outside of cognition, then it is reasonable to assume that there are
none. Or else it could be seen as amounting to a proof of vijñaptimātratā if one stipulates
that it is meant to be deployed in a specific dialectical situation: for the direct realist, if
what we are perceiving are not mind-independent objects, then there are none at all. On
neither of these readings would the proof of mere-cognition be a rigorously deductive one.
On the former, the conclusion that there are no uncognized objects is, in the final stage
of the argument, in effect offered as the best explanation for the lack of any evidence for
them. Thus, it is strictly something that could only be presumed, not asserted. On the latter,
insofar as the argument is context-bound, it will not be persuasive for all interlocutors (for
instance, non-Mīmāṃsaka representationalists).
What I would like to do, then, is develop a third reading of the SUN argument according to which it is rigorously deductive: that objects do not exist outside cognition will
follow immediately from the conclusion of the argument, that object and cognition are
not different. Some of Dharmakīrti’s opponents apparently perceived the argument in this
way, as sufficient by itself for proving cognition-only. Consider, for instance, Śaṅkara’s
presentation of it in his Brahmasūtrabhāṣya:
Moreover, a cognition arises having mere experience as its common nature.
The inclination (pakṣapāta) it has in regard to a particular object, so that it is
9
no argument proving the existence of an external object; or rather, the very demonstration that there is
no sādhaka whatsoever in favour of the external object constitutes a bādhaka, an argument refuting its
existence” (364). Kumārila seems to understand the strategy of the argument he is refuting similarly. See
ŚV, Śūnyavāda 7–9. If the ākāra we perceive belongs to the cognition, “then there is not the supposition
of an [external] object” (8d). In the context of the problematic of the Śūnyavāda that means that it does
not exist. See Taber 2010: 283.
Śāntarakṣita and Kamalaśīla would agree with this assessment, as Iwata I, 25–27 interprets them.
John Taber
445
the cognition of a pillar, the cognition of a wall, or the cognition of a pot – that
is not possible without some difference that belongs to the cognition [itself],
so that necessarily the cognition’s similarity with the object must be assumed.
And that being assumed, since the object-form is obtained (avaruddha) just
by the cognition, the postulation of an external object is pointless. Moreover,
the non-difference of object and cognition results from their invariable coapprehension; for it is not the case that there is the apprehension of the one
when there is no apprehension of the other. And this is not possible if there
is a difference of [their] nature, since there is no [other] cause [of their coapprehension] consisting in a connection. For that [reason] as well, there is no
[external] object.10
It is clear from this passage that Śaṅkara understands the proof of the fact that cognition
bears the form of the object, which is essentially dvirūpatā, as distinct from the proof of
the non-difference of object and cognition from their invariable co-apprehension. From
the cognition’s bearing the object-form within itself it follows only that an external object
is ‘pointless’ – that is, there is no reason to postulate one; but from the invariable coapprehension of object and cognition it follows that there is no external object.11
Therefore, one may even be historically justified in seeing the SUN argument as proving
vijñaptimātratā directly, without any additional, non-deductive step or specification of the
dialectical environment – hence as rigorously deductive. I shall now attempt to reconstruct
the argument explicitly along such lines.
The strongest formulation of the argument
We may take as our point of departure Dharmakīrti’s classic statement of the argument at
PVin 1.54ab:
Because they are invariably apprehended12 together, blue and the cognition of
it are not different.
sahopalambhaniyamād abhedo nīlataddhiyoḥ.
10
11
12
BSBh 544,1–6: api cānubhavamātreṇa sādhāraṇātmano jñānasya yo ’yaṃ prativiṣayaṃ pakṣapātaḥ
stambhajñānaṃ kuḍyajñānaṃ paṭajñānam iti nāsau jñānagataviśeṣam antareṇopapapdyata ity avaśyaṃ
viṣayasārūpyaṃ jñānasyāṅgīkartavyam. aṅgīkṛte ca tasmin viṣayākārasya jñānenaivāvaruddhatvād
apārthikā bāhyārthasadbhāvakalpanā. api ca sahopalambhaniyamād abhedo viṣayavijñānayor āpatati.
na hy anayor ekasyānupalambhe ’nyasyopalambho ’sti. na caitat svabhāvaviveke yuktaṃ pratibandhakāraṇābhāvāt. tasmād apy arthābhāvaḥ. The first sentence of this passage closely follows PV 3.302; the
penultimate sentence almost verbatim corresponds to PVin 1 40,2–4. Thanks to B. Kellner for pointing
this out.
But the final sentence of the passage, “For that [reason] as well (api), there is no external object,” may
suggest that Śaṅkara also thought that showing it is ‘pointless’ amounted to proving that it does not
exist. Such a proof, however, would involve an additional step. To say that something is pointless is
obviously not the same as saying it is nonexistent.
Although upalabdhi often means specifically perception, it also has the broader meaning of an apprehension by any pramāṇa, which I prefer to follow in my treatment because I think it is more suitable for
what I shall reconstruct as the first or major premise of the argument.
446
Philosophical Reflections on the sahopalambhaniyama Argument
I shall not be following Iwata’s method of examining the argument according to the terms
of an Indian anumāna: What is the dharmin, the hetu,13 and the sādhya, and so forth? I shall
rather be formulating the argument in the modern way, as a set of statements consisting
of premises and a conclusion following from them that could be translated into predicate
calculus if one wanted (though I shall not do that); for I believe that putting the argument
in this form makes it more accessible to philosophers. Nevertheless, what is normally
referred to as the subject of the argument (dharmin), that about which non-difference is
being demonstrated, should be clarified at the outset. Normally, blue and the cognition of
blue or, more generally, an object and the cognition of the object, are taken as the subject
of the argument, and what is being proved about them by the argument is that they are not
different.14 What is important to keep in mind when investigating the argument is that, in
order for it to be able to establish vijñaptimātratā, the status of blue, or the object, must be
undecided at the outset: it could be either not different from cognition or different from
it. That is to say, it must be left undecided whether blue is something occurring inside
or outside cognition. If one assumed at the outset that “blue” refers to an aspect of the
cognition itself, that is, to the so-called object-form (arthākāra or grāhyākāra), then one
would not prove anything very interesting or significant by demonstrating that it is not
different from the cognition.15 In fact, one might even in that case object that, if the subject
of the inference is interpreted in that way, and the purpose of the argument is to prove
vijñaptimātratā, then the argument would be assuming what it is supposed to prove. If, on
the other hand, the status of blue is left open, the argument immediately becomes much
more interesting. For it then establishes (if it is successful!) that what one might very well
have taken to be something external to cognition – a mind-independent object – turns out
in fact to be something that is not different from it. The achievement of the argument would
be, in effect, to bring everything that one is naively inclined to suppose exists outside of
cognition inside cognition itself, and that would be a rather spectacular result indeed.
Our initial task in evaluating the SUN argument is to formulate it in the best possible
way as a series of premises and a conclusion so as to avoid obvious prima facie objections.
Now some statements of the argument by Dharmakīrti himself suggest that what in Western
logic would be taken as the minor premise of the argument should be formulated as follows:
A. Whenever there is an apprehension of an object, there is an apprehension
of cognition and vice versa.
Thus, PV 3.387:
In what way can an object, which is invariably experienced at the same time
together with cognition, be shown to be different from that?16
13
14
15
16
He also asks, more specifically of the hetu, sahopalambhaniyamāt, what is the meaning of saha and
what is the meaning of niyama.
See Iwata I, 41–44.
Though Galen Strawson might disagree. See below. Precisely this point is made by Śubhagupta BSKā
87.
sakṛt saṃvedyamānasya niyamena dhiyā saha / viṣayasya tato ’nyatvaṃ kenākārareṇa sidhyati // Unlike
for Pramāṇaviniścaya 1, we do not have a critical edition of Pramāṇavārttika 3. I’ve used the edition of
Tosaki.
John Taber
447
Here, Dharmakīrti seems not to be talking about the cognition specifically of blue, as he
seems to be in PVin 1.54ab, but of cognition in general. That is, he is saying: whatever
object we happen to be aware of, that is not different from cognition.
Consider, however, a specific object, say blue, and the statement:
B. Whenever there is an apprehension of blue, there is an apprehension of
cognition and vice versa.
This statement, especially the conjunct implied by the phrase “and vice versa,” is clearly
false.17 We are conscious of many other things besides blue, namely, red, yellow, brown,
sweet, bitter, rough, smooth, and so on. Awareness in general is not invariably accompanied
by blue.
This point was very clearly stated by G. W. Moore in his famous essay “The Refutation
of Idealism:”
We all know that the sensation of blue differs from the sensation of green. But it
is plain that if both are sensations they also have some point in common. What
is it that they have in common? And how is this common element related to the
points in which they differ? I will call the common element ‘consciousness’
without yet attempting to say what the thing I so call is. We have then in
every sensation two distinct terms, (1) ‘consciousness,’ in respect of what all
sensations are alike; and (2) something else, in respect of which one sensation
differs from another. It will be convenient if I may be allowed to call this
second term the ‘object’ of a sensation: this also without attempting to say
what I mean by the word. (Moore 1922: 17)
Precisely because the object in consciousness varies while consciousness remains constant,
Moore goes on to say, we believe that consciousness is one thing, the object another.
In fact, we needn’t rely on a modern Western author to appreciate this point. It was
already effectively stated by Śaṅkara in his critique of the SUN argument:
Moreover, when there is a cognition of a pot [then] a cognition of a cloth,
there is a difference of the two qualifiers pot and cloth but not of the thing
qualified, the cognition; just as when there is a white cow [then] a black cow,
there is a difference of the whiteness and the blackness, but not of cowness.
And a difference of the one from the two is [thereby] established, and of the
two from the one. Hence, there is a difference of cognition and object.18
Thus, with B as a counterexample – the statement has now been shown to be false – it
seems, on the face of things, that A cannot be true.
17
18
That is to say, B is really the conjunction, “Whenever there is an apprehension of blue there is an
apprehension of cognition, and whenever there is an apprehension of cognition there is an apprehension
of blue.”
BSBh 550,1–3.
448
Philosophical Reflections on the sahopalambhaniyama Argument
On the other hand, it does seem plausible to suggest that blue and the cognition,
specifically, of blue are invariably apprehended together, as the wording of PVin 1.54ab
implies.
Now, if we take together all (presumably true) statements such as,
Blue is invariably apprehended together with the cognition of blue.
Red is invariably apprehended together with the cognition of red.
Etc.
then one can derive the universal generalization:
C. For every object X and every cognition of the object Cx, one apprehends X
if and only if one apprehends Cx,
which could also be stated as,
C1 . Any object and the awareness of that object are invariably apprehended
together.
And C1 looks very much like A above.19 Dharmakīrti does not make this clarification
himself, but it is not inconsistent with his statements. Indeed, when he says things like, “For
even though they have separate appearances, there is no nature of blue that is a different
thing from the experience,”20 (in his prose explanation of PVin 1.54ab), he could very well
mean by “experience” (anubhava) that particular type of experience, namely, an experience
of blue.
Thus, a general formulation of the minor premise of the SUN argument, with the terms
“object” and “cognition,” seems appropriate after all, such as PV 3.389:
Neither an object without awareness nor awareness without an object is ever
observed being experienced. Therefore, they are not distinct.21
Taking C1 as the minor premise of the argument, then, the following suggests itself as the
major premise:
D. Two things that are invariably apprehended together are not different.
Putting together D and C1 as the major and minor premises, we get the following formulation
of the argument.
19
20
21
Chakrabarti 1990: 32–33 comes up with virtually the same solution to this problem.
na hi bhinnāvabhāsitve ’py arthāntaram eva rūpaṃ nīlasyānubhavāt…, PVin 1 40,1. Cf. Iwata I,
110–111.
nārtho ’saṃvedanaḥ kaścit anarthaṃ vāpi vedanam / dṛṣṭaṃ saṃvedyamānaṃ tat tayor nāsti vivekitā //
John Taber
449
Premise 1 (major premise): Two things that are invariably apprehended
together are not different.
Premise 2 (minor premise): Any object and the cognition of that object
are invariably apprehended together.
Conclusion: Therefore, any object and the cognition of that object are
not different.
I maintain that the following statements are obvious corollaries of the argument:
In general, object and awareness are in some sense the same thing.22
There is no object external to/without cognition.
(For an object cannot exist unless there is awareness of it; hence, esse est
percipi.)
The world of objects that are experienced is “nothing but cognition.”
Thus, with the SUN argument formulated as suggested above, vijñaptimātratā follows from
it immediately.
Examination of premise 1
The SUN argument as formulated above is a logically valid argument: the conclusion
follows necessarily from the premises. To evaluate the argument further we must determine
whether the premises are true.
Now, one of the classic objections raised against the SUN argument is that what is
construed in Indian logic as the “reason” or hetu, namely, “because they are invariably
apprehended together,” is a “contradictory reason” (viruddhahetu); that is to say, it actually
proves the opposite of the desired conclusion. For when we say that things are apprehended
“together,” we obviously imply that they are different things. Thus,
I saw the ācārya walking together with his student,
and
I see the stars together in the sky,
prima facie suggest that one has to do, not with one thing, but with more than one thing.23
A related objection is that the reason “because they are invariably apprehended together”
22
23
For the purposes of my reconstruction of the SUN argument it is not necessary to specify whether “not
different” means merely not different or, taken as “non-different,” implies sameness. As far as I can
tell, either way the argument would go through and would amount to a proof of idealism. For Buddhist
defenders of the argument the debate about the precise meaning of abheda was driven by other doctrinal
issues, especially the problems of the unity and reality of the object-form. See Iwata I, 110–215.
See Iwata I, 93–94, referencing Jaina authors, e.g., Prabhācandra, and others (Iwata II, 84, n. 95).
The objection may have been raised first by the Vaibhāṣika (?) Śubhagupta; see BSKā 71, cited by
Kamalaśīla TSP 692,2–3. The example of the teacher and his student is from Yāmuna’s Ātmasiddhi; see
Mesquita 1990: 38. I regret that Saccone 2018 was not yet available for me to take into consideration in
this essay; to do so would have required substantial revisions.
450
Philosophical Reflections on the sahopalambhaniyama Argument
is inconclusive (anaikāntika); it does not reliably indicate that the things in question are
not different. However, it does not mean that they are different, either. An inconclusive
reason is not the same as a contradictory one. Thus, although the stars of a constellation
are invariably seen together, they are different.24
In my formulation of the argument these objections constitute challenges to the truth of
the first, major premise – which expresses what in Indian logic is the “pervasion” (vyāpti)
of the reason by the property-to-be-proved (sādhya). Thus, the question arises, How should
one interpret the first premise so that it is not vulnerable to these sorts of objections? Here,
there are at least two strategies one could follow. (1) One could stipulate a specific meaning
of the word “together.” (2) One could stipulate a specific meaning of the word “invariably.”
I shall discuss these suggestions in order.
Suggestion 1: The first strategy for defeating the sorts of objections raised above is to
suggest that one should understand “together” correctly. It does not mean simply juxtaposed
or simultaneous, but completely together, that is, inseparable like water and milk. Proposals
quite similar to this were in fact made by different Buddhist authors in response to the
objection that sahopalambhaniyamāt is a contradictory or inconclusive reason.25
This proposal is not without problems, however.26 For one thing, it would seem to
disqualify the example Dharmakīrti himself gives in support of his reason, “like two moons”
seen by a person afflicted by an eye disease. For the two moons, presumably, can be clearly
distinguished.27 Another problem with interpreting Premise 1 in this way, namely, as
saying that two things that are invariably apprehended as inseparable are not different, is
that it now sounds an awful lot like: two things that are invariably apprehended as not
different are not different. Thus, the non-difference of entities would already be given
with their invariable co-apprehension, so that by assuming, with Premise 2, the invariable
co-apprehension of object and cognition, one is assuming what one is supposed to be
proving – their non-difference. The argument becomes a petitio principii.28
More generally, however, it is difficult to come up with examples of two things invariably
seen together, as inseparable, that are the same thing. Consider:
the Commander-in Chief of the U.S. Armed Forces and the President of the
United States
24
25
26
27
28
See Iwata II, 84, n. 98; BSKā 68. The anaikāntikatā objection is generally thought to apply if saha is
understood specifically to mean “at the same time.”
Devendrabuddhi and Śākyabuddhi, for instance, interpret saha as eka, so that sahopalambha =
ekopalambha or abhedopalambha, an apprehension as united, or an apprehension as not different
(Iwata I, 67–71). Śāntarakṣita, Kamalaśīla, and Dharmottara interpret sahopalambha as eka evopalambhaḥ, a single apprehension [of both]. (See Iwata I, 84–88 on Dharmottara and 88–89 on Kamalaśīla.
Cf. Matsumoto 1980: 284–283. See n. 33 below.) For a schematic presentation of the different options
for construing saha see Iwata I, 66–67, and the entire section, 66–103, for a comprehensive discussion.
Bhāsarvajña neatly refutes most of the Buddhist maneuvers to evade these objections, NBhū 132–135.
Dharmakīrti himself says, PVin 1 40,1–2, that blue and its experience have separate appearances
“like two moons, etc.:” na hi bhinnāvabhāsitve ’py arthāntaram eva rūpaṃ nīlasyānubhavāt tayoḥ
sahopalambhaniyamād dvicandrādivat. I am not aware of this criticism being raised by a classical
author, but the literature is vast.
This is essentially the objection, raised by some classical opponents against the interpretation of
sahopalambhanaḥ as ekatvenopalambhanaḥ, that in that case the hetu would not be different from the
sādhya, resulting in siddhasādhanatā, “proof of what is already established.” See Iwata I, 94; BSKā 76.
John Taber
451
an Indian-Head Nickel and a Buffalo Nickel29
a duck-rabbit
It seems that the latter two cases in particular, where it is a question of a thing with different
visual aspects, are akin to the cognition of blue and blue. But are they seen together as
inseparable? Indeed, both aspects are not seen at the same time – seeing the Indian head,
one must turn the nickel over to see the buffalo and vice versa, and one can only shift from
seeing the duck to seeing the rabbit and back –, so it is difficult to understand how they
could be seen as inseparable or as one.
Thus, it becomes doubtful that Premise 1, even with a stronger construal of “together”
as “inseparable,” would have any basis in observation – unlike the major premise of most
syllogisms. We are confident in saying, “Where there is smoke, there is fire,” because
we have observed many times that something that is smoking is on fire and have never
observed that something that is smoking is not on fire, but that would not be the case here.
Suggestion 2: The second strategy for removing obvious challenges to the truth of
the first premise is to suggest that one has to understand “invariably,” which translates
niyama, correctly.30 Namely, one should take it modally, as meaning necessarily.31 Thus,
sahopalambhaniyamāt would mean “because they are necessarily apprehended together.”
One way of interpreting this would be that one cannot imagine apprehending one and not
apprehending the other.
I am not aware of any Buddhist authors making this move in response to the objections
of viruddhahetutva and anaikāntikahetutva.32 Dharmakīrti himself does not formulate
the relation between the apprehension of an object and the apprehension of a cognition
modally. Rather, he typically says things like, “For it is not the case that, when there is
not the apprehension of one form, there is apprehension of the other” (PVin 1 40,2–3). He
does not use modal language like “there could not be apprehension of the other.” We do,
however, find something like a modal formulation in the Tattvasaṅgraha:
That (X) whose awareness would necessarily (dhruvam) be the awareness of
that (Y), is non-different from that (Y) or does not differ from that.33
29
30
31
32
33
An American Indian-Head nickel was a five-cent coin minted in the United States from 1913 to 1933,
now a collector’s item. By definition, the nickel shows an Indian on the ‘heads’ side and a buffalo on
the ‘tails’ side.
Iwata translates niyama as “ausschließlich.”
Mesquita 1990 for instance translates it as “notwendigerweise,” Matsumoto 1980 as “necessarily.”
Iwata’s discussion of niyama is relatively short. It would seem that the interpretation of niyama was not
a matter of major concern in the debate over the SUN argument.
TS 2029: yatsaṃvedanam eva syād yasya saṃvedanaṃ dhruvam / tasmād avyatiriktaṃ tat tato vā na
vibhidyate // See Matsuoka, this volume, for a different translation. This verse, however, represents the
view that sahopalambha implies that there is just one apprehension of X and Y: eka evopalambhaḥ.
See TSP 692,23–24: na hy atraikenaivopalambha ekopalambha ity ayam artho ’bhipretaḥ. kiṃ tarhi?
jñānajñeyayoḥ parasparam eka evopalambhaḥ, na pṛthag iti. This is different from what I am suggesting,
and I think is attended by the sorts of problems brought out by Bhāsarvajña. See above, nn. 25 and 26.
452
Philosophical Reflections on the sahopalambhaniyama Argument
We also have the statement of Dharmottara:
Apprehension of the knowable [arises] only having the nature of the apprehension of the cognition, not otherwise; and apprehension of the cognition
has only the nature of the apprehension of the knowable.34
Let us, then, replace Premise 1 with
Premise 1’: Two things that are necessarily apprehended together are not
different.
Premise 1’ seems a priori justified. That is to say, the generalization expressed by Premise
1’ – the vyāpti of the traditional SUN argument – is not based on observation but is a truth
of reason, if you will.
In support of this claim I refer to a recent article by Galen Strawson, “What is the
Relation Between an Experience, the Subject of the Experience, and the Content of the
Experience?” from his collection Real Materialism.35 In this essay Strawson argues that
for any occurrent experience e, e cannot exist without a subject of the experience s; the
subject of the experience s cannot exist without an experiential content it is experiencing,
c; and the experience e also cannot exist without an experiential content c and vice versa.
Thus we get
e ⇔ s ⇔ c,
where “x ⇔ y” means “if x then necessarily y, and if y then necessarily x.”
Strawson does not appeal to invariable co-apprehension to support the equivalences e
⇔ s ⇔ c; rather, he notes certain conceptual relationships. For instance, in arguing for e ⇔
s he writes, “There cannot be an experience without a subject of experience simply because
experience is necessarily experience for – for someone-or-something. Experience involves
experiential ‘what-it-is-likeness,’ and experiential what-it-is-likeness is necessarily whatit-is-likeness for someone-or-something.”36 In arguing for the equivalence e ⇔ c Strawson
offers observations such as, “Evidently there can’t be concretely occurring experiential
content without there being an experience of some sort which the content is the content
of,”37 and, “For surely this very experience couldn’t have had a different content and still
be the experience it is?”38 Thus, Strawson’s method for establishing equivalences for e,
s, and c is different from that of the SUN argument – which of course is only concerned
34
35
36
37
38
jñeyopalambho jñānopalambhātmaka eva nānyathā / jñānopalambho ’pi jñeyopalambhātmaka eva //,
from Dharmottara’s Pramāṇaviniścayaṭīkā. See Iwata I, 107–108 and II, 89, n. 15.
Strawson 2008. Thanks to Mark Siderits for bringing this article to my attention.
Strawson 2008: 153. It should be noted that the type of subject Strawson conceives of as necessary
for experience is a ‘thin’ subject, which is not ontologically distinct from the experience but only its
subjective pole and which is different from one experience to the next. Thus, it is not incompatible
with a Buddhist view of consciousness which involves no self distinct from cognition but includes a
subjective factor, namely, the grāhakākāra or subject form. See esp. Strawson 2008: 182–183, also
167–168.
Strawson 2008: 153.
Strawson 2008: 177.
John Taber
453
with the necessary equivalence e ⇔ c. Moreover, Strawson takes c to be the content of
an experience, an “experiential content;” it is “internal,” “phenomenological.” Thus, it
belongs to the experience at the outset. Its status is not initially undetermined, as merely
the knowable that might or might not belong to the cognition, as it is on my interpretation
of the SUN argument. Nevertheless, Strawson’s reflections become particularly useful for
us when he suggests that one can move from
e⇔s⇔c
to
e=s=c
by force of the following principle, which he borrows from Descartes:
There is a real distinction between two phenomena … if and only if they can
possibly “exist apart,” and a merely conceptual distinction between them if
and only if they are conceptually distinct, like trilaterality and triangularity,
but cannot possibly exist apart.39
Strawson (and Descartes) accept this as an a priori principle. Following them, then, it seems
a priori justified – that is, not based on any observational data – that
Two things that are necessarily apprehended together are not different,
which is our Premise 1’. Of course, to say that two things are necessarily apprehended as
co-occurring is slightly different from saying that two things necessarily co-occur, but it
seems the difference is so slight as to be negligible. The only way to know that things are
a certain way is by apprehending them as being so. I shall, however, return to this point
below.
Before moving on to consider Premise 2, it may be helpful to clarify how Strawson is not
proving the same thing as the SUN argument in the way I am interpreting it. Indeed, it would
seem that by arguing for e = c Strawson is arguing for what the Buddhists referred to as
dvirūpatā, essentially, the non-difference of the object-form from the cognition (cognition,
that is, has two aspects: the cognition and the object), which I have distinguished from
vijñaptimātratā above. Strawson of course is not interested at all in the thesis of idealism;
however, he believes that e = s = c, hence e = c, is “a deep truth.”40 He believes it is
particularly significant that s and c cannot exist apart, hence that they must be identical in
some sense – the content of the experience is the subject. Why? The main reason appears
to be that Strawson, as a physicalist or materialist, believes that all of e, s, and c are not just
correlated with, but are physical entities: portions of “process-stuff” in the nervous system,
39
40
Strawson 2008: 168. Strawson cites Descartes 1985: I, 213–215 (Principles of Philosophy). See also
Strawson 2008: 181: “…Let me now formally endorse the principle that if there is at most a conceptual
distinction between two apparently distinct (concrete) particulars, if they cannot possible exist apart,
then they are not really two but only one…”
Strawson 2008: 180.
454
Philosophical Reflections on the sahopalambhaniyama Argument
which will eventually be pinpointed by neuroscience. The equivalence e = s = c, then,
predicts that the neuroscientists will someday discover that the neurological process-stuff
that is e ‘just is’ the neurological process-stuff that is independently identified as s, and
that s ‘just is’ the neurological process-stuff that is independently identified as c.41 Here we
have – though Strawson does not exactly say this – a priori philosophy proposing a specific
research program in brain science! Be that as it may, in the end Strawson’s essay offers
nothing for the Buddhist; for materialism is just as much a non-starter for the Buddhist – at
least for Dharmakīrti – as idealism is for the contemporary analytic philosopher.42
It seems that in light of this examination of Premise 1’ it is safe to say that the first
premise of the argument, for now, looks quite solid. Thus, having established that the
argument is valid, that is, commits no logical mistake, and that its first premise appears
true, it would seem that the best hope for defeating it is to show somehow that the second
premise is false.
Examination of premise 2
Is it true that object and cognition are necessarily apprehended together? Well, it seems
uncontroversial that one cannot possibly apprehend a cognition of blue without apprehending what it is a cognition of – blue. What seems, and was for Indian philosophers, much
more controversial is the other half of the equivalence: that one could not apprehend blue
without apprehending the cognition of it. Dharmakīrti is clearly attempting to establish
this with the second half of PVin 1.54:
The seeing of an object is not established for someone for whom the apprehension is not evident.43
The idea is that a cognition that does not reveal itself cannot reveal an object. Dharmakīrti
continues,
For there is not an apprehending of an object due to the presence of the object,
but rather due to the presence of the apprehension. And if that [presence]
is not known by a pramāṇa it does not attach itself to/conform to [verbal,
conceptual, and corporal forms of] behaviour that presuppose existence. If
that [apprehension] were not established, there would be no establishing of
the object either, so everything would disappear; for if it were not established,
even though it exists, it cannot be treated as existing. Therefore, someone who
does not apprehend the awareness of something is not aware of anything.44
41
42
43
44
Strawson 2008: 183–185.
See Taber 2003.
apratyakṣopalambhasya nārthadṛṣṭiḥ prasidhyati // Iwata I, 106.
PVin 1 41,1–5. na hi viṣayasattayā viṣayopalambhaḥ kiṃ tarhi tadupalambhasattayā. sā cāpramāṇikā
na sattānibandhanān vyavahārān anuruṇaddhi. tadaprasiddhau viṣayasyāpy aprasiddhir ity astaṅgataṃ viśvaṃ syāt. sato ’py asiddhau sattāvyavahārāyogyatvāt. tasmān nānupalabhamānaḥ kasyacit
saṃvedanaṃ vedayate nāma kiñcit. My translation follows Kellner 2011: 420 in part.
John Taber
455
Dharmakīrti seems to be saying that apprehending something, but not being aware that one
is apprehending it, is as good as not apprehending it at all.
Despite the strong prima facie appeal of this view it was aggressively attacked by
opponents of Yogācāra anti-realism. One of the most forceful critiques is by Kumārila,
who in the Śūnyavāda chapter of his Ślokavārttika is taking on an argument that anticipates
the SUN, which was already known to the Vṛttikāra: if the form (ākāra) that appears in a
cognition can be shown to belong to the cognition itself, then there is no reason to postulate
an object.45 In arguing for the possession of the form by the cognition the pūrvapakṣin
offers, among other reasons, that for something to illumine or manifest anything it has
to be apprehended itself: “… the object which is to be illumed is not ascertained when
the appearance of the cognition is not apprehended, because its illumination is dependent
on that, like a pot when there is the light of a lamp.”46 The opponent even says that the
cognition must reveal itself. We fail to be aware of something only because either a source
of illumination is lacking or there is something obstructing it. When a cognition arises,
however, neither is the case: nothing obstructs it, and it itself is the source of illumination.47
Moreover, if a cognition necessarily apprehends itself, it must apprehend itself as having
some form. Thus, the ākāra must belong to the cognition.48
Kumārila’s refutation of this argument in his siddhānta is extensive and complex. A
better understanding of it awaits a new edition and translation of the Śūnyavāda. Nevertheless, it is clear that it involves at least the following three elements. (1) In fact we are
not aware of both object and cognition at the same time. Usually we are only aware of the
object. “…When one cognizes the object (grāhya), the form blue and so forth, one does
not at that time ever observe an awareness having the form of the subject (grāhaka).”49
And sometimes one is just aware of the subject without the object, for instance when
one remembers experiencing something but cannot remember what it was.50 (2) It is not
necessary for the cognition itself to be apprehended in order for it to cognize an object.
The senses provide a counterexample: we apprehend things by means of them without
apprehending them themselves.51 (3) There is no example of something that illumines
itself as it illumines other things. Fire and so forth illumine other objects without their own
natures being revealed. When they are apprehended, it is the sense that apprehends them.52
In general, Kumārila seems to subscribe to the principle that something cannot act on itself:
an axe cannot cut itself, a finger cannot touch itself, and so on.
Obviously, we cannot adjudicate this dispute here – essentially, the debate about the
truth of the second premise of the SUN argument – which continued for centuries after
Kumārila and Dharmakīrti. We should note, however, that there is widespread support in
modern philosophy, in both the analytic and continental traditions, for the thesis that some
45
46
47
48
49
50
51
52
See Taber 2010.
ŚV, Śūnyavāda 22.
ŚV Śūnyavāda 23–24.
Taber 2010: 284–287.
ŚV Śūnyavāda 74.
ŚV Śūnyavāda 82cd–83.
ŚV Śūnyavāda 179cd–180ab.
ŚV Śūnyavāda 64–66.
456
Philosophical Reflections on the sahopalambhaniyama Argument
kind of self-awareness or reflexivity is essential to consciousness. Consider the following
précis of Sartre’s position by the contemporary phenomenologist Dan Zahavi.
Sartre, probably the best-known defender of a phenomenological theory of
self-consciousness, considered consciousness to be essentially characterized
by intentionality. He also claimed, however, that each intentional experience is
characterized by self-consciousness. Thus, Sartre took self-consciousness to
constitute a necessary condition for being conscious of something. To perceive
a withering oak, a dance performance, or a red pillow consciously without
being aware of it, that is, without having access to or being acquainted with
the experience in question was, for Sartre, a manifest absurdity. This line
of thought is elaborated in the important introduction to L’être et le néant,
where he claimed that an ontological analysis of intentionality leads to selfconsciousness since the mode of being of intentional consciousness is to be
for-itself (pour soi), that is, self-conscious. An experience does not simply
exist; it exists for itself, that is, it is given for itself, and this self-givenness
is not simply a quality added to the experience, a mere varnish, but it rather
constitutes the very mode of being of the experience. As Sartre wrote: “This
self-consciousness we ought to consider not as a new consciousness, but as the
only mode of existence which is possible for a consciousness of something.”53
Thus, it seems obvious that Premise 2, if not actually true, is at least defensible – many
philosophers of different traditions have held and defended it. The main point to be made
here in favor of the SUN argument is this, that if one accepts Premise 2, as it appears many
do, then, given that Premise 1 already looks quite strong, the argument will go through.
Further objections?
With the strategy of refuting the SUN argument decisively by contesting its premises not
looking very hopeful, one is challenged to come up with other ways to attack the argument.
(1) Perhaps one can ask, Does the argument really establish vijñaptimātratā as its
conclusion? Is the non-difference of object and cognition really tantamount to the idealist
thesis that esse est percipi – an object cannot exist unless there is an awareness of it – as I
maintained above?
Consider in this connection the one-liner the Vṛttikāra throws out at the end of his
refutation of the Buddhist argument that cognition is ‘empty,’ that is, without any objective
53
Zahavi 2005: 12. Note Zahavi’s claim about the ubiquitousness of this view: “One should not overestimate the homogeneity of the phenomenological tradition; like any other tradition, it spans many
differences. Although phenomenologists might disagree on important questions concerning method
and focus, and even about the status and existence of a self, they are in nearly unanimous agreement
when it comes to the relation between consciousness and self-consciousness. Literally all the major
figures in phenomenology defend the view that the experiential dimension is characterized by a tacit
self-consciousness” (11). Kellner 2017: 21, interestingly enough, also refers to Sartre’s view of consciousness when discussing the so-called saṃvedana argument and cites Zahavi’s treatment of it, but
she warns that the conclusion Dharmakīrti draws, that cognition cannot be of an external object, “is one
that phenomenologists shun.” My point is simply that the SUN argument would force them to accept it.
John Taber
457
support, because the form (ākāra) we are aware of belongs to the cognition, hence object
(artha) and cognition (buddhi) are one:
Moreover, if one likes, if there is sameness of nature [of the object and the
cognition], then it is the cognition that doesn’t exist, not the object, which is
[after all] perceptible.54
That is to say, if one proves the non-difference of object and cognition, then what is the
true nature of the (one) thing they both are? Is it the object or the cognition? It seems that
there is no way to decide this question. Hence, it would be illegitimate to draw the idealist
conclusion, that it is only the cognition that is real, and that the object, although it appears
to be something distinct from the cognition, is really just the cognition!55
The Vṛttikāra, in posing this question, seems to be presupposing that the object has some
determinate nature that the cognition could also turn out to have; presumably, it is physical
in nature. In doing so, however, he ignores the stipulation I made at the beginning of my
exposition of the SUN argument – but which should hold for any argument for idealism
– that the status of the object must be left undecided at the outset in regard to whether it
is an aspect of the cognition or not. For, once again, if one assumes that it is merely the
object-form, then the argument proves very little; but equally, one cannot assume that it is,
say, a physical object that could exist unperceived – the very opposite of what the argument
is supposed to prove. Given that the SUN argument establishes that object and cognition
are not different, the independent existence of the object – as, say, something physical –
is ruled out by the argument itself and the only remaining possibility is that it belongs
to the cognition. In other words, were it shown (by some other argument) that the object
does exist independently of the cognition, then one might conduct an investigation into its
nature, whether for instance it is something physical. Having proven, on the other hand,
that object and cognition are not different, and the nature of the cognition as something
internal and phenomenal being beyond doubt, the object acquires that nature as well.
(2) A different approach – one historically attempted by certain Brahmin authors – is
to claim that the very thesis (pratijñā) of the argument is contradicted by perception. This
objection probably originated with Kumārila but it is very forcefully stated, once again, by
Śaṅkara:
54
55
Frauwallner 1968: 30,10–11: api ca kāmam ekarūpatve buddher evābhāvo, na tv arthasya pratyakṣasya
sataḥ.
Ernst Steinkellner presented this objection to me after I delivered a preliminary version of this paper
at the Fifth International Dharmakīrti Conference, in Heidelberg, in August, 2015. He did not refer
to the passage in the Śābarabhāṣya, and I suspect he came up with it on his own. He went on to say,
with a mischievous twinkle in his eye, that whenever he tries to think about the philosophical problems
the Buddhist epistemologists are concerned with he finds himself faced with similar conundrums –
and that is why he prefers not to think about them. The Buddha, perhaps, would have said the same
thing. It was in order to avoid a similar problem that some Buddhist interpreters of the argument,
e.g., Dharmottara, suggested that “non-difference” (abheda) should be understood to connote just the
negation of difference and not also the affirmation of identity, that is, as a prasajyapratiṣedha rather than
a paryudāsa. Otherwise, if cognition and object were the same, then the object-form being unreal (as
some believed), the cognition would become unreal. See Matsumoto 1980: 281–278; Iwata I, 173–183.
458
Philosophical Reflections on the sahopalambhaniyama Argument
The non-existence of an external object certainly cannot be ascertained. Why?
Because it is apprehended (upalabdheḥ).56 For, for each cognition an external
object is apprehended, as [for instance] a post, a wall, a cloth, [or] a pot. And
something that is apprehended cannot not exist. Just as if someone who is
eating, experiencing the satisfaction produced by the food for himself, were
to say, “I am not eating,” or “I am not satisfied,” so if someone apprehending
for himself an external object by means of contact with his sense faculty were
to say, “I do not apprehend it, and it does not exist” – how could we take his
statements seriously?57
In other words, perception itself reveals to us objects outside consciousness. The externality
of the object, the fact that it exists independently of being perceived, is given as part of
the content of perception itself. As Kumārila puts it, the idea that arises, upon perceiving
something, that one’s cognition has an external support, is never overturned – unlike, say,
dreaming cognitions. Therefore, by implicit appeal to the principle of intrinsic validity that
a cognition is true unless and until it is overturned by another, that idea is correct.
Moreover, the notion that a cognition has an external object is true, for it is a
notion devoid of any contradiction, like the cognition [one has upon waking]
that contradicts a dream.58
This argument is more sophisticated than the mere prejudice against idealism that I dismissed as question-begging at the beginning of this essay, for it is an actual argument that
provides us with a reason for believing that our conviction that we are perceiving things
outside us is true.59 Nevertheless – and very briefly – I do not think this is a fatal objection
to the SUN argument. Even if our conviction in the externality and independence of the
objects we experience is firm and never overturned, there is still no explanation for how
such a conviction could be true. We would, somehow, have to be able to step outside our
minds – adopt a “God’s-eye view” – and observe that our perceptions are really being
caused by physical objects. Needless to say, that is impossible.60
56
57
58
59
60
This glosses BS 2.2.28: nābhāva upalabdheḥ. This by itself suggests that the Brahmanical understanding
of the purport of Buddhist arguments against the existence of external objects was that they do not exist
because they are not apprehended (anupalabdheḥ). They understood the Buddhist arguments against
the existence of a self in the same way. See Kellner/Taber 2014: 726–727; 731–732.
katham upādeyavacanaḥ syāt. BSBh 547,4–548,4. See the continuation of the passage. Śaṅkara points
out, among other things, that when the Buddhist says the object-form appears as if it were an external
object, he tacitly admits that perception itself reveals to us that the objects we experience are external.
bāhyārthālambanā buddhir iti saṃyak ca dhīr iyam // bādhakāpetabuddhitvād yathā svapnādibādhadhīḥ / ŚVT Nirālambanavāda 79cd–80ab. ŚVM reads svapnādibāhyadhīḥ in 80b.
Kumārila, ŚV Nirālambanavāda 79ab, introduces this as a counter-proof (pratisādhana) that renders
the reason or thesis of the idealist’s anumāna defective (he is considering a different one in the Nirālambanavāda). The reason would become, technically, a viruddhāvyabhicārihetu, or else the argument’s
thesis would be disproven by another inference (78cd).
Ratié 2010: 444–445, n. 23 and 2011: 484–486 shows how Utpaladeva and Abhinavagupta, and
apparently certain Vijñānavādins, used this type of argument to refute the Sautrāntika: we cannot infer
the existence of external objects as the cause of our perceptual experiences, because we cannot perceive
them independently of cognizing them, in order to establish a causal relationship between them and our
John Taber
459
(3) Having exhausted two initially promising avenues of attack against the SUN argument we are compelled to return to its premises and scrutinize them one last time. I do
think that Premise 2, though not proven, is certainly defensible, and so I shall not attempt to
investigate it further. There was, however, a point that came up at the end of my discussion
of Premise 1’ that I did not pursue.
It certainly seems indisputable that if one thing can possibly be apprehended without
the other, they are different. This is equivalent to the statement
If two things are not different, then they are necessarily apprehended together.
But is the converse, which is Premise 1’, true?
If two things are necessarily apprehended together, then they are not different.
I have offered in support of this principle the examples of the President of the United
States and Commander-in-Chief of the U.S. Armed Forces and an Indian Head Nickel
and a Buffalo Nickel, but ended up doubting that it could be established a posteriori. I
proposed that it be accepted as a priori justified instead, and appealed to the principle Galen
Strawson (relying on Descartes) articulates in asserting the identity of experience, subject,
and content of experience, namely (in one of its formulations):
If two things cannot possibly exist apart, so that there is only a conceptual
distinction between them, then they are one.
The wording of this principle, however, is slightly different from Premise 1’ above. Premise
1’ refers to the apprehension of two things, whereas Strawson’s principle refers to the
existence (or occurrence) of two things. Now, the question arises: Does this difference in
wording constitute an important difference?
It seems that an opponent of the SUN argument could claim that it does. It may be
the case that one never apprehends an object without apprehending the cognition that
apprehends it, and that it is even inconceivable that one could. But why, exactly? It could
indeed be because one can only apprehend an object by means of a cognition, and – as we
have seen Sartre argue – all cognitions are self-aware, so that whenever one is aware of
an object one is also aware of one’s cognition. Similarly, one can see an object by means
of light, and light also illumines itself; when one sees an object one is aware of both the
object and the illumination. Yet the object is different from the light that illumines it.
The word “apprehension” in Premise 1’, then, introduces a condition that guarantees
that object and cognition will seem invariably to co-occur. Precisely because it qualifies
the co-occurrence of two things as their apprehended co-occurrence, then, Premise 1’ does
not rule out categorically that an object could occur without a cognition. Indeed, that is
precisely the possibility that the realist thinks she envisions.
experiences. As Abhinavagupta puts it, “…Since the external object is not an object of perception, one
cannot establish any causal relation between this [external object] and anything else; as a consequence,
the cognition [of this or that object] is not a kāryahetu in regard to this [object] to be inferred” (Ratié
2011: 485, slightly adjusted).
460
Philosophical Reflections on the sahopalambhaniyama Argument
Again, it is Śaṅkara who states the problem about as succinctly as one could: “For
this reason also,61 it is to be assumed that invariable co-apprehension, too, is caused by
cognition and object being means and what is assisted (upāyopeyabhāvahetuka), not by
their non-difference.”62 Here he is taking advantage of an opening left by Dharmakīrti
himself. In his prose discussion of PVin 54ab Dharmakīrti clarifies that the invariable
co-apprehension of object and cognition establishes the non-difference of their nature,
“because there is no [other] cause [of their co-apprehension] consisting in a connection”
(pratibandhakāraṇābhāvāt).63 In the case of visible form and light, on the other hand, there
is a connection consisting of the object’s acquiring the capacity to produce a cognition
of visible form, or else the arising of a capacity of the sense to cognize it.64 It is this
connection, and not their non-difference, that accounts for their co-apprehension – the light
enables the apprehension of visible form.
What is most interesting is that Dharmakīrti does not think that there is any other
factor that determines the invariable co-apprehension of object and cognition other than
their non-difference. The question is, Why not? Why couldn’t, as Śaṅkara suggests, their
co-apprehension be due to the fact that the cognition is what enables us to apprehend the
object?
I believe that this is the most serious objection that can be leveled against the SUN
argument. Unfortunately, I do not have space to develop an adequate response to it, which
would take us deep into difficult philosophical issues (especially in philosophy of mind)
– and also, of necessity, back to the texts. For the Buddhists themselves, of course, were
aware of this problem, which may have been first formulated by Śubhagupta: the invariable
co-apprehension of object and cognition may be due, not to their non-difference, but to
the fact that the preceding causal complex causes them to arise at the same time as grāhya
and grāhaka, respectively.65 And Dharmakīrti already indicates the line of response later
Buddhist epistemologists will take: “The other object, however, which arises simultaneously
[with the cognition] does not shape [the cognition], because it does not influence it; how
would [the object] be manifest, which the cognition [in turn] does not shape with its own
form?”66 That is to say, simultaneously arising entities cannot stand in a causal relationship
to each other, any more than, say, “the right and left horns of a cow.”67 But more generally,
and independently of the historical development of the arguments, the question is: How can
there be any causal relation between cognition and physical object at all? That is, How can
61
62
63
64
65
66
67
The argument that precedes this was that the fact that the cognition is similar in form to the object does
not eliminate the object; to the contrary, it presupposes it, moreover the object is perceived externally
(bahirupalabdheś ca viṣayasya) (BSBh 549,4–5).
ata eva sahopalambhaniyamo ’pi pratyayaviṣayayor upāyopeyabhāvahetuko nābhedahetuka ity abhyupagantavyam, BSBh 549,6–550,1.
Iwata I, 221: “denn es gibt keine in einer Verbindung bestehende Ursache [für die Ausschließlichkeit
des Zusammenwahrgenommenwerdens].”
PVin 1 40,3–7. See Iwata I, 221–226. His presentation is rather dense and I am not sure how he
understands the passage.
See BSKā 66; Kamalaśīla presents Śubhagupta’s position at TSP 694,9–20 (ad TS 2031), citing BSKā
66 and 81.
nākārayati cānyo ’rtho ’nupakārāt sahoditaḥ / vyakto ’nākārayañ jñānaṃ svākāreṇa kathaṃ bhavet //
PV 3.420. Cf. Iwata I, 231.
Cf. TSP 695,4–5: nāpi tadutpatteḥ sahavedanam, sahabhūtayoḥ kāryakāraṇābhāvāt; Iwata I, 228–233.
The analogy is from Nyāyakumudacandra, cited by Iwata I, 232.
John Taber
461
consciousness illumine an object? Even if the (questionable) assumption that object and
cognition arise simultaneously is abandoned, the realist has a lot of explaining to do! The
defender of the SUN argument can defend the first premise simply by shifting the burden
of proof: Explain to me how cognition, which is experienced as being non-material and
non-spatial, can influence a physical thing to bring about the quality of its being manifest
to us, or how could a physical object impose its form upon cognition?68
In summary, the SUN argument appears to be a strong one. It is not invalid; there
is evidence in support of its premises; and initial, and even secondary, objections to the
premises can be answered. Even though one may not be convinced by it, it is not easy to
refute.69
References and abbreviations
Arnold 2008 D. Arnold, Buddhist Idealism, Epistemic and Otherwise: Thoughts on the
Alternating Perspectives of Dharmakīrti. Sophia 47 (2008) 3–28.
BS Brahmasūtra. See BSBh.
BSBh Brahmasūtrabhāṣya of Śaṅkara: Brahmasūtra Śāṅkara Bhāṣya, with the Commentaries Bhāmatī, Kalpataru and Parimala, ed. A. Śāstri. Varanasi 1982.
BSKā Bāhyārthasiddhikārikā of Śubhagupta, ed. N. A. Sastri. Bulletin of Tibetology 4/2
(1967) 1–96.
Chakrabarti 1990 A. Chakrabarti, On the Purported Inseparability of Blue and the Awareness of Blue: An Examination of sahopalambhaniyama. In: Mind Only School and
Buddhist Logic, ed. D. Tulku. Delhi 1990, 17–36.
Descartes 1985 R. Descartes, The Philosophical Writings of Descartes, trans. J. Cottingham et al. 2 vols. Cambridge 1985.
Foster 2008 J. Foster, A World for Us: the Case for Phenomenalistic Idealism. Oxford
2008.
Frauwallner 1968 E. Frauwallner, Materialien zur ältesten Erkenntnislehre der Karmamīmāṃsā. Vienna 1968.
Iwata 1991 T. Iwata, Sahopalambhaniyama: Struktur und Entwicklungen des Schlusses
von der Tatsache, daß Erkenntnis und Gegenstand ausschließlich wahrgenommen
werden, auf deren Nichtverschiedenheit. 2 vols. Stuttgart 1991.
Kellner 2011 B. Kellner, Self-Awareness (svasaṃvedana) and Infinite Regresses: A Comparison of Arguments by Dignāga and Dharmakīrti. Journal of Indian Philosophy
39 (2011) 411–426.
Kellner 2017 B. Kellner, Proofs of Idealism in Buddhist Epistemology: Dharmakīrti’s
Refutation of External Objects. In: Indian Epistemology and Metaphysics, ed. J.
Tuske. London 2017, 102–128.
68
69
This is essentially what Jñānaśrīmitra does in his Advaitabinduprakaraṇa. His line of investigation
strikes one as an elaboration of the thought behind what is known as the saṃvedana argument: awareness
is by nature just the “appearing in a certain way,” and like the awareness of awareness, can only be of
awareness. See Kellner 2017: 20–22.
I believe that Berkeley’s arguments for idealism, for instance, are easier to refute.
462
Philosophical Reflections on the sahopalambhaniyama Argument
Kellner/Taber 2014 B. Kellner and J. Taber, Studies in Yogācāra-Vijñānavāda Idealism I:
the Interpretation of Vasubandhu’s Viṃśikā. Asiatische Studien/Études Asiatiques
68/3 (2014) 709–756.
Matsumoto 1980 S. Matsumoto, Sahopalambha-niyama. Sōtōshū Kenkyūin Kenyūsei
Kenkyū Kiyō 12 (1980) 298–265.
Mesquita 1990 R. Mesquita, Yāmunācāryas Philosophie der Erkenntnis: eine Studie zu
seiner Saṃvitsiddhi. Vienna 1990.
Moore 1922 G. E. Moore, The Refutation of Idealism. In: Philosophical Studies. London
1922, 1–30.
NBhū Nyāyabhūṣaṇa of Bhāsarvajña: Nyāyabhūṣaṇam, ed. Yogīndrānanda. Varanasi 1968.
PV 3 Pramāṇavārttika of Dharmakīrti, Chapter Three, ed. H. Tosaki. Tōkyō 1985.
PVin 1 Dharmakīrti’s Pramāṇaviniścaya, Chapters 1 and 2, ed. E. Steinkellner. Beijing/
Vienna 2007.
Ratié 2010 I. Ratié, The Dreamer and the Yogin: On the relationship between Buddhist
and Śaiva idealisms. Bulletin of School of Oriental and African Studies 73/3 (2010)
437–478.
Ratié 2011 I. Ratié, Can One Prove that Something Exists Beyond Consciousness? A
Śaiva Criticism of the Sautrāntika Inference of External Objects. Journal of Indian
Philosophy 39 (2011) 479–501.
Ratié 2014 I. Ratié, On the Distinction Between Epistemic and Metaphysical Buddhist
Idealisms: A Śaiva Perspective. Journal of Indian Philosophy 42 (2014) 353–375.
Rickless 2013 S. Rickless, Berkeley’s Argument for Idealism. Oxford 2013.
Saccone 2018 M. S. Saccone, On the Nature of Things: A Buddhist Debate on Cognitions
and Their Object. Vienna 2018.
Strawson 2008 G. Strawson, What is the Relation Between an Experience, the Subject of
the Experience, and the Content of the Experience? In: Real Materialism, and Other
Essays. Oxford 2008, 151–188.
ŚVM Ślokavārttika of Kumārila. In: Ślokavārtikavyākhyā Tātparyaṭīkā of Uṃveka Bhaṭṭa,
ed. S. K. Ramanatha Sastri. Madras 1971.
ŚVT Ślokavārttika of Kumārila: Mīmāmsā Slokavārttika, with the Commentary Kasika of
Sucaritamiśra, ed. K. Sāmbaśiva Sāstrī. Parts 1 and 2. Trivandrum 1990.
Taber 2003 J. Taber, Dharmakīrti Against Physicalism. Journal of Indian Philosophy 31
(2003) 479–503.
Taber 2010 J. Taber, Kumārila’s Buddhist. Journal of Indian Philosophy 38 (2010)
279–298.
TS(P) Tattvasaṅgraha of Śāntarakṣita and Tattvasaṅgrahapañjikā of Kamalaśīla: Tattvasaṅgraha of Ācārya Shāntarakṣita, with the Commentary Pañjikā of Shri Kamalashīla,
ed. D. Shastri. Varanasi 1982.
Westerhoff 2009 J. Westerhoff, Nāgārjuna’s Madhyamaka: a Philosophical Introduction.
Oxford 2009.
Westerhoff 2010 J. Westerhoff, Twelve Examples of Illusion. Oxford 2010.
Zahavi 2005 D. Zahavi, Subjectivity and Selfhood: Investigating the First-Person Perspective. Cambridge, Mass. 2005.
The Truth, the Buddha’s Words, and Inference: Bhāviveka’s
Theory of Two Truths
by
Masaki Tamura
Introduction
The Buddhist truth, which was directly realized by the Buddha, is beyond verbalization
in itself. For it is precisely his personal, direct experience of reality. The Buddha, fully
aware that the truth is ineffable, taught it to us by resorting to words. No Buddhist can
deny the authority of the Buddha’s words. But the critical problem has arisen that they
are susceptible to divergent interpretations. The following serves as an example. In the
Daśabhūmikasūtra the Buddha states that the three realms are mind-only (cittamātra).1
According to Bhāviveka, the Yogācāra school interprets the statement to mean that an
external object does not exist independently of the mind,2 while the Mādhyamika school
interprets the same statement to mean that the self (ātman) serving as agent of an action
and as enjoyer of the fruit of the action does not exist independently of the mind.3 The
question comes up: What is the means for arriving at a real understanding of the Buddha’s
statement? According to Bhāviveka, it is scripturally based inference or the inference
(amunāna) which follows the Buddhist scriptures (āgamānuvidhāyin).4 Bhāviveka argues
that the truth is beyond the reach of inference and that inference plays the role of removing
the misconception about the truth which arises from the Buddha’s statement. The aim of
this paper is to show how Bhāviveka considers the determination of the truth to be linked
with its self-realization (pratyātmavedya) within the framework of the two truths theory.
It is to be noted that Bhāviveka uses the term anumāna “inference” without making
a clear distinction between svārthānumāna “inference for oneself” and parārthānumāna
“inference for others,” a distinction established by Dignāga. In this paper, accordingly, I
will also render the term anumāna used by Bhāviveka as “inference” without specifying
what the term means in a given context.
1
2
3
4
DBhS 98.8–9: cittamātram idaṃ yad idaṃ traidhātukam /
ViṃśV 3.2–4: mahāyāne traidhātukaṃ vijñaptimātraṃ vyavasthāpyate / cittamātraṃ bho jinaputrā
yad uta traidhātukam iti sūtrāt / … / mātram ity arthapratiṣedhārtham /
MHK V 28cd: sūtreṣu* cittamātroktiḥ kartṛbhoktṛniṣedhataḥ**// (*sūtreṣu ] L1, L2, Ec; sūtre ca H, S;
ś[ā]streva SG; śastreva Ms. **cittamātroktiḥ kartṛbhoktṛniṣedhataḥ ] L2, H, S, SG; cittamātroktikartṛbhoktṛniṣedhataḥ L1, Ec, Ms.)
It is well known that Dharmakīrti divides inference into two types: inference which functions by the force
of reality (vastubalapravṛttānumāna) and inference which is based on scripture (āgamāśritānumāna).
While the former is the inference whose object is imperceptible (parokṣa), the latter is the inference
whose object is completely imperceptible (atyantaparokṣa). What Bhāviveka means by the term
āgamānuvidhāyanumāna is the inference which functions only within the framework of the Buddhist
āgama and which is in perfect accordance with it. On Dharmakīrti’s concept of āgamāśritānumāna, see
Tillemans (1986; 1990: 23–35; 1999).
Birgit Kellner et al., eds., Reverberations of Dharmakīrti’s Philosophy: Proceedings of the Fifth International Dharmakīrti
Conference Heidelberg, August 26 to 30, 2014. Vienna 2020, pp. 463–472.
464
The Truth, the Buddha’s Words, and Inference: Bhāviveka’s Theory of Two Truths
1. When to use inference
1.1 Bhāviveka begins the fifth chapter of the Madhyamakahṛdayakārikā by giving the
outline of the Yogācāras’ view of the nectar-like truth as emptiness of the grasped-grasper
duality. After expounding the theories of “mind-only” (cittamātra), of asallakṣaṇānupraveśopāya, and of the three natures (trisvabhāva), Bhāviveka states the following kārikā.
MHK V 7: prajñāpāramitānītir iyaṃ sarvajñatāptaye /
na tūtpādanirodhādipratiṣedhaparo nayaḥ5 //
This doctrine propounded by the Prajñāpāramitāsūtras [as interpreted by the
Yogācāras]6 leads to the attainment of omniscience, while the doctrine as it is
taken to intend to deny arising, cessation, and so forth does not.
An important point to note here is: Bhāviveka states that the Yogācāras’ view of emptiness
and the Mādhyamikas’ view of emptiness are equally derived from the Prajñāpāramitāsūtras, which consist in the words of the Buddha. The Yogācāras might argue that only their
view of emptiness can lead to the attainment of omniscience and that the Mādhyamikas
misunderstand the Buddha’s words.
1.2
To this objection, Bhāviveka answers as follows:
MHK V 8–9: atrocyate pramāṇaṃ naḥ sarvaṃ tāthāgataṃ vacaḥ /
āptopadeśaprāmāṇyād bhadro hi pratipadyate //
nāgamāntarasaṃdigdhaviparyastamatiḥ paraḥ /
tasmāt tatpratipattyarthaṃ tanmṛgyo yuktimannayaḥ //
We will answer to this objection as follows. Since the teaching of a credible
person (āpta) is authoritative, all the words of the Tathāgata[, who is a credible
person,] are authoritative for us [Buddhists]. Indeed, a good person (bhadra)
accepts all the words of the Tathāgata, while our opponent does not. For, the
latter, under the influence of other traditions, holds that [what the Tathāgata
states is] doubtful and wrong.7 Therefore, the [good person] must seek for a
reasonable argument so that the opponent may accept the [Tathāgata’s words].
5
6
7
-pratiṣedhaparo nayaḥ ] H, S, SG, Ms; -pratiṣedhaparāyaṇā L1, L2, Ec.
The Tarkajvālā explains how the Yogācāras’ doctrine is derived from the Prajñāpāramitāsūtras. TJ ad
MHK V 7 [D202b1; P222b4–5]: sems de ni sems ma yin no zhes gsungs pa des ni gzung ba dang ’dzin
pa med par bstan to // (“The statement ‘That mind is not the mind’ (*tac cittam acittam) means that
there does not exist [the duality of] the grasper and the grasped (*grāhyagrāhakābhāva).”) APS 3.18:
tac cittam acittam /
Previous studies render the term āgamāntarasaṃdigdhaviparyastamatiḥ as follows. Iida 1966: 83:
“[Our] opponent, whose judgement is confused and perverted by other āgama(s).” Hoornaert 2000: 90:
“Others who have doubts and erroneous ideas from other scriptures.” Eckel 2008: 227: “the opponent,
whose mind is confused and misled by other traditions.” Krasser 2012: 546: “The other one, whose
mind is in doubt and confused by other scriptures.”
Masaki Tamura
465
There are a few points to note here. First, the Tathāgata is characterized as a credible person
(āpta) and hence his teaching must be authoritative.
Secondly, there are two types of hearers of the Tathāgata’s words: one is “a good
person” (bhadra) and the other is a person who is dubious of what the Tathāgata states and
who conceives of it as wrong. The latter is under the influence of āgamas other than the
Buddhist. It is important to note in this connection that Bhāviveka uses the term bhadra,
which refers to a Bodhisattva who confers a benefit on others.8 According to the Tarkajvālā,
Buddhist disciples are classified into four groups: (1) those who properly understand their
own āgamas; (2) those who doubt what they state; (3) those who incorrectly understand
what they state; (4) those who have insufficient intelligence to understand what they state.9
A Bodhisattva called bhadra belongs to the first group. Interestingly, Bhāviveka professes
to be a bhadra who is bound to lead the second and third groups to the Great Enlightenment
(mahābodhi), treating Yogācāra Buddhists as such groups of disciples.
Thirdly, the bhadra is said to be requested to seek for a reasonable argument (yuktimannaya) in order to lead the disciples of groups (2) and (3), referred to by the word
para “opponent” in the kārikā10 and characterized as those who are dubious of what
the Tathāgata states and who conceive of it as wrong, to a correct understanding of the
Tathāgata’s statement. The reasonable argument, according to the Tarkajvālā, consists in a
three-membered inference, comprised of a thesis (pakṣa), a reason (hetu), and an example
(dṛṣṭānta). The Tarkajvālā says:
TJ ad MHK V 9 [D203a1–2; P223a6–8]: de’i phyir de dag sgrub tu gzhug pa’i
phyir / smra ba po rigs11 pa dang ldan pa’i tshul phyogs dang / gtan tshigs
dang / dpe’i skyon med pa gzhan gyis smras pa’i nyes pa’i gnas su ma gyur
pa’i mtha’i rigs pa’i lam btsal bar bya’o //
8
9
10
11
See, for example, AvŚ I.184.2–4.
In MHK IX 155 Bhāviveka makes the same point. MHK IX 155: skyon med rjes su dpag pa yis // rigs
sam mi rigs ma dpyad par // blo ni gzhan gyis bskyed pa’i phyir // lung gzhan dag la the tshom za //
(Lindtner 1999: 295: “If one does not investigate what is logical and what is not logical by means of an
anumāna free from faults, then one’s understanding will be formed by other [traditions] and therefore
one will be in doubt about other traditions (āgama).”) TJ ad MHK IX 155 [D317a7–b2; P363a5–8]:
slob ma ni rnam pa gsum yod de / lung gzhan mthong na ’di ltar yin nam / ma yin zhes the tshom za
ba’i blo can dang / lung gcig la rab tu zhugs shing zhen par gyur nas gzhan la sems kyis (D; kyi P)
sdang (D; ldang P) bar gyur cing phyin ci log tu zhugs pa’i blo can dang / lung thams cad la rigs pa
yin nam / rigs pa ma yin zhes dpyod par mi nus pa’i rtogs pa dang bral ba’i blo can no // de la lung
gzhan gyis (D; gyi P) the tshom bskyed pa’am / phyin ci log tu zhen par gyur ba’i blo gang la yod pa de
dag ni rjes su dpag pa med par rigs sam / mi rigs zhes dpyod par nus pa ma yin no //
Krasser 2012 argues that the word para here refers to Dharmakīrti for the reason that the concept of
āgama introduced by Bhāviveka (i.e., āptopadeśaprāmāṇya, which echoes Dignāga’s definition of
āgama: āptavādāvisaṃvādasāmānyād anumānatā) is not shared by Dharmakīrti. I do not agree with
him. This is still being debated, though. See also Hoornaert 2000: 90, n.1. There is no question that in
the fifth chapter of the Madhyamakahṛdayakārikā Bhāviveka focuses on refuting the Yogācāra doctrine
propounded only by Asaṅga, Vasubandhu, and Dignāga. Incidentally, Krasser 2012: 546 gives the
following translation of the kārikās in question: “To this (pūrvapakṣa of the Yogācāra) we reply: All the
words of the Tathāgata are authoritative [pramāṇa] for us, because the teachings of a reliable person are
authoritative. A good one puts (these) into practice. The other one, whose mind is in doubt and confused
by other scriptures, does not. Therefore the path of reasoning [yuktimannaya] should be followed by
him in order to put these into practice.”
rigs ] D; rig P.
466
The Truth, the Buddha’s Words, and Inference: Bhāviveka’s Theory of Two Truths
Therefore, for an opponent to accept the Tathāgata’s words completely, a
proponent must seek for a reasonable argument, that is, a final logical method
which has no faults in the thesis, the reason, and the example, and with which
others find no fault.12
Needless to say, the argument in question is of the type parārthānumāna. The following is
an instance of such a reasonable argument:
MHK V 36: tasyālambanatā ceṣṭā tadābhamatihetutaḥ /
rāgavad bādhyate tasmāt pratijñā te ’numānataḥ //
[Thesis:] The [color and form in the form of an aggregation of atoms] are an
objective basis;
[Reason:] because they are the cause of a cognition in which they appear;
[Example:] like in the case of desire.
Therefore, your thesis is refuted through this inference.
Bhāviveka here intends to refute the theory of “mind-only” which the Yogācāra school
establishes by means of introducing counter-arguments against the atomic theory.13 In this
connection, it is important to note that, according to Dignāga, parārthānumana is what
makes what one has seen known to others (svadṛṣṭārthaprakāśana), and what serves to
give true information to others.14
2. Role of inference
Then what role can inference play in this context? First of all, Bhāviveka specifies that the
truth is beyond the sphere of inference.
MHK V 104: tattvasyātarkagamyatvāt 15 tadbodho nānumānataḥ /
nātas tarkeṇa dharmāṇāṃ gamyate dharmateti cet //
[Objection:] The truth is not realized through logical reasoning (tarka). [For]
it is said [in the āgama] that the real nature (dharmatā) of existential factors is
not realized through logical reasoning. Accordingly, there does not occur the
understanding of the [truth] through inference.
12
13
14
15
See also MHK V 109.
According to the Tarkajvālā, the theory is also contradicted by the Buddhist āgama. TJ ad MHK V 36
[D210a1; P232a4]: rnam par shes pa’i tshogs lnga’i gnas dang / dmigs pa ni bsags pa yin no zhes
gsungs pa’i lung dang yang ’gal lo // (“[Your thesis] contradicts the scripture also which says that
the supporting faculties and the objective basis of the five sense consciousnesses are aggregations [of
atoms].”)
PS III 1ab: parārtham anumānaṃ tu svadṛṣṭārthaprakāśanam /
-tvāt ] L1, L2, H, S, Ec; -tvā SG, Ms.
Masaki Tamura
467
According to Bhāviveka, the truth cannot be understood through inference. Nonetheless,
Bhāviveka does not absolutely deny that inference plays a certain role in understanding the
truth. Referring to the Buddhas, he states:
MHK V 105–106: ihānumānān nirdoṣād āgamānuvidhāyinaḥ /
kalpitāśeṣavividhavikalpaughanirākṛteḥ16 //
sakalajñeyayāthātmyam17 ākāśasamacetasaḥ /
jñānena nirvikalpena buddhāḥ paśyanty adarśanāt //
[Answer:] In our [Mādhyamikas’] view, the Buddhas, after negating a flood
of conceptions of what has been conceptually constructed, through inference
which has no fault and which accords with the Buddhist āgama, become those
who have a mind like space.18 Then they see, by way of non-seeing, the true
nature of all objects to be cognized, by means of a non-conceptual cognition.
Bhāviveka here shows the process through which the Buddhas come to get an intuitive
insight into the real nature of things. According to Bhāviveka, there are two stages. At the
first stage the Buddhas resort to inference in order to get rid of a flood of conceptualizations.
The inference is with no defects and accords with the Buddhist āgama. At the second stage
they, with minds pure as space,19 intuitively, by way of non-seeing, grasp the real nature of
things.
An important point to note is that Bhāviveka introduces inference in this way in the
context of the realization of the truth. Clearly, for Bhāviveka inference, which consists in
being svārthānumāna, has as its object not the truth itself but conceptualizations of what
has been conceptually constructed. Hence he says the following:
MHK V 107: ato ’numānaviṣayaṃ na tattvaṃ pratipadyate /
tattvajñānavipakṣo yas tasya tena nirākriyā //
For this reason, the truth is not understood to be within the sphere of inference.
That which is opposed to the knowledge of the truth is removed through
inference.
Bhāviveka here specifies the role of inference. The role of inference is to rule out the
opposite of the knowledge of the truth, that is, the conceptualizations of what has been
conceptually constructed.
Suppose that one makes the assertion “A certain thing is such and such” and another
the assertion “The thing is not such and such.” If the two assertions obtain, there arises a
doubt about the thing: Which assertion is true? Thus one must resort to inference.
16
17
18
19
-nirākṛteḥ ] L1, L2, H, S, Ec, SG; -nirākṛte Ms; bsal mdzad nas Tib (TJ: bsal bar mdzad nas de’i ’og
tu).
sakala- ] L1, L2, H, S, SG. Ms; akala- Ec.
The mind which is free from conceptual construction is likened to space in pureness. See Tamura 2013.
See the footnote above.
468
The Truth, the Buddha’s Words, and Inference: Bhāviveka’s Theory of Two Truths
MHK V 109: pratijñāmātrakeṇeṣṭā20 pratipakṣanirākriyā /
aniṣiddhe vipakṣe ca nirvikalpā matiḥ kutaḥ //
How can one admit that what is opposed to [the knowledge of the truth] is
negated by mere assertion? In addition, if what is opposed to [the knowledge
of the truth] is not negated, how can there arise a non-conceptual cognition
[of the truth]?
The structure envisaged by Bhāviveka is such that, when different understandings of the
truth arise from the Buddha’s words, its real understanding is arrived at by means of
removing its unreasonable understanding through inference; what is assumed to be a real
understanding of the truth is simply its conceptual understanding, so that it must be denied
to get an intuitive insight into the truth.
The following kārikā is interesting in that Bhāviveka explains why Buddhists, who
follow the Buddha’s teaching, can have different views concerning the truth.
MHK V 108: āgamāntarabhedena bhedāyātāsu buddhiṣu /
abhede ’py āgamasyānyaḥ21 kaḥ parīkṣākṣamo22 vidhiḥ //
Even if there is no difference in āgama among Buddhists, they have different
understandings by the influence of a variety of āgamas other than their own.
In this case, what else than inference could be the means to investigate the
truth?
Buddhists follow the Buddhist āgama. But, if they are influenced by āgamas other than
their own, they may have different views. In order to deny an opposite assertion, one must
resort to inference; otherwise, a doubt about the truth could not be banished. The inference
in question is one that is effective among those who accept the same āgama, that is, it is
the inference that is scripturally based.
3. The truth, the Buddha’s words, and inference
3.1 Two truths
In this way, an assertion different from one’s own is refuted by means of resorting to
inference, and thus one’s own assertion turns out to be justified. Even then, the assertion
obtains only in the sphere of conceptualization. For the assertion reflects only a conceptual
understanding of the truth. Therefore Bhāviveka continues to bring in the viewpoint of the
two truths theory.
MHK V 110: satyadvayam ataś coktaṃ muninā tattvadarśinā /
vyavahāraṃ samāśritya tattvārthādhigamo yataḥ //
20
21
22
pratijñāmātrakeṇeṣṭā ] em.; pratijñāmātrakā neṣṭā L1, L2, H, S, Ec, SG, Ms; dam bcas tsam gyis ji ltar
’dod (*pratijñāmātrakeṇeṣṭā kutaḥ) Tib.
-ānyaḥ ] L1, L2, H, S, Ec; -ādyaḥ SG, Ms.
-kṣamo ] L1, L2, H, S, Ec; -kṣayo SG, Ms.
Masaki Tamura
469
And, since the truth is realized in reliance on verbalization (vyavahāra), the
Muni, who sees the truth, taught two truths.
The given context clearly reveals that the term vyavahāra refers to the Buddha’s verbalization
of the truth, that is, the Buddha’s words about the truth, or the Buddha’s statement of the
truth.23 The Buddha taught two truths. This must mean that there is one truth beyond and
one within language.
Bhāviveka goes on to state that the understanding gotten from the statement of the truth
is false since it has the truth for its objective basis.
MHK V 111: sālambanatvād vitathā tathatālambanāpi dhīḥ24 /
svapnādidhīvat tadgrāhyaṃ nātas tattvaṃ ca yujyate //
[Thesis:] A cognition that has thus-ness as its objective basis (ālambana) is
erroneous;
[Reason:] because it has an objective basis (sālambanatvāt);
[Example:] like a dream-cognition and so forth.
Thus, it is not tenable that the truth is what is grasped by such [a cognition].
In Bhāviveka’s view, a cognition which has an objective basis is a conceptual cognition,
and is thus based on the dichotomy between a cognition and its object, and consists in
judging.25
The following kārikā is important in that Bhāviveka points out that the Buddha himself
denies the truth within language.
MHK V 112: agrāhyo ’nabhilāpyaś ca dhīpracāravivarjitaḥ /
dharma ukto munīndreṇa sa caivaṃ sati bādhyate //
The truth (dharma), which is beyond the sphere of cognitive activities, cannot
be [conceptually] grasped or expressed in words. But the great sage speaks of
the truth. This being the case, it is denied [by himself].
The reason that the Buddha denies the truth as it is spoken of is that the cognition stemming
from the verbalization of the truth is merely conceptual, so that it grasps the truth only
partially and not in its entirety. In reality the truth is beyond conceptualization and language.
3.2 The Buddha’s words and the Mādhyamikas’ understanding of the truth
The main intention of the fifth chapter of the Madhyamakahṛdayakārikā is to oppose the
Yogācāras’ understanding of the Buddha’s words about the truth. Bhāviveka ends the fifth
chapter as follows:
23
24
25
Hoornaert 2003: 168 and Eckel 2008: 296 render the term vyavahāra here as “conventional [truth]”
and “conventional usage,” respectively. I do not think that they catch the point.
tathatālambanāpi dhīḥ ] L1, H, S, Ec, SG, Ms; tathatālambanād api L2.
See Tamura 2011: 111–115.
470
The Truth, the Buddha’s Words, and Inference: Bhāviveka’s Theory of Two Truths
MHK V 113: ato yuktyāgamopetaṃ tattvaṃ yat prāgudāhṛtam /
parīkṣyamāṇaṃ yuktyaivaṃ tad evāvyāhataṃ sthitam //
For this reason, the truth which has been described before26 and which is
well founded on logical reasoning and āgama remains undenied even if it is
investigated through logical reasoning in this way.
Let us note that there are two different domains where we are involved in the understanding
of the truth. In one domain logical reasoning based on āgama works and in the other
the truth is beyond conceptualization and language and independently realized by way of
non-seeing. Bhāviveka states here that in the former domain the truth as conceived of by
the Mādhyamikas cannot be invalidated.
It is important to note in this connection that the Tarkajvālā cites MHK III 266 in
commenting on the present kārikā.
MHK III 266: jñeyasya sarvathāsiddher nirvikalpāpi yatra dhīḥ /
notpadyate tad atulyaṃ tattvaṃ tattvavido viduḥ //
Those who know the truth call “the unequalled truth” the truth with reference
to which there does not arise even a non-conceptual cognition since an entity
to be cognized is absolutely not established.
The Buddha verbalized the truth into which he had achieved intuitive insight. From the
Buddha’s statement referring to the truth the Mādhyamikas, though at a conceptual level,
correctly understand the truth, which is well founded on logical reasoning and āgama. To the
extent that the truth as the Mādhyamikas understand it accords with the truth as the Buddha
speaks of it, the Mādhyamikas’ view of the truth cannot be disputed by the Yogācāras.
Needless to say, the truth which the Buddha saw directly consists in svabhāvaśūnyatā,
namely that all existential factors are empty of an intrinsic essential nature.27
4. Conclusion
Buddhism originated in the Buddha’s verbalization of the truth that he directly experienced.
This is why for Buddhism the two truths theory, that there is one truth beyond and one
within language is a logical necessity.
In common to all Buddhists is the acceptance of the Buddha’s words. But it is not
always the case that the same statement brings about the same understanding of its meaning.
According to Bhāviveka, a Buddhist, even if following the Buddhist āgama, is subject to the
influence of āgamas other than the Buddhist. Thus Buddhists have different understandings
of what the Buddha stated. The Mādhyamika and Yogācāra schools hold different views of
the truth, emptiness, even though both base themselves on the Buddhist āgama. Bhāviveka
explicitly states that he feels it is necessary to give the Yogācāras a clear understanding
26
27
According to the Tarkajvālā, Bhāviveka describes the Mādhyamika truth in the third chapter of MHK.
See TJ ad MHK V 113 [D227a5–6; P253b4–6].
MHK III 115: svabhāvaśūnyatābodhān … /
Masaki Tamura
471
of what the Buddha states (pratipattyartham), and that there is no other means than inference for achieving this purpose. In his view, the function of inference is just to rule out
misunderstandings of the Buddha’s statements and not to give an intuitive insight into the
truth. Bhāviveka never says that inference is a means to know directly the Buddhist truth
of emptiness. If the Yogācāras’ understanding of the Buddha’s statement is refuted, the
Mādhyamikas’ understanding of it alone will remain as the only one true to the Buddhist
āgama. Furthermore, it is said that in Bhāviveka’s time a sectarian consciousness began
to develop among Buddhists. It is in this context that Bhāviveka brings in the concept of
āgamānuvidhāyyanumāna, the inference which works within the framework of the Buddhist
āgama.
References and abbreviations
APS Aṣṭasahasrikāprajñāpāramitāsūtra: Aṣṭasahasrikā Prajñāpāramitā with Haribhadra’s
Commentary Called āloka, ed. P.L. Vaidya. Darbhanga 1960.
AvŚ Avadānaśataka: Avadānaçataka, a century of edifying tales belonging to the Hīnayāna,
ed. J.S. Speyer. 2 vols. St. Petersburg 1902–1909 (repr. 1970).
Bahulkar 1994 S.S. Bahulkar, The Madhyamaka-Hṛdaya-Kārikā of Bhāvaviveka: A Photographic Reproduction of Prof. V.V. Gokhale’s Copy. Nagoya Studies in Indian
Culture and Buddhism: Saṃbhāṣā 15 (1994) 1–49.
DBhS Daśabhūmikasūtra: Daśabhūmīśvaro nāma Mahāyānasūtraṃ, ed. R. Kondō. Tokyo
1931 (repr. 1983).
Eckel 2008 M.D. Eckel, Bhāviveka and His Buddhist Opponents. Cambridge 2008.
Hoornaert 1999 P. Hoornaert, An Annotated Translation of Madhyamakahṛdayakārikā/
Tarkajvālā V.1–7. Studies and Essays, Behavioral Sciences and Philosophy 19 (1999)
127–159.
Hoornaert 2000 P. Hoornaert, An Annotated Translation of Madhyamakahṛdayakārikā/
Tarkajvālā V.8–26. Studies and Essays, Behavioral Sciences and Philosophy 20
(2000) 75–111.
Hoornaert 2003 P. Hoornaert, An Annotated Translation of Madhyamakahṛdayakārikā/
Tarkajvālā V.85–114. Studies and Essays, Behavioral Sciences and Philosophy 23
(2003) 139–170.
Iida 1966 Sh. Iida, Āgama (Scripture) and Yukti (Reason) in Bhāvaviveka. In: Kanakura
Hakushi Koki Kinen Indogaku Bukkyōgaku Ronshū (1966) 79–96.
Jiang 1991 Zh. Jiang, Sanskrit Tarkajvālā-Sūtra Manuscript Copy. (In Chinese.) In: Papers
in Honour of Dr. Ji Xianlin on the Occasion of his 80th Birthday. 1991, 111–117,
511–522.
Krasser 2012 H. Krasser, Bhāviveka, Dharmakīrti and Kumārila. In: Devadattīyam. Johannes Bronkhorst Felicitation Volume, ed. F. Voegeli, V. Eltschinger et al. Bern
2012, 535–594.
Lindtner 1995 Chr. Lindtner, Bhavya’s Madhyamakahṛdaya (Pariccheda Five) Yogācāratattvaviniścayāvatāra. The Adyar Library Bulletin 59 (1995) 37–65.
Lindtner 1999 Chr. Lindtner, Bhavya on Mīmāṃsā, Mīmāṃsātattvanirṇayāvatāraḥ. The
Adyar Library Bulletin 63 (1999) 243–302.
472
The Truth, the Buddha’s Words, and Inference: Bhāviveka’s Theory of Two Truths
Lindtner 2001 Chr. Lindtner, Madhyamakahṛdayam of Bhavya. Chennai 2001.
MHK Madhyamakahṛdayakārikā.
MHK (Ec) M.D. Eckel, ed. See Eckel 2008.
MHK (H) P. Hoornaert, ed. See Hoornaert 1999–2003.
MHK (L1) Chr. Lindtner, ed. See Lindtner 1995.
MHK (L2) Chr. Lindtner, ed. See Lindtner 2001.
MHK (Ms) Manuscript of MHK. See Jiang 1991.
MHK (S) A. Saito, ed. See Saito 2007.
MHK (SG) V.V. Gokhale’s Copy of MHK (Ms). See Bahulkar 1994.
PS Pramāṇasamuccaya: Sh. Katsura, Rediscovering Dignāga through Jinendrabuddhi. In:
Sanskrit Manuscripts in China: Proceedings of a panel at the 2008 Beijing Seminar
on Tibetan Studies, October 13 to 17, ed. E. Steinkellner, Duan Qing, and H. Krasser.
Beijing 2009, 153–166.
Saito 2007 A. Saito, “Chūganshinron” Madhyamakahṛdayakārikā oyobi “Ronrienron”
Tarkajvālā Daigosho ‘Yugagyōha no Shinjitsu[setsu] no [Hihanteki]kakutei’ Shiyaku.
In: Daijō-Bukkyō no Kigen to Jittai ni kansuru Sougouteki Kenkyū [Comprehensive
Research on Mahayana Buddhism: its Origins and Reality] (2007) 200–268.
Tamura 2011 M. Tamura, Bhāviveka’s criticism of the Yogācāra’s view of emptiness
(śūnyatā). (In Japanese.) South Asian Classical Studies 6 (2011) 93–117.
Tamura 2013 M. Tamura, Bhāviveka’s criticism of the Yogācāra’s concept of pariniṣpannasvabhāva: A simile of space (ākāśa) and the purity of the truth. (In Japanese.)
South Asian Classical Studies 8 (2013) 131–160.
Tillemans 1986 T. Tillemans, Dharmakīrti, Āryadeva and Dharmapāla on scriptual authority. Tetsugaku, The Journal of Hiroshima Philosophical Society 38 (1986) 31–47.
Tillemans 1990 T. Tillemans, Materials for the Study of Āryadeva, Dharmapāla and
Candrakīrti. Wien 1990.
Tillemans 1999 T. Tillemans, How much of a proof is scripturally based inference (āgamāśritānumāna)? In: Dharmakīrti’s Thought and Its Impact on Indian and Tibetan
Philosophy: Proceedings of the Third International Dharmakīrti Conference Hiroshima, November 4–6, 1997, ed. Sh. Katsura. Wien 1999, 395–404.
TJ Tarkajvālā. Tibetan Tripiṭaka. D3856, P5256.
ViṃśV Viṃśatikāvṛtti: Vijñaptimātratāsiddhi: deux Traités de Vasubandhu: Viṃśatikā (la
Vingtaine) accompagnée d’une Explication en Prose et Triṃśikā (la Trentaine) avec
le commentaire de Sthiramati, ed. S. Lévi. Paris 1925.
The Concept of sādhana in Chinese Buddhist Logic
by
Mingjun Tang
1. Introduction
In his article “More on parārthānumāna, theses and syllogisms” ([1991] 1999), Prof.
Tom J. F. Tillemans gave us a brief account of the interpretation of sādhana (means of
proof) in Dharmakīrti’s works as well as in the tradition following him. In this article, he
showed us, on one hand, the development concerning this concept in Dignāga’s thought
from the NMu to the PS, and on the other hand, insightfully explained the theoretical
significance of this development through a comparison with the Aristotelian syllogism.
In short, in Vasubandhu’s logical works as well as in Dignāga’s NMu, the sādhana was
identified with the linguistic expression of the three members, i.e., the thesis (pakṣa), the
reason (hetu) and the example (dṛṣṭānta). In Dignāga’s PS as well as in Dharmakīrti’s
tradition, it was identified only with the reason and the example. This exclusion of the thesisstatement from sādhana highlights a particular Buddhist conception of what is decisive for
the acceptability of an argument, one which contrasts with the principles governing the
Aristotelian syllogism. The decisive or probative factor in an argument, according to this
new conception of sādhana, is the truth of the premises, and not merely the logical form of
an inference.
The present paper, as an extended observation based on Tillemans’ abovementioned
article as well as on an earlier seminal article of Prof. Masahiro Inami,1 will show that
in the Chinese tradition of Buddhist logic, the concept of sādhana (neng li 能立) was
consistently interpreted as the reason-statement together with the statements of the positive
and negative example, or directly as the trairūpya, the triple characterization of a correct
reason. This interpretation of sādhana was explicitly ascribed to Dignāga himself as one
significant innovation over earlier masters. Although the Chinese tradition appears to have
proclaimed its theoretical exploration as exclusively based on the NP and the NMu, its novel
interpretation of sādhana in fact only finds support in the PS. Like the tradition following
Dharmakīrti, Chinese interpreters following Dignāga also took to various hermeneutic
strategies in order to harmonize their new interpretation with the old one from the NP and
the NMu, which is also found in various pre-Dignāgan Buddhist texts on logic.
In Chinese sources one also comes across the view that Indian Buddhist logicians
after Dignāga held the new interpretation instead of the old one. Moreover, these logicians
were reported in Chinese sources to have interpreted the “incompleteness” (nyūnatā) of
1
I would like to express my gratitude to my colleague, Dr. Liqing Qian, with whom I discussed the main
idea of this paper, and who kindly helped me correct my English, and also to Prof. Birgit Kellner, who
carefully read my paper and made many expert suggestions of various kinds towards its improvement.
Needless to say, I am responsible for all remaining mistakes.
Inami 1991. Here Inami explained the status of pakṣa in an argument, and the development of relevant
ideas from Dignāga to Dharmakīrti, in connection with corresponding developments in the theory of
pakṣābhāsa (pseudo-thesis).
Birgit Kellner et al., eds., Reverberations of Dharmakīrti’s Philosophy: Proceedings of the Fifth International Dharmakīrti
Conference Heidelberg, August 26 to 30, 2014. Vienna 2020, pp. 473–495.
474
The Concept of sādhana in Chinese Buddhist Logic
an argument as referring to the incompleteness of the three characteristics, and not as
indicating the incompleteness of the three statements that constitute a proof. This new
interpretation of nyūnatā together with the above new interpretation of sādhana was
inherited by Chinese logicians. In light of this new interpretation of nyūnatā, the present
paper tries to demonstrate once more, “from a slightly different angle” than Tillemans, that
it does not just reveal a terminological difference, but that it points to more fundamental
issues at stake and indicates “how logic works”2 for Buddhist thinkers.
2. The twofold meaning of sādhana in the NP and the NMu
The literal meaning of sādhana is “means of proof.” As we know, sādhana is one of the
eight topics in the basic framework of the NP. The eight topics are: (1) demonstration
(sādhana), (2) refutation (dūṣaṇa, neng po 能破), (3) pseudo-demonstration (sādhanābhāsa,
si neng li 似能立), (4) pseudo-refutation (dūṣaṇābhāsa, si neng po 似能破), (5) perception
(pratyakṣa, xian liang 現量), (6) inference (anumāna, bi liang 比量), (7) pseudo-perception
(pratyakṣābhāsa, si xian liang 似現量) and (8) pseudo-inference (anumānābhāsa, si bi
liang 似比量):
NP 1: sādhanaṃ dūṣaṇaṃ caiva sābhāsaṃ parasaṃvide / pratyakṣam anumānaṃ ca sābhāsaṃ tv ātmasaṃvide //
NPCh 11a28–29: 能立與能破
悟。
及似唯悟他,現量與比量
及似唯自
Demonstration (sādhana), refutation (dūṣaṇa) and their pseudo-forms (ābhāsa) are for the understanding of others. Perception (pratyakṣa), inference
(anumāna) and their pseudo-forms are for the understanding of oneself.3
The sādhana is the foremost among the eight topics. The sections on sādhana and sādhanābhāsa are the most extensive ones in the whole text of the NP. In this context, the
term sādhana refers to a three-membered argument and stands in contrast with dūṣaṇa (lit.
“means of refutation”). While the former is aimed at proving a view, the latter serves the
purpose of refutation. Therefore, we could translate the term sādhana when used in this
sense as “demonstration” or rather “argument,” i.e., the linguistic expression of a proof.4
2
3
4
Tillemans [1991] 1999: 78, 81.
Cf. Tachikawa 1971: 120.
The anumāna/sādhana distinction in the NP (cf. above NP 1) comes approximately to the inference/
argument distinction in today’s logic. I discussed this in Tang 2020: 414–416. In short, by inference we
mean nowadays “a process of linking propositions by affirming one proposition on the basis of one
or more other propositions.” By argument we mean “a structured group of propositions, reflecting an
inference” (Copi and Cohen 2005: 7). The working of an inference does not necessitate the medium of
certain linguistic expressions. An inference is only a process of pure thinking, regardless of whether
or not it is expressed linguistically. However, the working of an argument necessitates the medium of
certain linguistic expressions. An argument should spell out an inference in certain linguistic forms
which are suitable to be understood by others. Hence, I suggest “argument” as an alternative translation
of sādhana in addition to the traditional translation “demonstration.” A three-membered sādhana is
called a three-membered argument throughout this paper.
Mingjun Tang
475
The three members or statements making up a sādhana are the thesis (pakṣa, zong 宗),
the reason (hetu, yin 因) and the example (dṛṣṭānta, yu 喻). The last one normally consists
of two parts, the positive example (sādharmyadṛṣṭānta, lit. “example by similarity,” tong fa
yu 同法喻) and the negative example (vaidharmyadṛṣṭānta, lit. “example by dissimilarity,”
yi fa yu 異法喻):
NP 2: tatra pakṣādivacanāni sādhanam / pakṣahetudṛṣṭāntavacanair hi prāśnikānām apratīto ’rthaḥ pratipādyata iti //
NPCh 11b1–3: 此中宗等多言名為能立,由宗、因、喻多言開示諸有問
者未了義故。
Here [among the eight topics,] the sādhana is the [three] statements consisting
of the thesis and the other [two factors, i.e., the reason and the example],
because the object [yet] unknown to the questioners is made known by these
[three] statements consisting of the thesis, the reason and the example.5
NP 2.4: eṣāṃ vacanāni parapratyāyanakāle sādhanam / tadyathā / anityaḥ
śabda iti pakṣavacanam / kṛtakatvād iti pakṣadharmavacanam / yat kṛtakaṃ
tad anityaṃ dṛṣṭaṃ yathā ghaṭādir iti sapakṣānugamavacanam / yan nityaṃ
tad akṛtakaṃ dṛṣṭaṃ yathākāśam iti vyatirekavacanam // etāny eva trayo
’vayavā ity ucyante //
NPCh 11b19–23: 如是多言開悟他時,說名能立。如說聲無常者,是立
宗言;所作性故者,是宗法言;若是所作,見彼無常,如瓶等者,是
隨同品言;若是其常,見非所作,如虚空者,是遠離言。唯此三分,
說名能立。
The statements having these [factors, i.e., the thesis, reason and example, made]
on the occasion of convincing others, are the demonstration. For instance,
“sound is impermanent” is the statement of the thesis. “Because of being
produced” is the statement of the property of the subject [i.e., the statement of
the reason]. “Whatever is produced is observed to be impermanent, like a pot,
etc.” is the statement of the positive concomitance with the sapakṣa [, i.e., the
statement of the positive example]. “Whatever is permanent is observed to be
not produced, like ether, etc.” is the statement of the negative concomitance
(vyatireka) [i.e., the statement of the negative example]. Only these three
members are stated [to be the demonstration].6
A three-membered sādhana can consequently be written in its full form as follows:
5
6
Cf. Tachikawa 1971: 120.
Cf. Tachikawa 1971: 121–122. The phrase in the last square brackets is added in the Chinese translation.
476
The Concept of sādhana in Chinese Buddhist Logic
Proof (1)
Thesis:
Reason:
Positive example:
Negative example:
Sound is impermanent,
for sound is produced.
Whatever is produced is observed to be impermanent, like a pot;
Whatever is permanent is observed to be not
produced, like ether (ākāśa).
In the NMu, the three-membered sādhana together with its various pseudo-forms is
also the foremost topic. The same idea of sādhana as in the NP is found in the exposition
of the basic theoretical framework in NMu 1, together with the autocommentary:
NMu k. 1a: pakṣādivacanāni sādhanam
宗等多言說能立。
The sādhana is the [three] statements consisting of the thesis and the other
[two factors, i.e., the reason and the example].
NMu 1.1: 由宗、因、喻多言,辯說他未了義故,此多言於《論式》
等說名能立。又以一言說能立者,為顯總成一能立性 (sādhanam iti
caikavacananirdeśaḥ samastasādhanatvakhyāpanārthaḥ),7 由此應知隨有所
闕名能立過。
Since the object [yet] unknown to someone else is made evident [to that person]
by these [three] statements consisting of the thesis, the reason and the example,
these [three] statements are said to be the sādhana in the Vādavidhāna and
other [logical works of Vasubandhu]. Now, the expression “sādhanam” [here
in k. 1a] is in singular form so as to show that the sādhana is a united [whole,
even though it is comprised of three statements]. Thus it should be understood
that the lack [of any of these statements] is called a fault of the sādhana.8
In all the passages cited above, the grammatical phenomenon that the word vacana (statement, yan 言) appears in plural form (vacanāni/vacanair, duo yan 多言) betrays the view
of the NP and the NMu that a sādhana has more than two members, that is, specifically
three.
In both the NP and the NMu, the term sādhana is also used in a more limited sense
for the reason(-property), i.e., the predicate of the reason-statement, in the above proof
“producedness” or “being produced” (kṛtakatva). Now, the sādhana is in contrast with
sādhya (suo li 所立), the property to be proved or the inferable property, i.e., “impermanent”
in the above proof. The former property possesses the force of proving and the latter property
is to be proved by it to be present in the subject, i.e., “sound.” In this sense, the term sādhana,
when used as a substantive, can be translated as “means of proof;” when used as an adjective,
it can be rendered as “proving.” The translation of sādhana as probans and sādhya as
7
8
Inami 1991: 76, n. 33; cf. NPṬ 19,5–6.
Cf. Tucci 1930: 5–6; Katsura [1]: 109–111; Tillemans [1991] 1999: 85, n. 14; Inami 1991: 76–77.
Mingjun Tang
477
probandum, which is frequently encountered, is also suitable for the present context. This
meaning of sādhana can be found in the NP’s classification of four types of contradictory
reason (viruddha, xiang wei 相違), and as well as in that of ten types of pseudo-example
(dṛṣṭāntābhāsa, si yu 似喻):
NP 3.2.3: viruddhaś catuḥprakāraḥ / tadyathā / (1) dharmasvarūpaviparītasādhanaḥ / (2) dharmaviśeṣaviparītasādhanaḥ / (3) dharmisvarūpaviparītasādhanaḥ / (4) dharmiviśeṣaviparītasādhanaś ceti //
NPCh 12a15–16: 相違有四,謂法自相相違因,法差別相違因,有法自
相相違因,有法差別相違因等。
The contradictory [reasons] are of four kinds, as follows: (1) the [reason]
proving the opposite of the own form of the [inferable] property, (2) the
[reason] proving the opposite of [some] specific attribute of the [inferable]
property, (3) the [reason] proving the opposite of the own form of the propertypossessor, and (4) the [reason] proving the opposite of [some] specific attribute
of the property-possessor.9
NP 3.3–3.3.2: dṛṣṭāntābhāso dvividhaḥ / sādharmyeṇa vaidharmyeṇa ca // tatra sādharmyeṇa tāvad dṛṣṭāntābhāsaḥ pañcaprakāraḥ / tadyathā / (1) sādhanadharmāsiddhaḥ / (2) sādhyadharmāsiddhaḥ / (3) ubhayadharmāsiddhaḥ /
(4) ananvayaḥ / (5) viparītānvayaś ceti // … vaidharmyeṇāpi dṛṣṭāntābhāsaḥ
pañcaprakāraḥ / tadyathā / (1) sādhyāvyāvṛttaḥ / (2) sādhanāvyāvṛttaḥ / (3)
ubhayāvyāvṛttaḥ / (4) avyatirekaḥ / (5) viparītavyatirekaś ceti //
NPCh 12b1–4: 似同法喻有其五種:一、能立法不成,二、所立法不
成,三、俱不成,四、無合,五、倒合。似異法喻亦有五種:一、所
立不遣,二、能立不遣,三、俱不遣,四、不離,五、倒離。
The pseudo-example is of two kinds, by similarity or by dissimilarity. Of these,
first, the pseudo-example by similarity is of five kinds, as follows: (1) [an
example where] the proving property (sādhanadharma) is not established, (2)
[an example where] the inferable property (sādhyadharma) is not established,
(3) [an example where] both [properties] are not established, (4) [an example]
without [the statement of] positive concomitance and (5) [an example where]
the positive concomitance is reversed. … Second, the pseudo-example by
dissimilarity is of five kinds, as follows: (1) [an example where] the inferable
property is not excluded, (2) [an example where] the proving property is
not excluded, (3) [an example where] both [properties] are not excluded, (4)
[an example] without [the statement of] negative concomitance and (5) [an
example where] the negative concomitance is reversed.10
In the Chinese translation of the name for each contradictory reason, the word sādhana
is consistently rendered as “reason” (yin 因). The Indian commentator Haribhadra also
follows the same technique in glossing this word as hetu (reason). When commenting on
the first kind of contradictory reason, i.e., the dharmasvarūpaviparītasādhana, he says:
9
10
Cf. Tachikawa 1971: 125.
Cf. Tachikawa 1971: 126–127.
478
The Concept of sādhana in Chinese Buddhist Logic
NPṬ 39,4–5: atra dharmasvarūpaṃ nityatvam / ayaṃ ca hetus tadviparītam
anityatvaṃ sādhayati tenaivāvinābhūtatvāt /
Here the own form of the [inferable] property is permanence. Now, this reason (hetuḥ) proves (sādhayati) the opposite (viparīta) of that [own form of
the inferable property (dharmasvarūpa)], i.e., impermanence, because [it] is
invariably concomitant only with that [opposite property].
When commenting on the word sādhanadharmāsiddha as the name leading the NP list of
pseudo-examples, Haribhadra says:
NPṬ 44,5–11: sādhanadharmo hetur asiddho nāstīti bhaṇyate / tataś ca sādhanadharmo ’siddho ’smin so ’yaṃ sādhanadharmāsiddhaḥ / … evaṃ sādhyobhayadharmāsiddhayor api bhāvanīyam /
That is to say, the proving property, the reason, is not established, i.e., does
not exist. Hence, this sādhanadharmāsiddha is that in which the proving
property is not established. … With regard to the sādhyadharmāsiddha and
ubhayadharmāsiddha, it should also be thought in this manner.
Haribhadra analyzes the term sādhanadharmāsiddha as a bahuvrīhi compound, and identifies the sādhanadharma (proving property) with hetu (reason).11 On the term sādhanadharma, the NPVP explains further that: “This is both sādhana and property. Thus
sādhanadharma. What does it mean? The reason.”12 Here, the sādhanadharma is analyzed as a karmadhāraya compound. It refers to the property which is resorted to as the
means of proof (sādhana) in an argument and therefore possesses the force of proving.
When commenting on NP 3.3.1.(4) on ananvaya, Haribhadra directly glosses sādhana as
hetu:
NPṬ 46,7–9: vinānvayena vinā vyāptidarśanena sādhyasādhanayoḥ sādhyahetvor ity arthaḥ sahabhāva ekatravṛttimātram / pradarśyate kathyate ākhyāyate / na vīpsayā sādhyānugato hetur iti /
The meaning is: without [the statement of] positive concomitance, i.e., without the presentation of the pervasion, [merely] the co-occurrence, i.e., merely
the occurrence in one place, of the inferable and the sādhana, i.e., of the
inferable and the reason (hetu), is indicated, i.e., is stated or announced, [but]
not the reason as followed by the inferable in accordance with the requirement
of pervading.13
In the NMu classification of the pseudo-example, the name sādhanadharmāsiddha is
replaced by sādhanāsiddha, hence sādhanadharma by sādhana. Here, the sādhana is also
used in the sense of reason(-property):
11
12
13
See also NPṬ 47,9; 47,18: sādhanadharmo hetuḥ /
NPVP 109,21–22: sādhanaṃ cāsau dharmaś ca sādhanadharmaḥ / ka ity āha – hetur iti /
Cf. Tachikawa 1971: 127.
Mingjun Tang
479
NMu 5.3:「餘此相似」(k. 11d) 是似喻義。何謂此餘?謂於是處所立、
能立及不同品,雖有合、離而顛倒說。或於是處不作合、離,唯現所
立、能立俱有,異品俱無。如是二法或有隨一不成、不遣,或有二俱
不成、不遣。
That “all other [kinds of example] different from them are pseudo-[examples]”
means the pseudo-example. Which are those other [kinds] different from them?
They are [examples] where there is [the statement of] the positive concomitance
or [of] the negative concomitance with regard to sādhya, sādhana or asapakṣa
(i.e., an individual used as negative example),14 nevertheless, it is stated in
reversed manner; or [examples] where only the co-occurrence of sādhya and
sādhana or [only] the co-absence [of sādhya and sādhana] from vipakṣa is
indicated, [but] without the statement of the positive concomitance or of the
negative concomitance. [Pseudo-examples also include such cases where]
with regard to these two properties [i.e., the sādhya and the sādhana], either
(anyatara) [of them] is not established (asiddha) or not excluded (avyāvṛtta),
or both (ubhaya) are not established or not excluded.15
The word sādhana (sgrub pa/sgrub par byed pa/sgrub byed) does not occur in the corresponding PS IV kk. 13–14, nor in the PSV on these verses:16
The pseudo-form of that [i.e., example] is [an example where] the reason (gtan
tshigs), the sādhya or both are not established in or not excluded from the
asapakṣa (mi mthun phyogs), or [where] the concomitance is reversed in two
ways [i.e., in either positive or negative fashion], or [where] the concomitance
is absent. (k. 13)
[An example where] the [inferential] sign (rtags) is not found and so on, or
[where] the positive concomitance or the other [i.e., the negative concomitance] is reversed, is not a [correct] example. The [mere] aggregation [of two
properties in one place] is not a [logical] connection, because the [logical]
connection is [yet] not explicated. (k. 14)17
The term sādhana has here been completely replaced by the words gtan tshigs (hetu) and
rtags (liṅga = hetu); the above cited passage NPṬ 46,7–9 also demonstrates that these terms
are synonymous. Therefore, we can see that the terms sādhanadharma, sādhana and hetu
are interchangeable as referring to the reason-property.
14
15
16
17
Cf. Kitagawa 1965: 277–278, n. 615.
Cf. Tucci 1930: 40–41; Katsura [4]: 67–68.
See Kitagawa 1965: 527,12–529,9; 277–281.
K 152a5–6, 152b4–5: gtan tshigs bsgrub bya gñis ldan min // rjes ’gro ltog pa gñis dag ste // de’i
mi mthun phyogs bsal daṅ // rjes ’gro med pa der snaṅ ba’o // (k. 13) rtags med sogs daṅ rjes ’gro
sogs // phyin ci log pa dpe ma yin // ñe bar bsdu ba ma ’brel ba // ’brel pa rab tu ma bstan phyir // (k.
14); V 63a3–4, 63a7–b1: gtan tshigs bgrub bya gñis ka med // mi mthun phyogs las med ma byas // rjes
’gro phyin log rnam pa gñis // ltar snaṅ rjes ’gro med pa’aṅ yin // (k. 13) rtags med sogs daṅ dpe med
daṅ // rjes ’gro phyin ci log la sogs // ’brel par ma bstan pa yi phyir // ñer ’jal ’brel pa can ma yin // (k.
14) (Kitagawa 1965: 527,12–15; 529,5–8)
480
The Concept of sādhana in Chinese Buddhist Logic
In the NMu, the term *sādhanahetu (neng li yin 能立因) must be considered as another
relevant term relating to sādhana:
NMu 8:「餘所說因生」(k. 15b)18 者,謂智是前智餘。從如所說能立因
生,是緣彼義。
The sentence “[the inference (anumāna)] which is different [from perception]
is derived from the reason as presented [above in the discussion of sādhana]”
means that the [inferential] cognition is different from the above [perceptual]
cognition. It is derived from the *sādhanahetu as presented above. That is to
say, it is based on that [*sādhanahetu].19
Although I have found no Sanskrit material to confirm a karmadhāraya interpretation
of this term, it is highly probable that it, if it were in Sanskrit, must be construed in the
same way as the term sādhanadharma: the former refers to a reason which possesses the
force of proving, while the latter refers to a property which possesses the same force. Both
terms refer to the reason(-property). The PS has a corresponding definition of inference for
oneself (svārthānumāna):
PS II k. 1a–b: svārthaṃ trirūpāl liṅgato ’rthadṛk /20
[Inference] for oneself consists in observing an object through a triply characterized sign.21
Here, the term *sādhanahetu has been replaced by liṅga,22 and liṅga is just another name
of hetu. Now, we have a series of synonyms: sādhana, sādhanadharma, *sādhanahetu,
hetu and liṅga. All of them refer to the reason(-property) in this connection.
With this survey, we have exhausted almost all the occurrences of sādhana in the NP
and the NMu. In both texts, the term sādhana sometimes refers to a “demonstration,” i.e.,
a three-membered argument, and at other times the reason(-property). There is no third
option.
3. The new interpretation following the Pramāṇasamuccaya
Therefore, it seems surprising or even strange to some critical thinkers23 that Chinese
classical commentators consistently proclaim the sādhana to be the reason-statement
18
19
20
21
22
23
Cf. Katsura [5]: 84, n. 2: anyad nirdiṣṭalakṣaṇam.
Cf. Tucci 1930: 52; Katsura [5]: 91.
Katsura [5]: 92.
Cf. Hayes 1988: 231.
See also NP 4: anumānaṃ liṅgād arthadarśanam / liṅgaṃ punas trirūpam uktam / NPCh 12b29–c2:
言比量者,謂藉衆相而觀於義。相有三種,如前已說。“Inference is the observation of an object
through an [inferential] sign. The sign has been said [above] to have three characteristics.” Cf. Tachikawa
1971: 128. The word *sādhanahetu is also recurrent in NMu 10.14 on prāptyaprāptisama and ahetusama.
In the corresponding PSV passage, it has been replaced completely by gtan tshigs (hetu). Cf. Katsura
[7]: 46, nn. 3–4.
Cf. Chen [1945] 1997: 4–12; Zheng 1996: 29–32, 173–176.
Mingjun Tang
481
together with the positive and negative example-statements, or directly to be the trairūpya
(yin san xiang 因三相), the triple characterization of a correct reason. Since the statements
are three, and the characteristics of a correct reason are also three, the nature of sādhana
as being three-membered is still perfectly preserved in this interpretation. Moreover, this
interpretation of sādhana is ascribed to Dignāga himself as one significant innovation over
earlier masters. Kuiji 窺基 (632–682 CE) says:
YMDS 37–38 / 93a29–b2: 陳那能立,唯取因、喻,古兼宗等。…宗由
言顯,故名能立。
The sādhana of Dignāga only includes the reason and the example, while in
early times the thesis and others are also included. … The thesis is elucidated
through the statements [of the reason and two examples]. Therefore, [the
reason-statement and two example-statements] are named sādhana.
YMDS 50 / 93c28–94a3: 古師又有說四能立,謂宗及因、同喻、異喻。
世親菩薩《論軌》等說能立有三:一宗、二因、三喻。以能立者,必
是多言。多言顯彼所立便足,故但說三。
The early masters also talk about four [members of] sādhana. They are the
thesis, the reason, the positive example and the negative example. The Bodhisattva Vasubandhu in the Vādavidhi and other treatises says that there are
three [members of] sādhana, i.e., (1) the thesis, (2) the reason and (3) the
example. This is because the sādhana is necessarily comprised of more than
two statements, and [sādhana of] more than two statements is already adequate
for elucidating that which is to be proved (sādhya).24 Therefore, only three
[members] are asserted [by him].25
YMDS 52 / 94a14–17: 今者陳那因、喻為能立,宗為所立。自性、差別
二並極成,但是宗依,未成所諍。合以成宗,不相離性,方為所諍,
何成能立?故能立中,定除其宗。
Now, Dignāga [asserts that] the reason and the example are means of proof
(sādhana), and the thesis is what is to be proved (sādhya). Both the subject
(*svabhāva, zi xing 自性) and the property (viśeṣa, cha bie 差別) [i.e., the
qualificand and the qualifier in the thesis-statement,] have been well established
(prasiddha, ji cheng 極成) [i.e., accepted by both the proponent and the
opponent]. They are merely two substrata of the thesis-statement (*pakṣāśraya,
24
25
Note, Vasubandhu’s concept of sādhya is different from that of Dignāga in that only the inferable
property but not the whole thesis-statement is said to be what is to be proved. Moreover, Vasubandhu’s
concept of pakṣa is also slightly different from that of Dignāga in that only the subject is said to be
pakṣa. Cf. Frauwallner 1957: 33, frg. 1–3: pakṣo vicāraṇāyām iṣṭo ’rthaḥ. sādhyābhidhānaṃ pratijñeti
pratijñālakṣaṇam. me daṅ sa bon daṅ mi rtag pa ñid rnams rjes su dpag par bya ba ñid du dper brjod
pa’i phyir chos tsam rjes su dpag par bya ba ñid du mṅon par ’dod do źes rtogs par bya’o. “The pakṣa
is the object one wishes to investigate. The definition of proposition (pratijñā) is that the proposition is
the expression of what is to be proved (sādhya). Examples for the definition of what is to be inferred
(anumeya = sādhya) is said to be fire, seed and impermanence. Hence, it shall be understood that only
the property (dharma) is intended [here] as the definition of anumeya.” Cf. Frauwallner 1957: 16.
Cf. Frauwallner 1957: 16, n. 21.
482
The Concept of sādhana in Chinese Buddhist Logic
zong yi 宗依), but not [by themselves] the point under disputation. Only when
[they are] combined together so as to produce a thesis-statement, the invariable
concomitance (avinābhāva, bu xiang li xing 不相離性) [of the subject with
the property as expressed in the whole thesis-statement] then constitutes the
point under disputation. So, how can these [two substrata] be the sādhana?
Therefore, the thesis shall certainly be excluded from the sādhana.
Here, the reason-statement together with the example-statement is identified as sādhana.
This time, the sādhana is in contrast with sādhya, the thesis-statement, in that the reason
and the example are means of proof and the thesis is merely what is to be proved. Although
the sādhana here is also in contrast with sādhya, the sādhana and the sādhya here are
different from the interpretation of the NP and the NMu where they are taken only as
the reason-property and the inferable predicate. The hetu in Indian logic can mean either
the whole reason-statement or only the reason-predicate in that statement. Hence, the
exegetical movement from the reason-predicate to the whole reason-statement is not prima
facie breaking news. Nevertheless, the implied significance of this movement is very
important. It concerns not a mere terminological shift, but a shift of perspective in the
basic consideration of what makes a good argument good. When used in this new sense,
the term sādhana may be translated as “probative factor.”
In order to harmonize this new interpretation with the NP and NMu passages where
the sādhana is clearly said to have more than two statements – i.e., three members26
– the example is carefully counted as two members, i.e., the positive example and the
negative example. The reason together with these two example-statements can then easily
be interpreted as the three members of the sādhana. Kuiji continues:
YMDS 53 / 94a17–21: 問:然依聲明,一言云「婆達喃」,二言云「婆
達泥」,多言云「婆達」。27 今此能立,「婆達」聲說。既並多言,
云何但說因、喻二法以為能立?答:陳那釋云:因有三相,一因、二
喻,豈非多言?非要三體。由是定說宗是所立。
Question: However, according to Sanskrit grammar (śabdavidyā, sheng ming
聲明), a single statement is called vacanam, a pair of statements is called
vacane, [and] more than two statements are called vacanāni. Here, the sādhana
is mentioned in the form vacanāni. Since it is of more than two statements, why
do you only assert the reason-statement and the example-statement, these two,
as sādhana? Reply: Dignāga explains that the reason has three characteristics,
i.e., the reason and the two examples. Aren’t they three statements [in all]?
It is not required that there shall be three separate substrata (san ti 三體).
Hence, the thesis shall be definitely asserted to be [merely] what is to be
proved (sādhya).
Furthermore, when commenting on the last sentence in the NP 2.4 passage cited above (唯
此三分,說名能立), Kuiji says:
26
27
See the above cited passages NP 2, NMu k. 1a and NMu 1.1.
The transcription of vacanāni given by Zenju 善珠 (723–797 CE) is po da na ni 婆達那膩 (IRMS
237a28) instead of the po da 婆達 given in this passage. po da 婆達 must be a corruption of po da na
ni 婆達那膩. The IRMS is a running commentary on Kuiji’s YMDS.
Mingjun Tang
483
YMDS 304; 113b25–29:《理門論》云:「又比量中,唯見此理:若
所 比 處, 此 相 審 定(遍 是 宗 法 性 也); 於 餘 同 類, 念 此 定 有(同 品 定
有 性 也); 於 彼 無 處, 念 此 遍 無(異 品 遍 無 性 也)。 是 故 由 此 生 決 定
解。」(NMu 5.5) 即是此中唯舉三能立。
The NMu says: “and in an inference, only the following rule is to be observed: when the [inferential] sign (liṅga, xiang 相 = hetu) is ascertained
in the subject of inference (anumeya, suo bi 所比), that is, the reason is
pervasively a property of the subject (pakṣadharmatva, bian shi zong fa
xing 遍是宗法性), and in cases other than [the subject], we remember
its being [certainly] present in cases similar to that [subject in possessing
the inferable property], that is, the reason is certainly present in similar instances (sapakṣe sattvam, tong pin ding you xing 同品定有性), and its being
[pervasively] absent where that [inferable property] is absent, that is, the
reason is pervasively absent from dissimilar instances (vipakṣe ’sattvam, yi
pin bian wu xing 異品遍無性),28 then knowledge of this [subject] is generated.”29 This means the same as [when it] is [claimed] here that only three
[members of] a sādhana are presented.
Here, the three members of sādhana are further identified with the three characteristics
of a correct reason, the alleged basic criteria for a good argument in Buddhist logic. The
presupposition made here is that the reason-statement and especially the positive and
negative example-statements are nothing but the expression of the three characteristics,
in the sense that these three statements are true if and only if the three characteristics are
fulfilled.
This kind of interpretation of sādhana, though without being supported in the NP and
the NMu, can indeed be supported from the PS, Dignāga’s magnum opus and his final
work. Recent studies by Tom J. F. Tillemans have already showed that although in the NMu,
Dignāga did consider the thesis-statement to be a member of sādhana, “in PS Dignāga
did not consider the thesis-statement as being a sādhana, but nevertheless he most likely
allowed its presence in a parārthānumāna.”30 As pointed out by Tillemans, one passage
from PS fits quite well with the intention to exclude the thesis-statement from sādhana
while nonetheless letting it remain in the arrangement of a proof:
PSV ad PS III k. 1cd: tatrānumeyanirdeśo hetvarthaviṣayo mataḥ // (k. 1cd)
yan lag rnams la rjes su dpag par bya ba bstan pa gaṅ yin pa de ni kho bo cag
gi sgrub byed ñid du bstan pa ni ma yin te de ñid las the tsom skye ba’i phyir
28
29
30
NP 2: hetus trirūpaḥ / kiṃ punas trairūpyam / pakṣadharmatvaṃ sapakṣe sattvaṃ vipakṣe cāsattvam
iti // NPCh 11b6–7: 因有三相。何等為三?謂遍是宗法性,同品定有性,異品遍無性。 For
translation and discussion, see Tachikawa 1971: 121; Katsura 1985: 161–162.
Cf. Tucci 1930: 44; Katsura [4]: 74. The adverbs “certainly” (ding 定) and “pervasively” (bian 遍)
qualifying “being present” (astitva, you 有) and “being absent” (nāstitva, wu 無) respectively are
probably added in the Chinese translation. Cf. the parallel passage in PSV IV (K 150b5–7): rjes su
dpag pa la yaṅ tshul ’di yin par mthoṅ ste / gal te rtags ’di rjes su dpag par bya ba la ṅes par bzuṅ
na / gźan du de daṅ rigs mthun pa la yod pa ñid daṅ / med pa la med pa ñid dran par byed pa de’i
phyir ’di’i ṅes pa bskyed par yin no // (Kitagawa 1965: 521,8–13)
Tillemans [1991] 1999: 71.
484
The Concept of sādhana in Chinese Buddhist Logic
ro // ’on te gtan tshigs kyi yul gyi don yin pa’i phyir de ni de ma sgrub par
byed do (de ma sgrub par byed do K: des bsgrub par bya’o V) // (K 124b6–7,
Kitagawa 1965: 471,5–8)
In this regard, the presentation of what is to be inferred (anumeya) is held to
concern the goal of the reason. (k. 1cd)
Among the members, the presentation of what is to be inferred is not presented
by us to be the sādhana, because doubt will arise from it. However, because it
concerns the goal of the reason, it [i.e., the thesis,] is to be established by that
[i.e., the reason].31
Besides the exclusion of the thesis-statement from sādhana, the equivalence of the sādhana
with the expression of three characteristics can also be found in the PS:
PSV ad PS III k. 1: trirūpaliṅgākhyānaṃ parārthānumānam.32
Inference for others (parārthānumāna) is the communication of a triply characterized sign (liṅga).
The idea of assigning the reason together with the positive and negative examples to express
the three characteristics can be found in the NMu as well as in the PS:
NMu 5.6: 若爾喻言應非異分,顯因義故。事雖實爾,然此因言唯為
顯了是宗法性,非為顯了同品、異品有性、無性,故須別說同、異喻
言。
[Objection:] If so, then the example-statement must not be a separate member
[from that of the reason], because it is [designed] to express the implication
of the reason. [Reply:] Although the fact is actually so, yet the statement of
the reason is only meant to express [the reason’s] being a property of the
subject, but not to express [the reason’s] being present in similar instances and
being absent from dissimilar instances. Therefore, it is necessary to express
the positive and negative examples separately [from the reason-statement].33
PSV ad PS IV k. 7: ’on te de lta na dpe’i tshig kyaṅ tha dad par mi ’gyur te
gtan tshigs kyi don bstan pa’i phyir ro // … gtan tshigs ni mtshan ñid gsum
pa can yin la / bsgrub bya’i chos ñid ni gtan tshigs kyi tshig gis bstan pa yin
no // de las gtan tshigs lhag ma bstan par bya ba’i don du dpe brjod pa ni don
daṅ bcas pa yin no // (K 151a2–4, Kitagawa 1965: 522,7–523,2)
[Objection:] However, if so, even the example-statement will not be separated
[from the reason], because it is [designed] to express the implication of the
reason. [Reply:] … Since the reason possesses three characteristics, [only the
reason’s] being a property of the subject (sādhya = pakṣa) is expressed by the
31
32
33
Tillemans [1991] 1999: 71, translation slightly modified.
Kitagawa 1965: 126, n. 154. The parārthānumāna is a verbalized inference and corresponds to the
sādhana in the terminology of the NP and the NMu.
Cf. Tucci 1930: 45–46; Katsura [4]: 76–77.
Mingjun Tang
485
statement of the reason. In order to express the remaining [characteristics of
the] reason other than that [first characteristic], it is meaningful to express the
example.
Combining this idea with the PS’s claim that the sādhana is nothing but the expression of
the triple characterization of a correct reason, it is not difficult to arrive at the conclusion
that only the reason and example are sādhana, and not the thesis. To preserve the nature of
the sādhana as being three-membered, counting the example as two members may suggest
itself. The separation of positive and negative examples naturally results from the idea
that the second and third characteristics of a correct reason are to be expressed by them
respectively. At the same time, the statement of thesis is always preserved in an inference,
even though it is no longer recognized as a member of sādhana. This is also Dignāga’s
attitude as mentioned above.34
Therefore, the Chinese conception of sādhana as the reason together with two examples
or being exactly the triple characterization of a correct reason can be regarded as a natural
movement from Dignāga’s late thought, and may to some extent reflects Indian views
following Dignāga when Xuanzang was taught there. Although the Chinese tradition is
alleged to have only the NP and the NMu as its root texts, the ideas presented in the Chinese
commentarial literature on these two short treatises need not be limited to Dignāga’s early
views. On certain occasions, and to a certain extent, ideas in Chinese literature are even more
probably based on his later views, as well as on Indian interpretations that were produced
shortly after Dignāga and were not yet influenced by the revolutionary contributions from
Dharmakīrti (c. 600–660 CE). However, clues for gaining certainty on this hereto unknown
historical relation might always remain ambiguous, since no special reference to the subtle
differences between earlier and later stages in Dignāga’s thought could so far be found
in the Chinese tradition. In this regard, recent studies on Dharmakīrti and his successors
as well as on Dignāga himself will surely prove to be relevant to an improvement in our
understanding of Chinese hetuvidyā.35
34
35
Cf. n. 30.
Unlike Dharmakīrti (Tillemans [1991] 1999: 72–73), Chinese tradition consistently retains the thesisstatement. The idea that the thesis-statement can be known by “implication” (artha, yi 義) or by
“presumption” (arthāpatti, yi zhun 義准) is absent from Chinese sources. Nevertheless, like the tradition
following Dharmakīrti, the Chinese tradition also pays a lot effort and takes a roundabout hermeneutic
strategy so as to explain away the word pakṣa which always takes place at the beginning of the definition
of sādhana in the NP and the NMu (cf. n. 26). The gist of such a strategy is to say that this word is
so placed as to indicate exactly the aim or the object of sādhana. For details, see YMDS (54–56 /
94a21–b13) ad NP 1; YMDS (86–94 / 96c11–97b7) ad NP 2; For Wengui 文軌 (early 7th century)’s
similar discussion, see ZYS (1.4b–5b) ad NP 2 and ZYS (2.2a–3a) ad NP 2.4; For Kuiji’s ambiguous
comment on NMu k. 13cd = PS IV. k. 6cd, see YMDS (305 / 113c6–10) ad NP 2.4. Shentai 神泰 (early
7th century), the author of the only extant commentary on the NMu, says nothing on relevant passages in
the NMu. On one hand, he makes reference to his commentary on the NP, which however has been lost,
and on the other hand, he misleadingly ascribes this new interpretation even to Vasubandhu, see YZMS
(1.3b) ad NMu 1.1. Nevertheless, he has said something on nyūnatā (incompleteness), see below, n.
38. Furthermore, the author of the NP, Śaṅkarasvāmin, who was said to be a disciple of Dignāga, did
know the PS. There are certain elements in the NP which can be found only in PS, e.g. NP 3.1(9): eṣāṃ
vacanāni dharmasvarūpanirākaraṇamukhena NPCh 11c7–8: 如是多言,是遣諸法自相門故 ≈ PSV
ad PS III k. 2 (K 125a5–6): ’di yaṅ chos kyi raṅ gi ṅo bo daṅ ’gal bas sel ba’i sgo tsam źig bstan
486
The Concept of sādhana in Chinese Buddhist Logic
4. The “completeness” of an argument and the identification of the probative factors
As brought into light in Tillemans’ 1991 article, the above reinterpretations are not merely
terminological issues, but reach deeper into the nature of “how logic works.”36 The theoretical implication of these new interpretations can be clarified if we look at the matter from a
different angle, and take into account the reinterpretation of the fault called “incompleteness”37 (nyūnatā, que jian guo xing 缺減過性), which results from the new interpretation
of sādhana. In NMu 1.1 Dignāga defined sādhana to be a three-membered argument,
comprised of the thesis, the reason and the example. The lack of any one of these members
results in the fault called “incompleteness.”38 However, in the PS Dignāga says:
PSV ad PS III k. 1ab: ’dir yaṅ tshul gaṅ yaṅ ruṅ ba cig ma smras na yaṅ ma
tshaṅ ba brjod par ’gyur ro // (V 40b2, Kitagawa 1965: 470,7–8)
Here [in saying that inference for others is the communication of a triply
characterized reason], it shall also be called incomplete when any characteristic
[of the three characteristics] is unstated.39
36
37
38
39
pa yin la / (Kitagawa 1965: 472,14–15). Moreover, the theory of four kinds of contradictory reason
(viruddha, NP 3.2.3) can only be traced to the PSV ad PS III k. 26–27 (K 133b1–134a8), cf. Kitagawa
1965: 205–217. One aspect to consider in this context is the relationship between the composition of
popular manuals on logic on the one hand and more theoretical works on the other hand. Elements
of innovation might sometimes be diluted with more traditional viewpoint in these popular manuals
composed immediately or even several centuries after the emergence of that idea. The Hetutattvopadeśa
of Jitāri, which juxtaposes the NP with Dharmakīrti’s Nyāyabindu, may be thought of as one example
for this tendency.
Tillemans [1991] 1999: 81.
For this development, see Tillemans [1991] 1999: 75.
Cf. Tillemans [1991] 1999: 85, n. 14. When commenting on this passage, Shentai (YZMS 1.4a–b) gives
three different explanations of “incompleteness,” among which the first two correspond respectively to
that of Vasubandhu and that of scholars after Vasubandhu (but before Dignāga), and the last one to that of
Dignāga and his followers. However, he attributes the first two explanations both ambiguously to a “certain master,” cf. Tucci 1930: 6, n. 5. The definition of “incompleteness” in the NP is basically the same as
that in the NMu, see NP 6: sādhanadoṣo nyūnatvam / pakṣadoṣaḥ pratyakṣādiviruddhatvam / hetudoṣo
’siddhānaikāntikaviruddhatvam / dṛṣṭāntadoṣaḥ sādhanadharmādyasiddhatvam / tasyodbhāvanaṃ
prāśnikapratyāyanaṃ dūṣaṇam // NPCh 12c12–15: 謂初能立缺減過性、立宗過性、不成因性、不
定因性、相違因性及喻過性,顯示此言,開曉問者,故名能破。 For an English translation, see
Tachikawa 1971: 129. NPVP (124,8–12) ad NPṬ (54,12–13) on this passage: sādhanadoṣo nyūnatvaṃ
sāmānyeneti / nyūnatvaṃ pakṣādyavayavānāṃ yathoktalakṣaṇarahitatvaṃ pramāṇabādhitatvam iti
yāvat / ayam arthaḥ – sādhanavākye ’vayavāpekṣayā nyūnatāyā atiriktatāyāś ca sabhāsadaḥ purato
’bhidhānaṃ yat tat sāmānyena dūṣaṇam / viśeṣatas tu pakṣadoṣodbhāvanam asiddhaviruddhānaikāntikadoṣodbhāvanaṃ dṛṣṭāntadoṣodbhāvanaṃ vā dūṣaṇam iti / “The fault of the sādhana in general
(sāmānyena) is incompleteness. That is to say, incompleteness is the fact that the members including
the thesis and others lack [one or more of] the above mentioned definitions, or is contradicted by
[other] means of valid cognition (pramāṇa). This means: refutation is any expression in general of the
incompleteness or redundancy (atiriktatā) with regard to any member in the statement of a sādhana in
front of a witness (sabhāsad); in particular (viśeṣatas), however, refutation is either to point out the
fault of thesis, or to point out the fault of unestablished, contradictory or inconclusive [reason], or to
point out the fault of example.”
Tillemans [1991] 1999: 85, n. 15, translation slightly modified.
Mingjun Tang
487
This understanding changes not only how the fault called “incompleteness” as handed
down from the early phase of Indian logic is understood, but also the conception of what
kind of factor contributes to the “completeness” of an argument, in the absence of which
the argument becomes “incomplete” or unsound. We are now inclined to call them the
“probative factors.” Indeed, a number of elements can be regarded as being capable of
contributing to the “completeness” of an argument. At first, there should be certain linguistic
expression with certain ideas the proponent would like to communicate to the opponent.
This expression should be capable of explicating ideas in accordance with certain semantic
conventions. Even the intelligence of the opponent could be presupposed as part of the
necessary prerequisites for an argument to be practically adequate, since he should be
intelligent enough to pick out the meaning as being the one the proponent intended to
convey. In addition, one might also include the overall situation of the debate as being
one where arguments from each side are to be evaluated only according to principles for
thinking rationally. But not all factors which are necessary for a rational exchange to occur
contribute to the soundness of an argument in the same way. Therefore, by “probative
factors,” we do not mean all the necessary conditions for an argument to be “complete” in a
general sense, which are nearly infinite, but only those factors which were actually selected
by thinkers in the history of logic to be in the focus of their theorization of argument. The
notion of a “probative factor” is therefore just a meta-logical concept, not a logical one in
the usual sense. This is a concept only used to represent or recapture the main concern of a
logician in his theory of argument. In fact, we are bound to select only a limited number of
elements for reflecting on the soundness of a sound argument in a theoretical manner. This
does not prevent us from recognizing the fact that there must be other elements remaining
untheorized in our present framework, or even yet unobserved.
The mere illustration of probative factors contributes to a theory of argument just as
little as a good intuition of what a sound argument may look like; it is not yet a theory of
logic. The key feature of a theory is that the probative factors identified in it are at the same
time considered to be the criteria for discriminating in a general way a sound argument
from an unsound one. Theories of argument can be based on different approaches to the
identification of different kinds of probative factor, and as a result yield different systems
of criteria for sound argument. In short, different identifications of probative factors betray
different conceptions of argument, which may lead to different theories of argument, or
even different logical theories.
Now, let us return to the historical account as given in the Chinese literature, which is
in line with PS’s new interpretation of “incompleteness.”
YMDS 57 / 94b17–21: 世親菩薩,缺減過性,宗、因、喻中,闕一有
三,闕二有三,闕三有一。世親已後,皆除第七。以宗、因、喻三為
能立,總闕便非。既本無體,何成能立?有何所闕而得似名?
[According to] the Bodhisattva Vasubandhu, [there are seven cases of] incompleteness (nyūnatā). [Of them,] three are the lack of [only] one [statement]
among the thesis, the reason and the example, three are the lack of two [statements among them], and one is the lack of all the three [statements]. [Scholars]
after Vasubandhu all exclude the seventh case. Since the thesis, the reason
and the example, as three [members], form a sādhana, it is impossible for
488
The Concept of sādhana in Chinese Buddhist Logic
all of them to be lacking [and there is still an argument]. Since then, there
would be no substratum (wu ti 無體) at all, what could be an argument (sādhana) and what [kind of argument] could be called faulty on account of being
incomplete?
According to this view, logicians before Dignāga identified the probative factors with the
statements of the thesis, the reason and the example. When one of these factors is lacking,
the whole argument has the fault of incompleteness. A point that remains unclarified here
is that the linguistic expression itself could contribute to the “completeness” or soundness
of an argument in two ways. On one hand, the linguistic expression could be probative in
representing a certain form of valid reasoning. On the other hand, it could be probative
in that the reason-statement together with the example-statement is or is accepted to be
true. According to contemporary logical theory, an argument is sound if and only if all
its premises are true and the whole argument represents a certain form of valid reasoning.
Therefore, if the identification of probative factors with the linguistic expression itself does
not merely represent a good intuition of what a sound argument looks like, but is actually
taken as resting upon a solid theory, it could possibly provide the Buddhist logicians with
two different options in theorizing the “completeness” of an argument.
We name the first option the “formalist approach” in that the logical form itself is
identified as the probative factor, forms the focus of theorization, and becomes the criterion
for discriminating sound arguments from unsound ones. For the second option that takes the
truth of premises to be the foremost criterion for finding sound arguments, we will speak of
an “epistemic approach” or “dialectic approach,” depending on the interpretation of “truth”
that is chosen. If we interpret a true statement as being approved by ascertained evidences
(*niścayaprasiddha), we have an “epistemic approach.” If interpreting a statement’s being
true as merely being accepted to be true, i.e., being equally accepted by both sides in debate
(*abhyupagamaprasiddha, gong xu ji cheng 共許極成), we have a “dialectic approach.”
Our text continues:
YMDS 57–58 / 94b21–26: 陳那菩薩,因一喻二,說有六過,則因三相
六過是也。闕一有三,闕二有三,無闕三者。大師至彼六十年前,施
無厭寺有一論師,名為賢愛,精確慈悲,特以貫世,因明一論,時無
敵者,亦除第七。自餘諸師,不肯除之。因一喻二,即因三相。
[According to] the Bodhisattva Dignāga, six cases of the fault [of incompleteness] are mentioned concerning the reason, the first, with the examples, the
last two (yin yi yu er 因一喻二).40 They are the six faults concerning the three
characteristics of a correct reason. [Of them,] three are the lack of [only] one
40
Cf. RINM 30c29–31a2: 且「能立」者,即有二義:一一因二喻,二因一喻二。一因二喻,約
因三相也;因一喻二,約因二喻也。“Here, the sādhana has two meanings: (1) one reason with
two examples (yi yin er yu 一因二喻); and (2) the reason, the first, with the examples, the last two
(yin yi yu er 因一喻二). The ‘one reason with two examples’ concerns the three characteristics of a
correct reason. The ‘reason, the first, with the examples, the last two’ concerns the reason together
with two examples.” Gomyō 護命 (750–834 CE)’s division between 一因二喻 and 因一喻二 in this
passage is quite obscure. However, I have kept the room for such a subtle division in translating these
two expressions differently, though Kuiji in the above passage equated the three characteristics with 因
一喻二 but not 一因二喻.
Mingjun Tang
489
[characteristic], three are the lack of two [characteristics], and there is no case
where all the three [characteristics] are lacking. Sixty years before Xuanzang’s
arrival at the Nālandā Temple, there was a learned master (śāstrin, lun shi
論師) Bhadraruci (xian ai 賢愛), who was famous throughout the world for
his acute thinking and compassionate heart. No one could compare with him
in the discipline of hetuvidyā. He also excluded this seventh case. The other
masters, however, were not willing to exclude it. The reason, the first, with the
examples, the last two, is exactly the three characteristics of a correct reason.41
As is reported in the above passage, Buddhist logicians following Dignāga identified
the factors contributing to the “completeness” of an argument directly with the three
characteristics of a correct reason, the alleged basic criteria for a good argument in Buddhist
logic. As a detailed account of the trairūpya formulae according to the Chinese tradition is
not in place here, we can make no decision as to whether the “epistemic approach” or the
“dialectic approach” was actually adopted by those Chinese logicians following Dignāga.
Still, the evidence from our text clarifies that the “formalist approach” was not chosen.
Towards this end, we just point out that the “incomplete” or unsound arguments which are
to be ruled out by the three characteristics represent the same logical form with that of the
above proof (1), which is typically a sound argument in Buddhist logic, regardless of its
variety. Indeed, it is not difficult for us to abstract the following form from the above proof
(1):
Thesis:
Reason:
Positive example:
Negative example:
a
41
Sp
Hp
(x) (Hx → Sx)
(x) (¬Sx → ¬Hx)a
p = pakṣa the subject, “sound;” S = sādhyadharma the inferable property, “being impermanent;”
H = hetu the reason-property, “being produced.” Note that, the formalization here is just a
provisional one adjusted to the present purpose and ignores aspects that are irrelevant.
In fact, Kuiji is also unwilling to exclude the seventh possibility. Just a few lines after this passage,
he says: 又雖有言,三相並闕。如聲論師,對佛法者,立「聲為常,德所依故,猶如擇滅。
諸非常者,皆非德依,如四大種」。此「德依」因,雖有所說,三相並闕,何得非似?由此
第七亦缺減過。(YMDS 58–59 / 94b28–c3) “Again, there is also the case where the statement is
complete but all the three characteristics are lacking. For example, an upholder of the permanence of
sound (śābdika, sheng lun shi 聲論師), against a Buddhist, claims that ‘sound is permanent, because of
being a substratum of qualities (guṇāśraya, de suo yi 德所依, cf. NP 3.2.1(4)), like cessation through
analytical meditation (pratisaṃkhyānirodha, ze mie 擇滅). Whatever is impermanent is not a substratum
of qualities, like the four great elements (caturmahābhūta, si da zhong 四大種).’ Although there is an
argument from the reason ‘being a substratum of qualities,’ it lacks all the three characteristics. How
can it not be faulty? Hence, the seventh case should also be counted as incomplete.” That is to say,
there could be some linguistic expressions in which all the three characteristics are lacking, although
expressions of this kind have no probative force at all. Huizhao 慧沼 (650–714 CE) gives a much clearer
example for the seventh possibility: 如立「聲常,眼所見故」,虛空為同,盆等為異,三相俱
闕。(YMDS 753 / 141c21–22) “For example, ‘sound is permanent, because of being visible.’ [Here,]
ether is the similar instance. A dish, etc., are dissimilar instances. [The argument] lacks all the three
characteristics.”
490
The Concept of sādhana in Chinese Buddhist Logic
Moreover, if we regard the negative example as merely the contraposition of its positive
counterpart, we could just skip it. Then, the whole process of reasoning can be considered as
beginning from the statement of the positive example and ending with the thesis-statement.
If the formalization is correct, the above process clearly represents a valid reasoning.
When commenting on the NP passage on refutation (dūṣaṇa, NP 6),42 Huizhao43 has
provided each one of these seven possibilities with an example.44 Here, we just concern
the first three where only one characteristic is lacking respectively:
YMDS 752 / 141c12–16: 闕一有三者:如數論師,對聲論立:「聲是無
常,眼所見故」,聲無常宗,瓶、盆等為同品,虛空等為異品,此但
闕初而有後二;聲論對薩婆多立:「聲為常宗,所聞性故」,虛空為
共同品,瓶、盆等為異品,闕第二相;「所量性」因,闕第三相。
The three [kinds of incomplete argument where only] one [characteristic] is
lacking [respectively] are for example: (1) When a Sāṅkhya, against a Śābdika,
claims that “sound is impermanent, because of being visible (cākṣuṣatva, yan
suo jian 眼所見).” The thesis (pakṣa) that sound is impermanent has a pot and
a dish, etc., as similar instances (sapakṣa, tong pin 同品), and has ether, etc.
as dissimilar instances (vipakṣa, yi pin 異品). The [argument] lacks only the
first [characteristic] but has the last two [characteristics]. (2) When a Śābdika,
against a Sarvāstivādin (sa po duo 薩婆多), claims that “sound is permanent –
thesis, because of being audible (śrāvaṇatva, suo wen xing 所聞性).” [Here,]
ether is the similar instance equally [accepted by both sides] (gong tong pin
共同品). A pot and a dish, etc., are dissimilar instances. [The argument] lacks
[only] the second characteristic. (3) The reason “being cognizable (prameyatva,
suo liang xing 所量性)” [for the thesis “sound is permanent”] lacks [only] the
third characteristic.
If we present all the three arguments in a “syllogistic” manner with subject-predicate
statements, skip the negative example and all the individual cases cited and refer to the
“positive example” just as “example” for the sake of convenience, we end up with the
following three proofs:
Thesis:
Reason:
Example:
42
43
44
Proof (2)
Proof (3)
Proof (4)
Sound is impermanent,
for sound is visible.
Whatever is visible is impermanent.
Sound is permanent,
for sound is audible.
Whatever is audible is permanent.
Sound is permanent,
for sound is cognizable.
Whatever is cognizable is
permanent.
See n. 38.
The YMDS from the commentary on NP 3.3.1(1) to the end is actually written by Huizhao after the
death of his teacher Kuiji, see Zheng 2010: 605.
See YMDS 752–753 / 141c11–22.
Mingjun Tang
491
It is not a surprise to find out that all the unsound arguments illustrated here and the
above proof (1) as a sound argument share the same logical form. The differences are only
as follows:
In proof (2), the reason-statement “sound is visible” is not true, in that sound is clearly
not visible. Here, the proposition Hp is false. In this case, only the first characteristic
pakṣadharmatvam, i.e., the reason’s being (pervasively) a property of the subject (bian shi
zong fa xing 遍是宗法性), does not obtain or is not fulfilled.45
In proof (3), the example-statement “whatever is audible is permanent” is not true. It
cannot be instantiated in existent individuals apart from the subject “sound,” since only
sound is audible. Here, the positive example should be interpreted as a statement with
existential import, like (x) ((x ≠ p & Hx) → Sx) & (∃x) (x ≠ p & (Hx & Sx)).46 The whole
conjunction is false just because the last conjunct is false. In this case, only the second
characteristic sapakṣe sattvam, i.e., the reason’s being (certainly) present in similar instances
(tong pin ding you xing 同品定有性) is not fulfilled, since no similar instance or nothing
permanent apart from sound itself instantiates the reason-property “being audible.”47
In proof (4), the example-statement that “whatever cognizable is permanent” is also
not true. There certainly are impermanent things which are not only cognizable but also
different from sound, say a pot. Here, the first conjunct in the above conjunction is false.
Hence, the whole conjunction is false. In this case, only the third characteristic vipakṣe
’sattvam, i.e., the reason’s being (pervasively) absent from dissimilar instances (yi pin bian
wu xing 異品遍無性), is not fulfilled, since the dissimilar instances, things not “being
permanent” apart from sound itself, also have the reason-property “being cognizable,” like
a pot.48
In all the three cases, a valid form of reasoning does not perform a role in discriminating
sound arguments from unsound ones. Arguments are considered as unsound only on account
of the lacking of this or that characteristic. The “probative factor” is in this theory not
45
46
47
48
Cf. NP 3.2.1(1): śabdānityatve sādhye cākṣuṣatvād ity ubhayāsiddhaḥ // NPCh 11c12–13: 如成立聲為
無常等,若言是眼所見性故,兩俱不成。“When one is to prove that sound is impermanent, [the
reason] ‘because of being visible’ is not established for both (ubhayāsiddha).” Cf. Tachikawa 1971:
123.
Oetke 1994: 24, ES+eva 4.
Cf. NP 3.2.2(2): asādhāraṇaḥ śrāvaṇatvān nitya iti / tad dhi nityānityapakṣābhyāṃ vyāvṛttatvān
nityānityavinirmuktasya cānyasyāsaṃbhavāt saṃśayahetuḥ / kiṃbhūtasyāsya śrāvaṇatvam iti // NPCh
11c22–24: 言不共者,如說聲常,所聞性故,常、無常品皆離此因,常、無常外餘非有故是
猶豫因,此所聞性其猶何等?“An uncommon (asādhāraṇa) [reason] is: ‘[Sound is] permanent,
because of being audible.’ For, since this [reason] is [certainly] excluded from both the permanent and
impermanent kinds (pakṣa, pin 品) [apart from the subject ‘sound’], and since anything else which
is different from permanent and impermanent is impossible, this [reason] is a cause for doubt. [The
question remains:] ‘What kind of [thing] has audibility?’” For translation and discussion, see Tachikawa
1971: 124; Oetke 1994: 33–35.
Cf. NP 3.2.2(1): sādhāraṇaḥ śabdaḥ prameyatvān nitya iti / tad dhi nityānityapakṣayoḥ sādhāraṇatvād
anaikāntikam / kiṃ ghaṭavat prameyatvād anityaḥ śabda āhosvid ākāśavat prameyatvān nitya iti //
NPCh 11c19–22: 共者,如言聲常,所量性故,常、無常品皆共此因,是故不定。為如瓶等,
所量性故,聲是無常;為如空等,所量性故,聲是其常?“A common (sādhāraṇa) [reason] is:
‘Sound is permanent, because of being cognizable.’ For, since this [reason] is common to both the
permanent and impermanent kinds [apart from the subject ‘sound’], it is inconclusive (anaikāntika).
[The question remains:] ‘Is sound impermanent because of being cognizable, like a pot, or permanent
because of being cognizable, like ether?’” Cf. Tachikawa 1971: 124.
492
The Concept of sādhana in Chinese Buddhist Logic
the logical form, but the set of the three characteristics, just as the three characteristics
are proclaimed by logicians following Dignāga as sādhana, means of proof. Therefore, in
identifying the three characteristics as the probative factor, the “formalist approach” is not
the approach actually stepped on by Buddhist logicians.
Moreover, each case above where one characteristic is lacking or is not fulfilled can be
reduced to the scenario where one premise in the argument, either the reason-statement or
the example-statement, is not true. In this sense, the three characteristics concern nothing
formal. They are only the definition of the truths of the reason-statement and the examplestatement, i.e., the definition of the truths of the premises in an argument, in the sense that
all the premises are true if and only if all the three characteristics are fulfilled. Therefore,
in identifying the three characteristics as probative factor, the implied intention is that the
essential factors or criteria for discriminating a sound argument from unsound one should
be the truths of the premises. The theory of trairūpya is only a theorization of this implied
intention. It is only in this sense that the reason together with the positive and negative
examples is also proclaimed to be a “probative factor,” the sādhana. Whether the emphasis
is put on the three characteristics or on the reason and the examples is only a matter of the
level on which this implied intention is to be presented, the level of the meta-language or
that of the object-language.
However, one might argue in favor of a formalist interpretation of Buddhist logic that:
Since at least the statement of the reason and that of the example are recognized as sādhana
in this new interpretation following PS, there is certainly an awareness of the logical form
of an argument coming to the core in Buddhist theorization of argument. In this sense,
the “formalist approach” has not been totally neglected by Buddhist logicians following
this new interpretation of sādhana. As a matter of fact, what is actually at stake in this
new interpretation is not the form of these two members, but their truth. As we have said
above, on the one hand, a good intuition of what a sound argument looks like does not by
itself amount to a theory of argument, let alone to a “formalist” theory. The Buddhist view
of the three-membered argument is just a representation of this good intuition. Moreover,
there is only one form which is actually elaborated in this form of Buddhist theory of
three-membered argument.49 It is nothing but a linguistic standard for all the arguments
to follow. At any rate, a formal logic does not come about when there is only one form of
reasoning, which is neither compared with other equally valid forms of reasoning, nor with
other invalid ones. On the other hand, to adopt an approach other than a “formalist” one and
to take some factors other than the logical form itself as the theoretically most significant
does not necessarily imply that the other equally necessary factors, esp. the logical form,
are rejected or considered to be irrelevant to the “completeness” of an argument. To adopt
49
This is partly demonstrated by the constant practice of transforming a negative statement into its
affirmative counterpart, cf. NP 2.3: vaidharmyeṇāpi / … tadyathā / yan nityaṃ tad akṛtakaṃ dṛṣṭaṃ
yathākāśam iti / nityaśabdenātrānityatvasyābhāva ucyate / akṛtakaśabdenāpi kṛtakatvasyābhāvaḥ /
yathā bhāvābhāvo ’bhāva iti // NPCh 11b15–18: 異法者,…,謂若是常,見非所作,如虚空等。
此中常言表非無常,非所作言表無所作,如有非有說名非有。“[The example] by dissimilarity
[i.e., the negative example] is … for instance, ‘whatever is permanent is observed to be non-produced,
like ether.’ Here, the negation (abhāva) of being impermanent is said by the word ‘permanent,’ and the
negation of being produced is said by the word ‘non-produced,’ like non-being (abhāva) is the negation
(abhāva) of being (bhāva).” Cf. Tachikawa 1971: 121.
Mingjun Tang
493
one particular approach just concerns the focus of theorization and relegates other possible
candidates for a “probative factor” to a position at the edge of the horizon of a given
framework, not outside of it.50
Therefore, to interpret the Buddhist three-membered argument merely on its face
value as some Indian equivalence to the Aristotelian syllogism might well be an overinterpretation.51 For such an interpretation, the thesis or conclusion has to be taken into
account so that a form representing the complete process of reasoning can be available for
further considerations concerning its being valid or not. However, this is obviously not
the intention of logicians following Dignāga, in that the thesis is explicitly excluded by
them from “probative factors” and from their fundamental considerations concerning an
argument’s being tenable or not.
5. Conclusion
In the development from Vasubandhu to Dignāga and the latter’s Indian and Chinese
followers, and in the new interpretation of sādhana as the triple characterization of a correct
reason (trairūpya) instead of the linguistic expression of a three-membered argument, what
comes to the fore is a gradually clearer conception of what is essentially decisive for an
argument to be good or sound. In identifying the decisive factor as the trairūpya or the
truth of premises, Dignāga and his followers lead the Buddhist theory of argument to an
approach that sharply different from that of the formal logic of their European colleagues.
A crucial problem that remains undecided, however, is whether the “epistemic approach”
or the “dialectic approach” was adopted in further historical development after Dignāga
in India and China. As we have said, we leave the answer open at the present stage. I
believe that a solution will come about with a comparative study of the interpretation of
trairūpya by Chinese logicians and that by Indian Buddhist logicians after Dignāga, e.g.
Dharmakīrti.52
50
51
52
Indeed, there are other minor faults beyond the scope of trairūpya which concern the logical form of
a statement, cf. NP 3.3.1(5): viparītānvayo yathā / yat kṛtakaṃ tad anityaṃ dṛṣṭam iti vaktavye yad
anityaṃ tat kṛtakaṃ dṛṣṭam iti bravīti // NPCh 12b14–15: 倒合者,謂應說言,諸所作者,皆是無
常,而倒說言,諸無常者,皆是所作。“[An example where] the positive concomitance is reversed
is that: One states ‘whatever is impermanent is observed to be produced,’ when he should say ‘whatever
is produced is observed to be impermanent’;” NP 3.3.2(5): viparītavyatireko yathā / yad anityaṃ tan
mūrtaṃ dṛṣṭam iti vaktavye yan mūrtaṃ tad anityaṃ dṛṣṭam iti bravīti // NPCh 12b25: 倒離者,謂如
說言,諸質礙者,皆是無常。“[An example where] the negative concomitance is reversed is that:
one states ‘whatever is corporeal (mūrta, zhi ai 質礙) is observed to be impermanent,’ when he should
say ‘whatever is impermanent is observed to be corporeal’.” Cf. Tachikawa 1971: 127, 128.
This is exactly the conclusion of Tillemans’ 1991 article (Tillemans [1991] 1999: 78–81), where he
clearly shows that the thesis or conclusion constitutes “an integral part of the syllogism,” but not of
the Buddhist sādhana, and that this fact foreshadows the “fundamental incommensurability” between
syllogism and sādhana. To certain extent, my treatment of the concept of sādhana in Chinese literature
could be regarded as presenting the matter “from a slightly different angle.”
I have made some preliminary attempts in this direction, esp. on the concept of sapakṣa and vipakṣa
in Chinese Buddhist logic, and on a logical reading of the second characteristic under the Chinese
interpretation, see Tang 2015: 289–307, 323–336. For the most extensive and profound analysis of the
“epistemic operator” in Indian logic and of the trairūpya, see Oetke 1994.
494
The Concept of sādhana in Chinese Buddhist Logic
References and abbreviations
Primary sources
IRMS Inmyō ronsho myōtō shō 因明論疏明燈抄 (Zenju 善珠): T68, no. 2270.
NMu Dignāga, Nyāyamukha. See Katsura [1]–[7].
NP Śaṅkarasvāmin, Nyāyapraveśa. See Tachikawa 1971: 140–144; Chinese translation,
T32, no. 1630.
NPCh Chinese translation of NP: Yin ming ru zheng li lun 因明入正理論, T32, no. 1630.
NPṬ Haribhadra, Nyāyapraveśakaṭīkā. See Jambuvijaya 2009: 13–55.
NPVP Pārśvadeva, Nyāyapraveśakavṛttipañjikā. See Jambuvijaya 2009: 56–126.
PS(V) Dignāga, Pramāṇasamuccaya(vṛtti), Tibetan translation. See Kitagawa 1965.
RINM Daijō hossō kenjin shō 大乘法相研神章. Chapter 10 Ryakken inmyō nisshōri
mon 略顯因明入正理門 (Gomyō 護命): T71, no. 2309, 29a5–36b24.
T Taishō Shinshū Daizōkyō 大正新脩大藏經. Tokyo 1924–1935.
YMDS Kuiji 窺基, Yin ming da shu 因明大疏. See Zheng 2010; also T44, no. 1840.
YZMS Shentai 神泰, Yin ming zheng li men lun shu ji 因明正理門論述記. Woodprint
edition. Nanjing 1923.
ZYS Wengui 文軌, Yin ming ru zheng li lun shu 因明入正理論疏, abbreviated as Zhuang
yan shu 莊嚴疏. Woodprint edition. Nanjing 1934.
Secondary sources
Chen [1945] 1997 Daqi Chen 陳大齊, Yinming dashu lice 因明大疏蠡測 [Observations
on the Great Commentary on Hetuvidyā]. Tainan 1997.
Copi and Cohen 2005 I. M. Copi and C. Cohen, Introduction to logic. New Jersey 12 2005.
Frauwallner 1957 E. Frauwallner, Vasubandhu’s Vādavidhiḥ. Wiener Zeitschrift für die
Kunde Süd- und Ostasiens 1 (1957) 2–44.
Hayes 1988 R. P. Hayes, Dignāga on the interpretation of signs. Dordrecht 1988.
Inami 1991 M. Inami, On pakṣābhāsa. In: Studies in the Buddhist epistemological tradition. Proceedings of the second international Dharmakīrti conference, Vienna, June
11–16, 1989, ed. E. Steinkellner. Wien 1991, 69–83.
Jambuvijaya 2009 Nyāyapraveśakaśāstra of baudh ācārya Diṅnāga. With the commentary of ācārya Haribhadrasūri and with the subcommentary of Pārśvadevagaṇi, ed.
Muni Jambuvijaya. Delhi 2009.
Katsura [1]–[7] Shōryū Katsura 桂紹隆, Inmyō-shōri-mon-ron kenkyū 因明正理門論
研究 [A study of the Nyāyamukha]. Hiroshima Daigaku Bungakubu Kiyō, [1] vol.
37 (1977) 106–126; [2] vol. 38 (1978) 110–130; [3] vol. 39 (1979) 63–82; [4] vol.
41 (1981) 62–82; [5] vol. 42 (1982) 82–99; [6] vol. 44 (1984) 43–74; [7] vol. 46
(1987) 46–85.
Katsura 1985 Shōryū Katsura 桂紹隆, On trairūpya formulae. In: Buddhism and its
relation to other religions. Essays in honour of Dr. Shozen Kumoi on his seventieth
birthday. Kyoto 1985, 161–172.
Mingjun Tang
495
Kitagawa 1965 Hidenori Kitagawa 北川秀則, Indo koten ronrigaku no kenkyū: Jinna
no taikei インド古典論理学の研究–陳那(Dignāga)の体系 [A study of Indian
classical logic. Dignāga’s system]. Tokyo 1965.
Oetke 1994 C. Oetke, Studies on the doctrine of trairūpya. Wien 1994.
Tachikawa 1971 M. Tachikawa, A sixth-century manual of Indian logic. A translation of
the Nyāyapraveśa. Journal of Indian Philosophy 1 (1971) 111–145.
Tang 2015 Mingjun Tang, A study of Gomyō’s “Exposition of Hetuvidyā.” Text, translation and comments (1). In: Logic in Buddhist scholasticism. From philosophical,
philological, historical and comparative perspectives, ed. G. Paul. Lumbini 2015,
255–350.
Tang 2020 Mingjun Tang, Yin ming 因明 in Chinese Buddhism. In: Dao companion to
Chinese philosophy of logic, ed. Y. Fung. Springer 2020, 407–436.
Tillemans [1991] 1999 T. J. F. Tillemans, More on parārthānumāna, theses and syllogisms. In: Scripture, logic, language. Essays on Dharmakīrti and his Tibetan successors. Boston 1999, 69–87.
Tucci 1930 G. Tucci, The Nyāyamukha of Dignāga. Heidelberg 1930.
Zheng 1996 Weihong Zheng 鄭偉宏, Fojia luoji tonglun 佛家邏輯通論 [A general
introduction to Buddhist logic]. Shanghai 1996.
Zheng 2010 Weihong Zheng 鄭偉宏, Yinming dashu jiaoshi jinyi yanjiu 因明大疏校
釋、今譯、研究 [The Great Commentary on Hetuvidyā. Critical text with notes,
modern translation and investigation]. Shanghai 2010.
A Causeless Liberation? Kṣemarāja’s Response to
Dharmakīrti’s Critique of Initiation
by
Somadeva Vasudeva
1 Who are Dharmakīrti’s Śaivas?
Kṣemarāja (active around 1000–1050 CE), a Kashmirian exegete of the non-dualist Śaiva
Mantramārga, has supplied, as an appendix to his commentary on the Svacchandatantra,
a short Prakaraṇa style tract defending the theory that Śaiva ritual initiation, or dīkṣā, is
capable of bestowing final liberation. Kṣemarāja begins with a Buddhist pūrvapakṣa built
on ten and a half verses of Dharmakīrti’s Pramāṇavārttika, namely the Pramāṇasiddhi
257cd–267.1 Before proceeding to analyze Kṣemarāja’s work, it is helpful to establish
the identity of Dharmakīrti’s original opponent. What kind of Śaiva views is Dharmakīrti
attacking, and how do these views compare to what Kṣemarāja is defending?
As Acharya (2014) has shown, Dharmakīrti addresses the doctrines of the early dualist
Śaivasiddhānta, that is, groups who accepted the authority of revealed scriptures such as
the Niśvāsatattvasaṃhitā, the Kiraṇatantra, and the Svāyambhuvasūtrasaṃgraha. The
Niśvāsatattvasaṃhitā, the earliest work of this school that has come down to us, is tentatively
dated to around 450–550 CE. It is in this work that we see the first defense of the new
doctrine that was to become the hallmark of the Śaiva Mantramārga: The claim that dīkṣā,
or initiation, is sufficient cause for the end of rebirth. We are asked to believe that this is
possible because Śaiva mantras have an unimaginable power (acintyavibhava)2 capable
of excising the substantial defilements that lead to incarnation. The early Śaivasiddhānta
makes the claim that this excision can be empirically verified in a ritual of weighing, the
tulādīkṣā, or dhaṭadīkṣā, “Initiation with the scales.” Dharmakīrti responds to this kind of
a claim in the Pramāṇasiddhi section of the Pramāṇavārttika:
PV Pramāṇasiddhi 258cd–260ab:
nālaṃ bījādisaṃsiddho vidhiḥ puṃsām ajanmane //258//
tailābhyaṅgāgnidāhāder api muktiprasaṅgataḥ /
1
2
PV Pramāṇasiddhi, 257cd–267 (Edmi ).
For the Śaivas mantras are souls appointed by the cause deity Śiva to enable various effects. Mataṅgapārameśvaravṛtti of Rāmakaṇṭha to Vidyāpāda 7.40cd–42ab: …śabdātiriktā ātmaviśeṣā eva mantrāḥ
syuḥ…, “Mantras are specific souls, different from sounds.” Similarly, Mataṅgapārameśvara Vidyāpāda
7.42cd–43: vācyavācakabhedo ’yaṃ suprasiddho mahāmune / mokṣārthaṃ suniyuktānāṃ vācyānāṃ
kāraṇecchayā / niyuktā vācakatvena varṇāḥ śāstre śivodaye, “O great sage, this dichotomy of denoted
and denoter is well-known. In Śaiva scripture, [certain] phonemes have, by the will of the cause-deity
(Śiva = kāraṇa-), been prescribed as denoters of commissioned mantra[-souls], for the purpose of
liberation.” Scil., both the mantra souls that are the vācyas, and the phonemes used that are the vācakas
(sometimes also: śabdas), are considered to be mantras.
Birgit Kellner et al., eds., Reverberations of Dharmakīrti’s Philosophy: Proceedings of the Fifth International Dharmakīrti
Conference Heidelberg, August 26 to 30, 2014. Vienna 2020, pp. 497–512.
498 A Causeless Liberation? Kṣemarāja’s Response to Dharmakīrti’s Critique of Initiation
prāg guror lāghavāt paścān na pāpaharaṇaṃ kṛtam //259//3
mā bhūd gauravam evāsya na pāpaṃ gurv amūrtitaḥ /
A ritual admitted as effective for seeds and so on is not sufficient to end rebirth
for persons, because of the undesirable consequence that liberation could also
[be brought about] by rubbing with sesame oil, or by burning with fire, and so
on.
A removal of sin is not established by [the initiand’s] prior heavier weight and
subsequent lighter weight. Let him become even weightless [after initiation]!4
[It proves nothing:] Sin has no weight because it is formless.
Dharmakīrti objects here to the Saiddhāntika claim that mantras possess the power to
liberate. The intended inference is probably something like the following: “An initiatory
mantra-procedure is sufficient cause for the end of rebirth. Because sin-seeds are burnt.
Just as in the case of other burnt seeds.”5 Here the reason is a type of viruddhahetu called
dharmaviśeṣaviruddha. That is to say, it establishes an undesired additional property
(dharmaviśeṣa) for the probandum, namely that an end of rebirth can also be achieved by
other things that stop seeds from sprouting, such as heating or rubbing with oil. Both parties
actually accept that mantras can bring about other effects, such as destroying the capacity
of seeds to sprout and neutralizing snake-poison (perhaps the most common practical
application of mantras). For Dharmakīrti, as Eltschinger (2001: 86) explains: “…les mantra
possèdent, en vertu de leur propre nature (svabhāva, svarūpa), une efficacité naturelle
(*bhāvaśakti) dont le mode d’opération demeure inexplicable.” For the Śaivas, the efficacy
of mantras – understood as souls6 appointed to this high office by Śiva – derives from the
“virility of mantras” (mantravīrya).7 These minor supernatural effects supposedly prove
that mantras can indeed perform also their primary function of effecting liberation through
initiation. Dharmakīrti’s prasaṅga then asserts that if mantras can block rebirth, then other
methods used to prevent seeds from sprouting should also be capable of blocking rebirth.
3
4
5
6
7
Prajñākaragupta: atha tailābhyaṅgasya tāvanmātram eva sāmarthyaṃ evaṃ sati dīkṣāyām api samānaṃ
tathā / Manorathanandin: dīkṣāyāḥ prāg guroḥ paścāl lāghavād dīkṣayā na pāpaharaṇam asya dīkṣitasya kiṃ tu gauravam evāsya kṛtaṃ sat mā bhūd iti kasmān na kalpyate / lāghavaṃ hi gauravavirodhi
dṛśyamānaṃ tadabhāvam eva gamayen na pāpabhāvaṃ / pāpam eva gurv iti cen na pāpaṃ gurv
amūrtito mūrtatvābhāvāt /
For this translation see Acharya 2014.
Subject: [dīkṣā-/mantra-]vidhiḥ, property to be proven: alam ajanmane, reason: dagdhapāpa[bīja]tvāt,
example: dagdhabījavat.
Mataṅgapārameśvaravṛtti of Rāmakaṇṭha to Vidyāpāda 7.40cd–42ab: śabdātiriktā ātmaviśeṣā eva
mantrāḥ syuḥ, “Mantras are specific souls, different from sounds.”
For an authoritative, early, Saiddhāntika exposition of mantravīrya see Mataṅgapārameśvara Caryāpāda
5.13cd–16 with Rāmakaṇṭha’s commentary. The history of the concept of mantravīrya has not yet been
written. Kṣemarāja and his non-dualist co-religionists have settled on defining mantravīrya as pūrṇāhaṃtāvimarśa an awareness of the plenary self. See Śivasūtravārttika 1.20: mantravīryam iti proktaṃ
pūrṇāhaṃtāvimarśanam, also Tantrālokaviveka 5.137: mantrayati svābhedena viśvaṃ parāmṛśatīti
mantraḥ, paraḥ pramātā. For pūrṇāhaṃtāvimarśana, see Utpaladeva’s Ajaḍapramātṛsiddhi 22: yā
saṃvidaḥ svātmamātraviśrāntiḥ sa eva pūrṇāhaṃtāvimarśasvabhāvo ’haṃbhāvo ’rthavyavasthāpako
gīyate. On pūrṇāhaṃtā see Kṣemarāja’s Netroddyota 7.16 avataraṇikā: ahaṃ bhūtvā dehādipramātṛtāpraśamanena pūrṇāhaṃtām āviśyety arthaḥ.
Somadeva Vasudeva
499
Since other methods used to sterilize seeds are not capable of blocking rebirth, there is no
reason to believe that mantras are capable of doing so.8 The notion that mantras have the
power to nullify the ability of karmic seeds to sprout is also a topos of the early, dualist
Śaivasiddhānta Kiraṇatantra:
The action of many existences has its seeds burnt,9 so to speak (iva), by mantras
[in initiation]. Future [action] too is blocked; [but] that by which this [body is
sustained can be destroyed only] by experience. (trans. Goodall).10
Dharmakīrti then refutes the Śaiva claim that weighing the initiate before and after initiation,
and seeing that he weighs less afterwards, demonstrates that mantras have removed his
sins. Beyond the scriptural sources, the two earliest exegetes of this tradition to whose
ideas we still have access are Bṛhaspati (ca. 650–750, of whose work, however, only a few
fragments survive) and Sadyojyotis (ca. 675–725).11 Of these two, Sadyojyotis is explicit
on the probative value of the initiation with the scales (tulādīkṣā, dhaṭadīkṣā):
[By initiation] on the scales, [the initiand] is purified of great sins such as
brahminicide. One may know that his bonds have been destroyed through
direct evidence, just as the destruction of poison [by Gāruḍa-mantras is proved
by visible evidence].12
Very similarly, Sadyojyotis speaks of the ritual destruction of sin being “perceived” (dṛṣṭa) in
his Nareśvaraparīkṣā 3.83.13 Other scriptures, such as the Sārdhatriśatikālottara, classified
the ‘evidence’ of the ritual of the scales as the fourth of a list of eight pratyayas,14 which it
understood as faith-inspiring miracles,15 a position the Śaiva non-dualists were subsequently
to develop further.
The example of poison is possibly also what Dharmakīrti intended by the force of his
-ādi. Poison was a stock example for the Saiddhāntikas. In the Sarvajñānottara we read:
8
9
10
11
12
13
14
15
Dharmakīrti criticises the Saiddhāntika’s extension of the observable blocking of the power to sprout in
ordinary seeds, to invisible, karmic ‘seeds’ as unwarranted. If the two cases were really completely
congruent, then actions such as burning etc. should also have congruent effects. Pakṣa: [Dīkṣā-/mantra]vidhiḥ, sādhya: pāpaharaṇakṛt, hetu: prāg guror lāghavāt paścāt. Here, the idea that sin or karma has
weight is anyatarāsiddha; it is only accepted by the Saiddhāntika.
For the compound dagdhabījam see Svāyambhuvasūtrasaṃgraha 3.15a and Rauravasūtrasaṃgraha
4.51c.
Kiraṇatantra 6.19: anekabhavikaṃ karma dagdhabījam ivāṇubhiḥ / bhaviṣyad api saṃruddhaṃ yenedaṃ tad dhi bhogataḥ //
For the dates of these authors see Sanderson 2006.
Tattvasaṃgraha of Sadyojyotis 38: śuddhiṃ vrajati tulāyāṃ dīkṣāto brahmahatyato mukhyāt / pratyayato jānīyād bandhanavigamaṃ viṣakṣayavat //
Nareśvaraparīkṣā 3.83: śubhayā kriyayā vede kṣayaḥ pāpasya coditaḥ / dṛṣṭaś ca tatkṣayaḥ śaive
kriyayaiva tulādinā. Rāmakaṇṭha, ad loc., is perhaps also more cautious, as he calls this claim a
“supporting argument” (yukti) and not a “proof:” ihāpi dhaṭadīkṣāyām evaṃvidhā yuktir asty eveti na
niryuktikaḥ kriyayā karmasaṃkṣayo gadita iti.
According to Sārdhatriśatikālottara 21 these are: [1.] burning without fire, [2.] killing and reviving
trees, [3.] stunning, [4.] removing the major sins, proven by weighing on a scale, [5.] neutralizing poison,
[6.] rendering infertile, [7.] removing possession, [8.] quelling a fever.
See the entry to pratyaya 2 in TĀK III.
500 A Causeless Liberation? Kṣemarāja’s Response to Dharmakīrti’s Critique of Initiation
Just as the toxicologist16 (viṣavaidya), through the power of visualizations and
seed-mantras, effects the removal of poison, so the ācārya effects a disjunction
with the bonds through Śaiva [initiation] rituals.17 Just as a suppression of
poison [takes place] by the power of mantras and herbs, so there takes place a
suppression of all bonds through initiation.18
It is uncertain whether Dharmakīrti knew scriptures such as the Niśvāsatattvasaṃhitā19
or the Rauravasūtrasaṃgraha or, whether he was merely attacking a popular idea. It is
possible that he personally consulted these earliest Śaiva scriptures, though in view of their
archaic language it seems unlikely that he would have rated them as credible opponents.20
A more plausible candidate, however, is the post-Diṅnāga Mataṅgapārameśvara, which
aspires to a more śāstraic presentation and diction, and which engages in polemics. In its
yoga section, in the context of the rite of weighing, the Mataṅgapārameśvara does indeed
claim that dīkṣā removes pāpa, exactly the claim that Dharmakīrti attacks:
[The fire concentration (āgneyī dhāraṇā)],21 intense with a raging conflagration
of flames, deployed [through visualization] in the ritual of mounting the scales,
can render a person free from defilement, [because he is now]22 possessed of
a body from which sin (pāpa-) has been burnt away.23
Commenting on this, the tenth century Kashmirian Saiddhāntika commentator Bhaṭṭa
Rāmakaṇṭha defends the idea that sin (pāpa) has mass and that the “body really becomes
heavier, because sin is a function of tamas, and so has mass.” He corroborates this by citing
Īśvarakṛṣṇa’s Sāṃkhyakārikā 13c, where it is said that, “the factor tamas is heavy and it
conceals.”24 The Yuktidīpikā (ca. 680–720 CE) explains this heaviness as a dullness of the
16
17
18
19
20
21
22
23
24
This is how Aghoraśiva interprets the verse (…yathā viṣavaidyo mantrasāmarthyena dhyānādinā ca
karmaṇā viṣaharaṇaṃ karoti…). It would also be quite natural to take the ācārya as the subject. While
Aghoraśiva might not approve of the idea that an ācārya should engage in low status activities such as
curing snake-bite, it is plausible that this is the intention of the Sarvajñānottara. ayam abhiprāyaḥ /
prāyaścittakarmaṇā duṣkṛtebhyo mokṣaśravaṇād anaikāntikam etat / tataś ca yathā viṣavaidyo mantrasāmarthyena dhyānādinā ca karmaṇā viṣaharaṇaṃ karoti tathācāryo ’pi śivaśaktyupabṛṃhitena
dīkṣākhyena karmaṇā malādipāśaviśleṣaṃ karotīty avirodhaḥ.
Sarvajñānottara Vidyāpāda 8: viṣāpahāraṃ kurute dhyānabījabalair yathā / kurute pāśaviśleṣaṃ
tathācāryaḥ śivādhvaraiḥ //
Sarvajñānottara Vidyāpāda 9: mantrauṣadhabalair yadvat sannirodho viṣasya tu / tathā hi sarvapāśānāṃ sannirodhas tu dīkṣayā //
Niśvāsatattvasaṃhitā Mūlasūtra 7.15ab: tulayā śodhayet pāpam ātmanasya parasya vā /
Rauravasūtrasaṃgraha 8: 3: yathā sūryodayaṃ prāpya tamaḥ kṣipraṃ vinaśyati / evaṃ dīkṣāṃ samāsādya dharmādharmair vimucyate // “As darkness gets instantly destroyed after having reached the
time of sunrise, in the same way one is freed from merits and demerits as soon as he receives initiation.”
(Trans. Acharya 2014: 16.)
The fire concentration is one of the four (or five) dhāraṇās common in Saiddhāntika ṣaḍaṅgayoga, see
Vasudeva 2004: 297–299.
Read as a hetugarbhaviśeṣaṇa.
Mataṅgapārameśvara Yogapāda 2.48–9ab: tulārohavidhāne ca prayogārthe balotkaṭā / vegavadbhiḥ
karālāsrair mayūkhaiḥ saṃprayojitā // karoti vigatakleśaṃ dagdhapāpatanuṃ naram /
Sāṃkhyakārikā 13c: guru varaṇakaṃ tamaḥ. Yuktidīpikā ad loc.: tatra gurutvaṃ kāryasyādhogamahetur dharmaḥ, karaṇasya vṛttimandatā / varaṇam api kāryagataṃ ca dravyāntaratirodhānam /
karaṇagatā cāśuddhiḥ prakāśapratidvand[vī ]bhūtā /
Somadeva Vasudeva
501
internal instrument that causes downward motion of the effect (kārya). This shifts the focus
away from sin to the Sāṃkhya guṇa of tamas, a material constituent of primal matter; but
for Dharmakīrti this claim still would remain anyatarāsiddha.
A significant problem in assuming that Dharmakīrti knew the Mataṅgapārameśvara
as we now know it, is that the Mataṅgapārameśvara’s Vidyāpāda also teaches a more
developed theory of mala as opposed to the simpler idea of pāpa found in the Yogapāda.25
There is, however, no evidence that Dharmakīrti knew the theory of a material mala. As
Goodall (2013: s.v. pāśa) notes, it may be a later development:
In a number of the earliest surviving Siddhāntatantras, “impurity” (mala) is
noticeable by its absence: we find no mention of it in the Ni, in the non-eclectic
recensions of the Kālottara, such as the SārK, in what survives of the RauSS
or even in the SJU(G) (in which, pace Aghoraśiva, the word mala is used to
refer to karman). The first surviving Siddhānta in which “impurity” figures
may be the SvāSS.
It is conceivable, therefore, that Dharmakīrti knew only the Mataṅgapārameśvara’s Yogapāda, or, since this reference to sin (pāpa) in the Yogapāda is an archaic throwback, it is
also possible that Dharmakīrti had access to the archetype. It is relevant, in this context,
that the Mataṅgapārameśvara’s Yogapāda also elsewhere evidences Śaivas engaging with
the Buddhist Pramāṇa tradition. Mataṅgapārameśvara Yogapāda 4.15cd–16a paraphrases
Diṅnāga’s definition of pratyakṣa: anirdeśyam asaṃdigdhaṃ kalpanāpoḍhagocaram / pratyakṣam, “Sense-knowledge is unobjectifiable, free of doubt, and free of imagination.” If
the Mataṅgapārameśvara’s Yogapāda section itself appears to be in dialogue with the Buddhist Pramāṇa tradition, it becomes more likely that a Buddhist response can be expected.
Until a more reliable critical edition of the Mataṅgapārameśvara has been published, any
chronological layering of the text remains speculative. We may conclude that it is possible
that Dharmakīrti is responding to a text as late as the Mataṅgapārameśvara’s Yogapāda.
2 After Dharmakīrti
Dharmakīrti’s critique of initiation may have had a noticeable impact on the Saiddhāntikas.
Acharya (2014: 14–16) notes that Dharmakīrti’s Śaivas speak of initiation removing sin
(pāpa) or unseen, yet still weighty, demerit (adṛṣṭa), an idea that he traces back to the
Śaivasiddhānta’s foundational scripture, the Niśvāsatattvasaṃhitā. This he contrasts with
the post-Dharmakīrti Śaiva doctrine that initiation removes a defilement (mala), an imperceptible substance (dravya) that acts on the soul from outside.26 Acharya has posited that
this development was motivated by Buddhist criticism of the early Śaiva Mantramārga, and
that Dharmakīrti provides us with a concrete example of what this criticism looked like.27
25
26
27
Especially in chapter 6, the Puṃpāśeśvaraprakaraṇa.
See Sanderson 1992: 285.
“The Śaiva Mantramārga believed that all accumulated sin is removed by the rite of initiation, and also
that this rite alone can remove it. But already in an early phase of the development of Śaiva ideology,
the Pāśupata- Śaiva belief that initiation removes sin, permanently or otherwise, was met with strong
criticism from the Buddhists, and they had to readjust their theory. For this, the Śaivas apparently
502 A Causeless Liberation? Kṣemarāja’s Response to Dharmakīrti’s Critique of Initiation
3 The causes of liberation
If it is initiation that removes the defilements that cause bondage, then why does not
everyone seek Śaiva initiation? Is there some other factor that prompts individual to seek
Śaiva initiation. Both the dualists and non-dualists held that a salvific divine contact called
the descent of grace (śaktipāta) had to occur before a bound soul developed an interest in
seeking initiation. In a sense this was the most significant event in the spiritual career of a
Śaiva. But what, we may ask, caused the descent of grace?
Within the dualist fold two rival schools of thought emerged. One of the earliest ideas
was that the descent of grace was caused by a karmic equilibrium (karmasāmya) caused by
two simultaneously maturing karmas of equal strength. In the ensuing logjam, God needs
intervene to maintain the regular working of karmic retribution, but any divine contact
inevitably changes the individual in a profound way. A later idea, promoted by Sadyojyotis,
was the theory of the maturation of defilement (malaparipāka), a kind of ripening of mala
that renders it fit for removal.
The non-dualists rejected both of these narratives and held that God was completely
free in his bestowal of grace (nirapekṣa).
The history of these developments is somewhat difficult to trace in our available sources.
After Sadyojyotis and Bṛhaspati there follows a long gap of more than two hundred years in
our record of the Saiddhāntika theorists until their tradition was revived by later Kashmirians
such as Nārāyaṇakaṇṭha and Rāmakaṇṭha, authors who have been described as scriptural
fundamentalists who wished to return to the original positions of Sadyojyotis.28 In this
apparent interregnum, it appears that rival theories of the causes and processes that lead up
to liberation gained currency. We have evidence that there existed other early Saiddhāntika
exegetes who rejected Sadyojyotis’ interpretation of the processes that lead to initiation
and thence to liberation, objecting specifically to the idea that Śiva needs to take karma
into account, and espousing instead a radical karmanirapekṣavāda, as did the later nondualist Śaivas. Abhinavagupta, for example, reports, approvingly, that a certain Aniruddha,
apparently an early commentator of the Mataṅgapārameśvara, held that the descent of
liberating grace was in no way dependent on karma.29 Evidently, Bhaṭṭa Rāmakaṇṭha,
keen to reestablish Sadyojyotis’ theory of the ripening of impurity (malaparipāka) as the
decisive factor, would not have approved of this position. Above all, he wished to limit the
28
29
resorted to those Vedic texts from which their Pāśupata predecessors had adapted and adopted the rite
of initiation and related ideas. In these texts sin was depicted as innate impurity.” (Acharya 2014: 23)
“Bhaṭṭa Rāmakaṇṭha presents his exegesis in a fundamentalist spirit as a return to the original position
set out in an earlier time by the founding fathers of his tradition with the purpose of rescuing it from the
contamination it had suffered from attempts to assimilate its scriptures to alien perspectives, at one
extreme to that of orthodox Brahmanism and at the other to that of the Śākta Śaivas.” (Sanderson 2007:
427-428)
Tantrāloka 13.293cd–295ab: śrīmatāpy Aniruddhena śaktim unmīlinīṃ vibhoḥ // vyācakṣāṇena mātaṅge
varṇitā nirapekṣatā / sthāvarānte ’pi devasya svarūponmīlanātmikā // śaktiḥ patantī sāpekṣā na kvāpīti
suvistarāt. “The venerable Aniruddha too has taught that [Śiva’s liberating power] is autonomous when
commenting on the Lord’s ‘power that awakens’ in the Tantra of Mataṅga [the Mataṅgapārameśvara,
Vidyāpāda 4.44], explaining at great length that when the power of the Lord descends it is beholden
to nothing, being an unfolding of the nature of the self that can take place even in the extreme case of
immobile life-forms.” (Trans. Sanderson 2006: 81-82 n. 54)
503
Somadeva Vasudeva
Figure 1: The gradualist model
T+0
T+1
śaktisaṃbandha
⊕
Liberation through Dīkṣā
According to Abhinavagupta’s
Explanation of Mālinīvijayottara
1.42–45
ba = bound (baddha, sakala)
dī = initiated (dīkṣita)
mu1 = tīvratīvraśaktipāta
mu2 = sadyonirvāṇadīkṣā
mu2 = asadyonirvāṇadīkṣā
ajñāna
T+2
T+3
mantravīrya
dehapāta
⊕
⊕
dī
⊖
ba
mu3
dī
mu2
ba
⊖
yo$atā &
kālaviśeṣa
TLib
yiyāsā
ajñāna
dīkṣā
mu1
operation of Śiva’s descent of grace (śaktipāta) to what he considered meritorious human
beings.30
For Dharmakīrti, on the other hand, the cause of rebirth is fully explained by the
influence of two kinds of conscious intention (saṃcetanā), one being false knowledge
(mithyājñāna-) and the other being the craving that arises from it (tadudbhūtatarṣa-).31
PV Pramāṇasiddhi 2.260cd–262ab:
mithyājñānatadudbhūtatarṣasaṃcetanāvaśāt //
hīnasthānagatir janma tena32 tacchin33 na jāyate /
tayor eva hi sāmarthyaṃ jātau tanmātrabhavataḥ //
te cetane svayaṃ karmety34 akhaṇḍaṃ janmakāraṇam /
Birth is to enter an inferior mode of existence,35 under the influence of [two
kinds of] conscious intention, [one instigated by]36 false knowledge and [the
other instigated by] the craving that arises from it. Therefore, one who severs
these two is not born [again], for they alone, by their mere presence, are
sufficient [causes] for [re]birth. These two kinds of intention are of themselves
motivated action, and so constitute the complete cause of [re]birth.37
30
31
32
33
34
35
36
37
Kiraṇavṛtti on Vidyāpāda 6.11cd–12.
For a discussion of Dharmakīrti’s proof of rebirth see Franco 1997.
tena] Edmi Edpra , tatas Edma .
tacchin] Edma Edpra , tac chin Edmi (understand as: te (du.) chinattīti tacchid).
karmety] Edmi Edma , karmāty° Edpra .
A gati is a rebirth destination, see Franco 1997: 69.
Manorathanandin: tadudbhūtas tarṣo mithyājñānaprabhavā tṛṣṇā / tābhyāṃ saṃprayukte cetane tadvaśād yā hīnasthānagatis taj janmety uktaṃ.
See also the discussion of this passage in Bodhicaryāvatārapañjikā 9: nanu ca satyadarśanād avidyādi prahīyate, tatprahāṇāt saṃskārādiprahāṇakrameṇa tṛṣṇāpi prahīyate / tṛṣṇāviparyāsamatī ca
504 A Causeless Liberation? Kṣemarāja’s Response to Dharmakīrti’s Critique of Initiation
After Dharmakīrti, or, in different Śaiva traditions not known to him, the Śaivas had
developed the idea that initiation removed not pāpa, but the three substantial impurities
(mala): āṇava, māyīya, kārma.38 The non-dualists inherited this theory, but, on the authority
of the Mālinīvijayottara, they re-interpreted the defilements as three forms of ignorance.39
A standardized set of expansions for the three defilements becomes commonplace (see
e.g. Netratantroddyota 16.56): [1.] āṇava = apūrṇaṃmanyatā, erroneous belief that one is
incomplete, [2.] kārma = śubhāśubhādisaṃskāra, positive and negative karmic latencies,
[3.] māyīya = bhinnavedyaprathā, manifestation of differentiated objects of cognition.
It is not impossible, here (see table 1), to see a parallel between Dharmakīrti’s false
knowledge and the non-dualist’s erroneous belief that one is incomplete, and between
Dharmakīrti’s craving that arises from false knowledge and the non-dualists positive and
negative karmic tendencies. The third mala, that accounts for the differentiation of the
objects of cognition, is something Dharmakīrti might have included also under mithyājñāna.
It may be this closeness that leads Kṣemarāja to adopt at times a conciliatory tone, pointing
out that the only difference of opinion is that the Buddhists refuse to accept that mantras
have enough power to liberate.
Table 1: The causes of bondage
Early Śaivasiddhānta
Innate pāpa or adṛṣṭa
Dharmakīrti
Later Dualist
Śaivasiddhānta
Śaiva Non-dualists
Two kinds of conscious
intention:
1.
mithyājñānasaṃcetanā
2. tṛṣṇāsaṃcetanā
arising from
mithyājñāna
Three substantial
defilements (mala):
1. āṇava,
2. māyīya,
3. kārma
Three forms of
ignorance (mala):
1. āṇava =
apūrṇaṃmanyatā,
2. kārma = śubhāśubhādisaṃskāra,
3. māyīya =
bhinnavedyaprathā
4 The process of liberation
What happens to the initiand after the binding defilements are nullified with Śaiva mantras?
An overview of the possible processes and outcomes described in the Mālinīvijayottara is
given in figure 1. At time T+0 the bound soul is affected by a descent of grace (śaktisambandha). This occurs at a certain time of self-reflection (kālaviśeṣa) and is dependent
38
39
punarbhavotpattinimitte / tataś ca tayor abhāvāt tuṣarahitasya bījasyeva karmaṇaḥ sadbhāve ’pi na
kiṃcid vihanyate iti / taduktam – mithyājñānatadudbhūtatarṣasaṃcetanāvaśāt / hīnasthānagatir janma
tyaktvā caitan na jāyate // iti / The reading of 261b is quite different: “Having given up this (most
naturally etad here refers to janma, though logically we would expect a reference to ajñāna and tṛṣṇā)
one is not born again.”
See Vasudeva 2004: 164–166, and the respective entries in the Tāntrikābhidhānakośa for definitions of
these.
See Mālinīvijayottara 1.24ab.
CIPLUęUWB BOE DBVTBUJPO
505
Somadeva Vasudeva
Figure 2: The default case: liberation after death
EFBUI
o
o
o
EĕLŏĤ
o
o
o
CJSUI
⊕
"
⊕
#
1+
1"
#"
.V
1+ QBVSVŏBı K×ĤOBN TWBTWBSēQB
1" QBVSVŏBN BK×ĤOBN QVSVŏBOJŏńIBQĤUBLB
#" CBVEEIBN BK×ĤOBN CIFEBCVEEIJ
.V NVLUJ
'JHVSF 5IF %FGBVMU $BTF -JCFSBUJPO BûFS %FBUI
on the fitness of the candidate. If the descent of grace is of the rare, extremely intense
variety (tīvratīvra), souls are liberated immediately (mu1). For others, at T+1, ignorance
(ajñāna)
is loosened and there
go toWaŚaiva guru (yiyāsā). At T+2 the guru
ķto S
4WBDDIBOEBUBOUSPEEZPUB
arises
,FE aQdesire
performs the ritual intervention of initiation using efficacious Śaiva mantras. Liberation
UBUIĤ
IZ BZBN ĤUNĤ <> TBłLPDĤCIĤTBTBUBUUWĤQēSĸBNNBOZBUĤUNBOFIĤCIJMĤŏBħBCEPLUFOĤĸBWF
can take place either immediately afterward, or after death (mu2, mu3).40
OB NBMFOB
<> ħVCIĤħVCIBWĤTBOĤUNBOĤ WJWJEIBKBONĤZVSCIPHBEFOB LĤSNFĸB <> UBUQSBCIBWF
This account, however, glosses over an important additional event that takes place
OBduring
DB LB×DVLBQVSZBŏńBLBTUIēMBCIēUĤUNBOĤOĤKĤUJLBUSJWJEIBEFIBUBEĤħSBZBWJDJUSBCIVWBOBCIPLUB
initiation. It is not merely a ritual of eradication (kṣapaṇa), but also a ritual of
WZĤSUIBTĤSUIBQSBUĕUJCIĤKĤ
NĤZĤLIZFOB
NBMFOB
DB WBMJUBĴ ] ZBUBĴ
TBSWBTZBJWB
<>essential
TBłLVDJUP CIJ
conferring (dāna). The binding
defilements
(paśuvāsanāḥ)
are destroyed
and an
41
self-knowledge
is bestowed.
It is after elaborating
on this
dual power
ŏWBłHĤEJNBZP
<>(jñānasadbhāvaḥ)
OUBSVMMFLIBħBUĤLĕSĸBĴ
<> LŊħBHBVSĤEJSēQP
NVLBUSFEBı
KĤOĤNĕUZĤEJQSBUĕUJ
of
mantras
that
Kṣemarāja
cites
Pramāṇavārttika
2.258cd–259ab
and
charges
Dharmakīrti
TJEEIB FWĤZBN BSUIBĴ ]
with failing to take this aspect into account:
¡TBUBUUWĤ¡ > FN 4BOEFSTPO ¡UBUUWĤ¡ ,FEķ TBłLVDJUP > ,FE TBłLVDJUJ ߹ ߹ ߹ ķ
DPOK *TBBDTPO
NVUSFEBı ,5.88,
FE NVEEIFEBı
tu
Svacchandoddyota
S1 149v: ķyat
NVLBUSFEBı >
5P FYQMBJO UIF TFMG JT FOWFMPQFE CZ <> UIF EFmMFNFOU PG JOEJWJEVBUJPO EFTJH
42
vidhiḥ
OBUFE IFSF nālaṃ
CZ UIFbījādisaṃsiddho
XPSE ZFBSOJOH
XIJDIpuṃsām
JT UIF ajanmane
FSSPOFPVT/ DPOWJDUJPO UIBU POF JT JO
43
tailābhyaṅgāgnidāhādāv
api
muktiprasaṅgataḥ
// (Pramāṇa- <> CZ UIF EFmMF
DPNQMFUF XIJDI IBT BT JUT FTTFODF BO BQQFBSBODF PG DPOUSBDUJPO
vārttika 2.258cd–259ab)
NFOU PG LBSNJD SFUSJCVUJPO XIJDI JT NBEF VQ PG QPTJUJWF BOE OFHBUJWF MBUFOU JNQSFT
40
*O UIF
Abhinavagupta
interprets
this passagePO
slightly
differently
in his
Tantrāloka.
4WB5B
UIF UFYU CFJOH
DPNNFOUFE
$G 4WB5B
B
OJNJUUBN
BCIJMĤŏĤLIZBN
Kṣemarāja
in
Svacchandoddyota
5.88,
p.
76:
dīyate
jñānasadbhāvaḥ
kṣīyante paśuvāsanāḥ / dānakṣa 5BĘMB BQēSĸBıNBOZBUĤ DFZBı 5BĘM7JW BE MPD BQēSĸBıNBOZBUĤĸBWBNBMBMBLŏBĸĤ
paṇasaṃyuktā dīkṣā tenaeha kīrtitā.
41
42
43
°saṃsiddho Edmi Edma , °saṃsiddhau Ked S1 .
tailābhyaṅgāgnidāhādāv] Ked , tailābhyaṅgānidāhādāv S1 , tailābhyaṅgāgnidāhāder Edmi Edma .
506 A Causeless Liberation? Kṣemarāja’s Response to Dharmakīrti’s Critique of Initiation
ityuktaṃ tais tat teṣām evopahāsyatām āviṣkaroti / yataḥ sthāvarajaṅgamakṛtrimādyātmano viṣasya44 svakāryakaraṇaśaktipraśamanaṃ45 bījavad yan
mantraiḥ kriyate tatrāpy etad vaktuṃ prāptam / atha tailābhyaṅgāder bīja
eva śaktinirodhakṛttvaṃ na tu viśeṣakṛttvaṃ46 mantrāṇāṃ tu47 tatrāpīty ucyate / tarhy acintyaprabhāvatvāt teṣāṃ pāśapraśamane ’pi sāmarthyaṃ kiṃ
na sahyate /
But their claim: A ritual admitted as effective for seeds and so on is not
sufficient to end rebirth for persons, because of the undesirable consequence of liberation also [being brought about] by rubbing with sesame
oil, or by burning with fire, and so on, only makes them objects of derision.
For, in the case of a poison of either immobile, mobile, or artificial type,48
mantras quell its power to bring about its natural effect just as in the case
of the seed, and this objection [of rubbing with oil] can [therefore] also be
raised in that case.49 However (atha), anointing with oil etc. can only obstruct
a capacity of the seed, but it is not able to add anything distinct, but it is said
that mantras can do so. Therefore, because they have an inconceivable power,
why can you not accept that they are also capable of neutralizing the bonds.
What kind of knowledge is this special kind of gnosis that is conferred by initiation? To
determine this, Abhinavagupta proceeds with an heuristic evaluation in the opening section
of his Tantrasāra. Knowledge, he says, is the cause of liberation, because it is the opposite
of ignorance, which is the cause of bondage. Ignorance can be of two kinds, one present
in the intellect (buddhigata) and the other implicit in individuality itself (pauruṣa). The
former (buddhigata) is non-ascertainment (doubt) or wrong ascertainment (error). The
latter (pauruṣa) is individualized differential cognition (vikalpasvabhāva) which causes
saṃsāra. This pauruṣa form of ignorance is removed by initiation. But initiation cannot take
place while intellectual ignorance persists. Intellectual ignorance is removed by cultivating
the discernment (adhyavasāya) of what must be cultivated and what must be rejected.
Figure 2 shows these forms of knowledge and ignorance in relation to initiation and death.
To this can be added the idea of the jīvanmukta who develops gnosis before death, a theory
not discussed here.
44
45
46
47
48
49
viṣasya] Ked , viṣayasya S1 .
svakāryakaraṇa°] Ked , svakāryakāraṇa° S1 .
viśeṣakṛttvaṃ] Ked , viśeṣa° S1 .
tu] Ked , na tu S1 .
The three sources of poison are toxic plants, venomous animals, and artificial or processed derivatives
of these such as camphor. See Manoramāṭīkā to Tantrarāja 3.37.
The proposed inference is: [1. pakṣa:] A mantra-procedure [2. sādhya:] is sufficient to quell poison, [3.
hetu:] because it suppresses the innate power to produce the normal effect. [4. dṛṣṭānta:] just as in notes
case of a seed.” ([1] mantravidhiḥ [2] viṣapraśamanakṛt, [3] svakāryakaraṇaśaktinirodhakṛttvāt, [4]
bījavat. Here the reason should be again dharmaviśeṣaviruddha, yet Dharmakīrti does not disavow the
medical use of mantras. As we see Kṣemarāja adduces a different hetu: acintyaprabhāvatvāt, “because
they have inconceivable power.” This alludes to a famous list (maṇimantrauṣadha-) of non-ordinary
substances that can cancel the innate capacities of things. For example there exists a non-ordinary gem
(maṇi = asbestos?) that can cancel fire’s capacity to burn, etc.
Somadeva Vasudeva
507
2 Kṣemarāja and the Saiddhāntikas
Kṣemarāja’s own teacher Abhinavagupta had placed little value on such supernatural
displays as he does not subscribe to the probative view of initiation with the scales. He went
as far as claiming that such pratyaya-miracles occurring during initiation with the scales
are meant to give comfort to the simple-minded (mūḍhajana).50 While both the dualists
and the non-dualists accepted that a descent of grace was a prerequisite for liberation,
they differed fundamentally on its underlying cause. The non-dualists refused to admit
any causal factors at all, for to do so would have compromised their doctrine of divine
autonomy (svātantryavāda). Kṣemarāja therefore refutes the rival Saiddhāntika theories of
a karmic equilibrium and a maturation of defilement at great length. The Saiddhāntikas, in
turn, charged the non-dualists with a similarly specific catalogue of false doctrines:51
1. Īśvara would act without deliberation when he punishes and rewards souls without
cause or motivation.
2. Īśvara would have to create all realities at once.
3. Īśvara would have to bestow all karmic rewards to all at once.
4. Souls would lose their true form during liberation and would be reborn.
5. The scriptural statement that defilement is substantial would make no sense.
To get an overview of the structure of Kṣemarāja’s central argument we can look to his own
brief summary of it that he included in his commentary to Stavacintāmaṇi of Bhaṭṭanārāyaṇa
(late 9th to early 10th cent. CE).52 This pair of hymns reposed the question of grace as a
dilemma: does the individual need to pacify the mind before liberating grace can enter, or
is a calm mind the result of liberating grace?
Stavacintāmaṇi 117–118:53
I waver –
O Lord, do you enter54 awareness55 when it is pellucid,56
or does it become pellucid because you have entered?
50
51
52
53
54
55
56
Tantrāloka 20.1cd: atha dīkṣāṃ bruve mūḍhajanāśvāsapradāyinīm //
Svacchandoddyota p. 96.
Stavacintāmaṇivivṛti 117–118: vitatya caitan mayaiva śrīsvacchandavivṛtau pañcamapaṭalānte dīkṣāsamarthanāvasare vicāritam, “I myself have analyzed this in detail in my Svacchandavivṛti, at the end
of the fifth chapter in the context of justifying initiation.”
The verses are introduced with the assertion that the descent of divine grace is not beholden to anything
(nirapekṣa), Stavacintāmaṇivivṛti 117–118: idānīṃ nirapekṣo bhavacchaktipāta eva cidānandaghanasvātmasamāveśamayabhaktibhājo janān saṃpādayati, na tu karmasāmyamalaparipākādihetuko
’sau, nāpi bhavacchaktyadhiṣṭhānān malasya krameṇa paripāka ity etat ślokadvayena āha, “Now, with
two verses he states that your descent of grace, which is not dependent [on anything], directly perfects
persons who are filled with devotion, by which is meant an immersion into one’s own self, which is a
homogeneous bliss of consciousness.”
Kṣemarāja glosses this as “preside over with your power:” niviśase – svaśaktyadhiṣṭhānaṃ karoṣi.
Kṣemarāja: manasi – saṃvedane.
Kṣemarāja expands this to subsume the two theories of the Saiddhāntikas, the ripening of impurity
(malaparipāka) and karmic equilibrium (karmasāmya): prasanne – paripakvamale saṃjātakarmasāmye
vā sati (sati] Ed, manasi sati J1 mg R1 mg ).
508 A Causeless Liberation? Kṣemarāja’s Response to Dharmakīrti’s Critique of Initiation
But this is how the question can be resolved:
Your controlling presence is itself the pellucidity of awareness.
It is perfection, it is the highest level.57
Neither option spelled out in the first stanza is acceptable to the non-dualists, for in either
case Śiva remains beholden (sāpekṣa-) to some extrinsic factors or conditions necessary
to trigger the descent of grace. Bhaṭṭanārāyaṇa’s second stanza presents the non-dualists
solution: He identifies the active entrance (praveśa-) by Śiva’s liberating power of grace
as Śiva’s controlling presence (tvadadhiṣṭhāna-), which is then reified as nothing other
than the pellucid awareness that is the essence of the liberated state. Kṣemarāja explains
the cpd. as follows: “Your controlling presence is nothing other than the unveiling of your
autonomous powers by suspending immersion into the concealment of your true form.”58
Kṣemarāja’s commentary then rehearses his main attacks against the Saiddhāntikas. If
they argue that the descent of liberating grace is caused by either the ripening of impurity
(malaparipāka) or by a karmic equilibrium (karmasāmya) then we need to ask: What
causes either of these two phenomena?59 If the answer is that they are controlled by Śiva’s
superintending power, then they are redundant, and we should just admit that power as the
autonomous agency responsible for liberating grace. If the answer is that there is no cause,
then we need to ask why the descent does not take place for everyone all at once.60 If it is said
that it takes place after a specific amount of time for each person, then how is it not the same
as establishing a fixed time, since bondage61 is for everyone beginningless?62 If the answer
is that for any given soul it eventually takes place somehow, at some unspecified time, then,
unless we adduce some other dissimilar and sentient cause, we have the incoherence of
impurity – unequivocally insentient because its intrinsic power is that of obstruction –
producing a heterogeneous, sentient, effect in the form of maturation.63 If it is claimed that
the time we are talking about acts as an auxiliary cause, then, when asked what this time
might be, no answer is forthcoming.64 If it is argued that impurity gradually transforms into
a different, mature form, then, unless we adduce another dissimilar cause such as heat65
in the case of milk transforming into yoghurt, we have the incoherence of an impossible
57
58
59
60
61
62
63
64
65
Stavacintāmaṇi 117–118: prasanne manasi svāmin kiṃ tvaṃ niviśase kim u / tvatpraveśāt prasīdet tad
iti dolāyate janaḥ // niścayaḥ punar eṣo ’tra tvadadhiṣṭhānam eva hi / prasādo manasaḥ svāmin sā
siddhis tat paraṃ padam //
Kṣemarāja: tvadadhiṣṭhānaṃ – svarūpagopanānimajjanena tvatsvatantraśaktyunmīlanam (tvat-] Ed,
tat- J1 mg ) eva kevalaṃ manasaḥ saṃvedanasya prasādo, na tu malaparipākakarmasāmyādi kiṃ cit.
Stavacintāmaṇivivṛti 117–118: tathā hi yadi malaparipākaḥ śaktipātasya hetuḥ karmasāmyaṃ vā, tad
api tarhi kiṃhetukaṃ? samadhiṣṭhātrī bhagavacchaktir iti taddhetukam iti ced alaṃ tena, bhagavacchaktir eva svatantrānugrāhikā bhaviṣyati.
Ibid.: ahetukam iti cet sarvasya yugapat kiṃ na bhaviṣyati.
Most naturally “the beginning point of bondage,” cf. Alaka on Haravijaya 6.22: bandhakoṭiḥ – prakṛtyādibandhanadhārā.
Ibid.: kasmiṃś cit kāle bhavatīti cet sarveṣāṃ bandhakoṭeḥ anāditvāt kālaniyamaḥ kiṃ na kṛtaḥ.
Stavacintāmaṇivivṛti 117–118: tat kadā cit kasya cit kathaṃ cit bhavatīti cen malasya svaśaktyā
nirodhakatvenāvasthitasya jaḍasya vijātīyakāraṇānupraveśaṃ vinā paripākalakṣaṇavilakṣaṇakāryajananānupapattiḥ /
Ibid.: kaś cid eva kālo ’tra sahakārikāraṇam iti cet, ko ’sau iti praśne nottaraṃ labhyate /
Milk needs to be maintained for several hours at a temperature of 110–120 degrees F to produce dadhi.
See Vedamuthu (2006: 298ff.).
Somadeva Vasudeva
509
transformation into something heterogeneous.66 Kṣemarāja then makes the Saiddhāntika
approximate the non-dualist position:
If you [dualist Saiddhāntikas] say: “This [transformation] will happen just
in the same way as it does in the theory of [you] proponents of autonomous
descent grace, for whom the variety of experiences due to karma, and also
liberation, occur at a particular time, and for whom the relation of cause and
effect relates māyā, time etc. in the universe which has god as the only agent,”67
then an instance of the principle of tenuous connection is at hand.68
For Kṣemarāja the attempted rapprochement is unworkable, for the Saiddhāntikas do not
accept the identity of Śiva (the liberator) and the bound soul (the one to liberated):
Because in the non-dualist’s theory, the lord himself, freely assuming the
appearance of contraction, just as he appears as the various principles, so
he also appears as [their] relation of cause and effect, that is to say as their
fixed order, and also as the diversity of the respective experiences determined
by karma which thus appear to be temporally sequentialized, but in all of
this there is no individually [independent] reality for any of these temporal
moments or karmas etc. Since [Śiva] projects in this way both the diversity of
experience, as also, out of his autonomy, the diversity of [liberating] grace,
there is no other here who is being punished or favored, and so this theory is
faultless.69
Kṣemarāja subverts the charge of divine bias by explaining that in a non-dualist system
there exists no ‘other’ who is more or less favored or punished.70 A significant factor in
66
67
68
69
70
Stavacintāmaṇivivṛti 117–118: pariṇamanmalaḥ paripākātmakaṃ viśeṣam etīti ced atrāpi dugdhadadhipariṇāma iva uṣṇasparśasya virodhinaḥ kāraṇasya ananupraveśe vilakṣaṇapariṇāmānupapattiḥ /
Stavacintāmaṇivivṛti 117–118: yathā svatantraśaktipātavādipakṣe karmabhogavaicitryaṃ muktiś ca
kālaniyamena, bhagavadekakartṛke ca jagati māyākālādīnāṃ krameṇa kāryakāraṇabhāvaḥ, tathaiva
etad bhaviṣyatīti cec cāṣapañcāśanyāya āyātaḥ.
For the maxim of tenuous linkage see Kataoka 2010: 72–76 on Bhaṭṭa Jayanta’s expression: na cāṣeṇa
pañcāśad bhavitum arhati. He explains: 1. cāṣa and pañcāśad both share the same sound cāṣa (=cāśa).
2. pañcāśad means fifty in number 3. Therefore cāṣas, i.e. a flock of the blue-jay (cāṣāṇāṃ samūhaḥ),
are fifty in number. We could consider emending our passage to cāṣapañcāśanyāya, but the expression
cāśapañcāśanyāya with the palatal sibilant is well attested in Kashmirian works, so we may also rather
consider cāśa an orthographical variant for cāṣa.
Stavacintāmaṇivivṛti 117–118: yatas tatpakṣe bhagavān eva svātantryād gṛhītasaṃkocābhāso yathā
tattattattvātmanā bhāti, tathā tadgataniyatapaurvāparyātmakakāryakāraṇabhāvābhāsātmanāpi niyatakarmatatphalavaicitryātmanāpi kramābhāsamayena, na tv atra kālakarmādeḥ kasyāpi nijaṃ tattvam
astīti yathāsāv itthaṃ bhogavaicitryam ābhāsayati, tathā svasvātantryād anugrahavaicitryam apīti
nātra nigrahānugrahabhāk kaś cid anyo ’stīti niravadya evāyaṃ pakṣaḥ /
Ibid.: idam atra tattvaṃ, yac citiśaktir eva bhagavatī svasvātantryād gṛhītasaṃkocā cittabhūmiṃ saṃsāryātmarūpāṃ bahuśākhām ābhāsya, punaḥ svecchayaiva kvacit saṃkocaṃ praśamayya, pūrṇatayā
sphuratīty eva tat paraṃ padam, “This is the truth here: when the blessed power of consciousness out
of her own autonomy accepts contraction and diffuses into the level of citta and makes herself appear
to possess many offshoots, and then again out of her own autonomy loosens the contraction in some
region and flashes forth as full, that is the highest state.”
510 A Causeless Liberation? Kṣemarāja’s Response to Dharmakīrti’s Critique of Initiation
this perspective, is that causation in the world of transmigration is itself nothing but an
apparent phenomenon manifested by Śiva out of his free will. Abhinavagupta therefore
distinguishes the causal relationship into two types: an absolute, or non-artificial causal
relation (pāramārthikaḥ), and an artificial causal relation (kalpitaḥ). In the absolute causal
relation Śiva is the only existing cause and agent. This is so because true agency (kartṛtva)
can only be grounded in autonomy (svātantrya). The artificial causal relation of everyday
experience is merely an appearance, amounting to no more than a belief in personal agency,
as discussed in Vasudeva (2012–2014: 213–215). Even in the standard example of a potter
who makes pots, Abhinavagupta claims that Śiva is the actual, underlying agent. In the
sense that all artificial causation is merely apparent anyway, liberation is not really caused
by any of the factors adduced by the Saiddhāntikas. Or rather, it is caused by the same
absolute cause as everything else, Śiva’s will.
References and abbreviations
Acharya 2014 D. Acharya, On the Śaiva Concept of Innate Impurity (mala) and the
Function of the Rite of Initiation. Journal of Indian Philosophy 42 (2014) 9–25.
Ajaḍapramātṛsiddhi Ajaḍapramātṛsiddhi: Siddhitrayī and Pratyabhijñā-kārikā-vṛtti of
Rajanaka Utpala Deva, ed. M.K. Shāstrī. Srinagar 1921.
Bodhicaryāvatārapañjikā Bodhicaryāvatārapañjikā: Bodhicaryāvatāra of Śāntideva with
the Commentary Pañjikā of Prajñākaramati, ed. Vaidya, P.L. Darbhanga 1960.
Eltschinger 2001 V. Eltschinger, Dharmakīrti sur les mantra et la perception du suprasensible. Wien 2001.
Franco 1997 Eli Franco, Dharmakīrti on compassion and rebirth. Vienna 1997.
Goodall 2013 Tāntrikābhidhānakośa III. Ṭ–PH. Dictionnaire des termes techniques de la
littérature hindoue tantrique. A Dictionary of Technical Terms from Hindu Tantric
Literature. Wörterbuch zur Terminologie hinduistischer Tantren. Fondé sous la
direction de Hélène Brunner, Gerhard Oberhammer et André Padoux. Direction
éditoriale du troisième volume: Dominic Goodall et Marion Rastelli. Vienna 2013.
Kataoka 2010 Kei Kataoka, A Critical Edition of Bhaṭṭa Jayanta’s Nyāyamañjarī:
Jayanta’s View on jāti and apoha. Institute of Oriental Culture Bulletin 158 (2010)
168 (113)–220 (61).
Ked See Svacchandoddyota.
Kiraṇatantra Kiraṇatantra: Bhaṭṭa Rāmakaṇṭha’s Commentary on the Kiraṇatantra,Vol.
1, Chapters 1–6, critical edition and annotated translation by Dominic Goodall.
Pondicherry 1998.
Kiraṇavṛtti See Kiraṇatantra.
Mālinīvijayottara Mālinīvijayottara: Sri Mālinīvijayottara Tantram, ed. Madhusūdan Kaul
Shāstrī. Srinagar 1922.
Manoramāṭīkā See Tantrarāja.
Mataṅgapārameśvara Mataṅgapārameśvara: Mataṅgapārameśvara, Kriyāpāda, Yogapāda, and Caryāpāda, with the commentary (-vṛtti) of Bhaṭṭa Rāmakaṇṭha on Kriyāpāda 1.1–11.13, ed. N.R. Bhatt. Pondicherry 1982.
Mataṅgapārameśvaravṛtti See Mataṅgapārameśvara.
Somadeva Vasudeva
511
Nareśvaraparīkṣā Nareśvaraparīkṣā: Nareśvaraparīkṣā of Sadyojyotis with the commentary (-prakāśikā) of Bhaṭṭa Rāmakaṇṭha, ed. Madhusūdan Kaul Shāstrī. Srinagar
1926.
Netratantroddyota Netratantroddyota: The Netra Tantram, with Commentary by Kshemarāja, 2 vols., ed. M.K. Shāstrī. Bombay 1926 & 1939.
Netroddyota See Netratantroddyota.
Niśvāsatattvasaṃhitā Niśvāsatattvasaṃhitā: The Niśvāsatattvasaṃhitā. The Earliest Surviving Śaiva Tantra, Volume 1, A Critical Edition & Annotated Translation of the
Mūlasūtra, Uttarasūtra & Nayasūtra, edited by Dominic Goodall, in collaboration
with Alexis Sanderson & Harunaga Isaacson, with contributions of Nirajan Kafle,
Diwakar Acharya & others. Pondicherry 2015.
PV Edma Pramāṇavārttika: Dharmakīrti’s Pramāṇavārttika with a Commentary by Manorathanandin, ed. R. Sāṅkṛtyāyana. Journal of the Bihar and Orissa Reseach Society:
New Series 24–26 (1938) 349–84.
PV Edmi Pramāṇavārttika: Pramāṇavārttika-Kārikā of Dharmakīrti. Sanskrit and Tibetan,
ed. Yūsho Miyasaka. Acta Indologica 2 (1971/72).
PV Edpra Pramāṇavārttika: Pramāṇavārtikabhāshyam or Vārtikālaṅkāraḥ of Prajñākaragupta: Being a Commentary on Dharmakīrti’s Pramāṇavārtikam, ed. R. Sāṅkṛtyāyana. Patna 1953.
Rauravasūtrasaṃgraha Rauravasūtrasaṃgraha: Surviving Chapters are published as the
Vidyāpāda of the Rauravāgama In: Rauravāgama, Vol. I, ed. N.R. Bhatt. Pondicherry
1961.
S1 Svacchandatantroddyota, Śrīnagar acc. no. 1054-II. Śāradā. 411 folios.
Sāṃkhyakārikā See Yuktidīpikā.
Sanderson 1992 A. Sanderson, The Doctrine of the Mālinīvijayottaratantra. In: Ritual and
Speculation in Early Tantrism. Studies in Honour of André Padoux, ed. T. Goudriaan.
Albany 1992, 281–312.
Sanderson 2006 A. Sanderson, The Date of Sadyojyotis and Brhaspati. Cracow Indological Studies 8 (2006) 39–91.
Sanderson 2007 A. Sanderson, The Śaiva Exegesis of Kashmir. In: Tantric Studies in
Memory of Hélène Brunner, ed. Dominic Goodall & André Pardoux. Pondichéry
2007, 231-442.
Sārdhatriśatikālottara Sārdhatriśatikālottarāgama: Sārdhatriśatikālottarāgama avec le
commentaire de Bhaṭṭa Rāmakaṇṭha, ed. N.R. Bhatt. Pondicherry 1979.
Sarvajñānottara Sarvajñānottaratantra, NAK MS 1-1672. NGMPP Reel No. A 43/12.
Palm-leaf, early Nepalese script. Described by Śāstri (1905: lxxiv–lxxv and 85–86).
Also GOML MS D 5550 and IFP T. Nos. 334, 760, paper transcripts in Devanāgarī.
The verse and chapter numeration used is that of Goodall’s edition in progress.
Śivasūtravārttika Śivasūtravārttika: Śivasūtravārttika of Varadarāja, also called Kṛṣṇadāsa, ed. Madhusudan Kaul. Srinagar 1925.
Stavacintāmaṇi Stavacintāmaṇi: Stavacintāmaṇi of Bhaṭṭa Nārāyaṇa with the commentary
(-vivṛti) of Kṣemarāja, ed. Mukunda Rāma Śāstrī. Srinagar 1918.
Stavacintāmaṇi J1 Stavacintāmaṇi, Jammu ORL1329. Śāradā on paper. Contains marginal
glosses drawn from Kṣemarāja’s commentary.
512 A Causeless Liberation? Kṣemarāja’s Response to Dharmakīrti’s Critique of Initiation
Stavacintāmaṇi R1 Stavacintāmaṇi, Raghunath Temple MS. 9632. Śāradā on paper. Contains marginal glosses drawn from Kṣemarāja’s commentary.
Stavacintāmaṇivivṛti Stavacintāmaṇivivṛti: See Stavacintāmaṇi.
Svacchandoddyota Svacchandoddyota: The Svacchanda Tantram, with Commentary by
Kshemarāja, 6 vols., ed. M.K. Shāstrī. Bombay 1921–1935.
Svāyambhuvasūtrasaṃgraha Svāyambhuvasūtrasaṃgraha: Svāyambhuvasūtrasaṃgraha,
ed. Veṅkaṭasubrahmaṇyaśāstrī. Mysore 1937.
TĀK III Tāntrikābhidhānakośa: See Goodall 2013.
Tantrāloka Tantrāloka: Tantrāloka of Abhinavagupta with the Commentary (-viveka) of
Rājānaka Jayaratha, ed. M.K. Śāstrī. Bombay/Srinagar 1918–1938.
Tantrālokaviveka See Tantrāloka.
Tantrarāja Tantrarājatantra: Tantrarājatantra with Manoramāṭīkā, ed. L. Sastri. Calcutta
1929.
Tattvasaṃgraha Tattvasaṃgraha: Le Tattvasaṃgraha “Compendium des Essences” de
Sadyojyoti, ed. Pierre-Sylvain Filliozat. Bulletin de l’EFEO 77 (1988) 101–163.
Vasudeva 2004 Somadeva Vasudeva, The Yoga of the Mālinīvijayottaratantra. Pondicherry
2004.
Vasudeva 2012–2014 Somadeva Vasudeva, The Unconscious Experiencer: Bhoktṛtva in
the Pramātṛbheda of the Trika. Journal of Indological Studies 24 (2012–2014)
203–230.
Vedamuthu 2006 Ebenezer R. Vedamuthu, Other Fermented and Culture-Containing
Milks. In: Manufacturing Yogurt and Fermented Milks, ed. Ramesh C. Chandan.
Ames, Iowa 2006, 295–308.
Yuktidīpikā Yuktidīpikā: Yuktidīpikā: the most significant commentary on the Sāṃkhyakārikā, critically edited by Albrecht Wezler and Shujun Motegi. Stuttgart 1998.
Dharmakīrti and His Successors on āśrayāsiddha and
prasaṅgaviparyaya1
by
Toshikazu Watanabe
Introduction
For Buddhist philosophers the logical fallacy called āśrayāsiddha (a pseudo-logical reason
whose basis is not established) is an intractable problem when dealing with something
whose existence is not accepted by Buddhists as the subject of a thesis of their own, such
as the Sāṅkhya’s pradhāna (primordial matter) or the Vaiśeṣika’s eternal ākāśa (space) and
so on. Dignāga (ca. 480–540), the founder of the Buddhist logico-epistemological school,
presented different approaches to this issue in his two works, the Nyāyamukha and the
Pramāṇasamuccaya. In his earlier work, the Nyāyamukha, when proving the non-existence
of the Sāṅkhya’s pradhāna through the logical reason “not being perceived,” he permitted
pradhāna to be placed as the subject of the thesis by seeing pradhāna as a conceptually
constructed object (kalpita), in order to allow the logical reason to be a property of the
subject of the thesis (pakṣadharma). However, in the Pramāṇasamuccaya he does not use
this method, but addresses the problem in a different way. According to his new method,
the Buddhist proponent can put forth the Sāṅkhya’s pradhāna as the subject of the thesis
when he formulates a reductio ad absurdum kind of argument (prasaṅga) to refute (dūṣaṇa/
parihāna) the adversary’s tenets. For the sake of convenience I shall refer to this technique
for avoiding the problem of āśrayāsiddha as the “method of hypothetical assumption.”
As Tom Tillemans has pointed out,2 Dharmakīrti (ca. 600–660 or 550–650?) also
provides two ways to deal with this issue: The first is based on Dignāga’s solution found in
the Nyāyamukha. In his Svavṛtti on Pramāṇavārttika 1.205–206 Dharmakīrti explains the
word kalpita mentioned in the NMu as referring to an image that appears in the cognition
and that it thus can exist as the subject of the thesis. This first method is designated by
Tillemans as the “principle of conceptual subjects.” The second method is mentioned in PV
4.136–148, where Dharmakīrti comments on the word svadharmiṇi in Dignāga’s definition
of a thesis (pakṣa) given in Pramāṇasamuccaya 3.2. According to this second method,
when the Buddhist proponent places ākāśa, for example, as the subject of a thesis, what is
intended by the word ākāśa is not the eternal entity postulated by the Vaiśeṣika opponent,
but any other entity called ākāśa whose existence is recognized not only by the Buddhist
proponent, but also by ordinary people. Tillemans calls this “the method of paraphrase.”
1
2
Work on this paper has been generously supported by the Austrian Science Fund (FWF) in the framework
of the FWF project P23196–G15 “Buddhist literature and its context” and by the 2014 Bukkyo Dendo
Kyokai Fellowship for Japanese scholars for my project “Acceptance of logical thought in Mahāyāna
Buddhism and its development.” I would like to thank Ms. Katharine Apostle for correcting my English
and my colleague Dr. Pascale Hugon for giving me invaluable suggestions.
See Tillemans 1999: 174–176.
Birgit Kellner et al., eds., Reverberations of Dharmakīrti’s Philosophy: Proceedings of the Fifth International Dharmakīrti
Conference Heidelberg, August 26 to 30, 2014. Vienna 2020, pp. 513–528.
514
Dharmakīrti and His Successors on āśrayāsiddha and prasaṅgaviparyaya
In the third chapter of the Pramāṇaviniścaya, Dharmakīrti also adopts Dignāga’s second
method as presented in the PS, i.e., the “method of hypothetical assumption.” As a result, a
total of three methods – two from Dignāga and one developed by Dharmakīrti – coexist
within Dharmakīrti’s system of logic in order to prevent the problem of āśrayāsiddha. How
then does he harmonize these three, especially the two methods adopted from Dignāga? If
the “principle of conceptual subjects” is applied, the “method of hypothetical assumption”
seems to be no longer necessary. Moreover, it is reported by Takashi Iwata that Dharmakīrti’s
commentators, such as Dharmottara and Prajñākaragupta, have views on prasaṅga and
prasaṅgaviparyaya that are different from how they are discussed by Dharmakīrti in PVin 3.
The difference of their opinions depends on which of Dharmakīrti’s methods they ascribe
more importance to.
The aim of this paper is, therefore, to examine Dharmakīrti’s explanation of how to
avoid the problem of āśrayāsiddha and to clarify its historical development from Dignāga
to Dharmakīrti and on to his successors.
1. Dignāga
1.1 The Nyāyamukha: the principle of conceptual subjects
In the NMu, Dignāga explains the general rule that in an inference – be it an inferencefor-oneself (svārthānumāna) or an inference-for-others (parārthānumāna) – one property
of the subject (i.e., sādhya-dharma/a property to be proved) is known through another
property of the same subject (i.e., sādhana-dharma/a proving property). In this explanation,
Dignāga refers to an opponent (probably a Sāṅkhya) who insists that this rule cannot cover
every instance because in some cases the subject (dharmin) itself is proved to be existent
or non-existent. As an example this opponent gives the following reasoning, which uses
pradhāna as the subject of the thesis:3
hPrayoga 1i
[Thesis:] [pradhāna] does not exist;
3
Just prior to hPrayoga 1i, Dignāga refers to another reasoning (prayoga) that also has pradhāna as
the subject of the thesis: [Thesis:] pradhāna exists (asti pradhānam); [Reason:] because in various
individuals homology (anvaya) [concerning the three constituents, i.e., pleasure (sukha), pain (duḥkha),
and confusion (moha)] is observed (bhedānām anvayadarśanāt). (NMu [Katsura 1978: 110] T. 1628
1b29: 有成立最勝爲有。現見別物有總類故) His answer to this is: The thesis should be formulated
as “The various individuals certainly possess one and the same cause [i.e., pradhāna].” [If so,] the
[existence of] pradhāna is not established [directly]; hence, there is no error [of the violation of the
above-mentioned general rule]. (NMu [Katsura 1978: 110] T. 1628 1c1–2: 此中但立別物定有一因
爲宗、不立最勝、故無此失。) Dignāga criticizes this prayoga by pointing out that its thesis is not
formulated properly. The subject of the thesis should be “various individuals,” since otherwise the
logical reason “homology” cannot be a pakṣadharma. Although earlier studies (Tillemans 1999: 175 and
Yao 2009: 386–387) regard this explanation by Dignāga as similar to the method of paraphrase applied
by Dharmakīrti in PV 4.136–148, there is a fundamental difference. While Dharmakīrti’s method, as
will be explained below, § 2.2, just rephrases the subject of the thesis pradhāna into “pleasure, etc.,”
the existence of which is accepted by the Buddhists, Dignāga’s method mentioned above changes the
subject from pradhāna to “various individuals” in order to give pakṣadharmatva to the logical reason
“homology.” Thus Dignāga’s critique is not aimed at avoiding the problem of āśrayāsiddha.
Toshikazu Watanabe
515
[Reason:] because it is not perceived. (Cf. PVSV 105,15: na santi pradhānādayaḥ, anupalabdheḥ.)4
To this, Dignāga gives the following answer:
When the non-existence [of pradhāna] is proved [on account of its not being
perceived], “non-perception” is a property of a conceptually constructed object
(i.e., pradhāna) (假安立不可得法/*kalpitasyānupalabdhir dharmaḥ); hence,
there is also no error of [proving] the subject of the thesis [with the logical
reason].5
Here, in order to ensure the pakṣadharmatva for the logical reason “non-perception,”
Dignāga gives a certain status of existence to pradhāna by seeing it as a conceptual
construction (kalpita). In this way, any pseudo-entity can be accepted as a substratum of
the logical reason, and the fallacy of āśrayāsiddha can be thereby avoided. This strategy,
following Tillemans, is called the “principle of conceptual subjects.” At the time of the
NMu this was the only means for avoiding the problem of āśrayāsiddha. In this connection,
it should be noted that Dignāga shows here that using this principle enables Buddhists to put
pradhāna as a subject of the thesis in a proper proof (sādhana), i.e., a proof that is put forth
by Buddhist proponents themselves, even though they do not accept such pseudo-entities
in reality.
1.2 The Pramāṇasamuccaya: the method of hypothetical assumption
In the third chapter of the Pramāṇasamuccaya, while Dignāga deals with a reasoning that
has the same thesis as the hPrayoga 1i, i.e., nāsti pradhāna, he discusses it in a different
context, i.e., a discussion about the reductio ad absurdum kind of argument (prasaṅga).
4
5
See NMu (Katsura 1978: 110) T. 1628 1c1: 或立爲無。不可得故。
See NMu (Katsura 1978: 110) T. 1628 1c2–4: 若立爲無、亦假安立不可得法、是故亦無有有法
過。
516
Dharmakīrti and His Successors on āśrayāsiddha and prasaṅgaviparyaya
PS(V) 3.16cd–17ab:6 evaṃ kecid anyaprasaṅgaṃ kathayanti – nāsti pradhānam, bhoktṛbhogyayor guṇāguṇatvaprasaṅgāt … iti. sa katham āvīta iti cet.
na hy ayam āvītaḥ. yasmāt
prasaṅgo ’pakṣadharmatvād anyo hetupratijñayoḥ //16//
doṣoktyā dūṣaṇaṃ jñātaṃ pūrvatropagame sati /
[Objection:] Like [the Sāṅkhyas], some [Buddhists] also speak of another
reductio ad absurdum reasoning (prasaṅga) [as follows:]
hPrayoga 2i It is not the case that pradhāna exists, because [if the pervasion
(vyāpti) you presuppose in your proof of the existence of pradhāna were to be
accepted,]7 it would follow that both the enjoyer (i.e., puruṣa) and that which
6
7
Words in roman typeface are attested in Jinendrabuddhi’s commentary (PSṬ) or in fragments, whereas
those in italics have been reconstructed. The Sanskrit reconstruction of the PS and PSV used in this
paper is a result of the PSṬ seminar organized and led by Prof. Shōryū Katsura at Ryukoku University.
Through his kindness, I have been able to use their reconstruction. As for the number of kārikās in
PS 3, it is to be noted that recently the editorial team of the PSṬ 3 has adopted a different way of
counting them than was hitherto widely adopted by scholars, e.g., Kitagawa 1965 and Katsura 2009, etc.
While the earlier counting regards k.7 as consisting of only a half verse (i.e., 7ab), the new counting
includes the following half verse into k.7. That is, in the new counting k.7 consists of 7ab and 8ab of the
earlier counting. Therefore, the new number of the kārikās in PS 3 shifts back by half from k.7. In this
paper I use the new way of counting. The reading of k. 16cd–17ab I have adopted here is a modified
version of that presented in Katsura 2009: 158; there it reads as follows: prasaṅgo ’pakṣadharmatvāt
pūrvatropagame sati / hetupratijñayos teṣāṃ (or tatra) doṣoktyā duṣaṇaṃ gatam // Cf. K (P129b3–6):
de ltar na ’ga’ zhig gtso bo ni yod pa ma yin te / longs spyod pa po dang longs spyod par bya ba dag
yon tan dang yon tan med pa’i (yon tan med pa’i em. [see P129b8–130a1]: yon tan can gyi P) rang
bzhin du thal ba’i phyir … zhes thal bar ’gyur ba gzhan brjod par byed pa de ci ltar bsal te ’ongs
pa ma yin zhe na / ’di ni bsal te ’ongs pa ma yin te / gang gi phyir / thal ’gyur phyogs chos can min
phyir // sngon du khas blangs yod na ni // rtags dang dam bca’ gzhan dag la // skyon brjod sun ’byin
shes par bya //; V (D44b1–3, P47a8–47b2): yang ’dir thal ba gzhan brjod pa gtso bor yod pa ma yin
te / longs spyod pa po dang longs spyad par bya ba dag yon tan dang yon tan can nyid du thal bar ’gyur
ba’i phyir ro zhes pa … zhes bya ba de ji ltar bsal te ’ongs pa yin ’di ni bsal te ’ongs pa ma yin te / gang
gi phyir / thal ’gyur phyogs chos can min phyir // khas blangs sngon du song ba las // de gzhan (de
gzhan P Vṛtti : de bzhin D Vṛtti , D Kārikā ) rtags dang dam bca’ yi (yi D Vṛtti , P Vṛtti : yis D Kārikā ) // skyon
brjod sun ’byin du shes bya //
Dignāga refutes the Sāṅkhya’s proof of the existence of pradhāna through the following process: First
the Sāṅkhya puts forth hPrayoga Ai for proving the existence of pradhāna. hPrayoga Ai Pradhāna exists
because in various individual things homology (anvaya) [concerning three constituents, i.e., sukha,
duḥkha and moha] is observed. (asti pradhānam, bhedānām anvayadarśanāt.) (2) In order to ensure
the pakṣadharmatva of the logical reason “homology,” this hPrayoga Ai is rewritten as hPrayoga A’i.
Various individual things possess one and the same cause because homology is observed. (bhedānām
ekakāraṇatvam, anvayadarśanāt.) (3) The Sāṅkhya adds the following reasoning in order to prove
hPrayoga A’i, because otherwise the opponent, such as Buddhists, does not accept that pleasure etc. are
commonly shared by various individual things. hPrayoga Bi In various individual things homology is
observed because they have common effects such as joy (prīti). (4) Dignāga’s refutation: if hPrayoga
Bi is held to be true, the following vyāpti relation must be accepted. [Vyāpti:] Whatever brings about
common effects, such as joy, etc., has a homology concerning three constituents. (yasya prītyādi kāryaṃ
dṛṣṭam, tasya guṇānvayam.) (5) However, because even a soul (puruṣa), which cannot be accepted in
the Sāṅkhya system as consisting of three constituents, can bring about an effect such as joy if it is used
as an object of meditation by adherents of the Sāṅkhya, the logical reason “having common effects” in
hPrayoga Bi would be inconclusive (anaikāntika). (6) Suppose that, on the contrary, the Sāṅkhya does
Toshikazu Watanabe
517
is to be enjoyed (i.e., vyakti) consist of [three] constituents (guṇa), or do not
consist of [three] constituents.
… Why is this an āvīta reasoning? [Of course it is not.]
[Answer:] Actually, this is not an āvīta reasoning because,
since [its logical reason is] not the property of the subject of the thesis, the reductio ad absurdum reasoning is different [from the āvīta reasoning]. [Instead,
this reductio ad absurdum reasoning is] understood as a refutation (dūṣaṇa)
because [it] points out the failure of the [adversary’s] logical reason or of his
thesis after accepting [them].
Just before this passage, Dignāga rejects the Sāṅkhya view that in the āvīta reasoning –
a type of prasaṅga argument – the logical reason can prove the Sāṅkhya’s proposition
without its being a property of the thesis (pakṣadharmatva), and he explained that if the
pakṣadharmatva of the logical reason in the vīta reasoning for proving the existence of
pradhāna is accepted (at least) by the Sāṅkhya proponent, then the logical reason in the
āvīta reasoning can also be regarded as possessing pakṣadharmatva because it can be
reduced to the logical reason in the vīta reasoning.8
The Sāṅkhya raises an objection against this explanation: In spite of the fact that for
the Buddhists the hPrayoga 2i must consist in sound reasoning, the logical reason cannot
possess pakṣadharmatva because the existence of the subject of the thesis in hPrayoga 2i,
i.e., pradhāna, is not accepted by Buddhists. As a result, it cannot be held that every sound
logical reason must be a property of the subject of the thesis (pakṣadharma).
In his reply, Dignāga, unlike in the case of the NMu, does not try to ensure the pakṣadharmatva of the logical reason in the hPrayoga 2i. He rather agrees that the logical reason
does not possess pakṣadharmatva. This does not mean, however, that he relinquishes the
soundness of the hPrayoga 2i. He is able to do this by distinguishing the logical reason in a
proper proof (sādhana) from the logical reason in a refutation (dūṣaṇa). While the former
should possess pakṣadharmatva (and a vyāpti relation with the property to be proved), the
latter does not. According to Dignāga, a refutation is simply a means of denying the soundness of the adversary’s reasoning by pointing out the undesired consequences which occur
when the adversary’s thesis or logical reason (which necessarily includes a vyāpti relation)
is accepted.9 And since in the refutation what is to be investigated is the logical consequence
derived from the adversary’s thesis or reasoning, the existence or non-existence of the
subject of the thesis does not become a topic of discussion. Therefore, the problem of
āśrayāsiddha does not occur in the refutation. Since the prasaṅga argument is formulated
8
9
not accept the vyāpti relation in order to avoid the undesired consequence. In this case, because denying
the validity of the vyāpti relation results in the denial of the soundness of hPrayoga Bi, hPrayoga Bi
does not establish the pakṣadharmatva of the logical reason in hPrayoga Ai/hPrayoga A’i. And, as a
result, the logical reason in hPrayoga Ai/hPrayoga A’i does not prove the existence of pradhāna.
See Watanabe 2013.
See PS 3.13cd–14ab: hetupratijñādvāreṇa yatrāniṣṭiḥ prasajyate // taddvāreṇa prayogāt* sa parihāra
itīṣyate / (*sbyor phyir V: thal ba’i phyir K. Following K, Yoshimizu 2013: 433 adopts prasaṅgāt.)
518
Dharmakīrti and His Successors on āśrayāsiddha and prasaṅgaviparyaya
after hypothetically assuming the opponent’s system of philosophy, this method can be
called the “method of hypothetical assumption.”10
2. Dharmakīrti
2.1 Dharmakīrti’s version of the principle of conceptual subject – PV 1 and PVin 3
In PV 1.205–212 and his own commentary on these verses, Dharmakīrti explains Dignāga’s
previously mentioned hPrayoga 1i (“pradhāna does not exist because it is not perceived”)
presented in the NMu, developing Dignāga’s principle of conceptual subjects from the point
of view of the theory of apoha. In the following verses, he explains why in the hPrayoga 1i
the fallacy of āśrayāsiddha does not occur.
PV 1.205–206 = PVin 3.53–54:
anādivāsanodbhūtavikalpapariniṣṭhitaḥ /
śabdārthas trividho dharmo bhāvābhāvobhayāśrayaḥ //205//
tasmin bhāvānupādāne sādhye ’syānupalambhanam /
tathā hetur na tasyaivābhāvaḥ śabdaprayogataḥ //206//
The verbal object (śabdārtha), which is completely derived from conceptualisation proceeding from beginningless karmic tendencies, is a dharma of
three kinds: based on something existent, something non-existent or both.11
When this [verbal object, such as pradhāna, etc.], which is without any existent
substratum, is being proven, then the non-perception of this as being in such
a way [i.e. as existing externally] is the logical reason. The non-existence of
this very [śabdārtha] itself is not, for we do use words [like ‘pradhāna,’ etc.].
(Tillemans 1999: 176)
Following Dignāga’s description in the NMu, Dharmakīrti here explains that the subject of
the thesis pradhāna is a conceptual construction and that this pradhāna does not have any
external basis. But why can this conceptual construction be accepted as a proper subject of
the thesis? This is explained as follows:
PVSV 105,26–27 = PVin 3 68,1–3: vaktuḥ śrotuś ca tadvikalpabhājaḥ, yathāpratibhāsivastupratipādanasamīhāprayogāt, tadākāravikalpajananāc ca.
And both speaker and listener share such a conceptual cognition (i.e., a conceptual cognition coming from a beginningless imprint) because [the former]
10
11
This method corresponds to “the principle of propositional attitude,” as it has been called by Zhihua
Yao. In his article (Yao 2009: 393–396) he states that this principle was developed by Chinese Buddhist
logicians, such as Kuiji, and is merely implied by Dignāga and the Indian Buddhists. However, as we
have seen, this principle can clearly be traced back to Dignāga’s description in PS 3.
As Shinya Moriyama has pointed out in his presentation at the 17th IABS conference (21 Aug. 2014),
Dharmapāla (T. 1571 188b12–14) also applies the same triple typology. The difference between
Dharmapāla and Dharmakīrti is that the former applies it to the classification of the logical reason, but
the latter to the conceptual construction in general.
Toshikazu Watanabe
519
uses [words] according to the intention to convey a thing as it appears [to his
conceptual cognition] and because [the latter, by hearing the speaker’s words,]
brings about the conceptual cognition which has the same form [as that of the
speaker].
Although the conceptually constructed pradhāna does not have an external basis, it exists
as an image appearing to the cognition, that is, it has an internal basis. Moreover, since this
internal image is considered to be common to the speaker and listener,12 i.e, the proponent
and opponent, it can be a proper subject of the thesis.
In this way, Dharmakīrti develops Dignāga’s “principle of conceptual subjects” further
by providing an ontological basis. He first advocated this view in his PVSV, and in his later
work it is included in PVin 3 without any changes.
2.2 Method of paraphrase – PV 4
Dharmakīrti’s second method for avoiding the problem of āśrayāsiddha is, in the words of
Tillemans, the “method of paraphrase.” This method is employed by Dharmakīrti in his
PV 4.136–148, where he comments on the word svadharmiṇi that appears in Dignāga’s
definition of the thesis in PS 3.2. An example dealt with in the relevant passage of PV 4 is
a Buddhist proof of the impermanence of space and so forth (khādi or vyomādi) through
the logical reason “not producing sound, etc., all at once” (sakṛc chabdādyahetutva).13
The Vaiśeṣika opponent raises the objection that if this reasoning is held to be true and is
aimed at negating the Vaiśeṣika understanding of space as an eternal entity, it then follows
that, because this reasoning negates the nature of space as understood by the Vaiśeṣika and
hence the existence of that space cannot be accepted,14 the logical reason is categorized as
dharmisvarūpaviparītasādhana. Dharmakīrti replies to this objection by introducing the
distinction between svadharmin (the subject intended by the proponent) and kevaladharmin
(a thing that is not related to the subject intended by the proponent). In the case of the
reasoning of impermanence of space, the subject put forth by the Buddhist proponent
is not eternal space as postulated by the Vaiśeṣika opponent, but a certain entity called
“space” whose existence is widely accepted by ordinary people. Therefore, even though
the existence of eternal space is negated by the reasoning, no invalidation of the Buddhist
proponent’s thesis can occur.15 Although Dharmakīrti himself does not mention here the
problem of āśrayāsiddha directly, this method is applicable to our relevant problem. Hence
12
13
14
15
According to Dharmakīrti, although in reality the internal image held by the speaker must differ from
the internal image held by the listener, these two are regarded as the same. For this fundamental problem
of Dharmakīrti’s theory of apoha, see Kataoka 2010: 267–269.
See PV 4.141abcd’: yathā parair anutpādyāpūrvarūpaṃ na khādikam / sakṛc chabdādyahetutvād ity
ukte. “For example, when [the Buddhist] states that space, etc. do not have a novel nature unproduced
by other [conditions] because they are not causes for [producing their qualities such as] sound, etc. all
at once …” (Tr. Tillemans 2000: 202.)
See PV 4.141d’–142ab: prāha dūṣakaḥ // tadvad vastusvabhāvo ’san dharmī vyomādir ity api / “…
then the [Vaiśeṣika] adversary might say that in a similar way the subject, space, etc., would also not
have the nature of a real entity.” (Tr. Tillemans 2000: 202.)
See PV 4.142cd: naivam iṣṭasya sādhyasya bādhā kācana vidyate // “In this fashion [even though the
subject is invalidated], there is in fact no invalidation of the intended [proposition] to be proved (sādhya)
at all.” (Tr. Tillemans 2000: 202.)
520
Dharmakīrti and His Successors on āśrayāsiddha and prasaṅgaviparyaya
the logical reason “not producing sound, etc., all at once” is accepted to be a property of
the subject of the thesis because the svadharmin of this reasoning is a certain entity called
“space.”
2.3 Dharmakīrti’s version of the method of hypothetical assumption – PVin 3
In PVin 3, Dharmakīrti speaks about prasaṅga arguments formulated by relying on what is
accepted only by the opponent. There, in order to avoid the problem of āśrayāsiddha, he also
adopts the method of hypothetical assumption presented by Dignāga in PS 3. Dharmakīrti
begins his discussion as follows:
PVin 3 4,4–9: yas tu paraparikalpitaiḥ prasaṅgaḥ, yathā deśakālāvasthāviśeṣaniyataikadravyasaṃsargāvyavacchinnasvabhāvāntaravirahād anekavṛtter
ekasya na deśādiviśeṣavatānyena yogaḥ, tathābhūtasvabhāvasya virodhād
bhinnadeśādiyogena, sa ekadharmopagame ’paradharmopagamasandarśanārthaḥ. tadanabhyupagame cobhayanivṛttiḥ, vivekasya kartum aśakyatvāt,
tasyānyatra vastutaḥ pratibandhāt.
On the other hand, a reductio ad absurdum kind of argument (prasaṅga) [is
formulated] by means of [things] postulated by the opponent, as for instance:
hPrayoga 3i
[Thesis:] A single entity (eka), [although] it is [regarded by the opponent as]
occurring in several things, is not united with others which have differences in
terms of place, etc.
[Reason:] because it is devoid of another essential property which is not obstructed (avyavacchinna) by the unification (saṃsarga) with a single substance
restricted by a particular place, time and state.
For it is incompatible that a thing that has such a nature (i.e., singularity) is
united with those which are different in place, etc.
The purpose of such [an argument] (saḥ) is to show that when one property X
is accepted, the other property Y is [also] accepted. To the contrary, when Y
is not accepted, both are negated because [Y] cannot be distinguished [from
X]. This is because Y is in reality bound to the other (i.e., X).
I will discuss the entire structure of this hPrayoga 3i below. What I would like to focus
on here is the passage beginning with the relative pronoun saḥ. In this part Dharmakīrti
explains the purpose of the prasaṅga argument. According to him, the prasaṅga argument
is set forth in order to show that a vyāpti relation between the logical reason and the property
to be proved is established.
To this explanation, however, a Nyāya-Vaiśeṣika opponent raises an objection by regarding this hPrayoga 3i as a proper proof (sādhana). If hPrayoga 3i proves the non-existence of
a single entity, i.e., a universal (sāmānya) which is taken by the Nyāya-Vaiśeṣika opponent
as occurring in several things, then the logical reason of hPrayoga 3i cannot escape the
fallacy of āśrayāsiddha, because for the Buddhist proponent the existence of the subject,
Toshikazu Watanabe
521
i.e., the universal, cannot be accepted as real existence. Or if, relying on the Nyāya-Vaiśeṣika
theory, the Buddhist proponent accepts the existence of the universal as the subject of the
thesis, then, because the consequence of this hPrayoga 3i is incompatible with the NyāyaVaiśeṣika tenets, the invalidation of the thesis would occur. To this objection, Dharmakīrti
replies as follows:
PVin 3 5,1–2: nanu tathāpy asiddhir hetoḥ pratijñāyāś cābhyupetādibādhā,
svayam abhyupagamāntarāvasthānāt. na, parīkṣākāle kasyacid anabhyupagamāt.
[Objection:] Even then, the logical reason is unestablished and the thesis is
invalidated by that which is accepted [by the proponent], etc., because [the
proponent] bases himself on a different acceptance.
[Dharmakīrti:] This is not the case. For, at the time of the critical examination
(parīkṣā) [of dogmatic ideas], any particular [dogmatic views] are not accepted
[as its basis].
According to Dharmakīrti, a prasaṅga argument is used for the critical examination of
a dogmatically accepted notion and any critical examination should be done without
relying on particular tenets. In other words, a critical examination only concerns the logical
consequence that is necessarily derived from certain characteristics postulated by the
opponent as belonging to the subject of the thesis. Therefore, in a prasaṅga argument, the
ontological status of the subject of the thesis is left out of consideration. As a result, the
purpose of a prasaṅga argument is merely to show the vyāpti relation.
PVin 3 5,7–8: nāpy asiddhyādayaḥ, yady evam idam api syān na vobhayam
iti dharmayoḥ sambandhopadarśanāt. ekāntaparigrahe syād eṣa doṣaḥ.
Moreover, there is no [fault of] the unestablished etc. because [in the prasaṅga
argument] the relationship between two properties is shown as follows: if X
were the case, then Y would also follow; or [otherwise, if the latter is not
accepted], then both [X and Y] could not exist. [But if] the firm conclusion
(ekānta) is grasped [through the prasaṅga argument], such a fault would occur.
As far as the purpose of the prasaṅga argument is restricted to show the vyāpti relation,
there is no fault of āśrayāsiddha because in this case the first characteristics of a proper
logical reason, i.e., pakṣadharmatva, is not needed. But if the prasaṅga argument is set forth
for proving a property with regard to the subject of the thesis, then since the existence of the
subject of the thesis is not accepted by the proponent himself, the fallacy of āśrayāsiddha
cannot be avoided.
In this way, although he basically follows Dignāga’s method of hypothetical assumption,
Dharmakīrti, by restricting the purpose of the prasaṅga argument to present the vyāpti relation, can successfully explain the reason why the problem of pakṣadharmatva is eliminated
in the case of the prasaṅga argument.
522
Dharmakīrti and His Successors on āśrayāsiddha and prasaṅgaviparyaya
2.4 Dharmakīrti’s theory of prasaṅga and the relationship between his three methods
for avoiding āśrayāsiddha
Just after the passage in PVin 3 cited above, Dharmakīrti seems to link the principle of the
conceptual subject or the method of paraphrase to the prasaṅga argument.
PVin 3 5,8–6,1: na vā sati hetau, yuktiprāptasyāvaśyaṃ parigrahārhatvāt. …
asati tu hetau maulasya hetor vyāpyavyāpakabhāvasādhanaprakāra eṣaḥ, na
viparyayasādhanam, hetor apramāṇatvāt.
Or no [fault would occur] when the proper logical reason exists [in the
prasaṅga argument] because what is obtained by rational reasoning is necessarily grasped [through such a logical reason]. … On the contrary, when
the proper logical reason does not exist [in the prasaṅga argument], this (i.e.,
the prasaṅga argument) is a type of [reasoning] which proves, for the original
logical reason (i.e., the reason in the hPrayoga 3i), the relationship between
what is pervaded and what pervades. [But it is] not a proof of the opposite [of
the opponent’s view] because [its] logical reason is not [established by] valid
cognition.
In the first sentence of this passage, Dharmakīrti suggests that there is a possibility of the
transformation from the prasaṅga argument to a proper proof (sādhana).
When the vyāpti relation between the logical reason and the property to be proved
is acknowledged, one can turn the reasoning into a proper proof (sādhana), if both the
proponent and the opponent accept the following two things:
1. the existence of the subject of the thesis,
2. the logical reason’s being a property of the subject.
If one of these two conditions is not fulfilled, the reasoning remains a prasaṅga argument
that does not prove anything with regard to the subject, but simply demonstrates the vyāpti
relation. But if these two conditions are fulfilled, the reasoning becomes a proper proof
(sādhana) and hence no fault of asiddha occurs.
To fulfil the first of these two characteristics, the principle of conceptual subject or the
method of paraphrase must be used. When we take hPrayoga 3i as an example, the method
of paraphrase has probably been applied. That is, when the subject of the thesis “a single
entity” (eka) is not taken to mean the universal (sāmānya) which occurs in several things,
as the Nyāya-Vaiśeṣika opponent insists, but just a single entity like a form (rūpa), then the
existence of the subject of the thesis is acceptable for both the Buddhist proponent and the
Nyāya-Vaiśeṣika opponent. Therefore, it can be said that in Dharmakīrti’s system of logic,
the principle of conceptual subject and the method of paraphrase play an important role
also for the conversion from the prasaṅga argument to a proper proof.
In this connection, the relationship between the principle of conceptual subject and
the method of paraphrase must be considered. If we compare these two, it is obvious that
the former has a wider range of application. To wit, when the proponent proclaims the
non-existence of the subject (i.e., “X does not exist”), the latter cannot be applied to the
subject because in the case of the method of paraphrase the subject “X” is paraphrased
Toshikazu Watanabe
523
by a thing that is accepted by everybody as existing externally. Since the principle of
conceptual subject alone seems enough to avoid the problem of āśrayāsiddha, why then
does Dharmakīrti add the method of paraphrase? Although Dharmakīrti does not address
this problem, it is likely that he restricts the application of the principle of conceptual
subject to negating the existence of metaphysical things such as pradhāna, etc. in the form of
“pradhāna does not exist,” using the method of paraphrase as widely as possible. Otherwise
the inference-for-others (parārthānumāna) might lose touch with external objects.
3. Dharmakīrti and his successors on prasaṅgaviparyaya and another
way to avoid the problem of āśrayāsiddha
As I have mentioned above, a prasaṅga argument is formulated by the proponent on the
basis of the opponent’s assumption with regard to some subject. Therefore, even after the
existence of the subject of the thesis is accepted, the second condition for avoiding the
problem of asiddha, i.e., the logical reason’s being a property of the subject, should not be
fulfilled. However, in the case of hPrayoga 3i the logical reason “being devoid of another
essential property which is not obstructed (avyavacchinna) by the unification (saṃsarga)
with a single substance restricted by a particular place, time and state,” i.e., “being devoid of
multiplicity” (*anekatvaviraha) is accepted as a property of the subject of the thesis, i.e., a
single entity, by both the Buddhist proponent and the Nyāya-Vaiśeṣika opponent. Therefore,
it is understood that this hPrayoga 3i is more than just a normal prasaṅga argument. That
is, this prasaṅga argument has already undergone some alterations. This is the process, I
believe, that one can understand as Dharmakīrti’s formulation of prasaṅgaviparyaya.
Let’s present the vyāpti relation of property “A” by the property “B” as “A → B,” and
describe the establishment of the vyāpti with regard to a subject “S” as “S: A → B.” Although
Dharmakīrti himself did not explain the structure of prasaṅga and prasaṅgaviparyaya, we
can describe it, following his commentators, as follows (here the sign “¬” means negation):
prasaṅga S : A → B
prasaṅgaviparyaya S : ¬B → ¬A
If this description is applied to the hPrayoga 3i, the main structure of the hPrasaṅga 3i
would be expressed as follows:
hPrayoga 3–1i eka : *anekatvaviraha → nānyena yoga
This structure of hPrayoga 3i corresponds perfectly to the prasaṅgaviparyaya described
by Dharmottara, if the expressions anyena yoga (i.e., no negation of nānyena yoga) and
anekavṛtti are regarded as having the same meaning:16
hDharmottara’s prasaṅgai sāmānya : anekavṛttitva → anekatva
hDharmottara’s prasaṅgaviparyayai sāmānya : anekatvaviraha → nānekavṛttitva
16
See Iwata 1993: 50.
524
Dharmakīrti and His Successors on āśrayāsiddha and prasaṅgaviparyaya
Moreover, Prajñākaragupta also gives a similar interpretation of hPrayoga 3i as his second
interpretation:17
hPrajñākaragupta’s prasaṅgai sāmānya [= rūpādi] : vyāpitva → naikavyaktiniṣṭhatayopalambha
hPrajñākaragupta’s prasaṅgaviparyayai sāmānya [= rūpādi] : ekavyaktiniṣṭhatayopalambha → avyāpitva
Although these two commentators differ in how they express the logical reason in the
prasaṅgaviparyay, i.e., Dharmottara uses a negative expression and Prajñākaragupta uses
an affirmative expression, the contents conveyed by both are roughly the same. Therefore, it
can be said that both commentators understand Dharmakīrti’s hPrayoga 3–1i as an example
of prasaṅgaviparyaya. When we take their interpretations into consideration, it can be seen
that Dharmakīrti’s hPrayoga 3i consists of the following prasaṅga and prasaṅgaviparyaya:
hDharmakīrti’s prasaṅgai eka : anekavṛtti → [*anekatva]
hDharmakīrti’s prasaṅgaviparyayai eka : *anekatvaviraha → nānyena yoga (= hPrayoga 3–1i)
That is, hPrayoga 3i describes the following process of reasoning:
Step 1 (prasaṅga): Following the Nyāya-Vaiśeṣika view, the property anekavṛtti
is assumed with reference to the subject eka, and from it anekatva is logically
derived although it is not stated in hPrayoga 3i.
Step 2 (prasaṅgaviparyaya): But it is not the case that this anekatva is accepted by both the Buddhist proponent and the Nyāya-Vaiśeṣika opponent,
and it should be negated because it contradicts the subject eka. Therefore,
anekatvaviraha is assumed to be a property of the subject, as is accepted
by both. From this nānyena yoga, which for the opponent is an undesired
consequence, is proved.
Despite the fact that the main structure of the hPrayoga 3i is a prasaṅgaviparyaya, which
is in fact a proper proof (sādhana), Dharmakīrti presents this hPrayoga 3i as an example
of a prasaṅga argument. Perhaps the reason for this is that he thinks prasaṅgaviparyaya
can also be called prasaṅga in that both bring undesired consequences for the opponent.
Be that as it may, forming a prasaṅgaviparyaya is a method for ensuring that the
logical reason is a property of the subject of the thesis, enabling Dharmakīrti, sometimes
in connection with the principle of conceptual subject or the method of paraphrase, to
transform a prasaṅga argument into a proper proof.
Now I would like to mention briefly how Dharmakīrti’s successors try to solve the
problem of āśrayāsiddha. In commenting on hPrayoga 3i, Dharmottara presents another
way of avoiding the problem of āśrayāsiddha. According to him, if the logical reason
consists of non-existence (abhāva), it is established in the subject of the thesis, as for
17
See Iwata 1993: 72.
Toshikazu Watanabe
525
example the universal, whose existence is not accepted by the Buddhist proponent.18
This fourth way of avoiding the problem can be described as the “method of simple
negation.”19 As earlier research has shown, this method is also adopted by Devendrabuddhi
and Śākyabuddhi in their commentaries on PV 4.136–148.20 Dharmottara presents this
view on the basis of the following statement of Dharmakīrti:
PV 1.26ab (≈ PVin 2 100,8): tasmād vaidharmyadṛṣṭānte neṣṭo ’vaśyam ihāśrayaḥ /
Therefore, a (factual) basis (āśraya) [i.e., a real locus] in the case of the
dissimilar example is not assumed to be necessary here [in the tradition that
follows Dignāga]. (Steinkellner 2004: 236)
Since Dharmakīrti’s statement is based on Dignāga’s explanation of vaidharmyadṛṣṭānta in
PS(V) 4.3bcd,21 the “method of simple negation” can be traced back to Dignāga. However,
since neither Dignāga nor Dharmakīrti mention this method in the context in question,
it might be said that in some way this method was invented by the commentators to
solve the problem of āśrayāsiddha. In contrast to Dharmottara, etc., Prajñākaragupta,
when explaining hPrayoga 3i in his commentary on PV 4.12, does not adopt this method;
instead, he adopts the method of paraphrase, that is, he paraphrases the subject of the thesis
“universal” (sāmānya) as “form, etc.” (rūpādi).22 In his commentary on PV 4.136–148, he
uses the principle of conceptual subjects and then, as a second interpretation, again employs
the method of paraphrase.23 It is thus likely that he chooses the method of paraphrase as his
final view. Therefore, as far as the problem of āśrayāsiddha is concerned, Prajñākaragupta’s
view, when compared to that of the other commentators, is more similar to Dharmakīrti’s.
4. Concluding remarks
As seen above, whereas in the NMu Dignāga avoided the problem of āśrayāsiddha by
applying the principle of conceptual subject, in the PS he avoids the same problem by
employing the method of hypothetical assumption. This raises the question: Why doesn’t
Dignāga use the strategy employed in the NMu, i.e., the principle of conceptual subjects, in
the PS as well? It is likely that the difference in his two works of how the thesis is defined
has played a role in his shift of position. The two definitions run as follows:
18
19
20
21
22
23
PVinṬ (Ms 6a2–3): sa ca vyāpakābhāvaḥ sāmānyādāv abhāve siddha eva; ibid. (Ms 7a5–6): anekatvābhāvamātraṃ cāsaty api sāmānye siddham. See Iwata 1993: 54–56, Iwata 1999: 161.
Tillemans 1999 does not distinguish this method from the “principle of conceptual subjects.” However,
as Yao (2009: 391) has pointed out, these are two different things.
Tillemans 1999: 173–174 and Iwata 1999: 160–162.
PS(V) 4.3bcd: sādhye ’sati tv asat / (4.3b) vaidharmyeṇa tv avaśyam sādhyābhāve tasyaiva hetor abhāvo
darśaitavyaḥ. paryudāso niṣedhaś ca tathā sati vilakṣaṇe //3// tathā ca sati pūrvatra paryudāsaḥ,
aparatra tu prasajyapratiṣedha ity uktam. evaṃ ca nityānabhyupagame ’pi vaidharmyadṛṣṭāntaḥ
siddhaḥ. Cf. NMu §5.1 (Katsura 1981: 63): 由是雖対不立実有太虚空等、而得顕示無有宗処無因
義成。
See Iwata 1993: 69–73.
See Tillemans 1999: 177–180, Iwata 1993: 63–81, and Iwata 1999: 163–164.
526
Dharmakīrti and His Successors on āśrayāsiddha and prasaṅgaviparyaya
NMu: tatra svayam / sādhyatvenepsitaḥ pakṣo viruddhārthānirākṛtaḥ //24
PS 3.2: svarūpeṇaiva nirdeśyaḥ svayam iṣṭo ’nirākṛtaḥ / pratyakṣārthānumānāptaprasiddhena svadharmiṇi //25
Comparing these two, one of the most significant differences is the word svadharmin
in PS 3.2d. This word seems to stipulate that, in the proponent’s own thesis, not only
the property to be proved, but also the subject of the thesis (svadharmin) should not be
opposed. Therefore, once a thesis is defined in this way, a Buddhist proponent cannot place
pseudo-entities, such as pradhāna, into the position of the subject of his own thesis, even if
he intends to prove the non-existence of such a thing.
But this does not answer the question of why Dignāga does not repeat the principle
of conceptual subjects in the PS, for he could have also defined the thesis without using
the term svadharmin. Unfortunately, no decisive clue can be found to solve this question.
However, some possible reasons can be raised.
(i) Weakness of the NMu’s argument against the Sāṅkhya’s proof
In the NMu, even though the pakṣadharmatva of the logical reason in hPrayoga 1i is
ensured by employing the principle of the conceptual subject, the Sāṅkhya does not accept
the logical reason as necessarily negating the existence of pradhāna since for the Sāṅkhya,
pradhāna is, by definition, not perceived, but is to be inferred. For this reason, Dignāga
may have thought that to negate pradhāna, it is easier to refute the Sāṅkhya’s proof than to
explain hPrayoga 1i more convincingly to the Sāṅkhya.
(ii) Avoiding unnecessary expansion of his ontological framework
In the NMu, Dignāga posits a kind of existential status on pradhāna by using the principle
of conceptual subjects. But in the PS, he provisionally accepts the Sāṅkhya’s tenet of the
existence of pradhāna (along with the vyāpti relation). Basing himself on this provisionally
accepted tenet, he points out the defectiveness of the Sāṅkhya’s proofs. He may have thought
this strategy to be more preferable because, unlike the case of the principle of conceptual
subjects, it does not need any expansion of his own ontological framework. It is likely that
one or both of these reasons made Dignāga shift his position.
Unlike in the case of Dignāga, Dharmakīrti is able to overcome these two weak points
by his new anupalabdhi theory and apoha theory. Therefore, the principle of conceptual
subjects again becomes available as a way to avoid the problem of āśrayāsiddha. He
not only inherits two methods from Dignāga, but also adds a third, i.e., the method of
24
25
NMu (Katsura 1977: 109) T. 1628 1a8–9: 是中、唯隨自意樂爲所成立説名宗。非彼相違義能
遣。“Of them (viz. thesis, reason and example), a valid thesis (pakṣa) is [precisely] one which is
intended (īpsita) by [the proponent] himself as something to be proved (sādhya) and which is not
opposed (anirākṛta) by incompatible states of affairs.” (Tr. Katsura 2004: 119.)
“[A valid thesis] is one which is intended (iṣṭā) by [the proponent] himself (svayam) as something to be
stated (nirdeśya) in its proper form alone (svarūpeṇaiva) [i.e., as something to be proved (sādhya)]. With
regard to [the proponent’s] own subject (svadharmin), it is not opposed (anirākṛta) by perceptible objects
(pratyakṣārtha), by inference (anumāna), by authorities (āpta) or by what is commonly recognized
(prasiddha).” (Tr. Tillemans 2000: 4–5.)
Toshikazu Watanabe
527
paraphrase. Then he tries to connect these three methods in the discussion of prasaṅga and
prasaṅgaviparyaya. When the prasaṅga argument is used for refuting the opponent’s view,
thanks to the method of hypothetical assumption, the problem of āśrayāsiddha does not
occur. But when the proponent wants to formulate a proper proof, i.e., prasaṅgaviparyaya,
from the prasaṅga argument, the method of paraphrase or the principle of conceptual subject
is employed in order to avoid the fallacy of āśrayāsiddha in the case of the proponent
not accepting the existence of the subject of the thesis. After that, the contrapositive
(viparyaya) of the vyāpti relation in the prasaṅga argument is made in order to avoid
the logical reason being an asiddha. As I have shown elsewhere, Dignāga has already
indicated the possibility of reformulating a prasaṅga argument into prasaṅgaviparyaya.26
It is, however, Dharmakīrti who establishes this theory by using the methods for avoiding
the problem of āśrayāsiddha.
Note
After having finished my paper, I received Yoshimizu Chizuko’s article “Dharmakīrti’s
Statement of Consequence (prasaṅga) in the Third Chapter of the Pramāṇaviniścaya”
(Yoshimizu 2016). She takes, contrary to my understanding, the hPrayoga 3i to be a
prasaṅga argument. The difference between our interpretations comes from how we understand the role of anekavṛtti in the hPrayoga 3i. While Prof. Yoshimizu takes both eka
and anekavṛtti as being properties of the subject of the thesis (Yoshimizu 2016: 1248), I
understand eka as the subject of the hPrayoga 3i and regard anekavṛtti as just a subsidiary
element of eka that has been added to explain the opponent’s assumption. In the passage
discussed above, anekavṛtti is presented in apposition to eka. However, as Prof. Yoshimizu
has also pointed out (Yoshimizu 2016: 1253, note 5), Dharmakīrti refers to almost the same
argument with hPrayoga 3i in the subsequent part without referring anekavṛtti and gives
only eka as its subject. See PVin 3 6,6–7: ekasya tu yathoktasvabhāvāntaravirahopagamād
eva bhinnadeśādibhir yogābhāvaḥ.
References and abbreviations
D Derge edition of Tibetan Tripiṭaka.
Iwata 1993 T. Iwata, Prasaṅga und Prasaṅgaviparyaya bei Dharmakīrti und seinen Kommentatoren. Vienna 1993.
Iwata 1999 T. Iwata, On the interpretations of the subject (dharmin) of the inference
negating invariable entities in Dharmakīrtian logic. In: Dharmakīrti’s Thought and
Its Impact on Indian and Tibetan Philosophy, ed. S. Katsura. Vienna 1999, 156–172.
K PS(V) translated by Kaṇakavarman and Dad pa’i shes rab.
Kataoka 2010 K. Kataoka, Mittsu no apoha setsu: Dharmottara ni itaru model no hensen.
South Asian Classical Studies 5 (2010) 251–284.
Katsura 1977 S. Katsura, Inmyōshōrimonron Kenkyū (1). Hiroshima Daigaku Bungakubu
Kiyō 37 (1977) 106–126.
26
See Watanabe 2013.
528
Dharmakīrti and His Successors on āśrayāsiddha and prasaṅgaviparyaya
Katsura 1978 S. Katsura, Inmyōshōrimonron Kenkyū (2). Hiroshima Daigaku Bungakubu
Kiyō 38 (1978) 110–130.
Katsura 1981 S. Katsura, Inmyōshōrimonron Kenkyū (4). Hiroshima Daigaku Bungakubu
Kiyō 41 (1981) 62–82.
Katsura 2004 S. Katsura, Pakṣa, Sapakṣa and Asapakṣa in Dignāga’s Logic. Hōrin 11
(2004) 119–128.
Katsura 2009 S. Katsura, Rediscovering Dignāga through Jinendrabuddhi. In: Sanskrit
manuscripts in China, ed. E. Steinkellner. Beijing 2009, 153–166.
Kitagawa 1965 H. Kitagawa, Indo koten ronrigaku no kenkyū: Jinna (Dignāga) no taikei.
Tokyo 1965.
NMu Nyāyamukha. See Katsura 1977, 1978 and 1981.
P Peking edition of Tibetan Tripiṭaka.
PS(V) Pramāṇasamuccaya(vṛtti). See n. 6.
PSṬ Jinendrabuddhi’s Pramāṇasamuccayaṭīkā. See n. 6.
PV 1 Pramāṇavārttika, chapter 1, ed. R. Gnoli. Roma 1960.
PV 4 Pramāṇavārttika, chapter 4. See Tillemans 2000.
PVin 3 Pramāṇaviniścaya, chapter 3, ed. P. Hugon and T. Tomabechi. Beijing/Vienna
2011.
PVinṬ Dharmottara’s Pramāṇaviniścayaṭīkā.
PVSV See PV 1.
Steinkellner 2004 E. Steinkellner, The Early Dharmakīrti on the Purpose of Example. In:
The role of the example (dṛṣṭānta) in classical Indian logic, ed. S. Katsura and E.
Steinkellner. Vienna 2004, 225–250.
T Taishō Shinshū Daizōkyō. Tokyo 1924–1932.
Tillemans 1999 T. J. F. Tillemans, Scripture, Logic, Language: Essays on Dharmakīrti
and his Tibetan Successors. Boston 1999.
Tillemans 2000 T. J. F. Tillemans, Dharmakīrti’s Pramāṇavārttika: An annotated translation of the fourth chapter, vol. 1 (kk. 1–148). Vienna 2000.
V PS(V) translated by Vasudararakṣita and Seng rgyal.
Watanabe 2013 T. Watanabe, Dignāga on Āvīta and Prasaṅga. Journal of Indian and
Buddhist Studies 61:3 (2013) 1229–1235.
Yao 2009 Z. Yao, Empty Subject Terms in Buddhist Logic: Dignāga and his Chinese
Commentators. Journal of Indian Philosophy 37:4 (2009) 383–398.
Yoshimizu 2013 Ch. Yoshimizu, Reasoning-for-others in Candrakīrti’s Madhyamaka
thought. Journal of the International Association of Buddhist Studies 35:1–2 (2013)
413–444.
Yoshimizu 2016 Ch. Yoshimizu, Dharmakīrti’s Statement of Consequence (prasaṅga) in
the Third Chapter of the Pramāṇaviniścaya. Journal of Indian and Buddhist Studies
64:3 (2016) 1246–1254.
Another Look at avinābhāva and niyama in Kumārila’s
Exegetical Works
by
Kiyotaka Yoshimizu
Introduction
In one of my recent papers (K. Yoshimizu 2007a) I argued that in a fragment quoted by
Karṇakagomin from the Bṛhaṭṭīkā (BṬ), which I call the ‘avinābhāva fragment,’1 Kumārila
(ca. 560–620)2 states that avinābhāva, the inseparability of the reason (hetu) from the thing
to be inferred (sādhya), is not sufficient for a valid inference because it would even justify
an inference by asādhāraṇahetu, which has no positive examples. In another fragment
from the Bṛhaṭṭīkā, which I call the ‘niyama fragment,’3 Kumārila designates niyama,
natural restriction, as the foundation of inference, illustrating this by means of causality
(kāryakāraṇabhāva).
In the avinābhāva fragment, however, Kumārila does not deny the necessity of avinābhāva for a valid inference. In the niyama fragment, avinābhāva is considered to have
a sound footing in reality by means of niyama, which means that in the real world the
existence of one thing is spatiotemporally determined by another thing, and therefore the
existence of the latter is invariably inferred from the confirmation of the former’s existence. Inasmuch as the inseparability of one thing from another is considered to require
substantiation in reality, it follows that this inseparability is reflected in the consciousness
1
2
3
Pramāṇavārttikasvavṛttiṭīkā (PVSVṬ) 87,12–17: avinābhāvaśabdo ’py a[*darśanāt] sakalārthabhāk /
nānumāyogyasambandhapratipattiṃ karoti naḥ // yadi tāvad vinābhāvo na sa paścād viśiṣyate / tato
’sādhāraṇe ’py asti sa iti syād akāraṇaṃ // yo hy asādhāraṇo dharmaḥ sa tenaivātmasātkṛtaḥ / vinā na
bhavatīty eva jñāto hetuḥ prasajyate // For an English translation, see K. Yoshimizu 2007a: 1080.
Franco (2019) convincingly demonstrates that Dharmakīrti’s dates, which were assumed by Frauwallner
as ca. 600–660, cannot be moved back to the sixth century. I surmise that Kumārila passed away
about 620 CE (K. Yoshimizu 2015a: n. 1), before coming to know about young Dharmakīrti, whereas
Dharmakīrti wrote his first work, the first chapter of the PV refuting the Mīmāṃsā view of the intrinsic
relation between a word and its meaning, by thoroughly remodeling the idea of Kumārila advocated in
the BṬ that niyama is the foundation of avinābhāva. Because Dharmakīrti made use of this idea in order
to attack the Mīmāṃsā theory of word and meaning, we may rightly assume as follows: If Kumārila, a
hard-core Mīmāṃsaka, had first come to be informed of this idea through Dharmakīrti, he would have
held it too provocative to be adopted into his own philosophical system, contrary to the assumption of
Frauwallner (1962). See the appendix of the present article. Moreover, showing that PV 1.335cd is a
quotation from the BṬ, Franco (2019: 132–137) corroborates young Dharmakīrti’s acquaintance with
the BṬ. Following Frauwallner’s unacceptable assumption of the relationship between Kumārila and
Dharmakīrti, Kataoka (2011: 47–60) errs in assuming Kumārila’s date as ca. 600–650.
PVSVṬ 87,21–30: evam anyoktasaṃbandhapratyākhyāne kṛte sati / niyamo nāma saṃbandhaḥ svamatenocyate ’dhunā // kāryakāraṇabhāvādisaṃbandhānāṃ dvayī gatiḥ / niyamāniyamābhyāṃ syān niyamasyānumāṅgatā // sarve ’py aniyamā hy ete nānumotpattikāraṇam / niyamāt kevalād evaṃ na kiṃcin
nānumīyate // tasmān niyama evaikaḥ saṃbandho ’trāvadhāryate / gamakasyaiva gamyena sa ceṣṭaḥ
prāṅ nirūpitaḥ // niyamasmarataḥ samyag niyamyaikāṅgadarśanāt / niyāmakāṅgavijñānam anumānaṃ
tadaṅgiṣu // For an English translation, see Steinkellner 1997: n. 28.
Birgit Kellner et al., eds., Reverberations of Dharmakīrti’s Philosophy: Proceedings of the Fifth International Dharmakīrti
Conference Heidelberg, August 26 to 30, 2014. Vienna 2020, pp. 529–547.
530
Another Look at avinābhāva and niyama in Kumārila’s Exegetical Works
of human beings. Therefore, the avinābhāva in the BṬ might rightly be called ‘epistemic
inseparability’ in distinction from physical inseparability. In the present paper, I would like
to elucidate how, before composing the BṬ, Kumārila attempted to substantiate avinābhāva
through niyama4 by collecting from the Tantravārttika (TV) as many cases as possible
in which he presents the invariable transition from one concept to another on the basis
of the relation between two things in reality. In other words, I would like to reinforce the
conclusion of another paper of mine (K. Yoshimizu 2011), that Kumārila was not indebted
to Dharmakīrti for this idea, as assumed by Frauwallner (1962).
I. The conceptual transition modeled upon causality
1.1 Authorization of customs by assuming the existence of Vedic injunctions
In Mīmāṃsā, it is required that all provisions of man-made (pauruṣeya) scripture (smṛti)
as well as traditional customs (sadācāra), if they pertain to religious duties (dharma), be
based on the Veda. If the Vedic corpus one has inherited has no injunction that can be
deemed as testifying these smṛti or customs, then it is, unlike in modern philology, allowed
to assume the existence of corresponding Vedic injunctions in one way or another,5 insofar
as they are sanctioned by learned people (śiṣṭa), namely, by those who have received the
cultural inheritance of the Vedic tradition.6 The popular festivals celebrated by the common
peoples of different regions belong to those Mīmāṃsā topics where authority was discussed.
The opponent in Śabarasvāmin’s Bhāṣya (ŚBh) lists the names of three popular festivals
held in east, south, and north India7 and states that the authority of the Vedic injunctions
assumed to prescribe these festivals is limited to the respective regions, in the same way
4
5
6
7
For the chronological order of Kumārila’s works, see K. Yoshimizu 2007b: 213–219. We can outline the
change of Kumārila’s attitude towards Dignāga’s theory of logic as follows: In his early years, Kumārila
adopted it as a theory more excellent than those advocated in Brahmin schools. He proudly demonstrated
his proficiency in this innovative logic in the Anumānapariccheda of the Ślokavārttika (ŚV), which
holds formal logic to be neutral and separate from the religious conflict between Brahmanism and
Buddhism. In his middle years, however, having become active as a leading Mīmāṃsaka and defending
orthodox Brahmanism, in the Smṛtipāda of the TV he criticized Buddhism as a heresy (cf. K. Yoshimizu
2015a). In his later years, Kumārila finally directed his attack to Dignāga’s theory of logic, and in the
BṬ even attempted to find fault with “the three characteristics for a valid reason” (hetutrairūpya), the
basis of this logic, as I argue in K. Yoshimizu 2007a: 1084–1096.
In early Mīmāṃsā, this problem was solved by assuming that the Vedic text containing the needed
injunctions had disappeared. Kumārila, however, rejected this view in order to prevent heretics from
making use of this excuse to claim orthodoxy. Instead, he proposed assuming that a corroborating
injunction was preserved somewhere in other branches of the Veda. Cf. K. Yoshimizu 2012a: 650–654.
Cf. K. Yoshimizu 2015a, section III.
ŚBh 243,5–244,2: anumānāt smṛter ācārāṇāṃ ca prāmāṇyam iṣyate. yenaiva hetunā te pramāṇam,
tenaiva vyavasthitāḥ prāmāṇyam arhanti. tasmād holākāda[ya]ḥ prācyair eva kartavyāḥ, āhnīnaibukādayo dākṣiṇātyair eva, udvṛṣabhayajñādaya udīcyair eva. “Man-made scripture and traditional
customs are regarded as authoritative on the basis of the assumption [of corroborating injunctions of the
Veda]. On account of this very reason to make them authoritative, they (i.e., traditional customs) have
authority only in respective (regions). Therefore, Holākā, etc., are to be performed only by easterners,
Āhnīnaibuka, etc., only by southerners, and Udvṛṣabhayajña, etc., only by northerners.”
Kiyotaka Yoshimizu
531
as the authority of the Vedic injunctions assumed to prescribe the number of topknots
arranged on one’s head is limited to one’s own kinship group.8
1.2 Difference in authorization between regional festivals and the customs of kinship
group
Despite its conservative position, the Mīmāṃsā school never considers the Veda as forcing
human beings to comply with its injunctions unconditionally. Vedic injunction urges
only those who satisfy appropriate conditions (nimitta) of entitlement to perform duties.
With regard to these conditions, however, one’s living place does not matter at all if
one lives in the so-called “homeland of Aryans” (āryāvarta). No matter where someone
was born among the four geographical quarters, they are not endowed with a generic
property (jāti) distinguishing them from people who live in other quarters.9 The generic
properties recognized as real, at least by Mīmāṃsakas, such as humanness (manuṣyatva)
and Brahmin-ness (brāhmaṇatva) never change regardless of where one lives, even if one
moves to another country.10 Therefore, even if Vedic injunctions existed that were deemed
as testifying popular festivals, their authority could not be limited to a particular quarter
because one’s living place has no effect on one’s social affiliation.
In contrast, regarding the traditional customs of kinship groups, Kumārila adduces some
examples of ritual details prescribed in the Śrautasūtras specifically for people of particular
kinship groups, adding that it is actually observed that such rituals are performed by these
groups.11 On the evidence of these explicit regulations given in ritual tradition, Kumārila
8
9
10
11
ŚBh 244,2–3: yathā śikhākalpo vyavatiṣṭhate, kecit triśikhāḥ, kecit pañcaśikhā iti. “Just as [the number
of] topknots one arranges [on one’s head] is regulated in respective (families), for example, some
(families) keep three, some five.” For the śikhā to be arranged at the cūḍākaraṇa for the children of a
twice-born (dvija) family, see Pandey 2006: 94–101.
TV 246,10–11: na tāvat prācyatvadākṣiṇātyatvādijātiḥ pratīcyodīcyādivyaktivyāvṛttā sarvaprācyādivyaktiṣv anugatā kācid upapadyate, yadvacanam upapadaṃ holākādyadhikāraṃ viśiṃṣyāt. “First,
easterners or southerners do not have any generic properties which would be separate from every
westerner or every northerner and inhere in every easterner [or in every southerner], and which would
be denoted by an accompanying word, if any, that qualifies the entitlement to Holākā, etc.”
TV 246,12–13: yās tu manuṣyatvabrāhmaṇatvādijātayaḥ teṣu vidyante tāḥ sarvadeśavāsivyaktiṣv
aviśiṣṭā iti nācārānurūpaviśeṣaṇasamarthatvena jñāyante. “On the other hand, the generic properties
found in them, such as humanness, Brahmin-ness, and so on, are equally inherent in all individuals who
live in any region. Therefore, no [generic properties] are recognized as able to qualify [the assumed
Vedic injunction] to correspond to the [localized] customs.”
TV 248,23–25: yathaiva pañcāvattaṃ tu bhṛgūṇāṃ, vasiṣṭha-śunaka-atri-vadhryaśva-kāṇva-saṃkṛtirājanyānām “nārāśaṃso dvitīyaḥ prayājas tanūnapād anyeṣām” ity anvayato vyatirekataś copalakṣaṇasaṃbhavād vyavasthitavidhyavasānaṃ. “For example, the pañcāvatta is prescribed for the members
of the Bhṛgu family.* For the members of the families of Vasiṣṭha, Śunaka, Atri, Vadhryaśva, Kāṇva,
Saṃkṛti, and royal families, the second prayāja [verse that begins with] narāśaṃsa is prescribed
whereas that [which begins with] tanūnapāt is prescribed for the others.** As we observe [these details
performed] in this way through association (anvaya) [of practice with prescription] and dissociation
(vyatireka) [of practice with the absence of prescription], we conclude that there are [Vedic] injunctions
differentiated [by these kinship groups].” *Among the offerings in the new- and full-moon sacrifices,
the ājyabhāga (ghee-portion) scooped four times should be offered for Agni and Soma, respectively. But
in the case of the sacrifices of the Jamadagni family, the portion scooped five times (pañcāvatta) should
be offered. Cf. ĀpŚS 2.18.1; BhārŚS 2.17.7; HirŚS 2.2: 190; MānŚS 1.3.2.5. Jamadagni is compared to
Bhṛgu in Jaiminīyabrāhmaṇa (JB) 1.152. Cf. Bodewitz 1990: 85 and 256, n. 22; Atharvavedasaṃhitā
532
Another Look at avinābhāva and niyama in Kumārila’s Exegetical Works
concludes that with regard to traditional customs, if one observes a life-style restricted
(niyata) to a caste or Brahmin lineage (pratijātigotra), then one can infer the existence of
Vedic injunctions restricting each life-style to the corresponding kinship group.12
This argument reveals Mīmāṃsaka’s conservative view that one’s birth into a particular
kinship group is just as deeply engraved in one’s identity as the generic properties of a
natural species. This crystallization of one’s inborn social affiliation leads to the assumption
that the life-style of Aryan kinship groups is regulated by the Veda, which has no regional
difference in validity, just as the characteristics and the behaviors of natural species are
regulated by their nature and not by where they live. Considering that this argument is
modeled upon the laws of nature, Kumārila refers to a sort of rationale (nyāya)13 found
in causality, namely that effect is compliant to its cause (kāraṇānuvidhāyikāryanyāya),
and uses this rationale to infer that when observing an effect it must have a conformable
(anurūpa) cause (avaśyam upalabhyamānakāryānurūpakāraṇānumānair bhavitavyam).14
In this statement, the laws of nature regulating the behavior of individual entities through
their generic properties are said to have universal validity because generic properties are
not affected by the place where these entities are located or when they exist. Moreover,
the effect (kārya) is said to occur passively as regulated by the cause (kāraṇa); the effect
becomes the logical reason that actively leads to the existence of the cause.15 This reversal
12
13
14
15
(AV) 5.19.1. In the Mahābhārata (MBh 3.115.28), Jamadagni is said to have been born in the clan
of Bhṛgu (bhārgava). ** The hymn to be recited at the eleven fore-offerings (prayāja) of the animal
sacrifice is called Āprī. The most generally used Āprī hymn is RV 10.110, whose second verse begins
with tánūnapāt. Instead of RV 10.110.2, however, the Vasiṣṭa and related families used RV 7.2.2, which
begins with nárāśáṃsa, as the second verse of the Āprī hymn. Cf. ŚāṅkhŚS 1.7.2: tanūnapād agna
ājyasya vetv iti dvitīyaḥ; 3: narāśaṃso agna ājyasya vetv iti dvitīyo vasiṣṭhaśunakānām atrivadhryaśvānāṃ kaṇvasaṃkṛtīnāṃ rājanyānāṃ prajākāmānāṃ ca; ĀśvŚS 1.5.21–22: tanūnapād agna ājyasya vetv
iti dvitīyo ’nyatra vasiṣṭha-śunaka-atri-vadhryaśva-rājanyebhyaḥ; 22: narāśaṃso agna ājyasya vetv
iti teṣām; Schwab 1886: 90–91. Kumārila’s remark is much closer to ŚāṅkhŚS than to ĀśvŚS, since
ŚāṅkhŚS mentions the Kāṇva and the Saṃkṛti families. RV 10.110 is a hymn of Jamadagni, and RV 7
is the collection of Vasiṣṭha’s hymns. See Th. Aufrecht’s “Verzeichnis der angeblichen Hymnendichter
gemäß der Anukramaṇikā” appended to his edition of the RV.
TV 248,25–26: tathaiva pratijātigotraniyatatriśikhaikaśikhādikalpavyavasthitavidhiviśeṣānumānopapattir astīti. “In the same way, from the fact that whether one arranges three topknots or one topknot is
restricted (niyata) to each caste and each Brahmin lineage (pratijātigotra), it is appropriate to infer the
existence of a particular injunction differentiated [by the respective kinship group].”
The first of the proponent sūtras (Mīmāṃsāsūtra [MmS] 1.3.16: api vā sarvadharmaḥ syāt, tannyāyatvād
vidhānasya) asserts that a Vedic injunction, if it exists, must be a norm issued to all people (sarvadharma)
irrespective of where they live. To assert this, however, this sūtra merely adds the tautological reason that
it is reasonable (tannyāyatva) to consider that Vedic injunction works (vidhāna) in this way. Kumārila
uses the reason of causality to avoid this tautology.
TV 245,14–15: iha smṛtīr ācārāṃś copalabhya mūlaśrutiṣv anumīyamānāsu kāraṇānuvidhāyikāryanyāyenāvaśyam upalabhyamānakāryānurūpakāraṇānumānair bhavitavyam. “When one, having observed
some man-made scriptures and customs in this world, attempts to assume the corroborating injunctions
[of the Veda], one should invariably infer the cause that is conformable to the effect one has observed,
following the rationale that the effect is compliant to its cause.”
On another occasion, Kumārila remarks that unlike effect, cause does not work as an indicator (gamaka)
because in many cases it is inconclusive (vyabhicārabahutva), namely, when auxiliary conditions are
not fulfilled, it fails to indicate the occurrence of its effect. TV 544,20–21: kāryakāraṇayor api ca na
tathā kāraṇaṃ gamakaṃ vyabhicārabahutvād yathā kāryam. This remark is made immediately after
the argument that the inference from kṛtakatva to anitytva is not based upon causality. Cf. n. 41.
Kiyotaka Yoshimizu
533
of direction from cause to effect in reality and from effect to cause in cognition anticipates
the formulation given by Kumārila in the niyama fragment from the BṬ16 and reminds us
that Dharmakīrti assigns only the effect, not the cause, as the logical reason based on the
causality between a cause and its effect (PVSV 3,11–16 on PV 1.2abc).
II. The conceptual transition from species to genus
2.1 Various relations in reality between two epistemically inseparable things
In the TV we find various cases in which Kumārila assumes epistemic inseparability (avinābhāva) between two things. Illustrating the relation that makes one thing epistemically
inseparable from another, Kumārila adduces the relation between cause and effect (kāryakāraṇa), between owner and owned (svasvāmin), and the accompaniment (sahacarabhāva) of
two particular things.17 To illustrate the last, Kumārila says that if the constellation kṛttikā
appears in the night sky, the rohiṇī will soon appear.18 As regards the relation between
owner and owned, this might be considered a typical example of the relation between a
substrate and its properties, both essential ones and those that are accidental.19
16
17
18
19
PVSVṬ 87,29–30: niyamasmarataḥ samyag niyamyaikāṅgadarśanāt / niyāmakāṅgavijñānam anumānaṃ tadaṅgiṣu // “If one who correctly remembers the restriction sees one member [of the restrictionrelation], (namely) the restricted (member) (niyamya), the resultant cognition of the (other) member
which is the restricting (niyāmaka) is an inference in regard to such (entities) that possess these (two)
members [i.e. pakṣas].” (Tr. Steinkellner 1997: 634, n. 28.)
TV 139,13–14: atha vodāhṛtaviṣayahetulakṣaṇam etat*. avinābhāvo hy aneka-kāryakāraṇa-svasvāmisahacarabhāvādi-prabhedabhinnaḥ. “Or, this [merely] characterizes the logical reason given in the
quoted ‘example sentence of exegesis’ (viṣaya). For, the epistemic inseparability differentiates according
to the difference of the relation between cause and effect, that between owner and owned, the accompaniment, and so on.” Cf. Kanazawa 1983: 933; Harikai 2006: 307, n. 10. In view of the context of the
section in question (MmS 1.2.26–30), this portion of the TV is included in the part of the pūrvapakṣa
that construes śūrpeṇa juhoti, tena hy annaṃ kriyate as a reason justifying the use of a winnowing
basket (śūrpa) for an offering. For the contents of this section, see Harikai 2006. * This etat refers to the
kāryakāraṇabhāva in ŚBh 139,1: nanv aprasiddhe kāryakāraṇabhāve na hetūpadeśaḥ. “[Some people]
may contend that the presentation of a logical reason is made impossible unless a causal relationship is
established [between the reason and the thing to be proved].”
TV 139,8–9: tat* tv ayuktam. akāryakāraṇabhūtānām api kṛttikādīnām acirodgatarohiṇyādipratipattihetutvadarśanāt. Cf. ŚV, Anumānapariccheda, vv. 12d–13ab: kva cic cāpi viśeṣayoḥ // kṛttikodayam
ālakṣya rohiṇyāsattikḷptivat / *This tat refers to TV 139,7–8: kecid āhuḥ. kāryakāraṇayor evānumānam.
“For example, [hearing] that he has a vajra makes one infer that [the deity concerned is] Indra” (ṬṬ
1614,25–1615,15: yathā vajrītīndram anumāpayati) is used by Kumārila as a simple example of the
exegetical inference from a qualifier (viśeṣaṇa) mentioned in a text to the sacrifice qualified (viśeṣya) by
it. ṬṬ 1614,24–25: ahaḥśabdo yāgaviśeṣavacano ’śrutam api viśeṣyam anumāpayati. nedaṃ viśeṣaṇaṃ
yāgam antareṇa saṃbhavati. “The word ‘day,’ which expresses a particular sacrifice [in the text
concerned, that is, Pañcaviṃśabrāhmaṇa (PB) 22.9.1*], makes one infer the [sacrifice] qualified [by
it] even if [the sacrifice] is not directly expressed. This qualifier cannot occur [in the text] without the
sacrifice.” * Cf. PB 22.9.1: catvāri trivṛnty ahāny agniṣṭomamukhāni viśvajin mahāvrataṃ jyotiṣṭomo
’tirātraḥ “Four nine-versed (trivṛt) days, the first of which is an agniṣṭoma (the three others being
ukthyas); a Viśvajit-day; a mahāvrata-day, (and) a jyotiṣṭoma as over-night-rite.” (Tr. Caland 1982:
578.) This is the breakdown of a seven-day-sacrifice called janakasaptarātra (cf. Caland 1982: 578,
n. 1). Moreover, Kumārila considers the restriction of parts by the whole (samudāya) as the basis for
the conceptual transition from a piece of text to the entire text by which it is restricted (niyamya), in
534
Another Look at avinābhāva and niyama in Kumārila’s Exegetical Works
Moreover, Kumārila considers avinābhāva to include any kind of necessary condition
when he distinguishes avinābhāva from the relationship between the subordinate means
and the primary purpose (śeṣa-śeṣi-bhāva), one of the Mīmāṃsā topics that is exegetically determined from ritual texts.20 For example, the study of the Veda (adhyayana), the
installation (ādhāna) of sacred fires, the cultivation (kṛṣi) of crops for oblations, and the
earning of money (dravyārjana) for rewards to be paid to priests are all necessary for the
performance of sacrifices (kratu). But although a sacrifice cannot be performed without
them (vinā-asaṃbhava), they are not exegetically subordinate (śeṣa) to a sacrifice because
they are not prescribed in the text of the sacrifice.21
As a further instance of a necessary condition, Kumārila applies avinābhāva to his
theory of human action and Vedic injunction. In the Mīmāṃsā position, the ending of
a finite verb (tiṄ) denotes the general form of intentional action called “bringing into
being” (bhāvanā).22 Fully aware of the Pāṇinian rule that tiṄ in the active voice and the
middle voice denotes the agent (kartṛ) of an action,23 Kumārila maintains that the agent is
invariably cognized from bhāvanā on account of avinābhāva because if there is no agent,
no action can be performed.24
20
21
22
23
24
other words, to which it belongs. See n. 47. In addition, we may also classify the inference from an
accidental property (guṇa) to its substrate into this type. Kumārila offers the example that one can infer
the existence of a substance (dravya) from the cognition of color. Cf. ṬṬ 1368,15–18: yathā śuklādiguṇo
dravyeṇa vinā na bhavatīty avinābhāvād dravyaṃ lakṣayati, naivam iha; Kanazawa 1983: 932.
TV 654,10–15: ayuktam evaṃ śeṣatvam avinābhāvalakṣaṇam / vyabhicārāt tathāhīdam aśeṣeṣv api
dṛśyate // sarvadā hy avinābhūtā rūpasparśādayaḥ kṣitau / na caiṣāṃ tulyakalpatvād bhavaty anyonyaśeṣatā // svāminā ca vinā dāsā na bhavanti kadācana / tathā gṛhādayas teṣāṃ śeṣaḥ svāmī ca
neṣyate // “It is not right to say that subordination is defined as epistemic inseparability on account
of inconclusiveness, for we find this also in those cases that are not subordinate. As a matter of fact,
color, touch and other [qualities of elements] are always inseparable from one another in earth, but
among them, there is no subordination because they are equally arranged [in earth]. Without a master,
moreover, there can never be a slave, a house, and so on, but the master is not deemed to be subordinate
to them.” Cf. Kanazawa 1983: 932–933.
TV 654,20–21: etenādhyayanādhānakṛṣidravyārjanādiṣu // vācyaṃ prasaṅgiśeṣa(IO 2157, 6b6: śeṣa;
Ānandāśrama ed.: śeṣi)tvaṃ tair vinā kratvasaṃbhavāt / “By this (i.e., assuming avinābhāva to be
the definition of subordination), one would have to say that Vedic study, fire installation, cultivation,
earning money, and so on, come to be subordinate because without them the performance of a sacrifice
is impossible.” Moreover, Kumārila finds inseparability in the hierarchy of social order, saying that
when a king is invited somewhere, all the subordinates who are inseparable from him must follow him.
TV 486,23–26: loke ’pi ca yat pradhānaṃ rājaprabhṛti kvacin nīyate tadātmanāvinābhūtāny amātyādīny aṅgāny apy aśeṣāṇy ākṣipati. yadā tu kaścid bhṛtyānāṃ kvacid gacchati tadā tadasaṃbandhīni
bhṛtyāntarāṇi padam api na calanti. “Even in this world, a chief person, for example, a king, when
invited to some place, mobilizes all the subordinate people who are inseparable from him, for example,
ministers, etc. On the other hand, when one of the servants goes somewhere, the others do not take
even a single step because they are not connected to that (servant).”
For the reason why Mīmāṃsakas insist on denoting bhāvanā through verbal endings, in defiance of the
Pāṇinian rule, see K. Yoshimizu 2012b: 555–557.
A 3.4.67: kartari kṛt; 69: laḥ karmaṇi ca bhāve cākarmakebhyaḥ.
TV 914,1–2: yādṛśaś ca guṇabhūtaḥ kartātrāvagamyate na tādṛśena vinā bhāvanopapadyata ityarthāpattyānumānena vā śaktā gamayitum. “No matter in whichever manner an agent that is subordinate
[to the action] is cognized in this (finite verb), bhāvanā is impossible without an agent. In this way,
[bhāvanā] is able to make [an agent] known through logical derivation or through inference.”
Kiyotaka Yoshimizu
535
2.2 The inseparability of species from generic properties, and causality as its ontological background
Whereas an agent is indispensable for any action, action in many cases requires other
kārakas in addition to agent. According to Kumārila, bhāvanā can bear relations with
kārakas other than an agent without breaking the relation with the agent. To illustrate this,
he refers to the following simile: When one sees a tree and notices that it is a siṃśapā, one
can also optionally realize the same object in other ways – as a tree (vṛkṣa), an earthen thing
(pārthiva), a substance (dravya), or a being (sat) – without denying that it is a siṃśapā.25
This is because the specific property of a śiṃśapā, śiṃśapātva, exists only in an entity in
which generic properties of its genus such as tree-ness and so on invariably exist. Kumārila
expresses the inseparability (avinābhāva) of a species from each level of its genus by saying
that siṃśapātva is not inconclusive (na … anaikāntikatvam) in the sense that it never loses
its ability (pratyāyanaśakti) to make the existence of the generic properties known.26
Owing to this inseparability, śiṃśapātva can be compared to bhāvanā in the theory of
action. Upon hearing a finite verb, one may comprehend many things, such as bhāvanā,
action, agent, time and other ideas.27 If the verbal root or the tense changes, the action or the
time also changes; if the verbal ending remains the same, the comprehension with regard
to the agent remains the same. According to Kumārila, however, we do not have to assume
the agent to be the meaning of the verbal ending because, once bhāvanā is confirmed, it is
invariably known that there is an agent. This can be explained by analogy with deriving
the concepts of genus from the names of species; that is, we do not have to assume that
the word śiṃśapā denotes any generic properties of a genus, such as tree-ness, because a
tree is invariably known to be referred to when an object is called śiṃśapā, if it is accepted
that the word śiṃśapā denotes only śiṃśapātva.28 According to Kumārila, śiṃśapātva is
25
26
27
28
TV 914,7–8: yas tv anyatrāpi vartamānaḥ pūrvāvagatasaṃbandhyaparityāgenaiva vartate yathā śiṃśapātvaṃ vṛkṣatvapārthivatvadravyatvasattveṣu. “On the other hand, there is a case in which a thing does
not renounce the formerly cognized relation with another thing, even if it has come into the relation
with something else. For example, [when an object is cognized as śiṃśapā] the śiṃśapā-ness remains
[in one’s cognition] even if [the object is cognized as having] the generic properties of tree, earthen
thing, substance, and being.”
TV 914,8: na tasyānaikāntikatvaṃ, sarveṣu pratyāyanaśaktyavighātāt. “It (i.e., siṃśapātva) is not
inconclusive in the sense that it never loses its ability to make [the existence of] all of them known.”
TV 913,16–18: tad iha pacatyādiśabdoccāraṇād iyanto ’rthā gamyante, bhāvanādhātvarthakartṛtatsaṃkhyāpuruṣopagrahakālaviśeṣāḥ. teṣāṃ tu kaḥ śābdaḥ ko ’rthād iti vibhāgo na jñāyate. lakṣaṇaṃ
caitāvad yo ’rthān na gamyte sa śabdārtha iti. “Now, when hearing the word pacati, etc., pronounced,
one cognizes so many things, such as bhāvanā, the meaning of the verbal root (i.e., an action), agent,
its number, the grammatical person, grammatical voice, and a certain time. Among them, however, it
is not yet ascertained which one is known directly from the word and which one is from the meaning
[of the word], and the meaning of a word is merely defined as that which cannot be cognized through
anything else that is assumed to be the meaning [of that word].” Cf. Ogawa 1993: 1052–1053.
TV 940,12–14: śiṃśapāśabdo hi na tāvad vṛkṣatvapārthivatvadravyatvasattāprameyatvajñeyatvānām
abhidhāyako ’tha ca tadviśeṣam eva śiṃśapātvam(*-pātvam; IO 2158, 39(152) b2 & Ānandāśrama ed.:
-pām) abhidhatte. “For the word śiṃśapā does not denote tree-ness, nor earthen-ness, nor substanceness, nor being-ness, nor the property of being correctly known, nor the property of being known, but it
denotes only their particular form, namely, śiṃśapā[-ness].” Cf. TV 982,10–13; TV 1048,15–16.
536
Another Look at avinābhāva and niyama in Kumārila’s Exegetical Works
inseparable from tree-ness because śiṃśapātva is restricted by tree-ness (vṛkṣatvaniyata)
in its nature.29
Here, Kumārila seems to refer to the hierarchical relationship between two generic
properties, taking into account the Apoha chapter of the Pramāṇasamuccaya (PS)30 as well
as Patañjali’s Mahābhāṣya. Aṣṭādhyāyī (A) 2.2.30, upasarjanam pūrvam, lays down that
the subordinate item (upasarjana) forms the first part of a compound. Anticipating this
rule, A 2.1.57, viśeṣaṇaṃ viśeṣyena bahulam, lays down that whether a qualifier (viśeṣaṇa)
and the item qualified (viśeṣya) do or do not form a tatpuruṣa compound (2.1.22) including
karmadhāraya is determined according to the cases (bahulam).31 In his commentary on this
sūtra, Patañjali comments that the compound kṛṣṇatila (black sesame) has kṛṣṇa before tila
because kṛṣṇa is subservient to tila; kṛṣṇa qualifies the object as black sesame by excluding
white seasame and seasame of other colors.32 In this context, Patañjali additionally remarks
that the compound *vṛkṣaśiṃśapā is impossible because vṛkṣa does not qualify śiṃśapā as
a particular kind, for there is no śiṃśapā that is not a tree (avṛkṣa).33
In this argument, Patañjali merely deals with the semantic connotation of tree by
śiṃśapā without going into the ontological background, but we can say that Kumārila
implies an ontological relation between śiṃśapātva and vṛkṣatva in the phrase “śiṃśapātva
is restricted by tree-ness (vṛkṣatvaniyata),” for in his Ślokavārttika (ŚV) he has already
discussed the idea that a species is ontologically restricted by a generic property. In the
ŚV, Kumārila affirms that specific entities are endowed with generic properties through
their own nature (svābhāvika); to possess them they do not need any cause (na hetumān).34
29
30
31
32
33
34
TV 932,16–18: nābhidhātā kartṛkarmaṇoḥ pacatyādiśabdas tadatyantāvinābhūtārthābhidhāyitvāt.
yo yadatyantāvinābhūtārthābhidhāyī sa tasya na vācakaḥ, yathā śiṃśapāśabdo vṛkṣatvaniyataśiṃśapātvasya. “The word ‘pacati,’ etc., does not denote agent or object because it denotes a meaning
that is completely inseparable from them. If a word denotes something that is completely inseparable
from something else, then the word does not denote the latter, just as the word śiṃśapā [denotes only]
śiṃśapā-ness, which is restricted by tree-ness.” Cf. Kanazawa 1983: 929.
Kumārila may have collected the essential predicates applicable to a śiṃśapā from PS 5.35: vṛkṣatvapārthivadravyasajjñeyāḥ pratilomyataḥ / catustridvyeka sandehe nimittam niścaye ’nyathā // “‘treeness,’
‘earthen,’ ‘substance,’ ‘existent,’ and ‘knowable’ are [each] a cause of doubt, in reverse order, about
four, three, two, and one [properties]. In opposite order they serve the purpose of ascertainment.” (Tr.
Pind 2015: I 125.)
The case in which a compound is obligatory, such as kṛṣṇasarpa (cobra = “black snake”), is an exception
to the governing rule (adhikārasūtra), A 2.1.11: “optional” (vibhāṣā). Cf. Katre 1989: 121.
VMBh I, 399: 3–15: anyatarasya pradhānabhāvāt tadviśeṣakatvāc cāparasyopasarjanasaṃjñā bhaviṣyati. yadāsya tilāḥ prādhānyena vivakṣitā bhavanti kṛṣṇo viśeṣaṇatvena tadā tilāḥ pradhānaṃ kṛṣṇo
viśeṣaṇam. “(Therefore,) since one of the two is the main (word) and since the other functions as the
qualifier of that, the designation upasarjana will apply. When the sesame seeds are intended as the
main thing by somebody, and black as the qualifier, then (the word) tilāḥ becomes the main (word) and
(the word) kṛṣṇāḥ (becomes) the qualifying word.” (Tr. Joshi/Roodbergen 1971: 140.)
VMBh I, 399: 25–26: kathaṃ tarhīmau dvau pradhānaśabdāv ekasminn arthe yugapad avarudhyete
vṛkṣaḥ śiṃśapeti. naitayor āvaśyakaḥ samāveśaḥ. na hy avṛkṣaḥ śiṃśapāsti. “Then in what way can
these two main words vṛkṣaḥ śiṃśapā ‘the tree śiṃśapā’ be locked up together to refer to one and the
same object? The co-usage of these two words in the order vṛkṣaḥ śiṃśapā is not necessary, because
there is no śiṃśapā which is not a tree.” (Tr. Joshi/Roodbergen 1971: 147–149.)
ŚV, Ākṛtivāda, vv. 31cd–32: svābhāvikaś ca sambandho jātivyaktyor na hetumān // tenaitasya prasiddhyarthaṃ nānyat sāmānyam iṣyate / śaktisiddhivad etasya svabhāvo ’tra na vāryate // “The relation
between an individual and its generic property is based on its own natural disposition and has no [extraneous] cause. Therefore, for the purpose of the accomplishment of this (relation), no more universal is
Kiyotaka Yoshimizu
537
Kumārila regards generic properties as a kind of intrinsic ability (śakti) of an individual
entity.35 Generally speaking, entities of a certain kind are, by nature, invariably equipped
with their own particular abilities, which can never be shared with entities of other kinds.36
Moreover, because they are equipped with these abilities by nature, they start up their own
activity of their own accord once they have been brought into existence.37
Taking his own remarks into account, Kumārila maintains that there must be some
causes for an individual entity to be brought into existence, but in addition to these causes,
nothing is required for possessing generic properties. Once an entity is brought into existence
as an instance of a certain species (viśeṣeṣv eva labdheṣu keṣucit), no external condition is
required (nānyavāñchana) for its acquisition of the generic property of its genus. What is
required for this acquisition is only a cause to that extent (tāvanmātrapratīkṣaṇa), a cause
that brings about the individual as an instance of some species or other that belongs to the
genus.38 In other words, an animal has already been equipped with the generic property
of a cow when it is caused to be born as an instance of śābaleya, bāhuleya, or any other
specific kind of a cow, by a pair of cows, male and female, of the respective species. This
view is very close to Aristotle’s idea of “universals in the thing (in re)” (cf. K. Yoshimizu
2015b).
35
36
37
38
needed. It is not denied that an individual is innately related with this (particular generic property) in
the same way that [it is innately] equipped with some abilities.”
ŚV, Ākṛtivāda, vv. 17–18: bhinnā viśeṣaśaktibhyaḥ sarvatrānugatāpi ca / pratyekaṃ samavetā ca tasmāj
jātir apīṣyatām // tenātmadharmo bhedānām ekadhīviṣayo ’sti naḥ / sāmānyam ākṛtir jātiḥ śaktir vā so
’bhidhīyatām // “[This generic ability (śakti) of an individual] is different from specific abilities; [it]
recurs everywhere [in all individuals] and completely inheres in each individual, and thus should be
accepted as a generic property [called cow-ness]. Therefore, in our position, entities are endowed with
an essential property that is the object of the cognition of the same kind, and one may call it universal,
form, generic property, or ability.” For pratyekaparisamāpti as a characteristic of apoha in addition to
its ekatva and nityatva, see PS 5 44,11.
ŚV, Ākṛtivāda, vv. 28–29ab: bhinnatve ’pi hi kāsāñ cic chaktiḥ kāś cid aśaktikāḥ / na ca paryanuyogo ’sti
vastuśakteḥ kadā cana // vahnir dahati nākāśaṃ ko ’tra paryanuyujyatām / “Although [all individuals
are] different from one another, some of them have a particular ability and others do not, and one should
never bring the ability of a real thing into question. Who would be questioned about the fact that fire
burns but the ether does not?”
ŚV, Codanādhikaraṇa, vv. 47cd–48: na hi svato ’satī śaktiḥ kartum anyena śakyate // ātmalābhe ca
bhāvānāṃ kāraṇāpekṣatā bhavet / labdhātmanāṃ svakāryeṣu pravṛttiḥ svayam eva tu // “For any ability
that would not be intrinsic [in an entity] cannot be produced by something else. When coming into
existence, entities may call for [their own] cause. Once having been brought into existence, however,
they start up their own activity of their own accord.”
ŚV, Ākṛtivāda, v. 33: yad vā naimittikatve ’pi tāvanmātrapratīkṣaṇāt / viśeṣeṣv eva labdheṣu keṣucin
nānyavāñchanam // “Even if [assumed to be] dependent on a certain cause, [the relation between an
individual and its generic property] merely requires [a cause] to the following extent: once [an instance
of] any species has been brought into existence, nothing more is required.” With “a cause to the following
extent” (tāvanmātra), Kumārila may exclude the Vaiśeṣika’s idea that inherence (samavāya) must exist
between two things that are inseparable (ayutasiddha), such as substance and quality, individual and
universal. In ŚV, Pratyakṣasūtra, vv. 146–148, Kumārila criticizes this idea as follows: the assumption
that two things have already been established (niṣpanna) as inseparable would make inherence useless,
whereas the assumption that two things require inherence for their inseparability leads to a regress
(avyavasthā) of further intermediary (see Taber 2005: 108–109).
538
Another Look at avinābhāva and niyama in Kumārila’s Exegetical Works
2.3 Kumārila’s reference to the proof of the impermanence by nature
In criticizing the Jaina theory that the soul (jīva), that is, the self (ātman) in transmigration,
has the size of one’s body, Kumārila assumes the case that the soul consists of many
parts (avayava), and from this assumption, he leads to the absurd consequence that the
soul (jīva) would perish like a pot (ghaṭa) at some time because, he says, all forms of
conjunction (saṃyoga) are inseparable from disjunction (viyogāvinābhūta).39 Considering
the production and the perishing of a pot adduced as an example, we can substitute producedness (kṛtakatva) for the conjunction of the parts that form a whole, and impermanence
(anityatva) or perishing (vināśa) for the disjunction of conglomerated parts from one
another.40
In another place (MmS 2.2.25), Kumārila briefly discusses the ontological relation
that enables the inference of impermanence (anityatva) from produced-ness (kṛtakatva).
Here Kumārila holds this type of inference to be different from those based on causality.41
Presenting this inference, he remarks that even between two things that are not cause and
effect (akāryakāraṇabhūta), it is well known that in some cases one thing indicates the
other (gamyagamakabhāvaprasiddhi).
Moreover, in the ŚV chapter on the permanence of words (Śabdanityatādhikaraṇa),
Kumārila enumerates three kinds of permanent elements (asaṃskṛta-dharma) in the Buddhist Abhidharma, namely, “cessation [of the occurrence of a dharma] by means of reflection” (pratisaṃkhyānanirodha), “cessation [of the occurrence of a dharma] without
39
40
41
TV 403,1–5: sarve hy avayavasadbhāvādayaḥ pratyakṣādyaviṣayāḥ santa utprekṣāmātreṇa kalpyante. te
śabdavac cehāpi pratyākhyeyāḥ. śarīramātrāvadhikānāṃ caiṣām utprekṣitum apy ānantyād aśakyaṃ, na
caiṣāṃ dravatvādinā vinā kaścit saṃśleṣahetur vidyate. na cāsaṃśliṣṭānām ekajīvārambhasāmarthyam.
sarvasaṃyogānāṃ ca viyogāvinābhūtatvād ghaṭāditulyaḥ kadācij jīvasya pradhvaṃso bhavet. “Since
all these (assumed things, cf. TV 402,27–30), namely, the existence of the parts [of the soul], etc., are
not the object of perception, they are assumed only by imagination and are to be rejected here just like
[the parts of] a word [in the sense of phonemes (varṇa) as rejected in Sphoṭavāda vv. 11–13 of the
ŚV]. Because they are boundless, one can never imagine that they are limited to the extent of the body.
Without some fluid substances, moreover, there cannot be any ground for their conglomeration, and
they are unable to bring about a life unless conglomerated. Besides, because all sorts of conjunction are
epistemically inseparable from disjunction, the soul would suffer perishing at some time in the same
way as a pot.”
The logical inseparability of kṛtakatva and sāvayavatva from vināśatva has already been mentioned in
ŚV, Anumānapariccheda vv. 20cd–21ab: kṛtasāvayavatvādiprayuktā ca vināśitā // prayatnānantarajñānasadṛśair na prayujyate / “Perishability is caused [to be known] by produced-ness, composite-ness,
etc., but it is not caused [to be known] by the fact of being cognized immediately after exertion, etc.”
TV 544,18–20: ato vacanāt* kāryakāraṇayor eva gamyagamakabhāvo bhāṣyakārābhipreta iti kecin
manyante. tat tv ayuktam. akāryakāraṇabhūtānām apy anityatvakṛtakatvādīnāṃ gamyagamakabhāvaprasiddheḥ. “Some people assume from this statement that the author of the Bhāṣya has the idea in
mind that only a cause and its effect have the relation between them of logical indicator and the thing to
be indicated. This is, however, not correct. Even between two things that are not a cause and its effect, it
is well known that in some cases one thing indicates the other, for example, between impermanence and
produced-ness.” * This vacana refers to ŚBh 544,2–4: yad dhi yasya kāraṇabhūtaṃ dṛṣṭaṃ siddhe, tac
cet sādhye ’pi kāraṇabhūtam ity avagamyate. bhavati tat tasya sādhakam. “As a matter of fact, when A
has already been found to be the cause of B in a well-known example, and if in the subject of inference,
too, A is found to be the cause, then A is the logical reason for inferring [the existence of] B [in the
subject of inference].” As regards causality, Kumārila, unlike Śabara and like Dharmakīrti, remarks
that an inference from cause to effect is inconclusive. Cf. n. 15.
Kiyotaka Yoshimizu
539
reflection” (apratisaṃkhyānanirodha), and the ether (vyoman).42 Thereafter, without distinguishing between Sarvāstivāda and Sautrāntika, Kumārila refers to the Buddhist theory that
things are destined to perish (vināśa) through their own nature (svabhāvasiddha) without
any external cause (ahetuka).43 Then adducing two examples, he compares generation with
perishing in terms of whether or not an external cause is needed:44 When something new is
immediately generated after the perishing of an old thing, an extraneous cause is necessary.
For example, a piece of fuel is changed into ashes when connected to fire, and a pot is
changed into fragments when struck by a hammer. However, perishing is not produced
(akriyamāna) within individual entities afterwards; rather it has been in progress from the
beginning in a subtle form (sūkṣma) without being noticed; it manifests (abhivyajyate) in
a gross form (sthūla) upon meeting a heterogeneous cause (vilakṣaṇahetu) such as fire
or a hammer.45 In short, perishing is the essential nature of an entity (svābhāvika) and is
already established insofar as the entity is brought into existence (jātamātrapratiṣṭhita).
42
43
44
45
ŚV, Śabdanityatādhikaraṇa, vv. 22–23ab.
ŚV, Śabdanityatādhikaraṇa, vv. 23cd–24: tau ca dvāv apy anāśitvād iṣṭāv akṛtakāv api // āhuḥ svabhāvasiddhaṃ hi te vināśam ahetukam / hetur yasya vināśo ’pi tasya dṛṣṭo ’ṅkurādivat // “And both of
them (i.e., pratisaṃkhyānanirodha and apratisaṃkhyānanirodha) are considered not to be produced
because of their imperishability. For they (i.e., Buddhists) say the perishing [of an entity] is established
by nature without [extraneous] cause because anything that has a cause is also observed to be perishable,
just like sprouts, etc.” I am indebted to Dr. Taisei Shida for telling me about this reference by Kumārila
to the Buddhist proof of impermanence without extraneous cause in the Śabdanityatādhikaraṇa.
ŚV, Śabdanityatādhikaraṇa, vv. 25cd–29: bhavati hy agnisambandhāt kāṣṭād aṅgārasantatiḥ // mudgarādihatāc cāpi kapālaṃ jāyate ghaṭāt / svābhāviko vināśas tu jātamātrapratiṣṭhitaḥ // sūkṣmaḥ
sadṛśasantānavṛtter anupalakṣitaḥ / yadā vilakṣaṇo hetuḥ patet sadṛśasantatau // vilakṣaṇena kāryeṇa sthūlo ’bhivyajyate tadā / tenāsadṛśasantāno hetoḥ sañjāyate yataḥ // tenaivākriyamāṇo ’pi
nāśo ’bhivyajyate sphuṭaḥ / sa mudgaraprahārādiprayatnānantarīyakaḥ // “The temporal succession
of embers proceeds from the fuel if it is connected to fire; and when a pot is struck by a hammer, a
fragment is born from that (pot). The perishing [of the fuel and the pot], however, is based on their
own natural disposition and established insofar as they are produced. [The inborn perishing] in subtle
form is imperceptible on account of their existence in homogenous succession. When a heterogeneous
cause happens to the homogenous succession, then [the inborn perishing] is manifest in gross form by
means of heterogeneous efficacy. In this manner, there occurs a heterogeneous succession on account
of the (heterogeneous) cause. By means of this (cause), the perishing, despite not being produced, is
manifest in explicit form. This [explicit perishing] appears as soon as one makes an effort, for example, to hit a hammer [on a pot].” Kumārila’s explanation of spontaneous perishing seems to be based
on Abhidharmakoṣabhāṣya (AKBh) 193,5–7: saṃskṛtasyāvaśyaṃ “vyayāt” (AK 4.2d). ākasmiko hi
bhāvānāṃ vināśaḥ “[The momentariness of everything is established] because the conditioned entity
necessarily ‘perishes.’ For the destruction of things is spontaneous” (tr. von Rospatt 1995: n. 397),
because the simile of the contact of the fuel with fire is applied in AKBh 194,1–6: yadi ca kāṣṭhādīnām
agnyādisaṃyogahetuko vināśaḥ syād evaṃ sati pākajānāṃ guṇānāṃ pakvataratamotpattau “hetuḥ
syāc ca vināśakaḥ” (AK 4.3b). hetur eva ca vināśakaḥ syāt. “And if the destruction of firewood etc.
was caused by the contact with fire etc., this being so, in the case of the more and more progressed
origination of the qualities borne from burning (pākaja) ‘the cause [for these qualities] would be [the
factor] destroying [them],’ [that is,] precisely [their generative] cause would be the [factor] destroying
[them].” (Tr. von Rospatt 1995: n. 402.)
Kumārila insists, however, that the proof of impermanence cannot be applied to words (śabda) because
words are not produced but only made manifest by means of an utterance. Cf. ŚV, Śabdanityatādhikaraṇa, v. 39ab: vyañjakābhāvataḥ śabde ’py abodho badhirādivat / “As regards words, too, if there
is no cause of manifestation, one cannot perceive them just like deaf people;” vv. 40cd–41: sann eva
sādhanābhāvāc chabdo naivopalabhyate // kṣaṇikaṃ sādhanaṃ cāsya buddhir apy anuvartate / meghāndhakāraśarvaryāṃ vidyujjanitadṛṣṭivat // “In spite of existing, a word is never perceived insofar
540
Another Look at avinābhāva and niyama in Kumārila’s Exegetical Works
The phrase jātamātrapratiṣṭhita, which conforms to tāvanmātrapratīkṣaṇa (ŚV, Ākṛtivāda,
v. 33), anticipates Dharmakīrti’s definition of the svabhāva to be proved by a svabhāvahetu,
namely, bhāvamātrānurodhin (PV 1.2d & 39b).
III. Limits of logical formulation in exegetical works
From the abovementioned examples, we may say that in the TV Kumārila has already
conceived the gist of the niyama fragment from the BṬ as stating that if one thing is restricted
(niyamya) in nature, namely, brought into existence under the control of another thing
(niyāmaka) in reality, then the confirmation of the former becomes the logical indicator
(gamaka) of the existence of the latter thing (gamya) in cognition. He also had the idea of
the invariable one-way transition in the concept from an effect to its cause, and from the
existence of a thing to its generic properties on the basis of causality. He was even aware
that the inference of impermanence from produced-ness is an example of the latter type of
conceptual transition.46 However, because every section of the TV deals with exegesis of
scriptures, Kumārila refers only sporadically to the invariable transition from one concept
to another on the basis of the natural restriction of various types.47 Unlike Dharmakīrti,
he never attempts to reduce various types of niyama to fundamental ones and relate them
to each other. Moreover, in another exegetical work by Kumārila, the Ṭupṭīkā (ṬṬ), we
find some cases in which the object to be known from an indicator that is inseparable
(avinābhūta) from it is called “restricted” (niyata), instead of “the agent of restriction”
(niyāmaka).
When one hears the Vedic injunction “One who wishes for heaven should perform
a sacrifice” (svargakāmo yajeta), one may first, urged by the verb in the optative, feel
obliged to perform a sacrifice. However, one usually performs an action in order to achieve
a purpose one has set up in advance. The general form of action called “bringing into being”
(bhāvanā) is, therefore, said to be inseparable from the motivation to achieve a purpose
(prayojyādibhir avinābhūtā). Then, paying attention to another word in this injunction
that describes the entitled person (adhikārin), “one who desires heaven” (svargakāmaḥ),
46
47
as the means [of its perception, i.e., an utterance] is absent. And the cognition of that (word) conforms
to the means that remains only for a moment, just like the sight caused by a flash of lightning during a
dark night covered with clouds.” Cf. ŚV, Śabdanityatādhikaraṇa, vv. 424–425; von Rospatt 1995: n.
369. Kumārila also denies the momentariness of things for everyday use. ŚV, Śabdanityatādhikaraṇa, v.
426: kṣaṇabhaṅgo ghaṭādīnāṃ vāryas tair eva sādhanaiḥ / tathaiva pratyabhijñānād, yāvad dṛṣṭaṃ na
bādhate // “The momentary perishing of a pot, etc., is to be rejected for the same reasons [as those given
in vv. 416–421 in order to disprove the momentary perishing of words]. Because [a pot] is recognized
to be the same [as a few moments before], [the continuation of its identity] is not denied insofar as it
remains to be perceived.”
Cf. n. 41.
Based on a holistic view that the whole pervades (vyāpaka) its parts and parts are restricted (niyamya)
by the whole they form, Kumārila maintains an exegetical principle that a word that directly refers to a
Ṛgveda verse (ṛc) indirectly indicates the whole text of the Ṛgvedasaṃhitā. Cf. TV 807,17–18: vedaś
ca vyāpakatvād ahetulakṣaṇayuktaḥ san na śaknoty evāvayavaṃ lakṣayitum. ṛgādayas tu niyamyatvāt
samudāyaṃ lakṣayanti. “Because [an entire text of] the Veda pervades [all of its parts], it is not fit
for the definition of a logical reason. Therefore, it cannot indirectly denote its part. The verses of ṛc,
etc., on the other hand, indirectly denote the whole [text of the Ṛgvedasaṃhitā] on account of being
restricted [by the whole].”
Kiyotaka Yoshimizu
541
one may come to be aware that the purpose to perform the sacrifice is a particular kind
of something valuable for a human being (puruṣārtha), namely, heaven (svarga).48 In this
way, focused on the verb in the optative at first, one may simply feel one’s duty to perform
a sacrifice (yāgasya kartavyatā), but this sense of unconditional duty is suspended and
replaced by the motivation to achieve heaven by performing a sacrifice.49
However, accepting that the performance of a sacrifice is motivated by something
valuable for human beings, Kumārila calls this goal of conceptual transition “restricted”
(niyata-puruṣārtha-prayojyatva), instead of calling it the agent of restriction (niyāmaka)
in view of the causa finalis. Nor does he consider the performance of a sacrifice to be the
niyāmaka in view of the causa efficiens, for he assumes that the efficacy of a sacrifice is
based on nothing other than the Vedic injunction (vidhi).50
In the case of Dharmakīrti’s logic, the arising of an effect (kāryotpāda) is inferred from
the “complete cause” (kāraṇasāmagrī ), in other words, when all the necessary conditions
are fulfilled in addition to the existence of the main cause (PVSV 6,22–7,12 on PV 1.7–8).
Although Kumārila analogically compares the performance of a sacrifice to ordinary
activities such as farming for a harvest,51 it is impossible to consider the performance
of a sacrifice a “complete cause” (kāraṇasāmagrī ) for attaining a result such as heaven
48
49
50
51
ṬṬ 1352,16–18: ānantaryaśrutyā yāgasya kartavyatā. pratyayaḥ punaḥ svārthe bhāvanāyāṃ puruṣaṃ
pravartayati. sā ca prayojyādibhir avinābhūtā, tena yan niyamitaṃ tad eva puruṣārthaṃ prayojyatvenākāṅkṣati, na yāgam. “No sooner than one has heard the injunction, one recognizes that the sacrifice
is to be performed. The ending [of the verb in the injunction], however, prompts one to exert oneself
for bhāvanā, its own meaning. And this (bhāvanā [of one’s activity]) is inseparable from its purpose,
etc. [that is, purpose (sādhya), means (sādhana) and the mode of performance (itikartavyatā)]. For
its purpose (prayojya), [the bhāvanā] requires only the things valuable for a human being insofar
as they are restricted (niyamita) by that (injunction), not the sacrifice itself.” Kumārila distinguishes
two kinds of bhāvanā (cf. TV 378,20–21). The bhāvanā discussed in this context is the “intentional
force for activity” (arthātmabhāvanā/arthībhāvanā), the general form of one’s action that requires a
purpose to be achieved by the one who performed the action, but not the “verbal force for activity”
(abhidhābhāvanā/śabdībhāvanā), which is the function of the verbal ending of an injunction to urge
one to exert one’s bhāvanā.
ṬṬ 1352,23–25: tasmān naiva kadācid yāge pratyaya upanipatati saty evānantarye. ato yatra kartavyatāvacanas tatra puruṣārthaḥ sākṣāt pāramparyeṇa vāvinābhūtaḥ. tasmād yāgasya kartavyatā prāptā,
sā bhāvanāśabdasya niyatapuruṣārthaprayojyatvād bādhyate. “Therefore, the verbal ending never
happens to be directed to the sacrifice even though it is immediately cognized. Thus, no matter in
what statement a duty is enjoined, something valuable for a human being is inseparably connected
[to that duty], whether directly or indirectly. Therefore, even if it is assumed that the sacrifice is to be
performed [for its own sake], this assumption is given up because the word that [urges the hearer to
exert his] bhāvanā has [the hearer’s] purpose [in performing the sacrifice] restricted [by the injunction]
to something valuable for a human being.”
In his commentary on the ṬṬ, Pārthasārathimiśra remarks that what Kumārila refers to by the pronoun
tena, by which the purpose in performing the sacrifice is restricted to a particular puruṣārtha (ṬṬ
1352,17: tena yan niyamitaṃ tad eva puruṣārthaṃ prayojyatvenākāṅkṣati), is not the action of sacrifice
but the injunction (vidhi) that prompts human beings by nature (puruṣapravartanātmaka). Cf. TR
390,10–13.
TV 395,3–6: laukikaṃ cāpi yat karma phale kālāntarodgatau / tatrāpi śaktir evāste na tv apūrvam
iheṣyate // yāny api ca laukikāni kṛṣighṛtapānādhyayanaprabhṛtīti karmāṇi kālāntaraphalatveneṣyante
teṣāṃ api svarūpāvasthānāsaṃbhavāt saṃskārair eva tiṣṭhadbhir vyavahārasiddhiḥ. “Even in worldly
activities, there must remain an ability that brings about its result later on, but [such ability] is not
recognized as apūrva in this (science of exegesis, mīmāṃsā). Such worldly activities as cultivation,
drinking ghee, and study are accepted to have their result later on. Because they do not continue to
542
Another Look at avinābhāva and niyama in Kumārila’s Exegetical Works
because we cannot clarify why or how a sacrifice can bring about a result by empirically
investigating its mechanism. This is because in Mīmāṃsā it is firmly presupposed that
the Veda alone has authority over religious duties (dharma); there is no need to resort to
empirical knowledge.
Conclusion
Despite assuming human rationality to be subordinate to Vedic authority as far as religious
duties are concerned, Kumārila never advocates an extreme fundamentalism in which
human beings are incapable of acquiring any knowledge without Vedic revelation. On the
contrary, Kumārila clearly states that there are many fields to be investigated on the basis of
empirical observation, giving many examples concerning the man-made scriptures (smṛtis)
and the sciences ancillary to the Veda (vedāṅgas).52 Accordingly, we may safely conjecture
that after having completed the Ślokavārttika, Kumārila continued to study logical reasoning
while compiling such exegetical works as the Tantravārttika and the Ṭupṭīkā. Finally, in
the Bṛhaṭṭīkā, he declared natural restriction (niyama) to be the foundation of epistemic
inseparability (avinābhāva).53
References and abbreviations
A Aṣṭādhyāyī.
AK Abhidharmakośa. See AKBh.
AKBh Abhidharmakośabhāṣya, ed. P. Pradhan. Patna 2 1975. [1st ed. 1967].
ĀpŚS Āpastambaśrautasūtra, ed. R. Garbe. New Delhi 2 1983. [1st ed. 1882–1902].
ĀśvŚS Āśvalāyanaśrautasūtra, ed. Ranbir Singh Bawa, et al., part 1. Hoshiarpur 1986.
AV Atharvavedasaṃhitā, ed. R. Roth and W. D. Whitney. Berlin 2 1924.
BhārŚS Bhāradvājaśrautasūtra, ed. C. G. Kashikar, part 1. Poona 1964.
Bodewitz 1990 H. W. Bodewitz, The Jyotiṣṭoma Ritual: Jaiminīya Brāhmaṇa I, 66–364.
Leiden 1990.
52
53
exist in their own form [until the time of the appearing of the result], these ordinary activities are
accomplished through certain dispositions that continue [until then].”
Cf. TV 166,27–28: tatra yāvad dharmamokṣasaṃbandhi tad vedaprabhavam. yat tv arthasukhaviṣayaṃ
tal lokavyavahārapūrvakam iti vivektavyam “Among those (smṛti texts), the portions have their origin in
the Veda insofar as they pertain to duties and emancipation, but those concerned with profit and pleasure
are to be distinguished because they are based on worldly activities;” 166,29–167,1: yat tu pṛthivīvibhāgakathanam tad dharmādharmasādhanaphalopabhogapradeśavivekāya kiṃcid darśanapūrvakaṃ
kiṃcid vedamūlam, vaṃśānukramaṇam api brāhmaṇakṣatriyajātigotrajñānārthaṃ darśanasmaraṇamūlam “The reports of the various parts of the earth are partly based on observance and partly on the Veda,
for the purpose of distinguishing the regions appropriate for receiving the reward for having performed
dharma and adharma. In addition, the chronicles of genealogies, which serve to know the caste and
lineage of Brahmins and Kṣatriyas, are based on observance and legends;” 167,4: aṅgavidyānām api
kratvarthapuruṣārthapratipādanaṃ lokavedapūrvakatvena vivektavyam “Also in the ancillary sciences
[of the Veda], one should distinguish between the portions based on the Veda for informing what is
valuable for the sacrifice, and the portions based on the ordinary world for informing what is valuable
for human beings.”
See appendix for a reply to Ernst Steinkellner.
Kiyotaka Yoshimizu
543
BṬ Bṛhaṭṭīkā.
Caland 1982 W. Caland, Pañcaviṃśa-brāhmaṇa: the Brāhmaṇa of Twenty Five Chapters.
Calcutta 1982. [1st ed. 1931].
Ch. Yoshimizu 2011 Chizuko Yoshimizu, What makes all that is produced impermanent?
The proof of impermanence and the theory of causality. In: Religion and Logic in
Buddhist Philosophical Analysis. Proceedings of the Fourth International Dharmakīrti Conference. Vienna, August 23–27, 2005, ed. H. Krasser, H. Lasic, E. Franco,
and B. Kellner. Vienna 2011, 491–506.
Franco 2019 E. Franco, Xuanzang’s Silence and Dharmakīrti’s Dates. Wiener Zeitschrift
für die Kunde Südasiens 56–57 (2019) 117–141.
Frauwallner 1962 E. Frauwallner, Kumārila’s Bṛhaṭṭīkā. Wiener Zeitschrift für die Kunde
Süd- und Ostasiens 6 (1962) 78–91.
Harikai 2006 K. Harikai, The Mīmāṃsā Interpretation of the Particle Hi in Brāhmaṇa
Literature: Is Anumāna Possible in the Veda? In: Expanding and Merging Horizons.
Contributions to South Asian and Cross-Cultural Studies in Commemoration of
Wilhelm Halbfass, ed. K. Preisendanz. Vienna 2006, 303–313.
HirŚS Hiraṇyakeśiśrautasūtra (= Satyāṣāḍhaśrautasūtra). Poona 1907.
IO Manuscript of the Tantravārttika, preserved in the British Library, London. Eggeling
No. 2157 & 2158.
JB Jaiminīyabrāhmaṇa, ed. Raghu Vira and Lokesh Chandra. Delhi 1986. [1st ed. 1954].
Joshi/Roodbergen 1971 S.D. Joshi and J.A.F. Roodbergen, Patañjali’s Vyākaraṇa-Mahābhāṣya: Karmadhārayāhnika (P. 2.1.51–2.1.72). Poona 1971.
K. Yoshimizu 2007a Kiyotaka Yoshimizu, Reconsidering the fragment of the Bṛhaṭṭīkā
on inseparable connection (avinābhāva). In: Pramāṇakīrtiḥ, Papers Dedicated to
Ernst Steinkellner on the Occasion of his 70th Birthday, ed. B. Kellner, H. Krasser,
H. Lasic, M.T. Much, and H. Tauscher. Vienna 2007, 1079–1103.
K. Yoshimizu 2007b Kiyotaka Yoshimizu, Kumārila’s Reevaluation of the Sacrifice and
the Veda from a Vedānta Perspective. In: Mīmāṃsā and Vedānta, Interaction and
Continuity, ed. J. Bronkhorst. Delhi 2007, 201–253.
K. Yoshimizu 2011 Kiyotaka Yoshimizu, Reconsidering the fragment of the Bṛhaṭṭīkā
on restriction (niyama). In: Religion and Logic in Buddhist Philosophical Analysis.
Proceedings of the Fourth International Dharmakīrti Conference. Vienna, August
23–27, 2005, ed. H. Krasser, H. Lasic, E. Franco, and B. Kellner. Vienna 2011,
507–521.
K. Yoshimizu 2012a Kiyotaka Yoshimizu, Kumārila and Medhātithi on the Authority of
Codified Sources of dharma. In: Devadattīyam. Johannes Bronkhorst Felicitation
Volume, ed. F. Voegeli, V. Eltschinger, and D. Feller. Bern 2012, 643–681.
K. Yoshimizu 2012b Kiyotaka Yoshimizu, Tradition and Reflection in Kumārila’s Last
Stand against the Grammarians’ Theories of Verbal Denotation. In: Saṁskṛtasādhutā: Goodness of Sanskrit. Studies in Honour of Professor Ashok N. Aklujkar,
ed. Ch. Watanabe, M. Desmarais, and Y. Honda. New Delhi 2012, 552–586.
K. Yoshimizu 2015a Kiyotaka Yoshimizu, Kumārila’s Criticism of Buddhism as a Religious Movement in his Views on the Sources of dharma. Acta Asiatica 108 (2015),
43–62.
544
Another Look at avinābhāva and niyama in Kumārila’s Exegetical Works
K. Yoshimizu 2015b Kiyotaka Yoshimizu, Kumārila ni okeru kotai to huhen no hi-betsui:
Aristotle tono hikaku. (Kumārila on the Non-difference between an Individual and
Its Generic Property: A Comparison with Aristotle). Indo Ronrigaku Kenkyū (Indian
Logic) 8 (2015) 5–30.
Kanazawa 1983 A. Kanazawa, Kumārila to <avinābhāva>, Journal of Indian and Buddhist
Studies 31(2) (1983) 928–933.
Kataoka 2011 K. Kataoka, Kumārila on Truth, Omniscience, and Killing. Part 2. Vienna
2011.
Katre 1989 S. M. Katre, Aṣṭādhyāyī of Pāṇini. Delhi 1989. [1st ed. 1987].
MānŚS Mānavaśrautasūtra, ed. J. M. van Gelder. Delhi 1985. [1st ed. 1961].
MBh Mahābhārata, critical ed. V. S. Sukthankar, et. al. Poona 1933–1966.
MmS Mīmāṃsāsūtra. See TV.
Oetke 1992 C. Oetke, Pragmatic Implicatures and Text-Interpretations. Studien zur Indologie und Iranistik 16 (1992) 185–233.
Ogawa 1993 H. Ogawa, Kumārila no śabdārtha ron. (Kumārila’s Theory of śabdārtha:
abhidheya and gamya). Journal of Indian and Buddhist Studies 41(2) (1993)
1051–1055.
Pandey 2006 R. Pandey, Hindu Saṁskāras: Socio-Religious Study of the Hindu Sacraments. Delhi 2006. [1st ed. 1969].
PB Pañcaviṃśabrāhmaṇa, ed. A. Ch. Vedantavagisha, 2 vols. Varanasi 1989. [1st ed.
1870–1874].
Pind 2015 O. H. Pind, Dignāga’s Philosophy of Language. Dignāga. Pramāṇasamuccayavṛtti V on anyāpoha. Part I: Text. Part II: Translation and Annotation, ed. E.
Steinkellner. Vienna 2015.
PS 5 Pramāṇasamuccaya, the fifth chapter. See Pind 2015: Appendix 1.
PV 1 Pramāṇavārttika, the first chapter. See PVSV.
PVSV Pramāṇavārttikasvavṛtti, ed. R. Gnoli. Rome 1960.
PVSVṬ Pramāṇavārttikasvavṛttiṭīkā, ed. R. Samkrityayana. Kyoto 1982.
RV Ṛgvedasaṁhitā, ed. T. Aufrecht, Wiesbaden 1968. [1st ed. Bonn 1877].
ŚāṅkhŚS Śāṅkhāyanaśrautasūtra, ed. A. Hillebrandt, vol. 1. New Delhi 2002. [1st ed.
1888–1899].
ŚBh Śābarabhāṣya. See TV.
Schwab 1886 J. Schwab, Das altindische Thieropfer: mit Benützung handschriftlicher
Quellen bearbeitet. Erlangen 1886.
Steinkellner 1971 E. Steinkellner, Wirklichkeit und Begriff bei Dharmakīrti. Wiener Zeitschrift für die Kunde Südasiens 15 (1971) 179–211.
Steinkellner 1991 E. Steinkellner, Dharmakīrti on the Inference of Effect (kārya). In:
Papers in Honour of Prof. Dr. Ji Xianlin on the Occasion of His 80th Birthday (II),
ed. L. Zheng, J. Zhongxin, D. Qing, and Q. Wenzhong. Beijing 1991, 711–736.
Steinkellner 1997 E. Steinkellner, Kumārila, Īśvarasena, and Dharmakīrti in Dialogue, A
New Interpretation of Pramāṇavārttika I.33. In: Bauddhavidyāsudhākaraḥ, Studies in
Honour of Heinz Bechert. ed. P. Kieffer-Pülz and J.-U. Hartmann. Swisttal Odendorf
1997, 625–646.
Kiyotaka Yoshimizu
545
Steinkellner 2007–2008 E. Steinkellner, Further Remarks on the Compound avinābhāvaniyama in the Early Dharmakīrti. Wiener Zeitschrift für die Kunde Südasiens 51
(2007–2008) 193–205.
Steinkellner 2013 E. Steinkellner, Dharmakīrtis frühe Logik: Annotierte Übersetzung der
logischen Teile von Pramāṇavārttika 1 mit der Vṛtti. Tokyo 2013.
ŚV Ślokavārttika, ed. S. Dvarikadāsa Śāstrī. Varanasi 1978.
Taber 2005 J. Taber, A Hindu Critique of Buddhist Epistemology: Kumārila on Perception.
London/New York 2005.
TR Tantraratna, ed. Ganganath Jha. Varanasi 1963.
ṬṬ Ṭupṭīkā. See TV.
TV Tantravārttika. In: Mīmāṃsādarśanam, ed. Subbāśāstrī, 6 parts. Poona 1929–53.
VMBh Vyākaraṇamahābhāṣya, ed. F. Kielhorn, 3 parts. Revised by K.V. Abhyankar. Poona
4
1985. [1st ed. 1880].
von Rospatt 1995 A. von Rospatt, The Buddhist Doctrine of Momentariness. A Survey of
the Origins and Early Phase of this Doctrine up to Vasubandhu. Stuttgart 1995.
Appendix: A reply to E. Steinkellner (2007/2008)
In his PVSV on PV 1.1, after having concisely explained the three kinds of logical reason
(PV 1.1b: hetus tridhaiva saḥ), namely, effect (kārya), essential property (svabhāva), and
the absence of the perception (anupalabdhi) of a perceptible thing, Dharmakīrti summarizes
that the first two are reasons for affirmation (vastusādhanau) and the last one for negation
(pratiṣedhahetuḥ) (PVSV 2,19). Then, in order to briefly answer why logical reasons must
be one of these three kinds, he introduces svabhāvapratibandha, which takes either one of
the two forms: tādātmya (PVSV 2,21: tadātmatva), that is, the relation between an essential
property (bhāva) and another essential property (svabhāva) of one and the same thing; or
tadutpatti (PVSV 3,4), that is, causality between two different (anya) things.
Then, after a short interlude, he briefly explains how an inferential cognition can be a
pramāṇa on account of its non-deviation (avyabhicāra) from its object, saying “Because,
between the [two] things one of which is not dependent on the other, there is no restriction
[that would account] for an invariable occurrence [of the latter] with [the former]” (PVSV
3,9: anāyattarūpāṇām sahabhāvaniyamābhāvāt). This phrase can be considered the logical
inverse (¬P→¬Q) of the general law (P→Q) given in PVSV 2,19–20: “Because, if [one
thing] is bound [by the other] by nature, the former does not deviate from the latter”
(svabhāvapratibandhe hi saty artho arthaṃ na vyabhicarati). If P and Q are logically
equivalent, the conditional sentence “P→Q” as well as its inverse “¬P→¬Q” holds good
(cf. Oetke 1992: 198–199). What is called āyattarūpa is something A whose essence (rūpa,
i.e., svabhāva) is dependent on (āyatta), in other words, bound (pratibaddha) by, the other
thing B.54 Sahabhāva, that is, the invariable occurrence (bhāva) of B with (saha) A, is
logically equivalent to its contraposition, namely, avinābhāva, that is, the never-occurrence
(a-bhāva) of A without (vinā) B, which can be paraphrased as artho (A) arthaṃ (B) na
vyabhicarati. Thus, sahabhāvaniyama is said to be the restriction of A’s nature by B for
making the conceptual occurrence of B with A invariable; but it is unlikely that in this
54
For the synonymy between āyatta and pratibaddha, see Y. Fukuda’s article in the present volume.
546
Another Look at avinābhāva and niyama in Kumārila’s Exegetical Works
context sahabhāva is said to be restricted (*niyata) to the three modes in accord with
the three kinds of logical reason including non-perception (anupalabdhi) that leads to
non-existence, because what is argued about is the cognition of the two kinds of object to
be affirmatively inferred (PVSV 3,5: etāv dvāv anumeyapratyayau), namely, svabhāva and
cause (kāraṇa).
In the PVSV on PV 1.2a–c, in which kāryahetu is defined, niyama is the name of the
state of real affairs in which the production of effect is regulated by its cause (3,11–12:
tadutpattiniyama; 3,16: tatkāryatvaniyama). Later, concerning an inference of one thing
from another different (anya) thing that is neither its effect nor its cause (PVSV 7,12:
akāryakāraṇabhūta, cf. Steinkellner 2013, vol. I: 19), Dharmakīrti repeats the content of
the phrase anāyattarūpāṇām sahabhāvaniyamābhāvāt in PVSV 8,12–13: apratibaddhasvabhāvasya^avinābhāvaniyamābhāvāt, in which sahabhāvaniyama is explicitly replaced
with avinābhāvaniyama. As regards the immediately preceding sentence, “Hence there is
no other logical reason that makes [it] known” (PVSV 8,12: tena nānyo hetur gamako ’sti),
Karṇakagomin glosses “no other” (nānyo) with “nothing other than a hetu of [either one
of] the three kinds” (PVSVṬ 49,14: trividhād dhetor nānyo; 50,6: hetutrayavyatirekeṇa
nānyo). But this gloss is off the mark. What is intended here with “no other” is “nothing
other than a kāryahetu,” because this sentence is Dharmakīrti’s own comment on the last
part of PV 1.10, tat kāryaliṅgajā, which concludes that the inferential cognition from a
particular taste (rasa) to a particular color (rūpa) arises from the epistemic application of
a special type of causality that brings about two effects simultaneously. In the rest of his
commentary on PV 1.10 and from PV 1.11 onwards, Dharmakīrti goes on arguing about
how to apply causality in inference.
Moreover, in the PVSV on PV 1.2cd, which defines svabhāvahetu, it is said that what
is not spatiotemporally present (B) in the presence of something (A) is not restricted to
its later presence (4,3: tadbhāve ’bhūtasya paścād bhāvaniyamābhāvāt), in other words,
will not necessarily appear later on. With this lack of paścādbhāva-niyama in case of the
absence of something B when another thing A is present (tadbhāve ’bhūtasya), Dharmakīrti
formulates the contraposition (¬Q→¬P) of the general law (P→Q), svabhāvapratibandhe
hi saty artho (A) arthaṃ (B) na vyabhicarati, in the case of pseudo tādātmya. On the
other hand, as mentioned above, he formulates the inverse (¬P→¬Q) arguing the lack of
avinābhāvaniyama in the case of pseudo tadutpatti in PVSV 8,12–13: apratibaddhasvabhāvasya^avinābhāvaniyamābhāvāt.
In all these cases of PVSV 3,9, 4,3 and 8,12–13, the “restriction” (niyama) that enables
sahabhāva, tadbhāve-bhāva, and even avinābhāva between two concepts is completed
within either one of the two binominal forms of svabhāvapratibandha, tādātmya or tadutpatti, respectively. For the spatiotemporal occurrence in reality, the nature of one thing
is determined by another thing that is either its own essential property or its cause. Later,
Dharmakīrti refers to this ontological restriction with the term svabhāvaniyama (cf. Steinkellner 1971: 188f.). In all of these three cases, however, niyama does not restrict the total
number of the types of epistemic inseparability: It is not discussed how many modes there
are of the invariable transition from one concept to another.
Seeking Dharmakīrti’s own paraphrase of avinābhāvaniyamāt (PV 1.1c) in his PVSV
on PV 1.1, we find only anāyattarūpāṇām sahabhāvaniyamābhāvāt (PVSV 3,9) as an
approximate phrase in the inverse form. If the avinābhāvaniyama in PV 1.1c is paraphrased
Kiyotaka Yoshimizu
547
as sahabhāvaniyama in PVSV 3,9, and this sahabhāvaniyama is replaced with avinābhāvaniyama in PVSV 8,12–13, then the avinābhāvaniyama in PV 1.1c, too, cannot be the
restriction “of” avinābhāva in three modes in accord with the three kinds of logical reason,
because anupalabdhihetu does not require its own ontological restriction for conceptual
transition. Dharmakīrti does not separately assume a third form of svabhāvapratibandha
for anupalabdhihetu because this kind of logical reason is based on the tādātmya of the
complete cause (kāraṇasāmagrī ) of the perception of a visible thing (cf. PVSV 105, 1–3;
Steinkellner 1991: 712). The state of real affairs presupposed in applying an anupalabdhihetu is that the complete cause of perception including the existence of a visible thing is
restricted by, or, naturally equipped with, their essence (svabhāva), which is the fitness
(yogyatā) to bring about its perception as their effect. It is only on an epistemic level
that one can, seeing an existing empty purse, infer the non-existence of money from its
non-perception on the grounds of the non-existence of the impediments of its perception,
because the existence of money in a purse without impediments naturally possesses the
fitness to be perceived.
That being the case, we may construe even the avinābhāvaniyama in the c-pāda of
PV 1.31, which Dharmakīrti inserts as one of the two “intermediate (or, transitive) verses”
(antaraślokau) after finishing a discussion about anupalabdhihetu (PVSV 19,23–20,13 on
PV 1.29–30). He does this in order to introduce a new section, referring to the ontological
restriction in either one of the two positive forms, but not the epistemic restriction “of”
avinābhāva to the three modes of conceptual transition. This is because in the ab-pādas of
the same verse Dharmakīrti declares avinābhāvaniyama to be brought into existence in
“either” (vā) one of the two positive forms: kāryakāraṇabhāvād vā svabhāvād vā niyāmakāt,
as I argued in K. Yoshimizu 2007a: n. 62.55 Accordingly, as far as PV 1 and the PVSV
are concerned, we may conjecture that Dharmakīrti uses niyama in the compound avinābhāvaniyama “[natural] restriction [that accounts] for [epistemic] inseparability” as an
ontological term allied with svabhāvaniyama and svabhāvapratibandha when he concisely
asserts that the restriction of the nature of things in either one of the two positive forms
brings about epistemic inseparability.
Finally, let me quote the final phrase of K. Yoshimizu 2007a: 1100:
Dharmakīrti revises Kumārila’s concept of niyama as advocated in the BṬ
by ontologically confining it to two kinds, tādātmya and tadutpatti, both of
which he later asserts not established between a word and its meaning (PV I v.
336) in order to disprove the Mīmāṃsā doctrine that a word is intrinsically
(autpattika) related to its meaning.
55
See also Ch. Yoshimizu 2011, Appendix 1.
Index
Abhayākaragupta: 307
Abhidhammatthasaṅgaha: 31
Abhidhammāvatāra: 31
Abhidharmahṛdaya: 28
Abhidharmakośa: 22–24, 28, 91, 95, 169,
302, 365, 366, 539
Abhidharmakośabhāṣya: 27, 28, 87, 169,
348, 349, 539
Abhidharmakośavyākhyā: 22, 87, 170,
302, 348
Abhinavagupta: 364, 369, 443, 502,
505–507, 510
Abhisamayālaṃkāra: 301, 307
Advaitabinduprakaraṇa: 461
Aghoraśiva: 500, 501
Ajaḍapramātṛsiddhi: 498
Akalaṅka: 111, 113, 114, 117, 126, 422
Alaka: 508
Ālambanaparīkṣā: 203
Ānandapūrṇa: 71
Anekāntajayapatākā: 160, 340
Anekāntajayapatākāsvopajñavyākhyā:
160, 167
Aṅguttaranikāya: 365
Aniruddha: 502
Antarvyāptisamarthana: 182, 186–188,
307, 375
Anyāpohavicārakārikā (Śubhagupta): 340
Āpastambaśrautasūtra: 531
Arcaṭa: 10, 375–377, 380, 381, 383, 384,
386, 411, 421, 422, 432
Āryavimuktisena: 307
Asaṅga: 465
Aṣṭādhyāyī: 7, 11, 317–325, 327, 356, 357,
536
Aṣṭasahasrikāprajñāpāramitāsūtra: 464
Āśvalāyanaśrautasūtra: 532
Atharvavedasaṃhitā: 531
Ātmasiddhi: 449
Atthasālinī: 31
Avadānaśataka: 465
Bahirarthaparīkṣā: 204
Bāhyārthasiddhikārikā: 215, 446, 449,
450, 460
Bāhyārthavādin: 260
Berkeley: 461
Bhāradvājaśrautasūtra: 531
Bhartṛhari: 314–318, 321, 322, 324, 325,
327, 345, 348
Bhāsarvajña: 391, 401, 404, 405, 407, 450,
451
Bhaṭṭanārāyaṇa: 507, 508
Bhāvanākrama: 31
Bhāvaśuddhi: 71
Bhāviveka: 42, 43, 203, 397, 463–471
Blo ldan shes rab: 354, 357
Bodhicaryāvatārapañjikā: 503
Brahmasiddhi: 43, 47, 53, 62
Brahmasiddhivyākhyā: 62
Brahmasūtrabhāṣya: 444, 445, 447, 460
Bṛhadāraṇyakopaniṣad: 63
Bṛhaspati: 499, 502
Bṛhatī: 46
Bṛhaṭṭīkā: 42, 45, 46, 52, 53, 55, 56, 58,
64, 65, 204, 529, 530, 540, 542,
547
Bsdus chen gyi rnam bzhag: 306
Bu ston: 133
Byang chub shes rab: 354
Candrakīrti: 395
Caraka: 393
Carakasaṃhitā: 391–396, 407
Cārvāka: 21, 27, 31–33, 83, 84
Cittamātra: 305
Dar ma rin chen: 307
550
Daśabhūmikasūtra: 463
Daśavaikālikasūtrahāribhadrīvṛtti: 412
Dbu ma rgyan zin bris: 301, 304–306
Dbu ma snang bshad: 149
Dbu ma tshig gsal gyi ti ka: 133
Descartes: 453, 459
Devendrabuddhi: 25, 131, 159, 160, 268,
277, 345, 350–352, 450, 525
Dge lugs pa: 301, 307, 309, 310
Dharmakīrti: 1, 3–5, 8, 9, 27, 29, 32,
41, 42, 46–48, 51–54, 59, 62,
63, 81–89, 91–95, 111–120,
124–127, 129, 130, 132–137,
140, 143, 147, 148, 155–158,
160, 163–173, 177, 181–183,
195, 196, 203–205, 207–211,
214–216, 251–257, 259–264,
267–271, 275, 276, 283–289,
291–295, 297, 298, 301–304,
309, 310, 332, 340, 343,
344, 346, 349–353, 355, 358,
363–368, 370, 375, 376, 378,
383, 384, 386, 391, 397–401,
403, 405, 407, 411, 412, 416,
422, 429–438, 441–448, 450,
451, 454, 455, 460, 463, 465,
473, 485, 486, 493, 497–501,
503–506, 513, 514, 518–527,
529, 530, 538, 540, 541, 545–547
Dharmapāla: 148, 167, 518
Dharmaśrī: 28
Dharmottara: 92, 96, 130–135, 139, 148,
177, 208, 224, 227, 234, 238,
239, 263, 267, 271–278, 350,
375–387, 450, 452, 514, 523–525
Dharmottarapradīpa: 252, 272–274
Dignāga: 5, 25, 41, 112, 130, 131,
168–170, 173, 191–193, 195,
197–199, 201, 203, 204, 206,
215, 216, 257–259, 261,
263, 268–270, 276, 284, 303,
313–317, 319–321, 327, 328,
343–346, 348–353, 358, 375,
411, 412, 415–423, 429, 435,
437, 441, 463, 465, 466, 473,
Index
481–483, 485, 486, 488, 489,
492, 493, 500, 501, 513–521,
525–527, 530
Dīpaṃkararakṣita: 354
Dka gnas: 135, 142, 144
Durvekamiśra: 96, 272, 273, 276, 384
Gcig du bral gyi rnam bzhag: 305, 306
Glo bo mkhan chen: 301, 310
Haravijaya: 508
Haribhadrasūri: 160, 167, 307, 477, 478
Helārāja: 56, 317, 321, 324, 325
Hetubindu: 80, 164, 165, 182, 376, 386,
412, 430, 432–435
Hetubinduṭīkā: 10, 167, 183, 376, 384,
411, 418, 421, 432
Hetubinduṭīkāloka: 10, 384
Hetumukha: 416, 437
Hetuprakaraṇa: 52
Hetutattvopadeśa: 486
Hiraṇyakeśiśrautasūtra: 531
Huizhao: 489, 490
Inmyō ronsho myōtō shō: 482
Īśvarakṛṣṇa: 500
Īśvarapratyabhijñāvivṛtivimarśinī: 254
Īśvarasena: 87, 435
Jaimini: 4, 43, 56
Jaiminīyabrāhmaṇa: 531
Jain: 42, 111
Jayanta: 55, 56, 65, 346, 347, 354, 358,
509
Jinendrabuddhi: 148, 170, 191–197, 199,
200, 254, 259, 314, 316, 318, 320,
321, 324, 326, 345, 346, 349, 358,
411–415, 417–420, 422, 423, 516
Jitāri: 1–11, 16, 42, 486
Jñānagarbha: 132
Jñānaśrībhadra: 133
Jñānaśrīmitra: 42, 177–188, 375, 386,
441, 461
Jñānaśrīmitranibandhāvali: 177–185, 187,
188
Kaiyaṭa: 344, 356, 357
Index
Kamalaśīla: 31, 32, 42, 43, 52, 56, 64,
130–132, 148, 149, 167, 177,
204, 210, 212, 215, 331–334,
337–340, 444, 449, 450, 460
Karṇakagomin: 80, 96, 284, 285, 291, 340,
356, 529, 546
Kāśikā (Sucaritamiśra): 206–209, 254
Kāśikāvivaraṇapañcikā: 11
Kāśikāvṛtti: 7, 11
Kiraṇatantra: 497, 499
Kiraṇavṛtti: 503
Kṣaṇabhaṅgādhyāya: 177, 178
Kṣaṇabhaṅgasiddhi (Dharmottara): 376,
377, 380, 382, 384–386
Kṣaṇabhaṅgasiddhi (Ratnakīrti): 185
Kṣemarāja: 363, 364, 368–370, 497, 498,
504–509
Kuiji: 481, 482, 485, 488–490
Kumārila: 42, 43, 45, 46, 52–54, 56,
58, 61, 64, 71, 87, 203, 204,
207–211, 214–216, 349, 444,
455, 529–532, 535–542, 547
Laghīyastraya: 111, 114
Lalitavistara: 344
Mādhava: 191–195, 197, 198, 200, 201
Madhyamaka: 130–132, 305
Madhyamakahṛdayakārikā: 43, 463–471
Madhyamakālaṃkāra: 131, 301, 305
Madhyamakālaṃkārapañjikā: 131, 148
Madhyamakāloka: 130, 132, 148, 149
Madhyamakāvatāra: 395
Mādhyamika: 260, 309, 463, 464,
469–471
Mahābhārata: 532
Mahābhāṣya: 301, 302, 310, 318–320,
322, 344, 345, 357, 536
Mahābhāṣyapradīpa: 344
Mahābhāṣyapradīpoddyota: 344
Mahāvibhāṣā: 28
Mahāvyutpatti: 91, 95, 356
Mahāyānasūtrālaṃkāra: 302
Mahāyānasūtrālaṃkārabhāṣya: 302, 344
Majjhimanikāya: 29, 86
551
Mālinīvijayottara: 504
Mānavaśrautasūtra: 531
Maṇḍana: 42–44, 52–57, 59, 62–64, 66
Māṇikyanandi: 111, 114, 115, 117–120,
125–127
Manoramāṭīkā: 506
Manorathanandin: 25, 258, 259, 271, 291,
364, 370, 443, 498, 503
Mantramārga: 497, 501
Mataṅgapārameśvara: 497, 498, 500–502
Mataṅgapārameśvaravṛtti: 497, 498
Mi pham zhal lung: 309
Mīmāṃsā: 3, 4, 42, 89, 529–531, 542, 547
Mīmāṃsaka: 42, 43, 54, 56, 57, 444,
529–532
Mīmāṃsāsūtra: 42, 56–58, 532, 538
Mīmāṃsāsūtrabhāṣya: 3, 4
Mokṣākaragupta: 375
Mun sel: 130, 137–140, 142–144,
146–149
Nāgārjuna: 395, 397, 442
Nāgeśa: 344
Naiyāyika: 57, 118, 132, 135, 188, 271,
392, 401, 407, 416–418
Nārāyaṇa: 344, 348
Nārāyaṇakaṇṭha: 502
Nārāyaṇīya: 344
Nareśvaraparīkṣā: 499
Nayacakravālavṛtti: 326
Netratantroddyota: 504
Netroddyota: 498
Nirukta: 345
Niśvāsa: 363, 367
Niśvāsatattvasaṃhitā: 363, 364, 367, 497,
500, 501
Niyogakāṇḍa: 62
Nyāya: 260, 391, 404, 520–524
Nyāyabhāṣya: 270, 348, 391, 393,
395–397, 401, 407
Nyāyabhūṣaṇa: 391, 401, 404, 405, 407,
450
Nyāyabindu: 47, 49, 51, 53, 54, 59, 62,
85, 92–94, 96, 111, 116, 117, 119,
552
120, 126, 129, 252, 263, 267–270,
272, 418, 486
Nyāyabinduṭīkā: 92, 96, 252, 271–275,
376, 418
Nyāyakaṇikā: 42–44, 46–48, 63, 386
Nyāyakumudacandra: 460
Nyāyamañjarī: 55, 56, 254
Nyāyamukha: 131, 148, 180, 353,
412, 473, 474, 476, 478–480,
482–486, 513–515, 517, 518,
525, 526
Nyāyapraveśa: 148, 416, 473–478, 480,
482–486, 489–493
Nyāyapraveśakaṭīkā: 476, 478, 479, 486
Nyāyapraveśakavṛttipañjikā: 478, 486
Nyāyaratnākara: 206, 208
Nyāyasūtra: 43, 270, 391–393, 395–398,
400, 401, 403, 404, 407, 416, 418,
419
Nyāyavārttika: 391, 396, 397, 401, 416
Nyāyavārttikatātparyapariśuddhi: 403
Nyāyavārttikatātparyaṭīkā: 43, 386, 391,
401, 404, 416
Nyāyāvatāra: 422
Nyāyāvatāravivṛti: 422
Od zer: 130, 133, 136, 137, 139, 140,
142–145, 148
Paddhati: 324
Pakṣilasvāmin: 348
Pañcavastukavibhāṣāśāstra: 22
Pāṇini: 318
Parahitabhadra: 132
Parīkṣāmukha: 111, 117, 119, 120, 125,
126
Pārthasārathi: 206, 541
Pāśupata: 367
Patañjali: 44, 301, 319, 320, 344, 356, 357,
536
Pātrasvāmin: 421
Phya pa chos kyi seng ge: 130, 132–151
Prabhācandra: 111, 114, 115, 117, 449
Prabhākara: 46
Pradīpa (Kaiyaṭa): 318
Index
Prajñākaragupta: 25, 42, 59, 64, 130, 131,
133, 135, 155, 156, 160, 161, 172,
177, 208, 221–247, 257–260,
267, 271, 274–278, 344, 346,
347, 350, 352–358, 498, 514,
524, 525
Prajñāpāramitāsūtra: 464
Prakāśa (Helārāja): 56, 317, 321, 322, 324,
325, 327
Pramāṇanayatattvālokālaṃkāra: 111, 115
Pramāṇasaṃgraha: 422
Pramāṇasamuccaya: 5, 24, 131, 148, 168,
169, 180, 191, 193, 198–201,
206, 211, 213, 258, 259, 263,
268–270, 313–321, 324–326,
328, 343, 348, 349, 352, 353, 412,
413, 415–417, 420, 422–424,
466, 473, 479, 480, 483–486,
492, 513–517, 519, 520, 525,
526, 536, 537
Pramāṇasamuccayaṭīkā: 5, 169, 170, 180,
191–193, 196–198, 201, 254,
314–316, 318, 320, 324, 345,
346, 349, 350, 411, 413–415,
417–424, 516
Pramāṇasamuccayavṛtti:
191–193,
196–198, 200, 201, 206, 303,
348, 349, 352, 353, 411–413,
415, 417–420, 483, 516
Pramāṇavārttika: 1, 3, 4, 8, 9, 21, 25,
27–32, 41, 46, 47, 49, 51–55, 59,
60, 62–65, 77, 79–86, 93, 120,
129, 131, 135, 147, 155–160,
164, 169–173, 203, 212,
214, 221–225, 229–236, 239,
242, 243, 245–247, 251–254,
256–263, 267–271, 274–277,
283–298, 301, 302, 304, 305,
309, 332, 345, 350–352, 354,
363–370, 416, 430, 434, 436,
442–446, 448, 460, 497, 503,
505, 513, 514, 518, 519, 525,
529, 540, 541, 545–547
Pramāṇavārttikālaṃkāra: 64–66, 135,
157, 158, 160, 161, 163, 168, 172,
Index
208, 223, 230–235, 237–246,
258, 260, 271, 274–277, 346,
347, 352–355
Pramāṇavārttikapañjikā: 157, 159, 160,
271, 278, 345, 346, 350, 351
Pramāṇavārttikasvavṛtti: 47, 51–54,
77–81, 85, 87–89, 91–95,
111, 117, 131, 157, 163, 167,
168, 171, 180, 212, 284–287,
292–295, 332, 340, 355, 416,
430, 431, 433, 434, 436–438,
515, 518, 519, 541, 545–547
Pramāṇavārttikasvavṛttiṭīkā: 80, 93, 94,
96, 157, 164, 285–287, 290,
292–296, 340, 356, 434, 529, 546
Pramāṇavārttikaṭīkā: 93, 95, 157, 164,
278, 287, 289, 332, 340, 345, 350,
351, 356, 434
Pramāṇavārttikavṛtti: 4, 9, 258, 271, 291,
364, 370
Pramāṇavārttikavṛtti (Ravigupta): 157,
160
Pramāṇaviniścaya: 3, 5, 10, 41, 47,
54, 62, 63, 80, 86, 129–136,
142, 145, 147–149, 180, 203,
207, 208, 210–212, 251–257,
259–263, 267–270, 353, 376,
378, 383, 386, 433, 442, 444, 445,
447, 448, 450, 451, 454, 460, 514,
518–522, 525, 527
Pramāṇaviniścayaṭīkā: 133, 134, 148,
376–380, 382–386, 452, 525
Prāmāṇyaparīkṣā (Dharmottara): 350
Prameyakamalamārtaṇḍa: 111
Pratyabhijñā: 364
Prayogahṛdaya: 391
Prayogasāra: 391
Pudgalavādin: 24
Puṃpāśeśvaraprakaraṇa: 501
Puṇyarāja: 323
Rāmakaṇṭha: 497–500, 502
Ratnākaraśānti: 3, 182
Ratnakīrti: 42, 55, 178, 185, 375, 386
553
Ratnakīrtinibandhāvali: 4, 178–180, 183,
185
Rauravasūtrasaṃgraha: 499, 500
Ravigupta: 25
Ṛgveda: 532, 540
Rgyal tshab dar ma rin chen: 94, 95, 133
Rigs gter: 303
Rnam bshad snying po rgyan: 307
Rngog blo ldan shes rab: 132, 135, 142,
144, 145
Ryakken inmyō nisshōri mon: 488
Sa skya paṇḍita: 25, 301, 303, 310
Śabara: 4, 206, 211, 530, 538
Śābarabhāṣya: 4, 205, 206, 211, 530, 531,
538
Śabdārthaparīkṣā: 331
Śābdika: 490
Ṣaḍdarśanasamuccaya: 398
Sadyojyotis: 364, 499, 502
Saiddhāntika: 498–502, 507–510
Śaiva: 363, 364, 366, 368, 369, 497–502,
504, 505
Śaivas: 367, 443
Śaivasiddhānta: 497, 499, 501, 504
Śākyabuddhi: 93–96, 131, 276, 277, 284,
289, 332, 340, 350, 351, 355, 358,
450, 525
Śākyaśrībhadra: 25
Śālistambasūtra: 366
Sāmānyaparīkṣāvyāsa (Dignāga): 191
Sāmaṭa: 43
Sambandhaparīkṣā: 172, 437
Sāṃkhya: 27, 57, 130, 191, 194, 198–201,
490, 501, 513, 514, 516, 517, 526
Sāṃkhyakārikā: 500
Sāṃkhyanāśaka: 191
Sāṃkhyaparīkṣā (Dignāga): 191
Sāṃkhyas: 516
Sāṃkhyavaināśika: 192
Saṃyuttanikāya: 23
Śaṅkara: 442, 444, 445, 447, 460
Śaṅkaranandana: 42, 55
Śaṅkarasvāmin: 485
Śaṅkhapāṇi: 62
554
Śāṅkhāyanaśrautasūtra: 532
Śāntabhadra: 133
Santānāntarasiddhi: 340, 341
Santānāntarasiddhiṭīkā: 340, 341
Śāntarakṣita: 32, 41–43, 52, 58, 131, 204,
207, 209–216, 305, 331–334,
337–340, 397, 407, 421, 431,
444, 450
Sārdhatriśatikālottara: 499
Sarvajñānottara: 499, 500
Sarvajñasiddhi (Jñānaśrīmitra): 42
Sarvajñasiddhi (Ratnakīrti): 42, 48, 55
Sarvajñasiddhi (Śaṅkaranandana): 42, 55,
297
Sarvajñasiddhi (Śubhagupta): 42
Sarvajñasiddhisaṃkṣepa: 42
Sarvāstivāda: 27, 539
Sarvāstivādin: 490
Satyadvayavibhaṅgakārikā: 132
Satyadvayavibhaṅgavṛtti: 132
Sautrāntika: 2, 9, 27, 28, 251–254, 258,
259, 261–263, 269, 305, 309,
442–444, 539
Shentai: 485, 486
Siddhāntatantra: 501
Siddhiviniścaya: 422
Śikṣāsamuccaya: 354
Śivasūtravārttika: 498
Ślokavārttika: 42, 50, 52, 54–56, 58, 59,
61, 64, 65, 203–209, 211, 214,
254, 444, 455, 530, 536–540
Sras bsdus grwa: 306
Stavacintāmaṇi: 507, 508
Stavacintāmaṇivivṛti: 507–509
Sthiramati: 302
Śubhagupta: 3, 42, 204, 215, 340, 446,
449, 460
Sucarita: 208
Sumati: 354
Supariśuddhā: 66, 208
Sūtrālaṃkārabhāṣya: 302
Sūtrālaṃkāravṛttibhāṣya: 303
Svacchandatantra: 363, 497
Svacchandatantroddyota: 367–370
Svacchandoddyota: 505, 507
Index
Svāyambhuvasūtrasaṃgraha: 367, 497,
499
Tantrāloka: 369, 502, 505, 507
Tantrālokaviveka: 498
Tantrarāja: 506
Tantraratna: 541
Tantrasāra: 506
Tantravārttika: 530–532, 535, 536, 538,
540–542
Tarkabhāṣā: 2, 183, 254, 375
Tarkajvālā: 43, 464–466, 470
Tarkarahasya: 157, 159
Tātparyaṭīkā (Umbeka): 59, 206–208
Tattvasaṃgraha: 2, 3, 5, 6, 32, 42,
43, 45, 50, 52–56, 58, 64, 65,
167, 172, 203–206, 208–215,
331–334, 336–339, 451, 460
Tattvasaṃgrahapañjikā: 2, 3, 6, 43, 56,
58, 167, 206, 208–210, 212–215,
331–339, 421, 451, 460
Tattvasamīkṣā: 387
Ṭhāṇaṃgasutta: 114
Thar lam gsal byed: 304
Thub bstan lhun poi mdzes rgyan: 305,
306
Ṭīkā (Punyarāja): 323
Tshad ma bsdus pa: 144
Tsong kha pa: 301, 304–307, 310
Ṭupṭīkā: 540–542
Udayana: 392
Udayananirākaraṇa: 157
Uddyotakara: 204, 391
Umbeka: 59, 207, 210, 216
Upāyahṛdaya: 391, 393, 395–397, 407
Utpaladeva: 498
Uttaramīmāṃsā: 71
Uttaratantra: 302
Vācaspati: 42–44, 53, 63, 386, 401,
403–405, 407
Vādanyāya: 51, 80, 165–167, 171, 376,
386, 391–393, 395, 397–405,
407, 430, 431, 433–436
Vādanyāyaṭīkā: 166, 183, 421
555
Index
Vādanyāyaṭīkā Vipañcitārthā: 163, 166,
167, 171, 172, 397–400, 407, 432
Vādarahasya: 157
Vādavidhāna: 476
Vādavidhi: 481
Vādi Devasūri: 111, 114, 115, 117
Vaibhāgika: 191–193, 201
Vaibhāṣika: 3, 9, 27, 28, 305, 449
Vaināśika: 191, 192, 201
Vaiśeṣika: 113, 132, 135, 147, 148, 150,
157, 305, 322, 513, 519–524, 537
Vaiśeṣikasūtra: 114
Vaiyākaraṇasiddhāntakaumudī: 318
Vākyapadīya: 56, 157, 315–318, 321–325,
327
Valmīkisutta: 86
Vārttika (Kātyāyana): 344
Vasubandhu: 22, 24, 28, 87, 302, 348, 397,
443, 465, 473, 476, 481, 485–487,
493
Vasumitra: 22
Vātsīputrīya: 304
Vātsyāyana: 270, 391, 396, 403
Vibhāṣā: 169, 170
Vibhūticandra: 346, 356, 357
Vibhūticandra (Notes): 157, 159, 160,
346, 350, 357
Vidhiviveka: 42–44, 46, 47, 51–54, 56–59,
61, 62, 65, 66, 71
Vijñānavāda: 257
Vijñaptimātratāsiddhi: 1–11, 16
Viṃśatikāvṛtti: 463
Vinītadeva: 133, 166, 252, 340
Visuddhimagga: 31
Viśvanātha: 419
Vṛttikāra: 455
Vyākaraṇamahābhāṣya: 536
Vyāpticarcā: 188
Wengui: 485
Xuanzang: 485, 489
Yajñaṭa: 43
Yamāri: 66, 208, 275, 346, 354–358
Yāmuna: 449
Yāska: 344
Yaśomitra: 22, 170, 302
Yin ming da shu: 481–483, 485, 487–490
Yin ming ru zheng li lun shu: 485
Yin ming zheng li men lun shu ji: 485,
486
Yogācāra: 9, 258, 260, 305, 309, 444, 455,
463–466, 469–471
Yuktidīpikā: 419, 500
Zenju: 482
Zhang thang sag pa: 133