Chinese Translations of Pratyakṣa

Chinese Translations of Pratyakṣa

Funayama Toru

The theory of direct perception was one of the most fundamental topics in Buddhist epistemology. It became more important particularly with Dignāga (ca. 480-540 C.E.), who advocated the ground-breaking theory of pramāṇa (lit. “measure”, “scale”, or “standard”) or the “means of valid cognition”. As is well known, pratyakṣa (“direct perception”) and anumāna (“inference”) constitute the two most important elements of this theory.

It was Xuanzang (玄奘, 600/602-664) who laid the foundation for the study of pramāṇa in China. In the Chinese context this study was called yinming (因明, “science of logic”, *hetuvidyā), which is often counted as one of five sciences (Skt. pañcavidyā, Ch. wuming 五明), by contrast to pramāṇavāda (“the theory of pramāṇa”) which belongs to the Sautrāntika and/or the Yocācāra position in Buddhist logic and epistemology. Although Xuanzang was not the first person to bring Dignāga’s views to

China, it was with him that study of this topic in China began in earnest.

Xuanzang translated pratyakṣa as xianliang (現量). In this paper I would like to consider the historical situation before and after Xuanzang’s adoption of this translation. The first half of this paper will be spent examining earlier appearances of the term in question, before Xuanzang’s time. As I will show, this topic is important for at least two reasons. First, the term xianliang is, strictly speaking, not a literal translation of pratyakṣa, despite Xuanzang’s general tendency to give fully literal translations. Second, it is not yet clear who first used this term. It is clear, however, that Xuanzang was not the first person. In the second half of this paper, I will shift to the post-Xuanzang period, paying special attention to some Chinese interpretations of xianliang. Developments in this period are possibly related to the process of the “sinification of Buddhism”, in the sense that Chinese works during the Tang and Ming dynasties showed some serious discrepancies from the Indian Buddhist tradition, and began to develop peculiar Chinese interpretations of this word.

1 Xianliang as translation

I will begin with some observations about Xuanzang’s usage of xianliang. This term is used in some important translations by Xuanzang, such as the Yinming zhengli men lun (因明正理門論, Nyāyamukha of Dignāga), the Yinming ru zhengli lun (因明入正理論, Nyāyapraveśa[ka] of Śaṅkarasvāmin) and the Apidamo jushe lun (阿毘達磨倶舍論, Abhidharmakośa[bhā-

ṣya] of Vasubandhu), in which xianliang is obviously used as a translation of pratyakṣa. For example, Dignāga’s well-known definition of direct perception, viz., pratyakṣaṃ kalpanāpoḍham (“Direct perception is free from conceptual construction”), is rendered by Xuanzang as xianliang chu fenbie (現量除分別).

Modern readers of Xuanzang’s translations generally take xianliang as an unproblematic translation of pratyakṣa. Strictly speaking, however, the term xianliang is not a literal translation, because liang (量), which literally means “measure”, “measurement”, “scale”, or “amount/quantity”, as either a verb or a noun, corresponds to pramāṇa. According to the Index to the Abhidharmakośabhāṣya compiled by Hirakawa et al., the word pratyakṣa corresponds in some of Xuanzang’s texts to other Chinese terms as well: e.g., xianjian (現見). The Xianyang shengjiao lun (顯揚聖教

論), one of Xuanzang’s earliest translations, is consistent in the usage of xianliang for pratyakṣa. On the other hand, yet another important text, the Yuqie shi di lun (瑜伽師地論, Yogācārabhūmi), the search for which inspired Xuanzang’s journey to India, occasionally has xian (現) and xianzheng liang (現證量) as translations of pratyakṣa, although in many cases it prefers xianliang. These examples show that for Xuanzang, though xianliang was certainly a well-established translation of pratyakṣa, it was not necessarily the only Chinese translation, as modern scholars often assume.

The matter of wording with or without liang, per se, is not at all a serious problem, but it is interesting to imagine why Xuanzang might have wanted to add it to his translation of pratyakṣa. As a rigorous Sanskritist, Xuanzang by and large preferred literal translation; he did not like adding words and thereby embellishing the meaning of the original term. In this sense, it is worthwhile to ask whether xianliang was a translation newly created by Xuanzang himself. As it turns out, it was not. There are some noteworthy examples of xianliang before Xuanzang’s time.

Before Xuanzang came back from India, Prabhākaramitra (波羅頗蜜多羅, 565-633, also called 波羅頗迦羅蜜多羅) had employed xianliang (現量) and yanliang (驗量) for pratyakṣa and anumāna, respectively, in his

translation of Bhāviveka’s commentary, Prajñāpradīpa, on the Middle Treatise. Even prior to this time, at the end of the sixth century, members of the Dilun (地論) school had already used the term. A typical example is found in Jingying Huiyuan’s (淨影寺慧遠, 523-592) doxographical compendium entitled Dasheng yi zhang (大乘義章). Huiyuan summarizes the theory of the threefold classification of pramāṇa as follows:

First, the Exposition of Names [as follows:] The doctrine of the three measurements [of valid cognition] comes from the Saṃdhinirmocana-sūtra. When the mind of wisdom grasps entities, each has its own portion. Therefore it is called “measurement” [or “amount”]. There are manifold ways of distinguishing [types of] “measurement”: one says that there are three: first, direct measurement; second, measurement as inference; and third, measurement as teaching. (第一釋名。三量之義出於《相續解脱經》中。慧心取法，各有分限，故名為量。量別不同，一門説三：一是現量，二是比量，三是教量; T44:1851.670c7-9) It is clear here how Huiyuan understands xianliang. Further, the term fenxian 分限, which appears here in the explanation of liang, is also interesting, because fenxian means “portion” or “amount”, which reminds us more of Sanskrit words like parimāṇa, rather than pramāṇa. At least, it seems true that by liang Huiyuan took the term to mean “measure” in the broad sense, and not specifically “a means of valid cognition”, as it was defined in later texts composed by Dignāga and his followers.

Moreover, members of the Dilun school used xianliang in other texts too. For example, the Dunhuang manuscript S.613v mentions, as a Dilun theory, a fourfold classification of pramāṇa: xianliang (現量), biliang (比 量), xinyanliang (信言量), and jiaoliang (教量). Further, in the above

quotation, Huiyuan refers to the Saṃdhinirmocana-sūtra, but this sūtra’s wording is not the same as his. Guṇabhadra’s (求那跋陀羅, 394-468) translation has xianqian (現前) for *pratyakṣa (Tib. mṅon sum) and xianqian liang (現前量) for *pratyakṣaṃ pramāṇam (Tib. mṅon sum gyi tshad ma).8 Bodhiruci’s (菩提流[/留]支, d. 527) translation, on the other hand, has xianqianjian (現前見) and xianjian (現見) for pratyakṣa.9 Returning to Xuanzang, we may assume that his use of xianliang was influenced by the preceding scholastic tradition, as found in Huiyuan’s compendium, and also in Prabhākaramitra’s translation.

Here a question arises. Though it is evident that the Dilun employed the term xianliang in their writings, what kind of textual basis did they have in translated texts? This is an interesting question because, as stated just above, in spite of Huiyuan’s explicit reference to the Saṃdhinirmocana-sūtra as the source for the theory of the threefold classification, neither Guṇabhadra’s nor Bodhiruci’s translations have xian-

(i.e., pratyakṣa “direct perception”, anumāna “inference”, āptavacana (or āgama) “words of a reliable person”, and upamāna “analogy”). It is a peculiarity of the Dilun fourfold classification that they distinguish xinyanliang and jiaoliang, which are usually regarded as identical. This view is criticized even within the Dilun School. See Dasheng yi zhang 10 (T44:1851.671b4-6; pointed out in Funayama, op. cit.: 153 n. 39). According to Aoki Takashi (Aoki, 2000: 194, 198-201), the date of S.613v is around 560 or later. As pointed out in Funayama, op. cit., the Dilun classification of pramāṇa into these four kinds is also found in S.4303, another important Dilun text, which was, according to Aoki (loc., cit.), composed later than S.613v and before ca. 585 CE. 8 Guṇabhadra: 一切行無常，一切行苦，一切法無我，若世間現前得，如是等名現前

得相 (T16:679.719a14-16). Tib: de la ’du byed thams cad mi rtag pa ñid daṅ | ’du byed thams cad sdug bsṅal ba ñid daṅ | chos thams cad bdag med pa ñid ’jig rten na mṅon sum du dmigs pa daṅ | de lta bu daṅ mthun pa gaṅ yin pa de ni de mṅon sum su dmigs pa’i mtshan ñid yin no || (Lamotte, 1935: 156). Cf. Xuanzang: 一切行皆無常性，一切行皆是苦性，一切法皆無我性，此為世間現量所得 (T16:676.709b25-26). Guṇabhadra: 若此助成如是現前量比量信言量，是名五種快淨相 (T16:679.719b5-6). Tib: de ltar ’thad pa’i sgrub par pa’i rigs pa de ni mṅon sum gyi tshad ma daṅ | rjes su dpag pa’i tshad ma daṅ | yid ches pa’i luṅ gi tshad mas mtshan ñid lṅa po dag gis yoṅs su dag pa yin no ||

(Lamotte, 1935: 157). Cf. Xuanzang: 如是證成道理，由現量故，由比量故，由聖教量故。由五種相，名為清淨 (T16:676.709c28-710a1). 9 Bodhiruci: 一切有為行無常，一切有為行苦，一切法無我，世間現前見法，如是等是名彼現前見相 (T16:675.686b7-8); 此依生成相應，現見相應，量相應，比智相應，聖人説法相應，知五種相，是名清淨相 (686c2-4). See also the previous note. liang. Therefore, it is still unclear at this juncture what source Huiyuan drew upon in employing xianliang. In what follows, I will try to survey translations of pratyakṣa in the Six Dynasties period.

The earliest texts in which I have identified translations of pratyakṣa and pramāṇa are attributed to Kumārajīva. Investigation into the question of whether there exist any earlier translations than Kumārajīva will be a task for future research. (a) Kumārajīva (鳩摩羅什, ca. 350-409) – xin (信) or kexin (可信) for *pramāṇa; xianshi (現事) or xianzai (現在) for *pratyakṣa 中論》 3, 觀法品（青目釋）: 信有四種。一現事可信。二名比知可信，如見烟知有火。三名譬喩可信，如國無鍮意志，喩之如金。四名賢聖所説故可信。如説有地獄，有天，有欝單曰，無有見者， 信聖人語，故知 (T30:1564.24a28-b3). 《成實論》 8: 有三種信。見名現在信。聞名信賢聖語。知名比知… (T32:1646.304a7-8). (b) Tanwuchen (曇無讖, 385-433) – xianzhi (現智) for pratyakṣa 《菩薩地持經》 1, 真實義品: 云何學所知真實義耶。如世智人依現智比智及從師聞思量修學，彼決定智所行處事，結集建立。是名學所知真實義也 (T30:1581.893a1-3). Bodhisattvabhūmi, Tattvārthapaṭala (Wogihara, 1936: 37, 22-38, 1; Dutt, 1966: 25, 17-21): yuktiprasiddhatvaṃ katamat. satāṃ yuktārthapaṇḍitānāṃ vicakṣaṇānāṃ tārkikāṇāṃ mīmāṃsakānāṃ tarkaparyāpannāyāṃ bhūmau sthitānāṃ svayaṃ pratibhānikyāṃ pārthagjanikyāṃ mīmāṃsānucaritāyāṃ pratyakṣam anumānam āptāgamaṃ pramāṇaṃ niścitya suviditasuviniścitajñānagocarajñeyeyaṃ [D: suviniścitajñānagocaro jñeyeyaṃ W] vastū-

papattisādhanayuktyā prasādhitaṃ vyavasthāpitam, idam ucyate yuktiprasiddhaṃ tattvam. Cf. 玄奘譯《瑜伽師地論》 36, 菩薩地・真實義品: 云何道理極成真實。謂諸智者有道理義。諸聰叡者諸黠慧者。能尋思者能伺察者。住尋伺地者具自辯才者。居異生位者隨觀察行者。依止現比及至教量極善思擇決定智所行所智事。由證成道理所建立所施設義。是名道理極成真實 (T30:1579.486b27-c3). Cf. 求那跋摩 (367-431) 譯《菩薩善戒經》2: 云何名方便流布。如世智人先以籌量，然後造作經書論義，是名方便流布 (T30:1582. 968b7-9). 菩薩地持經》 3, 力種性品: 現智比智，從師具聞而為人説，非不思量 (T30:1581.904c23-24). Bodhisattvabhūmi (Wogihara, 1936: 106, 24-25; Dutt, 1966: 75, 10-11): pratyakṣānumānāptāgamayuktāṃ ca kathāṃ karoti, nāpramāṇayuktām. Cf. 玄奘譯《瑜伽師地論》38, 菩薩地力種姓品: 又依現比至教道理而説正法，非不依彼三量道理 (T30:1579.503b10-11). No corresponding passage exists in Guṇavarman’s translation (cf. T30: 1566.78c). (c) Guṇabhadra (求那跋陀羅, 394-468) – xianqian (現前) for *pratyakṣa; xianqianliang (現前量) for *pratyakṣaṃ pramāṇam. See the above paragraph. (d) Jijiaye (吉迦夜, ca. 472) – xianjian (現見) for pratyakṣa 方便心論》: 凡欲立義，當依四種知見。何等為四。一者現見，二者比知，三以喩知，四隨經書 (T32:1632.23c29-24a2).

Ibid.: 知因有四。一現見，二比知，三喩知，四隨經書。此四知中，現見為上 (T32:1632.25a26-27). (e) Bodhiruci (菩提流[/留]支, d. 527) – xin (信) for pramāṇa; xianjian (現見) for pratyakṣa; xianxin (現信) for pratyakṣaṃ pramāṇam 唯識論》: 問曰。依信説有。信者有四種。一者現見，二者比知，三者譬喩，四者阿含。此諸信中，現信最勝 (T31:1588.68b25-27). See also (f) Paramārtha’s translation of the same text below. (f) Paramārtha (真諦, 499-569, also called Kulanātha 拘羅那他) – liang (量) for pramāṇa; zheng (證) or zhengliang (證量) for pratyakṣa; zhengliang (證量) for pratyakṣaṃ pramāṇam 大乘唯識論》: 一切量中，證量最勝 (T31:1589.72c22-23). Viṃśikā/Viṃśatikā: sarveṣāṃ ca pramāṇānāṃ pratyakṣaṃ pramāṇaṃ gariṣṭham iti (Lévi, 1925: 8.23). Cf. 玄奘譯《唯識二十論》: 一切量中，現量為勝 (T31:1590.76b15-16). Thus, it is clear that none of the translators above uses xianliang as the translation of pratyakṣa. Among the translations we do encounter in these texts, Paramārtha’s term zhengliang (證量) is peculiar. It translates either pratyakṣa or pratyakṣaṃ pramāṇam, in which zheng signifies “something vivid, clear or direct”. As mentioned above, this wording is some-

times, though not very often, found even in Xuanzang’s translation of the Yogācārabhūmi, in the form of xianzheng liang (現證量). There is, however, a text in which the term xianliang does appear as a translation – the Huizheng lun (迴諍論) or Vigrahavyāvartanī of Nāgārjuna. According to the Translation Record attached to the head of the translation (T32:1631.13b11-21), the Chinese translation was made in the third year of Xinghe 興和 of the Eastern Wei dynasty, i.e. 541 CE. The translators were Pimuzhixian 毘目智仙 and Qutan liuzhi 瞿曇流支 (alias Prajñāruci 般若流支). The relevant section is as follows, shown together with the corresponding Sanskrit text: (g) 《迴諍論》: 又復有義。偈言: 若彼現是有，汝可得有迴。 彼現亦是無，云何得取迴。此偈明何義。若一切法有現可取，汝得迴我諸法令空。而實不爾。何以知之。現量入在一切法數，則亦是空。若汝分別依現有比，現比皆空，如是無現比，何可得現之與比，是二皆無，云何得遮。汝言一切諸法空者，是義不然。 若汝復謂: 或比或喩，或以阿含，得一切法。如是一切諸法自體，我能迴者。此我今説。偈言: 説現比阿含 譬喩等四量 現比阿含成 譬喩亦能成 此偈明何義。比喩阿含現等四量若現能成，比阿含等皆亦能成，如一切法皆悉是空，現量亦空… (T32:1631.16a5-21). kiṃcānyat.

---

na) as conducted on the basis of conceptual cognition (vikalpa, kalpanā, Ch. fenbie 分別).

pratyakṣeṇa hi tāvad yady upalabhya vinivartayasi bhāvān | tan nāsti pratyakṣaṃ bhāvā yenopalabhyante || [5] yadi pratyakṣataḥ sarvabhāvān upalabhya bhavān nivartayati – śūnyāḥ sarvabhāvā iti, tad anupapannam. kasmāt. pratyakṣam api hi pramāṇaṃ svabhāvāntargatatvāc chūnyam. yo ’pi sarvabhāvān upalabhate so ’pi śūnyaḥ. tasmāt pratyakṣeṇa pramāṇena nopalambhabhāvaḥ. anupalabdhasya ca pratiṣedhānupapattiḥ - śūnyaḥ sarvabhāvā iti, tad anupapannam. syāt te buddhir anumānenāgamenopamānena vā sarvabhāvān upalabhya sarvabhāvavyāvartanaṃ kriyata iti. atra brūmaḥ.

anumānaṃ pratyuktaṃ pratyakṣeṇāgamopamāne ca |

anumānāgamasādhyā ye ’rthā dṛṣṭāntasādhyāś ca || [6] anumānam apy upamānāgamāś ca pratyakṣeṇa pramāṇena pratyuktāḥ. yathā hi pratyakṣaṃ pramāṇaṃ śūnyaṃ sarvabhāvānāṃ śūnyatvāt… (Yonezawa, 2008: 228, 230 with stylistic modification by FT). Here, it is evident that xianliang (現量) appears as the translation of pratyakṣaṃ pramāṇam, and not pratyakṣa. This distinction is obvious because pratyakṣa is translated as xian (現). As far as I know, the Huizheng lun is the only translation text prior to Huiyuan which employs xianliang. However, the exact reason for the Dilun school’s preference for the term xianliang is still not entirely clear, because the Huizheng lun is not a major text for the Dilun school, and it is scarcely cited in Dilun compositions. In any case, I find it interesting that even though the actual usage of xianliang is rather limited in translations, the Dilun school preferred it over other options, and that it was then adopted even by Xuanzang, in spite of the fact that the element liang is superfluous as a literal translation of pratyakṣa. In fact, the use of this term meant that Xuanzang could not distinguish between pratyakṣa and pratyakṣaṃ pramāṇam in his translation, since both are translated in the same way as xianliang. This is certainly a small point, but I think it is noteworthy when we take into account the generally rigorous character of Xuanzang’s translations.

In ending this first section, I would like finally to refer to the usage of xian liang in Guṇabhadra’s translation of the Laṅkāvatāra-sūtra (Lengqie jing 楞伽經, T670). In fact, we find that the phrase zixin xian liang (自心現量) appears forty times in the Lengqie jing. In addition to this, a similar expression, zixin xian (自心現, without liang), also appears many times in the same text. When we compare Guṇabhadra’s translation with the extant Skt. text, we can identify zixin xian liang with svacitta-dṛśya-mātra, which means “nothing but what is experienced by one’s own mind”. That is, xian liang is not a single term in this case; xian signifies “to appear, manifest itself” and liang is a translation of mātra “merely, only, nothing but”, often rendered in other translations as wei (唯).

The usage of xian liang in the following passage from the Chinese Laṅkāvatāra is also noteworthy in connection with pramāṇa theory: O Mahāmati! The Nirvāṇa that I teach means the following: one understands with one’s superior awareness what is none other than the manifestation of one’s mind. (Note: Measurement is of four kinds: one, direct perception; two, inferential understanding; three, analogy; and four, what is transmitted by distinguished predecessors. All those non-Buddhists [mentioned previously in the sūtra] are not trustworthy on [the topic of] the four [types of] mensuration.) One does not become attached to the external nature [of ontic entities]; is free from [[[Wikipedia:denial|denial]] by the method of] the tetralemma (catuṣkoṭi); perceives the condition of what exists in accordance with reality, without falling into the two delusory extremes manifested by one’s own mind, so that neither cognizer nor cognized is apprehended; and the nonperception of all [types of] mensuration is accomplished. When one is ignorant of true reality, one is not able to apprehend [it] at all. When one discards [the above-stated erroneous condition of mind], one attains the truth of self-awakened sages; understands the two kinds of no-self; transcends the two kinds of affliction; purifies the two obstacles and removes them; and becomes eternally free from the two

deaths. [Being equipped with] numerous kinds of profound samādhi such as the shadow-and-illusion-like [[[samādhi]] that are acquired in] higher and higher grounds (or stages, bhūmi) [of bodhisattvas and the final] ground of the Thus Come One, one becomes entirely free from mind (citta), mentation (manas) and mental consciousness (manovijñāna). This is called “Nirvāṇa”.23 (大慧，如我所説涅槃者[1]，謂善覺知自心現量[2]〈量[3]有四種。一現見，二比知，三譬喩，四先勝相傳。彼外道，於四度量[4]，悉皆不成[5]也〉，不著外性，離於四句，見如實處，不墮24自心現妄想[6]二邊，攝所攝不可得，一切度量[7]不見所成[8]。愚於真實，不應攝受。棄捨彼已，得自覺聖法，知二無我，離二煩惱，淨除二障，永離二死，上上地如來地如影幻等諸深三昧，離心意意識，説名涅槃, T16:670.505a8-15; underlining and index numbers [1]-[8] added by FT for convenience.) There is an interlinear note (shown by the angle brackets: 〈…〉) after zixin xian liang (自心現量)[2]. This note is a reference to a fourfold classification of pramāṇa (liang 量[3]) which is most probably maintained by

23 This is a tentative translation of the Chinese translation, which is not exactly the same as the Sanskrit text. A closer examination, as well as a comparison of the Sanskrit and the Chinese versions, must await future research. 24 I take the variant duo (堕) in the “Gong (宮, Palace)” edition (i.e., the Kaiyuan si 開元寺 edition, alias the Pilu dazang jing 毘盧大藏經), which fits āpatana in Skt. The Korean edition (i.e., both the first and the second editions) has sui (隨). the Nyāya school.27 First of all, I do not find any strong reason to regard this note as a later interpolation. In other words, in my view, this is probably a kind of commentary added by the translators themselves, i.e., Guṇabhadra et al.28 In this case, a question will naturally follow: Did the translators mistake zixin xian liang “what is none other than the manifestation of one’s mind” for “direct perception of one’s own mind”? To this question, my answer is definitely, “No.” We should not confuse the meanings of the two occurrences of the character liang, i.e., liang for Skt. mātra in the translation and the four kinds of liang for pramāṇa in the interlinear note. This passage corresponds to the following Sanskrit passage, though the two are not entirely identical: anye punar Mahāmate varṇayanti – sarvajñasiṃhanādanādino yathā svacittadṛśyamātrā[2]-vabodhād bāhyabhāvābhāvānabhiniveśāc cātuṣkoṭikarahitād yathābhūtāvasthānadarśanāt svacittadṛśyavikalpasyā[6]-ntadvayāpatanatayā grāhyagrāhakānupalabdheḥ sarvapramāṇā[7]-grahaṇāpravṛttidar-

27 Namely, xianjian (現見) for pratyakṣa, bizhi (比知) for anumāna, piyu (譬喩) for upamāna, and xiansheng xiangchuan (先勝相傳) for āptavacana, respectively. The interlinear note clearly says that these four kinds of “measurement” represent a theory promulgated by non-Buddhists (waidao 外道). 28 Guṇabhadra’s translation of the Laṅkāvatāra also has other interlinear notes, some of which reveal that the person(s) who wrote them had knowledge of the Sanskrit text. For example, a note at T16:670.483b17-18 on the word xin (心) in the term diyi yi xin (第一義心) distinguishes two Skt. words for Ch. xin, i.e., ganlida (肝栗大, for Skt. hṛdaya, lit. “heart”) and zhiduo (質多, for Skt. citta, lit. “mind”), and clearly states that the word xin corresponds to hṛdaya in Sanskrit. This indication coincides with the reading in the extant Skt. text. It would be natural, in light of this evidence, to consider such comments as due to the translator(s); and this is to say nothing of a more general tendency, whereby such interlinear notes are usually due to the translator(s) in any case. śanāt[8] tattvasya vyāmohakatvād agrahaṇaṃ tattvasya, tadvyudāsāt svapratyātmāryadharmādhigamān nairātmyadvayāvabodhāt kleśadvayavinivṛtter āvaraṇadvayaviśuddhatvād bhūmyuttarottaratathāgatabhūmimāyādiviśvasamādhicittamanomanovijñānavyāvṛtter nirvāṇaṃ kalpayanti (Vaidya, 1963: 75.3-8; cf. Nanjio, 1923: 184.15-185.6. The index numbers [2], [6], [7] and [8] correspond to the same index numbers in the above Chinese translation). Obviously svacittadṛśyamātra- is translated as zixin xian liang in Chinese, and there is no note in Skt. I direct the reader’s attention to the term pramāṇa (Index Number [7]) which is translated as duliang (度量) in Chinese. The context reveals that pramāṇa here clearly signifies a foil theory of pramāṇa (which the author will ultimately reject). The Chinese wording bujian suocheng (不見所成[8] in the translation) and xijie bu cheng ye (悉皆不成[5]也 in the note) also seems worthy of our attention. Both have the negative particle bu and the verb cheng. Taking it into consideration that these appear after the reference to pramāṇa/liang, I think that the Chinese interlinear note is placed in the wrong position: it must be a brief commentary on the word pramāṇa (Index Number [7]), and not svacittadṛśyamātra. The mislocation of the note may also possibly be caused by the fact that the word liang appears twice in Chinese ([2] and [7]), as does the wording zixin xian/svacittadṛśya- ([2] and [6]). To sum up, in my view, in this passage of the Laṅkāvatāra-sūtra, zixin xian liang has nothing to do with xianliang as pratyakṣa, though at first glance it appears that it does, because the interlinear note on the fourfold classification of pramāṇa was put in the wrong place.

2 The sinification of the concept of xianliang

So far we have examined earlier usages of xianliang and reached the following conclusions: that xianliang had already been used before Xuanzang in texts by the Dilun school; and that as a translation, xianliang corresponds to pratyakṣaṃ pramāṇam, and not to pratyakṣa in the strict sense.

In what follows, I would like to consider how this term was construed by Chinese scholar-monks, focusing on the “evolution” of their interpretations. The Sanskrit word pratyakṣa is a compound which consists of two elements: prati- (“towards”, “in the vicinity of”, or “with regard to”) and akṣa (“the eye”, or “the sense organ” in the broad sense); hence pratyakṣa means either “perception” as a neuter compound noun pratyakṣam, signifying a type of cognition (Skt. jñāna, vijñāna, etc.), or “perceptible” as an adjective, or even “that which is perceptible”, namely “the object of perception” (in the form of nt. pratyakṣam, m. pratyakṣaḥ, or f. pratyakṣā). Of these two, the former case, viz., a neuter noun meaning a kind of cognition, is predominant over the latter; the former usage as a neuter compound noun is much more popular in many texts. In other words, it is a common, generic word for perception or the object of perception. In the context of the Buddhist Pramāṇa school (i.e., the Sautrāntika and/or Yogācāra), Dharmottara (ca. 740-800), in his Nyāyabinduṭīkā ad Nyāyabindu I 3, explicates direct perception as pratigatam āśritam akṣam (“that which depends on – namely is based on – the sense organ”). He also states that the term can be taken as any gender (sarvaliṅgaḥ pratyakṣaśabdaḥ). In his commentary on Dharmottara’s Nyāyabinduṭīkā, Durvekamiśra explicitly states that akṣa here means the sense organ (akṣam indriyam).

In spite of its frequent use among modern Buddhist scholars, the Chinese term xianliang is a strange word. Once we have been informed that it is a translation of Skt. pratyakṣa, or that it means “perception” in English, we usually do not inquire any further. But in fact, as a Chinese word, the meaning of xianliang as it is remains totally unclear. One of the problems lies in the fact that it was created artificially, most probably for the purpose of translation, and no actual usage can be found in pre-Buddhist Chinese literature. Another difficulty for us in understan-

ding the term is that both elements, xian and liang, each being a single graph, have various senses; for example, it is not entirely clear whether liang in this term is a verb or a noun, and in fact it can be used in both ways as a Chinese word, as we will see in some of the examples below. Moreover, the exact relationship between xian and liang is also not selfevident. As indicated above, Xuanzang often employed the term as a single equivalent to pratyakṣa. Probably he had no difficulty in understanding the term himself, because he knew the meaning and the usages of pratyakṣa in Sanskrit very well. However, the technical term xianliang started its own journey when Chinese scholar-monks began to comment on it.

We start our examination from the pre-Xuanzang period. In his Dasheng yi zhang, Jingying Huiyuan of the Sui dynasty, one of the earliest scholars of Chinese pramāṇa theory, gives the meaning of xianliang as follows: What is called “direct measurement” [is the following]: Direct understanding (xianzhi 現知) of entities is termed “direct measurement”. Further, [another explanation is the following:] Understanding of direct (i.e., present) entities (xianfa 現法) is also called “direct [measurement]”. (言現量者，現知諸法，名為現量。又知現法，亦名為現, T44:1851.670 c11-13.） Here, Huiyuan gives two different interpretations to the term. One is to take xian in the sense of xianzhi 現知, “to understand directly”, or alternatively, “to understand vividly”. The other is to understand xian in the sense of xianfa 現法, “present entities (dharmas)”. The former means that xian is an adverb (or alternatively, that xianzhi is a single verb), and the latter that it is an adjective. Giving two different meanings to a single notion looks ambiguous, but it is a typical working method in commentarial literature to present as many interpretations as possible. Further, in this context, Huiyuan pays attention to xian only; he does not give any comment on liang.

Another notable text which reports the meaning of xianliang is the She dasheng lun chao (攝大乘論抄, T2806). This text is a commentary on Vasubandhu’s She dasheng lun shi lun (攝大乘論釋, *Mahāyānasaṃgrahabhāṣya, T1595) translated by Paramārtha. It belongs to what is called the Shelun (攝論) school. Although the exact date of the activity of the author of this text is not certain, it probably belongs to either the late Sui or the early Tang dynasty. This anonymous commentary has recently been re-edited by Ikeda Masanori. Consulting his new edition, let me refer to a line which gives the literal meaning of xianliang. It says: Names and Bodies (Essences) of the Three Measurements – [First,] Name(s). 1) Present (or Direct) Measurement. What ordinary and holy people cognize is neither [entities] in the past, nor in the future, [hence] it is termed “present”. “Measurement” (liang) means precisely “mensuration” (liangdu). [Thus the word xianliang] is so called in terms of the time of engagement...

(三量名體。名。一現量。凡聖所知，事非過未，名現。量即量度。從事時義為名… T85:2806.1003c20-21; Ikeda, 2009: 18.) It is remarkable that the author explicates the meaning of xian by saying that it refers to present entities to be cognized by ordinary and holy persons, distinguishing them from past (guo 過) and future (wei 未) entities. This interpretation of xian as referring to the present object corresponds to the second of Huiyuan’s interpretations. The other element, liang, is defined as “mensuration” (liangdu 量度). Huiyuan and the author of the She dasheng lun chao were aware of the significance of the threefold classification of pramāṇa as found in Vasubandhu’s texts. They were not, however, in any position to get access to the new theory of Dignāga, who opened up a new era of pramāṇa in Indian Buddhism; Dignāga’s views on pramāṇa were first translated into Chinese only later, by Xuanzang. Xuanzang’s translations of the Nyāyamukha and the Nyāyapraveśa, in particular, were the most important texts for East Asian studies of yinming. Both the Nyāyamukha and the Nyāyapraveśa give the same etymology of pratyakṣa, that is: akṣam akṣaṃ prati vartata iti pratyakṣam, “Because it occurs in connection with each sense organ, it is (called) direct perception.” Xuanzang translated this passage, xian xian bie zhuan, gu ming xianliang (現現別轉，故名現量).33 If we compare the Chinese translation with Sanskrit, we can safely conclude that xian (現) is a translation of akṣa. Namely, xian xian bie zhuan gu ming xianliang

現 現 別 轉 故 名 現量

akṣam akṣam prati vartate iti φ pratyakṣam It is evident that the expression xian xian is a translation of akṣam akṣam (that is, xian means akṣa), and that xianliang is used for pratyakṣa. On the other hand, as we have verified in “I. Xianliang as translation” above, in his translation of the Yogācārabhūmi, Xuanzang sometimes uses xian (and sometimes even xianjian) for pratyakṣa. Moreover, he also uses xianliang as a translation of pratyakṣaṃ pramāṇam in his translation of the Twenty Stanzas. Here we encounter a somewhat confusing result of Xuanzang’s terminology: the word xian can be used for either akṣa or pratyakṣa and xianliang for either pratyakṣa or pratyakṣaṃ pramāṇam. Thus, in sum, Xuanzang translated Skt. pratyakṣa in two ways: as xian and xianliang. It seems to be the case that this ambiguity opened the way to a new phase of interpretation, which guided later scholars in the direction of philosophical developments different from those seen in Indian Buddhism.

33 T32:1630.12b29, T32:1628.3b17. See Katsura, 1982: 84; Funayama, 1992: 89 n. 182; and Taber, 2005: 191 n. 71. Taber also introduces some different views on the etymology of pratyakṣa in Brahmanical (Hindu) texts, such as Praśastapāda’s Padārthadharmasaṃgraha (which employs a similar, but slightly different wording, akṣam akṣaṃ pratītyotpadyata iti pratyakṣam), Vātsyāyana’s Nyāyabhāṣya, and others. His elucidation is partially based on Sharma, 1985. It is noteworthy that, according to Sharma, some grammatical authors, such as Haradatta (a commentator on the Kāśikāvṛtti), construe the meaning of prati in the sense not of akṣam akṣaṃ prati but akṣi akṣi prati, but others, such as Vātsyāyana, Uddyotakara and Vācaspatimiśra (all three authors belonging to the Nyāya school), prefer akṣa to akṣi. Among Buddhist authors, Sharma takes up Dharmottara, but he does not mention Dignāga. In his commentary on the Nyāyapraveśa, Kuiji (窺基, alias Dashengji 大乘基 or Ji 基, 632-682) gives an intricate explanation of the passage in question in terms of the fourfold classification of non-conceptual cognition (wushishen 五識身, wujuyi 五倶意, zhuzizheng 諸自證, and xiudingzhe 修定者) and five kinds of sensory cognition (pañca-vijñāna).36 Further, in his commentary on the Viṃśikā/Viṃśatikā (The Twenty Stanzas), the same author gives a different type of elucidation in terms of differences between schools. According to him, there were at least eight ways of interpreting xianliang: the views of Vasumitra of the Sarvāstivāda school; of Dharmatrāta of the same school; of Ghoṣa of the same school; of the Saṃmitīya/Saṃmatīya school; of the Sautrāntika school; of the non-Buddhist Vaiśeṣika and Sāṃkhya schools; and finally, of the Mahāyāna. In briefly introducing the gist of each idea, Kuiji even gives a formal analysis of xianliang as a compound, after the manner of Sanskrit grammar, using terminology such as “karmadhāraya compound” (chiye shi 持業釋) and “tatpuruṣa compound” (yishi shi 依士釋, also called yizhu shi 依主釋).37

36 The original passage in Kuiji’s Yinming ru zhengli lun shu (因明入正理論疏, 下) reads as follows: 論。現現別轉，故名現量。述曰。此顯名也。此四類心，或唯五識，現體非一，名為現現。各附境體，離貫通縁，名為別轉。由此現現各別縁故，名現量。故者，結上所以，是名現量 (T44:1840.139c1-4). 37 The original passage in Kuiji’s Weishi ershi lun shuji (唯識二十論述記) reads as follows: 言現量者，諸部説異。且薩婆多，用世友説：以根名見，根體是現量。以顯現義是根義故。此能量境，故名現量。是持業釋。法救説：識名見。能量境故。識名現量。持業如前。妙音：慧名見，能量法，勝慧名現量。正量部説：心心所法和合名見，心心所法合名現量。經部師説：根識和合，假名為見，假能量境，假名現量。吠世史迦、徳句義中覺為現量。數論師説：十一根中五根是現量。若歸於本，自性是現量。大乘師説：根名為現，依・發・屬・助・如根五義勝餘故。然是色法不能量境。唯心心所能量度故。心心所法，正是量體。依現之量，名為現量。此依士釋 (T43:1834.999a11-24; translation omitted). This is Kuiji’s explanation of the word xianliang as it is used in the Twenty Stanzas in the form yiqie liang zhong, xianliang wei sheng (一切量中，現量為勝, T31:1590.76b15-16; Skt. sarveṣāṃ ca pramāṇānāṃ pratyakṣaṃ pramāṇaṃ gariṣṭham iti). In other words, in this context, xianliang is used primarily as a translation of pratyakṣaṃ pramāṇam. In the above commentary, it seems certain that Kuiji takes xian (現) in the sense of jian (見) and liang (量) in the sense of nengliang (能量, “to measure” as a verb). The critical point here is that Kuiji uses terms such as xian, liang, and xianliang as explanations of different views held in the Indian context by Vasumitra (Shiyou 世友), Dharmatrāta (Fajiu 法救) and Ghoṣa (Miaoyin 妙音) of the Sarvāstivāda school (sapoduo 薩婆多), the *Sammitīya school (zhengliang

I am not able to give here a precise survey of these detailed discussions, but Kuiji’s exposition is fascinating, inasmuch as it is, to some extent, certainly based on contemporaneous philosophical development in India, that is, on information most probably stemming from Xuanzang’s oral instruction.38 However, at the same time, we should not overlook the Chinese flavor of his exposition. I want to draw special attention to one point: Kuiji reports that Indian followers of the Mahāyāna take xianliang as a tatpuruṣa compound, construing xianliang as yi xian zhi liang (依現之量), “measure (liang) which is based on the actual (xian)”.39 Throughout the same discussion, Kuiji paraphrases “measure[ment]” (liang) by “mensuration” (liangdu), and explains the meaning of “the actual (xian)” as “the sense organ (gen 根)”.40 The first point is in line with the view expressed in the She dasheng lun chao, and the second point is probably based on Xuanzang’s own view, which was based on the pramāṇa-tradition founded by Dignāga. These points reveal without any doubt that Kuiji takes xian and liang as a compound(!) consisting of two elements, and understands xian as

bu 正量部), Sautrāntika (jingbu shi 經部師), the Vaiśeṣika school (feishishi jia 吠世史迦), the Sāṃkhya school (shulun shi 數論師) and the Mahāyāna school (dasheng shi 大乘師). Therefore, it is evident that Kuiji uses the term xianliang as an Indic word, referring to the Indian context, and not as a Chinese word. 38 For example, parallel discussion regarding differences of opinion between Vasumitra, Dharmatrāta and Ghoṣa is found in Xuanzang’s translation of the Mahāvibhāṣā (Da piposha lun 大毘婆沙論, T27:1545.61c, 63b, 489c). 39 The explanation “yi X zhi Y”, as in yi xian zhi liang (依現之量), signifies that the term X-Y is a tatpuruṣa compound. This point is clear from Kuiji’s exegesis. In his Cheng weishi lun shuji (成唯識論述記), Kuiji explains yiye (意業 for Skt. manaskarman) as yi yi zhi ye (依意之業), stipulating that it is a tatpuruṣa compound (yishi shi, 依士釋) (T43: 1830.276a3). Regarding this, see also n. 44 below. Further, some texts explain that yanshi (眼識 for Skt. cakṣurvijñāna) is a tatpuruṣa compound by using the expression yi yan zhi shi (依眼之識; e.g., Kuiji’s expression “…依主釋…如眼識等…依眼之識，故名眼 識” in T43:1830.377b24-26) or yan zhi shi (眼之識) without resorting to yi (依) in the same sense (e.g., Kuiji’s expression “(彼云如)眼之識故名眼識” in T43:1830.416b

10-11).

40 See Kuiji’s passage in n. 36 above.

corresponding to pratyakṣa, and liang to pramāṇa.41 We know this because reference to a kind of compound makes sense only if the word xianliang is interpreted as a single compound noun. It looks as though Kuiji assumed an underlying form such as *pratyakṣapramāṇa, as a single compound. Now, such a form is not attested in any Sanskrit Buddhist texts on pramāṇa: we have many examples of pratyakṣaṃ pramāṇam as two nouns, but, as far as I know, the form *pratyakṣapramāṇa as a tatpuruṣa compound is unattested. On the other hand, Kuiji’s statement does not make sense if we assume that the form pratyakṣa was the compound Kuiji meant, for in that case, we cannot find any reason why he would add the interpretation(s) of pramāṇa (liang). Thus, we might suppose that Kuiji is mixing up two different things: the analysis of pratyakṣa (< prati+akṣa) and the analysis of xianliang (< xian+liang). I do not mean that Kuiji’s endeavor is nonsense. Rather, I would like to see this complicated exegesis as a new idea, which makes sense only in the Chinese language, and not in Sanskrit. If this is the case, we have here an example of the “sinification” of Buddhist terms.

Another interesting interpretation of the above-mentioned passage, common to the Nyāyapraveśa and the Nyāyamukha, is found in Jingyan’s (淨眼) commentary entitled Yinming ru zhengli lun hou shu (因明入正理論後疏, Pelliot chinois No. 2063). This text was edited by Takemura Shōhō (武邑尚邦) and subsequently studied by Shen Jianying (沈劍英).

41 Against my interpretation, some people might claim that Kuiji used the term yi xian zhi liang only to explain the Chinese word xianliang, and that it had nothing to do with the explanation of the Skt. term *pratyakṣapramāṇa. However, I do not think that this idea is tenable, because, as I pointed out in n. 37 above, in the case of his commentary on the Twenty Stanzas, Kuiji did not intend to explain a Chinese scholarly situation – i.e., how Chinese scholar-monks construed the Chinese word xianliang – but rather, introduced various Indian interpretations, referring to the word yi xian zhi liang as a Mahāyānic interpretation current in India. This implies that Kuiji tried to analyze the relationship between xian and liang in the Indian context, namely, in terms of the relationship between pratyakṣa and pramāṇa in the Sanskrit language. In fact, however, we do not find this type of discussion attested in extant Sanskrit texts. All that we can actually confirm is that there existed different interpretations of the relationship between praty- (prati) and akṣa, as shown, for example, in Sharma, 1985 and Taber, 2005: 191 n. 71.

According to Takemura, Jingyan is a commentator who lived between

Wengui (文軌, d.u.) and Huizhao (慧沼, d. 714).43 Regarding the latter, Wei Jen Teng (Teng, 2011: 117) has recently pointed out that Huizhao explains that xianliang is a tatpuruṣa compound in his Dasheng fayuan lin zhang bu que (大乘法苑林章補闕) 8.44

Jingyan first introduces the three interpretations of the passage in question already mentioned. Though I cannot describe them all here, the second interpretation is similar to Kuiji’s, construing xianliang as a tatpuruṣa compound. The phrase yi xian zhi liang appears in that context. According to the third interpretation, the term xianliang should be interpreted as a karmadhāraya compound (chiye shi 持業釋), in the sense that xian itself is liang. After introducing these different types of interpretation, Jingyan further proceeds to present his own view (jin jie 今解) in two ways: namely that it can be either a tatpuruṣa- or a karmadhāraya compound. Here we can see a further development in interpretation. Next, in the second half of the eighth century, Tankuang (曇曠) composed the Dasheng bai faming men lun kaizong yiji (大乘百法明門論開宗義 記) at Dunhuang. In this work, he describes the meaning of xianliang as follows: What is called “measurement regarding the eight kinds of consciousness” is of three kinds in total. The first is “direct measurement”. “Di-

43 Takemura, 1986: 36. 44 “The compound [is analyzed as follows:] An object [of the mind] is called xian. The mind is called liang. It is the liang regarding that xian. [Therefore] it is called xianliang.

It is a tatpuruṣa compound.” (合釋者。若境名現，心名為量，即現之量，名為現量，依主釋也, Z1, 2, 3, 1, 30, verso, b8-10; X55:882.159b22-23). rect” (or “real”, xian 現) means that which is really existent in front of one (xianqian 現前). “Measurement” (liang) means “mensuration” (liangdu). That is to say, when colors and so forth are clear and determinate and directly exist in front of one, one is not deluded by an erroneous form, attains a determinate cognition, and is free from the conception of various designations, species and classes, [thus] illuminating [the object] clearly. Therefore this is called “direct measurement”. It is “direct”, on the one hand, and at the same time it is “measurement”, on the other. [Hence] it is an action-carrying (karmadhāraya) compound.

(謂八識量，總有三種。一者現量。現謂現前。量謂量度。謂於現前明了色等，不迷亂相，而得了知，離諸名言種類分別，照鏡明白， 故名現量。現即是量，持業釋也, T85:2810.1053a8-11.) Tankuang construes the term as a karmadhāraya compound, and states that xian means xianqian 現前, “(that which is) before one[’s eyes]”, which is a vividly manifesting, non-erroneous object of cognition; and that liang means liangdu “mensuration”. Thus, Chinese scholar-monks developed the interpretation of the term xianliang by considering the relationship between xian and liang. This viewpoint would not have been possible in Sanskrit literature, because *pratyakṣapramāṇa is not a common compound, even if it is not entirely impossible; and because the normal form pratyakṣaṃ pramāṇam is not a compound at all.

Finally, let me introduce a Chinese attempt to associate xianliang as pramāṇa with the teaching of the Laṅkāvatāra-sūtra. Such an idea is found in the commentary entitled Lengqie abaduoluo bao jing xuan yi (楞伽阿跋多羅寶經玄義), composed by Zhixu (智旭, 1599-1655). Very interestingly, this important monk of the Ming-Qing period explicates the meaning of xianliang in yet another way: What is called “direct measurement” [means the following:] “Direct” means direct manifestation. “Measurement” means amount. This implies that, regarding all entities such as the five entities, the three kinds of intrinsic nature, the eight kinds of consciousness, the two kinds of no-self, and so on, to [as many as] ten realms (jie 界), a hundred realms, or a thousand, all of these various entities [that are mentioned in the Laṅkāvatāra-sūtra] are, as a whole, nothing but what is directly manifested (xianxian 顯現) by one’s own mind, and do not go beyond [what the mind can] discern and measure (fenliang 分量). Therefore it is called “direct measurement (xianliang 現量)”.

(言現量者，現謂顯現，量謂分量。言一切五法三自性八識二無我，乃至十界百界千，如等種種諸法，總是唯心所現，不出心之分量， 故名為現量也, Z1,1,26,1, 49, recto, b15-18; X17:328.484b.) Here, Zhixu states that xian signifies “directly manifest” (xianxian 顯現) as a verb, and liang means “[what the mind can] discern and measure” (fenliang 分量) as a noun. As a whole, he claims, the term xianliang signifies that all the mental categories, such as the five entities (pañcadharma), the three natures, and so forth, are nothing but the manifestation of one’s own mind, and they all remain within the scope of the mind. Although similar wording regarding xian in the sense of xianxian had existed previously, such a combination of pramāṇa theory and Mind-Only theory is a unique result of the Chinese Buddhist exegetical tradition, and cannot be found in Indian literature. We should also bear in mind that the idea developed here has a special connotation because, as a commentator on the Laṅkāvatāra, Zhixu needed to integrate the tenets of this sūtra with the notion of pramāṇa. His unique exegesis of liang as fenliang is influenced by the notion of liang, which is a translation of Skt. mātra in the sūtra in question (as we saw above). Conclusion

In the Six Dynasties period, pratyakṣa was translated by various words, such as xian, xianqian, xianjian, zheng, and so forth. The earliest reference to xianliang meaning “direct perception” is found in the Huizheng lun (Skt.

Vigrahavyāvartanī) of Nāgārjuna, in which xianliang is a translation of pratyakṣaṃ pramāṇam, and not pratyakṣa as a single word. Soon thereafter, at the end of the Six Dynasties period, the Dilun school started to use xianliang as a technical term. This is probably what influenced Xuanzang’s usage of xianliang. He employed it alternately as a translation of both pratyakṣa and of pratyakṣaṃ pramāṇam. To put it another way, the term xianliang has a double meaning in Xuanzang’s translations, and without consulting the original Sanskrit text, it is impossible to determine in which of these two senses each instance of xianliang is being used.

We also saw that it is hard to imagine that any term like *pratyakṣapramāṇa, as a single compound, prevailed in the Indian Buddhist world. In Sanskrit texts, the most popular form is undoubtedly pratyakṣaṃ pramāṇam, i.e. two words in apposition. Though I hesitate to say that the form *pratyakṣapramāṇa never existed in Sanskrit, even if, hypothetically, the term did exist, I do not think it would be possible to regard such a compound as a tatpuruṣa, for as long as *pratyakṣapramāṇa is intended as a synonym of pratyakṣaṃ pramāṇam, the compound would have to be a karmadhāraya. However, after Xuanzang, Chinese scholar-monks began to analyze the meaning of xianliang by dividing it into xian and liang, and in so doing, they made use of their knowledge of classes of Sanskrit compounds like tatpuruṣa and karmadhāraya. The results of their analyses look somewhat unacceptable to the eyes of anyone familiar with Sanskrit grammar, because these Chinese scholars conflated the construction of xianliang with that of *pratyakṣapramāṇa. However, it would not be correct to criticize their views only with reference to Indian modes of analysis, because these Chinese monks used the Chinese language and thought in Chinese. Here, we should bear in mind that Chinese scholar-monks after Xuanzang applied such terms as yishi shi (tatpuruṣa) or chiye shi (karma-

dhāraya) to Chinese terms or Chinese translations, and not directly to original Sanskrit terms or their phonetic transcriptions. Under such circumstances, earlier monks such as Kuiji probably had sufficient knowledge of the Sanskrit equivalents of the Chinese words they analyzed, while later - for example, in the Ming - scholar-monks no longer had any interest in, or knowledge of, the original Sanskrit terms, nor of the relationship between a Chinese translation and its Sanskrit equivalent. This implies that many of these later figures just applied Sanskrit words such as yishi shi or chiye shi to the explanation of Chinese terms in the Chinese context. In other words, they used the names of Sanskrit compounds to talk about the Chinese language. This being the case, it would be beside the point or meaningless to ask whether such Chinese interpretations make sense from a Sanskrit point of view. Rather than harshly criticizing those Chinese views, it would be better to take them differently; such Chinese interpretations look extremely attractive when we view them in a different light, as a matter of the Chinese language.

It is almost meaningless to say, on the basis of Indic language, that the Chinese way of understanding xianliang was a mistake. Rather, it can be evaluated as a new type of development. In this sense it is an interesting example of what is called the “Sinification of Buddhist Concepts”.

Bibliography

Aoki Takashi (青木隆) (2000). “Jiron shū no yūsoku ron to engi setsu” (地論宗の融即論と縁起説). In Hokuchō Zui Tō Chūgoku Bukkyō shisō shi (北朝隋唐中國佛教思想史), edited by Aramaki Noritoshi (荒牧典 俊). Kyoto: Hōzōkan, pp. 179-204.

Dutt, Nalinaksha, ed. (1966). Bodhisattvabhūmiḥ: Being the XVth Section of Asaṅgapāda’s Yogācārabhūmiḥ. Tibetan Sanskrit Works Series Vol. 7. Patna: K. P. Jayaswal Research Institute. Funayama, Toru (1992). “A Study of Kalpanāpoḍha: A Translation of the Tattvasaṃgraha vv. 1212-1263 by Śāntarakṣita and the Tattvasaṃgrahapañjikā by Kamalaśīla on the Definition of Direct Perception.” Zin-

bun (Annals of the Institute for Research in Humanities, Kyoto University) 27, pp. 33-128. ―― (船山徹) (2000). “Jiron shū to nanchō kyōgaku” (地論宗と南朝教學). In Hokuchō Zui Tō Chūgoku Bukkyō shisō shi (北朝隋唐中國佛教思想史), edited by Aramaki Noritoshi (荒牧典俊). Kyoto: Hōzōkan, pp. 123-153. ―― (2010). “The Work of Paramārtha: An Example of Sino-Indian Cross-cultural Exchange.” Journal of the International Association for Buddhist Studies 31:1-2 (2008), pp. 141-183. Ikeda Masanori (池田將則) (2009). “Tonkō bon Shō daijō ron shō no genpon (Moriya korekushon bon) to kōzoku bubun (Sutain 2554) to ni tsuite: honkoku to kenkyū (zenpen)” (敦煌本『攝大乘論抄』の原本

守屋コレクション本）と後續部分（スタイン 2554）とについて──翻刻と研究（前篇）). Bukkyō shi kenkyū (佛教史研究) 45, pp. 1-75. Hirakawa Akira et al., eds. (1973). Index to the Abhidharmakośabhāṣya (P. Pradhan Edition). Part One: Sanskrit-Tibetan-Chinese. Tokyo: Daizō shuppan. ―― (1977). Index to the Abhidharmakośabhāṣya (P. Pradhan Edition). Part Two: Chinese-Sanskrit. Tokyo: Daizō shuppan. Johnston, E. H. and Kunst, Arnold, eds. (1986). The Dialectical Method of Nāgārjuna: Vigrahavyāvartanī. Second Edition (Revised and Enlarged). Translated from the original Sanskrit with Introduction and Notes by Kamaleswar Bhattacharya. Text Critically edited by E. H. Johnston and Arnold Kunst. Delhi: Motilal Banarsidass.

Kano, Kazuo (2008). “Two Short Glosses on Yogācāra Texts by Vairocanarakṣita: Viṃśikāṭīkāvivṛti and *Dharmadharmatāvibhāgavivṛti.” In Manuscripta Buddhica, Vol. I: Sanskrit Texts from Giuseppe Tucci’s Collection, Part I, edited by Francesco Sferra. Serie Orientale Roma. Roma: IsIAO, pp. 343-380. Katsura Shōryū (桂紹隆) (1982). “Inmyō shōri mon ron kenkyū [go]” (因明正理門論研究 [五]). Hiroshima daigaku bungakubu kiyō (廣島大學文學部紀要) 42, pp. 82-99. Lamotte, Étienne (1935). Saṃdhinirmocana Sūtra: L’explications des mystères. Louvain: Bureaux du recueil, Bibliothèque de l’Université. Lévi, Sylvain (1925). Vijñaptimātratāsiddhi, Deux traités de Vasubandhu, Viṃśatikā (la Vingtaine) accompagnée d’une explication en prose et Triṃśikā (la Trentaine) avec le commentaire de Sthiramati, 1ère partie, texte. Paris: H. Champion. Malvania, Dalsukhbhai, ed. (1955). Paṇḍita Durveka Miśra’s Dharmottarapradīpa [Being a Sub-Commentary on Dharmottara’s Nyāyabinduṭīkā, a Commentary on Dharmottara’s Nyāyabindu]. Patna: Kashiprasad Jayaswal Research Institute. Nagao Gadjin (長尾雅人) (1987). Shō daijō ron: Wayaku to chūkai (ge) (攝大乘論 和譯と注解[下]). Tokyo: Kōdansha. Nakamura Hajime (中村元), ed. (1975). Bukkyō go dai jiten (佛教語大辭典). Tokyo: Tōkyō shoseki. Nanjio, Bunyiu, ed. (1923). The Laṅkāvatāra sūtra. Kyoto: Otani University Press. Radich, Michael (2008). “The Doctrine of *Amalavijñāna in Paramārtha (499-569), and Later Authors to Approximately 800 C.E.” Zinbun (Annals of the Institute for Research in Humanities, Kyoto University) 41, pp. 45-174. Sasaki Gesshō (佐々木月樵) (1931). Kan’yaku shi hon taishō Shō daijō ron (漢譯四本對照攝大乘論). Tokyo: Hōbunsha. Sharma, Sarveswara (1985). “What Kind of Compound is the Word Pratyakṣa?” Adyar Library Bulletin 49, pp. 14-29. Shen Jianying (沈劍英) (2008). Dunhuang yinming wenxian yanjiu (敦煌因明文獻研究). Shanghai: Shanghai guji chubanshe. Taber, John (2005). 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: RoutledgeCurzon. Takasaki Jikidō (高崎直道) (1980). Ryōga kyō (楞伽經). Tokyo: Daizō shuppan. Takemura Shōhō (武邑尚邦) (1986). Inmyō gaku: kigen to hensen (因明學起源と變遷). Kyoto: Hōzōkan. (Cf. Yang Jinping [楊金萍] and Xiao Ping [肖平], tr. (2008). Yinming xue de qiyuan yu fazhan [因明學的起源與發展]. Peking: Zhonghua shuju.) Teng, Wei Jen (2011). Recontextualization, Exegesis, and Logic: Kuiji’s (632-682) Methodological Reconstruction of Chinese Buddhism. PhD dissertation, Harvard University. Ueyama Daishun (上山大峻) (1990). Tonkō Bukkyō no kenkyū (敦煌佛教の研究). Kyoto: Hōzōkan. Vaidya, P. L., ed. (1963). Saddharmalaṅkāvatārasūtram. Buddhist Sanskrit Texts No. 3. Darbhanga: Mithila Institute of Post-Graduate Studies and Research in Sanskrit Learning. Wogihara, Unrai (1936). Bodhisattvabhūmi: A Statement of Whole Course of the Bodhisattva (Being Fifteenth Section of Yogācārabhūmi). Tokyo: Sankibo Buddhist Book Store, 1971 [reprint of the 1st ed. published in 1936]. Yanagi Mikiyasu (柳幹康) (2011). “Ryōga kyō to Ninyū shigyō ron: ‘Ryōga shū’ no shisō to soko ni shimeru Ryōga kyō no ichi” (『楞伽經』と『二入四行論』──「楞伽宗」の思想とそこに占める『楞伽經』の位 置). Indo tetsugaku Bukkyō gaku kenkyū (インド哲學佛教學研究) 18, pp. 71-85. Yokoyama Kōitsu (横山紘一) and Hirosawa Takayuki (廣澤隆之), eds. (1996). Yuga shi ji ron sō sakuin (瑜伽師地論總索引). Tokyo: Sankibō. Yonezawa, Yoshiyasu (2008). “Vigrahavyāvartanī: Sanskrit Transliteration and Tibetan Translation.” Naritasan Bukkyō kenkyūjo kiyō (成田山佛教研究所紀要) 31, pp. 209-333.