A STUDY OF THE ŚARĪRĀRTHAGĀTHĀ IN THE YOGĀCĀRABHŪMI

A STUDY OF THE ŚARĪRĀRTHAGĀTHĀ IN THE YOGĀCĀRABHŪMI

A dissertation presented by Hsu-Feng Lee

to The Department of Indian Subcontinental Studies School of Languages and Cultures

Faculty of Arts and Social Sciences

A thesis submitted in fulfilment of requirements for the Degree of Doctor of Philosophy at The University of Sydney March 2017

Abstract

The Śarīrārthagāthā (Tǐyì qiétā 體義伽他;‘dus pa’i don gyi tshigs su bcad pa) is a collection of canonical verses with accompanying commentary in the Yogācārabhūmi (Yúqié shī dì lùn 瑜伽師地論; rnal 'byor spyod pa'i sa), an encyclopedic text of India’s major Mahāyāna philosophical school. To date the

Śarīrārthagāthā has not attracted much scholarly research and many interesting aspects have hitherto gone unnoticed that are worthy of further investigation.

Some researchers have identified the sources of these verses, and a study by Enomoto (1989) is the most complete. In this dissertation, I have carried out further analyses based on the results found by these researchers. The initial topics are the place of the Śarīrārthagāthā verses in the formation of Buddhist texts (especially, aṅga classification) and the reason why early verses in particular were collected in the Śarīrārthagāthā. The work of Yìnshùn has provided significant information for the investigation of the above issues. He investigated the development and relationship between aṅga and Āgamas from texts during the period of early Buddhism to Mahāyāna.

Moreover, the distinctive characteristics of the Śarīrārthagāthā verses have been investigated through a comparison with their parallels in other texts, with the aim of assessing the school affiliation of these texts. Also, three aspects pertaining to the development of the Śarīrārthagāthā have been discussed: (1) the status of the Śarīrārthagāthā in the development of the Yogacārabhūmi, (2) the reason why verse classification (uddāna) is treated differently in the Tibetan

version compared to the Chinese and Sanskrit versions, and (3) why were these particular early Buddhist verses selected for inclusion in the Śarīrārthagāthā and not others.

Key words: Śarīrārthagāthā, geya, vyākaraṇa, Mūlasarvāstivāda.

Table of Contents

Abstract i

Acknowledgements viii

Abbreviations xii

Chapter one: Introduction 1


1.1. Three introductory aspects of the Śarīrārthagāthā 1
1.1.1. The place of the Śarīrārthagāthā in the Yogacārabhūmi 2
1.1.2. The classification of verses in the Śarīrārthagāthā 7
1.1.3. The meaning of the title Śarīrārthagāthā 15
1.2. Literature review 18
1.2.1. Sources of the Śarīrārthagāthā verses 21
1.2.2. Relevant information regarding the Śarīrārthagāthā 18
1.2.3. Materials pertaining to aṅga classification 21
1.2.4. Textual formation 22
1.2.5. Other relevant studies 24
1.3. Aim and significance of this study 24
1.3.1. Aim 24
1.3.2. Significance 26
1.4. Methodology 27
1.4.1. Aṅga classification and Śarīrārthagāthā verses 28
1.4.2. Sources of the Śarīrārthagāthā verses 29
1.4.3. Analysis of the Śarīrārthagāthā verses 29
1.4.4. Textual development and stratification 30
1.4.5. Principles of selecting verses 30
1.5. Structure of this study 31
1.6. Sanskrit manuscripts, Chinese and Tibetan translations and exegetical literature 33


Chapter two: Aṅga classification and the Śarīrārthagāthā verses 39


2.1. Aṅga classification and the Yogācārabhūmi 39
2.1.1. Aṅga classification and the formation of early Buddhist texts 39
2.1.2. The first three aṅgas and the Yogācārabhūmi 45
2.1.3. The fourth aṅga, gāthā, and the Śarīrārthagāthā verses 52
2.1.4. Summary 56
2.2. The second aṅga, geya, and the Group A verses of the Śarīrārthagāthā 57

2.2.1. The development of geya 57
2.2.2. The source of the Group A verses in the Śarīrārthagāthā 66
2.2.3. Summary 70
2.3. The third aṅga, vyākaraṇa, and the verses of Group B of the Śarīrārthagāthā 70

2.3.1. The development of vyākaraṇa 71
2.3.2. The source of Group B verses in the Śarīrārthagāthā 78
2.3.3. Summary 84
2.4. Conclusion 85


Chapter three: The Group A verses 87


3.1. The three versions of the Group A verses with corresponding translations, and the narratives of corresponding sūtras and parellels 87
3.1.1. The first set: evil (è 惡, pāpa, sdig pa); 4 pādas / 1 verse 88
3.1.2. The second set: what can be expressed (shuō 說, ākhyeya, brjod bya); 16 pādas / 4
verses 92

3.1.3 The third set: lust (tān 貪, rāga, 'dod chags); 20 pādas / 5 verses 96
3.1.4. The fourth set: flood (liú 流, ogha, chu bo); 12 pādas / 3 verses 101
3.1.5. The fifth set: being frightened (bù 怖, utrasta, ‘jigs); 14 pādas / 3 verses 106
3.1.6. The sixth set: the caste system (lèi 類 , abhivarṇṇatā, rigs); 12 pādas / 3 verses 109
3.1.7. The seventh set: obtaining fame (yù 譽, yaśasvin, grags ldan); 8 pādas / 2 verses 113
3.1.8. The eighth set: current (chí 池, sara, mtsho); 12 pādas / 3 verses 116
3.1.9. The ninth set: (two) floods (liú 流, ogha, chu bo gnyis); 8 pādas / 2 verses 120
3.1.10. The tenth set: (two) floods (liú 流, ogha, chu bo gnyis); 12 pādas / 3 verses 123
3.1.11. The eleventh set: lust and hatred (tān 貪 , rāgadveṣa, ‘dod chags zhe sdang); 16 pādas / 4 verses 128
3.1.12. The twelfth set: what ought to be done (zuò 作, kāryatā, bya ba); 20 pādas / 5 verses 133
3.1.13. The thirteenth set: struggle (qú láo 劬勞, āyūhana, ‘dzin pa); 4 pādas / 1 verses 137
3.1.14. The fourteenth set: attainment of the aim (dé yì 得義, arthaprāpti, don thob); 12 pādas
/ 3 verses 141

3.2. The analysis of Group A verses 145
3.2.1. The distinctive characteristics of the Chinese Group A verses 146
3.2.2. The features of the Group A verses of the Śarīrārthagāthā 146
3.2.3. Other issues 149

3.2.4. Summary 153

3.3. Conclusion 153

Chapter Four: The Group B verses 154

4.1. The three versions of the Group B verses with their English translations, and comparison of the verse with other texts 154
4.1.1. The first set: impermanence (mi rtag pa); 4 pādas / 1 verse 155
4.1.2. The second set: carefulness (bag yod); 4 pādas / 1 verse 159
4.1.3. The third set: strong lust (’dod chags shas che); 4 pādas / 1 verse 160
4.1.4. The fourth set: dwelling in the dharma (chos la gnas); 4 pādas / 1 verse 162
4.1.5. The fifth set: contemplating the truth (nges rtag); 4 pādas / 1 verse 163
4.1.6. The sixth set: good saying (legs par smras pa); 4 pādas / 1 verse 164
4.1.7. The seventh set: faith (dad pa); 4 pādas / 1 verse 166
4.1.8. The eighth set: learning (thos pa); 4 pādas / 1 verse 167
4.1.9. The ninth set: being like the sky (nam mkha' mtshungs); 4 pādas / 1 verse 169
4.1.10. The tenth set: physical appearance (gzugs); 20 pādas / 5 verses 171
4.1.11. The eleventh set: king (rgyal); 4 pādas / 1 verse 176
4.1.12. The twelfth set: city of bone (rus pa'i ra ba); 4 pādas / 1 verse 177
4.1.13. The thirteenth set: turtle (rus sbal); 4 pādas / 1 verse 179
4.1.14. The fourteenth set: equality (mtshungs); 4 pādas / 1 verse 180
4.1.15. The fifteenth set: mud (’dam); 4 pādas / 1 verse 182
4.1.16. The sixteenth set: two dwellings (gnas gnyis); 4 pādas / 1 verse 183
4.1.17. The seventeenth set: two dwellings (gnas gnyis); 4 pādas / 1 verse 185
4.1.18. The eighteenth set: having cleaning up (bsal gyur); 4 pādas / 1 verse 186
4.1.19. The nineteenth set: giving (byin pa); 4 pādas / 1 verse 188
4.1.20. The twentieth set: all evils (sdig pa kun); 4 pādas / 1 verse 189
4.1.21. The twenty first set: falling down (ltung ba); 4 pādas / 1 verse 191
4.1.22. The twenty second set: skilled (mkhas); 4 pādas / 1 verse 192
4.1.23. The twenty third set: craft (bzo); 4 pādas / 1 verse 194
4.1.24. The twenty fourth set: travelling far away (ring du 'gro); 4 pādas / 1 verse 195
4.1.25. The twenty fifth set: obscuring (bsgribs); 40 pādas / 10 verses 196
4.1.26. The twenty sixth set: sensual pleasures (’dod); 24 pādas / 6 verses 204
4.1.27. The twenty seventh set: auspicious (bzang po); 6 pādas / 1 verse 208
4.2. 1. The analysis of Group B verses 211

4.2.1. The distinctive nature of the Chinese Group B verses 211
4.2.2. The features of the Group B verses of the Śarīrārthagāthā 212
4.2.3. Other issues 215

4.2.4. Summary 218

4.3. Conclusion 219

Chapter five: Texual issues of the Śarīrārthagāthā 221

5.1. The formation of the Śarīrārthagāthā 221
5.2. Differences in the summary verse (uddāna) between the Tibetan and Sanskrit/Chinese versions 224
5.3. Potential principles of selecting verses in the Śarīrārthagāthā 231

5.4 Conclusion 234

Chapter six: Conclusion 236

6.1. Summary of previous chapters 236
6.1.1. Relationship between the two aṅgas of geya and vyākaraṇa and the Śarīrārthagāthā verses 236
6.1.2. Distinctive characteristics of Śarīrārthagāthā verses 237
6.1.3. Textual development of the Śarīrārthagāthā 240
6.2. Contribution 241

6.3. Future investigation 243
Table 2.1 Geya function 245
Table 2.2 Geya function by Saṃyutta 255
Table 2.3. The findings of four scholars (Group A) 256
Table 2.4 The correspondence between the verses of Group A and SĀ-G and SĀ-U 257
Table 2.5 The findings of four scholars (Group B) 259
Bibliography 263

I. Primary Sources 263
II. Dictionaries, Indexes, and Catalogues 266

Ⅲ. Secondary works 268
 

Acknowledgements

The origins of this study on the Śarīrārthagāthā can be traced back to the time when I was studying for a master degree at the Chung-Hwa Institute of Buddhist Studies (the predecessor of the Dharma Drum Institute of Liberal Arts). My supervisor, Bhikkhu Huimin, provided me with this topic and his inspiring guidance, for which I am grateful. During this research, I identified the relationship between the Śarīrārthagāthā verses and aṅga classification, which was discussed in my master thesis and also forms part of chapter two of this doctoral work. This was the first time that I felt I had the ability to undertake academic research. During my continuing study at The University of Sydney,

my doctoral supervisor, Mark Allon, played a significant role. Without his sound guidance of my work, and also reading tirelessly over many drafts and providing invaluable corrections, this doctoral thesis could not have been completed. His patience and support over the last few years have made my research experience a very pleasant and fruitful one. I am profoundly grateful that the invisible hands of causes and conditions have led me to study under his supervision.

Besides this, I have been supported and encouraged by many other individuals. I would like to thank a number of helpful mentors. Prof. Fumio Enomoto provided me with significant information relevant to the Śarīrārthagāthā. Prof. Fritz Grohmann (Gāo míngdào 高明道) spent his
valuable time contributing his knowledge for solving issues related to Chinese texts. Prof. Peter Skilling was willing to read some of my work. Andrew
 
McGarrity gave important feedback at an early stage of this project, and assisted me in working on Tibetan texts. Chiew-Hui Huo patiently discussed and read Chinese texts with me and improved my Chinese-English translation skills.

Bhikkhunī Dhammadinnā unconditionally helped me with the Tibetan translations, pointed out the issues of my work, and provided me inspiring moral support. Ulrich Timme Kragh supplied me some relevant information and helped me make contact with Prof. Lambert Schmithausen.
Being brought up with a mother tongue that is not English, I was fortunate to have received help in this regard from a number of colleagues. Chris Clark has dedicated time and energy to assist me with Sanskrit-English translation, proofread my draft chapters and provided many useful comments, thereby helping me avoid potential confusion caused by my clumsy expressions.

Similarly, in the early period of my study, Blair Silverlock helped me with Sanskrit-English translation, proofread my work, and also gave me a lot of spiritual support. Elizabeth McDougal (Tenzin Chozom) has kindly spent her time checking my Tibetan-English translation. Erick Neiss provide assistance with translation from Tibetan to English.

I am also grateful that the Faculty of Arts and Social Sciences provides its postgraduate students with a congenial research centre (PGARC), where I have spent much of my time over the last few years. Due to this wonderful place, I made a lot of friendships with other PhD colleagues. I need to express deep appreciation to the Hwa-Tsang Buddhist Monastery for offering me secure and peaceful accommodation, and welcoming me into their saṃgha while I studied in Sydney from 2011 to 2015.
 
While studying in Australia with limited resources and information, I sometimes needed assistance from overseas, therefore I also need to express my gratitude to the following individuals. Guō jiélì 郭捷立 always actively helped
me obtain resources and contact relevant scholars for the issues I faced, as well

as provided me with his moral support. Xǔ tiánzhì 許恬智 efficiently helped me look for relevant information. Venerable Siǒuhuà 脩 化 法 師 assisted with finding Tibetan texts. Venerable Xiǎnzhào 顯照法照 helped me obtain relevant information in Chinese. Hú huìtíng 胡慧婷 helped edit some parts of the Tibetan Śarīrārthagāthā commentary. Zhào shūhuá 趙淑華 spent a lot of time discussing the issue of the uddāna with me. Importantly, I am deeply indebted to the

Chung-Hwa Institute of Buddhist Studies for providing a scholarship for four years, and many other monks and lay Buddhist disciples for financial support.
Finally, I would like to acknowledge the behind-the-scenes support I received from my monastery (Seeland Monastery 西蓮淨苑), my parents, my family

members, many Dharma brothers and friends, such as Venerable Xiūxián 修賢,
Changtzu 長慈, Huorong 皓融, Xìngkōng 性空, Zōnghào 宗皓, Fachi 法持, Cílóng 慈隆, Cílǜ 慈律, Zizhao 子釗, Nengrong 能融, Wang Bin 王彬, Klaus Radityo Tan, William (Liu Qingjun), Liào Xiùfèng 廖秀鳳, Amy Lin, Charlotte, Jannette, etc.

My gratitude to those individuals and all the visible or invisible conditions is impossible to express fully. May the virtue and merits lead the world to peace and happiness and may all beings always stay in the dharma, as indicated in the Śarīrārthagāthā:

於心相善知 能餐遠離味 靜慮常委念 受無染喜樂
 

cittanimittasya kovidaḥ, pravivekasya ca vindate rasaṃ |

dhyāyī nipakaḥ pratismṛto, bhuṃkte prītisukhaṃ nirāmiṣaṃ ||




/ sems kyi mtshan ma mkhas pa dang / / rab tu dben pa'i nyams thob dang /

/ rtag 'grus dran ldan bsam gtan pa / / dga' bde zang zing med pa spyod /




One [who] is skilled in the signs of the mind and knows the flavour of seclusion, meditating, wise [and] mindful,
enjoys spiritual joy and happiness.
 
Abbreviations



Abhidh-k-bh Abhidharmakośabhāśya (ed. P. Pradhan, 1975)

Abhidh-k-vy Abhidharmakośavyākhyā (ed. U. Wogihara, 1932–1936)

Abhidh-k-ṭ Abhidharmakośaṭīkā (Peking ed. No. 5595)

AN Aṅguttaranikāya (eds. R. Morris & E. Hardy, 1885–1900)

Ap Apadāna (ed. M. E. Lilley, 1925–1927)

BHSD Buddhist Hybrid Sanskrit Dictionary (ed. Franklin Edgerton,
1953)
DĀ Chinese Dīrghāgama (T 1)

Dhp Dhammapada (eds. von Hinüber and K. R. Norman, 1994)

DN Dīrghanikāya (eds. T. W. Rhys Davids & J. E. Carpenter,
1890–1911)

EĀ Chinese Ekottarikāgama (T 125)

GDhp Gāndhārī Dhammapada (ed. J. Brough, 1962)
Ibh Isibhāsiyāiṃ (ed. W. Schubring, 1969)
Jā Jātaka (ed. V. Fausbøll, 1877–1896)
JH-Eng Jeffrey Hopkins’ Tibetan-English Dictionary
MĀ Chinese Madhyamāgama (T 26)
Mbh Mahābhārata (eds. V. S. Sukthankar et al., 1933–1966)
MPS Mahāparinirvāṇasūtra (ed. E. Waldschmidt, 1950–1951)
Mvu Mahāvastu (ed. E. Senart, 1882–1897)
MN Majjhimanikāya (eds. R. Chalmers & V. Trenckner, 1977–
1979)
MW A Sanskrit-English Dictionary (ed. Monier Monier-
Williams, 1899)
Nett Nettipakaraṇa (ed. E. Hardy, 1902)
PDhp Patna Dhammapada (ed. G. Roth, 1980)
Peṭ Peṭakopadeśa (ed. A. Barua, 1949)
PSVy Pratītyasamutpādavyākhyā (ed. G. Tucci, 1930)
 
PTS Pāli Text Society
PTSD Pali-English Dictionary (eds. T. W. Rhys Davids & William
Stede, 1921–1925)
RY Rangjung Yeshe Tibetan-English Dictionary
SĀ-G Guṇabhadra’s Chinese translation of the Saṃyuktāgama, Zá
āhánjīng 雜阿含經, (T 99)

SĀ-U Chinese translation of the Saṃyuktāgama by an unknown
translator, Biéyì zá āhánjīng 別譯雜阿含經 (T 100)
SHT Sanskrithandschriften aus den Turfanfunden (eds. E.
Waldschmidt et al., 1965ff)

SN Saṃyuktnikāya (ed. M. Léon Feer, 1884–1898)
SNN Saṃyuktnikāya (ed. G. A Somaratne, 1998)
Sn Suttanipāta (eds. D. Andersen & H. Smith, 1913)
Spk Sāratthappakāsinī (ed. F. L.Woodward, 1929– 1937)
ŚrBh Śrāvakabhūmi (ed. K. Shukla, 1973)

T Taishō Shinshū Daizōkyō 大正新脩大藏經 (eds. Takakusu J.
and Watanabe K., 1924–1934)

Th Theragāthā (eds. K. R. Norman & L. Alsdorf, 1966)
Thī Therīgāthā (eds. K. R. Norman & L. Alsdorf, 1966)
Ud Udāna (ed. P. Steinthal, 1885)
Uv-B Udānavarga (ed. F. Bernhard, 1965)
Uv-S Udānavarga (ed. L. Schmithausen, 1970)
YBh Yogacārabhūmi (T 30 no. 1579; Derge 4035; Peking 5536)
 
Chapter one: Introduction

This introductory chapter is divided into six parts. The Śarīrārthagāthā will be introduced in the first part, including its overall position within the Yogacārabhūmi. A literature review will be provided in the second part in order to summarise previous research, which will assist in avoiding potential mistakes and obtaining the most comprehensive findings possible. In the third part, this study’s aims and significance will be described. The fourth part will explain the methodology by which the aims of this study will be reached. The overall structure of this dissertation will be outlined in the fifth part, in order to help readers easily grasp the points of each chapter. In the final part, the primary sources relevant to the Śarīrārthagāthā will be introduced, that is, the Sanskrit manuscripts, and Chinese and Tibetan translations and exegetical literature.

1.1. Three introductory aspects of the Śarīrārthagāthā


In this section, three aspects of the Śarīrārthagāthā will be discussed.

Because of the wide range of materials that make up the Yogacārabhūmi, the first aspect to be presented will be the place, role and function of the Śarīrārthagāthā in the Yogacārabhūmi. The second aspect to be discussed will be the classification of the verses of the Śarīrārthagāthā. The Śarīrārthagāthā is composed of forty one verse sets with commentary on each verse set. The classification of Śarīrārthagāthā verses constitutes an overview of the content of each verse set. The third aspect is to explore the meaning of the title “Śarīrārthagāthā” and the relationship between the title and content.
 
1.1.1. The place of the Śarīrārthagāthā in the Yogacārabhūmi


As shown in the chart below, the Śarīrārthagāthā is located in the Cintāmayībhūmi, the eleventh section (bhūmi) of the Yogacārabhūmi. The Cintāmayībhūmi (Chin. Sīsuǒchéng dì 思所成地, Tib. bsams pa las byung ba'i
sa)1 consists of three parts:Svabhāvaviśuddhita, Jñeyapravicaya and

Dharmapravicaya, which are the three works or tasks that followers of Yogācāra (practice of spiritual discipline), should perform. The first, Svabhāvaviśuddhita, involves the purifying of self-nature: a Yogācāra follower should take nine preparatory steps towards purifying his/her thought.2 Once thought has been purified, the follower’s self-nature is considered clean. The second part, Jñeyapravicaya, is the examination of observed object(s) (parīkṣya- artha),3 which are defined as “the existence of objects that exist and the non-
existence of objects that do not exist.”4 The third and last part, Dharmapravicaya, entails the examination of the dharma, which takes two forms—the meaning of the discourses (sūtra-artha) of the Buddha and the meaning of the verses

(gāthā-artha). The discourses (sūtra) here are Vastusaṃgrahaṇi, one of the five major divisions of the Yogacārabhūmi which briefly comments upon the canonical text of the Saṃyuktāgama, and the instruction of the bodhisattvapiṭaka, which refers to Mahāyāna texts.5 The verses (gāthā) are grouped as follows: the


1 Hereafter, the following abbreviations will be used: Chin. = Chinese, Tib. = Tibetan, Skt. = Sanskrit.
2 Śrāvakabhūmi (ŚrBh) 23b4M-4R; Derge Tibetan translation (D) 200b5–201a1; Taishō Chinese translation (T) 30.361b21–29.
3 ŚrBh 23b6L; D 201a1; T1579, 30.361c1.

4 謂於有法了知有相 於非有法了知無相 如是名為所觀察義 (T1579, 30.361c2–3); sac ca sato 'sac cāsathḥ parīkśyo 'rtha ity ucyate (ŚrBh 23b6M); yod pa la yang yod par, med pa la yang med par brtag pa ni, brtag par bya ba'i don ces bya'o, brtag par bya ba yod pa'i dngos (D 201a2).
5 Cf. Wayman 1961: 163.
 
Paramārthagāthā (verses on the supreme meaning), the Ābhiprāyikārthagāthā (verses on the intended meaning) and the Śarīrārthagāthā (verses on the meaning of “relics” or “corpus” or “collection”).6 The difference between these three groups of verses is that the Śarīrārthagāthā verses are all quoted from Śrāvakayāna texts, rather than from Mahāyāna ones. But the origins of the Paramārthagāthā and Ābhiprāyikārthagāthā have not as yet been identified with certainty.7 The structure of the Cintāmayībhūmi and the place of the Śarīrārthagāthā in it as the last section of the dharma-examination (dharmapravicaya) can be clearly seen in the following schema.


Cintāmayībhūmi:

1. Svabhāvaviśuddhita

2. Jñeyapravicaya

3. Dharmapravicaya

(1) Sūtrārthapravicaya

(2) Gāthārthapravicaya

A. Paramārthagāthā

B. Ābhiprāyikārthagāthā

C. Śarīrārthagāthā


As shown in the schema below, the Cintāmayībhūmi is the eleventh ground or level (bhūmi)8 of the Maulyo bhūmayaḥ/Maulī bhūmi9 in the Yogacārabhūmi.
6 For the details, see §1.1.3.

7 Ui (1958: 316–319) surmised that the verses in the Paramārthagāthā and Ābhiprāyikārthagāthā were canonical verses, though he was unable to locate them in the surviving canons. Wayman (1961: 164) conjectured that the author(s) utilized components of canonical verses to create their own verses. Schmithausen (1987: 223) suggested tha
t the verses in the Paramārthagāthā, which are not entirely original compositions, may have been quoted from paracanonical sources.
8 The definition of bhūmi does not imply a “stage” of spiritual progress, but refers to various “levels” or ”grounds” of spiritual cultivation. For example, the Śrāvakabhūmi and Pratyekabuddhabhūmi are followed by the Bodhisattvabhūmi, but that does not mean that the meditator must reach the previous two stages in order to have the ability to study the works of a
 
While there are seventeen grounds (saptadaśa bhūmayaḥ) in the Maulyo bhūmayaḥ, they are independent and separate systems that are interrelated.10 The Yogācāra follower does not need to begin his/her practice with the first ground. The Cintāmayībhūmi emphasizes the conditions, methods and contents of contemplation or reflection.


Bodhisattva. Deleanu (2006: 48) sometimes used the term bhūmi and at other times renders it as “textual unit”, “ground” or “foundation.” But he chose “textual unit” as the best description of bhūmi on the basis of the occurrence of the three expressions Gotrabhūmi, Avatārabhūmi, and Naiṣkramyabhūmi in the Śrāvakabhūmi. These three examples, Deleanu (2006: 36) claimed, can also apply to other meanings such as “step” or “position.” However, I prefer to take bhūmi as “ground” because the meaning “textual unit” is not common in the Yogacārabhūmi, its Indian commentaries, or even Sanskrit texts generally. Also, three examples of such usage are not sufficient to generalize or apply to the titles of the seventeen bhūmis in the Maulyo bhūmayaḥ. The meaning “ground” can be seen in Jinaputra’s Yogacārabhūmivyākhyā (瑜伽師地論釋, T1580,

30.884c25–26): “the meditator depends on this place to boost the white dharma [positive
spiritual benefit], so it is called bhūmi, like farming land” (瑜伽師依此處所增長白法, 故名為地， 如稼穡地).
9 Modern scholars are still unsure as to whether the entire first part was called Maulī bhūmi (singular form) or Maulyo bhūmayaḥ (plural form), because we do not possess the complete Sanskrit text. But Deleanu (2006: 46) concluded that Maulyo bhūmayaḥ seems appropriate given the evidence from the Tibetan text. See Hakamaya 1982: 59–60, Schmithausen 2000: 245 note 3
and Deleanu 2006: 44–46.

10 The seventeen grounds are: 1. Pañcavijñānakāyasaṃprayuktābhūmi (Wǔshí shēn xiàngyīng dì 五識身相應地; rnam par shes pa’i tshogs lnga dang ldan pa’i sa), 2. Manobhūmi (Yì dì 意地; yid kyi sa), 3. Savitarkāsavicārābhūmi (Yǒu xún yǒu sì dì 有尋有伺地; rtog pa dang bcas shing dpyod pa dang bcas pa’i sa), 4. Avitarkāvicāramātrābhūmi (Wú xún wéi sì dì 無尋唯伺地; rtog pa med

cing dpyod pa tsam gyi sa), 5. Avitarkāvicārābhūmi (Wú xún wú sì dì 無尋無伺地; rtog pa yang med la dpyod pa yang med pa’i sa), 6. Samāhitābhūmi (Sānmóxìduō dì 三摩呬多地; mnyam par bzhag pa’i sa), 7. Asamāhitābhūmi (Fēi sānmóxìduō dì 非三摩呬多地; mnyam par ma bzhag pa’i sa), 8. Sacittikābhūmi (Xīn dì 心 地 ; sems yod pa’i sa ), 9. Asacittikābhūmi (Wú xīn dì 無 心 地 ; sems med pa’i sa), 10. Śrutamayībhūmi (Wén suǒ chéng dì 聞所成地; thos pa las byung ba’i sa),

11. Cintāmayībhūmi (Sī suǒ chéng dì 思 所 成 地 ; bsams pa las byung ba’i sa), 12. Bhāvanāmayībhūmi (Xiū suǒ chéng dì 修所成地; bsgoms pa las byung ba’i sa), 13. Śrāvakabhūmi (Shēngwén dì 聲聞地; nyan thos kyi sa), 14. Pratyekabuddhabhūmi (Dújiào dì 獨覺地; rang sangs rgyas kyi sa), 15. Bodhisattvabhūmi (Púsà dì 菩 薩 地 ; byang chub sems dpa’i sa), 16. Sopādhikābhūmi (Yǒuyú yī dì 有餘依地; phung po dang bcas pa’i sa ) and 17. Nirupādhikābhūmi (Wúyú yī dì 無餘依地; phung po med pa’i sa).
 

The structure of the Yogacārabhūmi


1. Maulyo bhūmayaḥ (fascicle 1–50)
(1) Pañcavijñānakāyasaṃprayuktabhūmi
(2) Manobhūmi
(3) Savitarkāsavicārābhūmi
(4) Avitarkavicāramātrābhūmi
(5) Avitarkāvicārābhūmi
(6) Samāhitābhūmi
(7) Asamāhitābhūmi
(8) Sacittakābhūmi
(9) Asacittikābhūmi
(10) Śrutamayībhūmi
(11) Cintāmayībhūmi
(12) Bhāvanāmayībhūmi
(13) Śrāvakabhūmi
(14) Pratyekabuddhabhūmi
(15) Bodhisattvabhūmi
(16) Sopādhikābhūmi
(17) Nirupādhikābhūmi

2. Viniścayasaṃgrahaṇī (fascicle 51–80)

3. Vyākhyāsaṃgrahaṇī (fascicle 81–82)

4. Paryāyasaṃgrahaṇī (fascicle 83–84)

5. Vastusaṃgrahaṇī (fascicle 85–100)


The Maulyo bhūmayaḥ (Chin. Běn dì fèn 本地分; Tib. sa'i dngos gzhi), which
constitutes half (50 fascicles) of the Yogacārabhūmi, is the first and indispensable core section of Yogacārabhūmi. At the beginning of the Yogacārabhūmi, seventeen grounds (saptadaśa bhūmayaḥ) are regarded as Yogacārabhūmi (Yogacārabhūmiḥ katamā / sā saptadaśa bhūmayo draṣṭavyā
 
/ ).11 The Yogacārabhūmi can be established because of these seventeen grounds. The second division, Viniścayasaṃgrahaṇī (Shè juézé fèn 攝 決 擇 分 ; rnam par gtan la dbab pa bsdu ba) comprising 30 fascicles, is a detailed discussion or
commentary on the Maulyo bhūmayaḥ. The third division, Vyākhyāsaṃgrahaṇī12 (Shè shì fèn 攝 釋 分 ; rnam par bshad pa bsdu ba), consisting of 2 fascicles, explains the exegetical principles for interpreting the various Śrāvakayāna sūtras. The fourth division, Paryāyasaṃgrahaṇī (Shè yìmén fèn 攝異門分; rnam grangs bsdu ba), which contains 2 fascicles, collects, classifies and expounds
upon the synonyms found in the Śrāvakayāna sūtras.13 The final division, Vastusaṃgrahaṇī (Shèshì fèn 攝 事 分 ; gzhi bsdu ba) comprising 14 fascicles, summarizes the crucial parts of the Buddha’s teachings as found in the Tripiṭaka.

According to modern research,14 the contents of the Yogacārabhūmi are heterogeneous. The different layers are identifiable and reflect different views, approaches and developments of thought.15 The general opinion is that the Maulyo bhūmayaḥ contains the oldest material. In particular, Schmithausen (1987: 14) identified the oldest layers of the Yogacārabhūmi on the basis of the absence of reference to ālayavijñāna.



11 Yogacārabhūmi manuscript 3a01L.

12 The order of the remaining three divisions in the Tibetan version differs from that found in the Chinese version. Also, there is an extra division in the Tibetan version called the Vinayasaṃgrahaṇī which has been separated from the Vastusaṃgrahaṇī. Therefore, the order of the last three Tibetan divisions is Vastusaṃgrahaṇī (and Vinayasaṃgrahaṇī), Paryāyasaṃgrahaṇī, and Vyākhyāsaṃgrahaṇī. See Suguro 1989: 280–283 and Deleanu 2006: 46.
13 See Jinaputra's Yogacārabhūmivyākhyā (瑜伽師地論釋; T1580, 30.884c25–26). But Yìnshùn
(1988: 241–2) said that both the Vyākhyānasaṃgrahaṇī and Paryāyasaṃgrahaṇī explain the doctrines of the Āgamas and the various synonyms used in the Āgamas.

14 Mukai 1981, Schmithausen 1987: 14, Yìnshùn 1988b: 241–2, Aramaki 2000, Lin 2002, Kritzer
2005: xviii, Dhammajoti 2006, 2007a, Bronkhorst 2009: 158.
15 See Schmithausen (2007). Schmithausen also stated that “the compilation of the text seems to have taken place in several steps, perhaps in the fourth century A.D., but the chronology of these texts is fraught with uncertainty.”
 
1.1.2. The classification of verses in the Śarīrārthagāthā


The Śarīrārthagāthā is composed of forty-one verse sets with its commentary. The classification of the Śarīrārthagāthā verses, as explained in the uddāna found at the end of this text, is as follows:

evil, what can be expressed, lust, flood, being frightened, caste, obtaining fame,16 current, (two) floods,17 lust and hatred, what ought to be done, struggle, attainment of the aim, and specification; [this is a total of] fourteen [kinds].”18
The Śarīrārthagāthā verse sets may be assigned with these titles from this classification:

The first set: evil (è 惡, pāpa, sdig pa); 4 pādas / 1 verse.
The second set: what can be expressed (shuō 說, ākhyeya, brjod bya); 16
pādas / 4 verses.

The third set: lust (tān 貪, rāga, 'dod chags); 20 pādas / 5 verses. The fourth set: flood (liú 流, ogha, chu bo); 12 pādas / 3 verses.
The fifth set: being frightened (bù 怖 , utrasta, 'jigs); 14 pādas / 3 verses. The sixth set: caste (lèi 類 , abhivarṇṇatā, rigs); 12 pādas / 3 verses.

The seventh set: obtaining fame (yù 譽, yaśasvin, grags ldan); 8 pādas / 2

16 In Chinese, the word yǔ 與 appears to be a scribal error. Judging by the content, it perhaps ought to read yù 譽. Yù 譽 is also consistent with the Sanskrit and Tibetan versions.

17 The Chinese version does not include the word “two,” but rather simply states liú 流 (flood). But “two floods” fits better because the Sanskrit term is oghau (two floods) and there are two
verse sets about floods in this division.

18 The content in Chinese is 惡說貪流怖 類與池流貪 作劬勞得義 論議十四種 (T1579, 30.388a29–b1; FS 18.162a20). The content in Sanskrit is pāpam ākhyeyarāgaś ca oghotrastābhivarṇṇatā / yaśasvī sara oghau rāgadveṣau ca kāryatā / āyūhanam arthaprāpti upadeśaś caturddaśaḥ // (Enomoto 1989: 21). However, the structure of the Tibetan uddāna is

different, which I will explain later in this section. The first thirteen divisions in the Tibetan uddāna are sdig dang brjod bya 'dod chags dang / / chu bo dang ni 'jigs dang rigs / / grags ldan mtsho dang chu bo gnyis / / 'dod chags zhe sdang bya ba dang / / 'dzin pa dang ni don thob dang
/.
 
verses.

The eighth set: current (chí 池 , sara, mtsho); 12 pādas / 3 verses. The ninth set: flood (liú 流, ogha, chu bo gnyis); 8 pādas / 2 verses. The tenth set: flood (liú 流, ogha, chu bo gnyis); 12 pādas / 3 verses.
The eleventh set: lust and hatred (tān 貪 , rāgadveṣa, 'dod chags zhe sdang); 16 pādas / 4 verses.
The twelfth set: what ought to be done (zuò 作, kāryatā, bya ba); 20 pādas

/ 5 verses.

The thirteenth set: struggle (qú láo 劬勞, āyūhana, 'dzin pa); 4 pādas / 1 verse.
The fourteenth set: attainment of the aim (dé yì 得義, arthaprāpti, don thob); 12 pādas / 3 verses.
The fifteenth set: specification (lùnyì 論議, upadeśa)19; 182 pādas / 45 verses.


The ninth and tenth sets share the one title, “flood,” so there are fourteen kinds/divisions in total, as mentioned in the uddāna.
However, the fourteenth division (the fifteenth verse set) in the Chinese and Sanskrit versions called “specification” is not included in the Tibetan version.
The details of the Tibetan uddāna are as follows:20


sdig dang brjod bya 'dod21 chags dang / / chu bo dang ni 'jigs dang rigs / / grags ldan mtsho22 dang chu bo gnyis / / 'dod chags zhe sdang bya ba dang
19 This is not mentioned in the Tibetan uddāna.
20 See Peking 5536.304b3–b6; Derge 4035.261a7–b2; Golden (TBRC) 138 756.6–757.3.
21 Golden: ’dong.
22 Golden: mtshon.
 
/ / 'dzin pa dang ni don thob dang / / mi rtag23 pa24 dang bag yod dang / / 'dod chags shas che chos la gnas / / nges rtag25 legs par26 smras pa dang / / dad pa dang ni thos pa dang / / nam mkha'27 mtshungs dang gzugs dang rgyal / / rus pa'i ra ba rus28 sbal dang / / mtshungs dang 'dam dang gnas gnyis dang / / bsal29 gyur dang ni byin pa dang / / sdig pa kun dang ltung ba dang / / mkhas dang bzo30 dang ring du 'gro / / bsgribs dang 'dod dang bzang po yin // (Peking 5536.304b3–b6; Derge 4035.261a7–b2; Golden 138 756.6–757.3).
Evil, what is expressed, lust, flood, being frightened, caste, famous, current, two floods, lust and hatred, what is to be done, grasping, attainment of the aim, impermanence, carefulness, strong lust, dwelling in the dharma, contemplating the truth, good speech, faith, learning, being like the sky, physical appearance, king, city of bone, turtle, equality, mud, two dwellings, having cleaned up, giving, all evils, falling down, skilled, craft, travelling far away, obscuring, sensual pleasures, and auspicious.


The Tibetan version is longer since twenty seven individual divisions replace the single division in the Chinese/Sanskrit version called “specification.”31 However, these twenty seven divisions in the Tibetan version help us understand the internal structure of the “specification” division in Chinese/Sanskrit version.


23 Golden and Peking: rtags.

24 Golden: ba.

25 Golden: rtags.

26 Golden: bar.

27 Peking: kha'.

28 Golden: ru.

29 Golden and Peking: brtsal.

30 Golden and Peking: gzo.

31 The sixteenth and seventeenth sets of Group B share the title “two dwellings” (gnas gnyis).
 

In order to better understand the Śarīrārthagāthā verses, it is best to divide them into two groups. The first thirteen divisions could be classified as “Group A” and the last division “specification” (or twenty seven divisions in Tibetan) as “Group B.”32 This is reasonable due to the fact that the characters of these two groups are different. The Group A verses are attributed to the geya aṅga and are all cited from the Saṃyuktāgama, while the Group B

verses belong to the vyākaraṇa aṅga and are cited from the Udānavarga, Pārāyaṇa, Arthavargīya and Madhyamāgama.33 Also, the number of verses in each group is similar (Group A contains 170 pādas / 42 verses and Group B includes 182 pādas / 45 verses). The identification of these two discreet groups, (A and B) which has not been made before, shows a deliberate structure by the author(s) of the Śarīrārthagāthā.
Three circumstances of the Śarīrārthagāthā verses need to be noticed.

Firstly, the Tibetan title “auspicious” (bzang po) of the last (twenty seventh) set in Group B cannot be found in the content of the verse. The uddāna terms have generally been taken from the first or second verse pāda of each division in the Śarīrārthagāthā. The term “auspicious” (bzang po) possibly implies that this verse set is cited from the sūtra called “auspicious night” (’phags pa mtshan mo bzang po zhes bya ba'i mdo).34 Secondly, in the thirteenth set, the final verse of the parallels35 is not quoted in the Śarīrārthagāthā.36 Instead, only the prose can


32 For details, see the tables below.

33 The details can be seen in chapter two.

34 This Tibetan title might correspond to the Sanskrit “Bhadrakarātrīnāma-ārya‐sūtra.” The corresponding sūtra is MĀ 165. See Skilling 1997: 585, Anālayo 2012: 422.

35 SĀ-G: “久見婆羅門 逮得般涅槃 一切怖已過 永超世恩愛 (T99, 2.348b19–20); SĀ-U: “我昔已曾見 婆羅門涅槃 久捨於嫌怖 能度世間愛 (T99, 2.438c27–28); SN: cirassaṃ vata passāmi, brāhmaṇaṃ parinibbutaṃ. appatiṭṭhaṃ anāyūhaṃ, tiṇṇaṃ loke visattikan ti (SN I 1.21–22 = SNN
I 2.12–13).
 
be seen. Lastly, there are slight variations in the number of verse pādas in Group A between the three versions (see the below table).37 The differences are the second set (Sanskrit version), the eleventh set (Tibetan version), and the twelfth and fourteenth sets (Chinese version).


Group A

set Chinese Sanskrit Tibetan
Verse Pāda Verse Pāda Verse Pāda
1 1 4 1 4 1 4
2 4 16 4 17* 4 16
3 5 20 5 20 5 20
4 3 12 3 12 3 12
5 3 14 3 14 3 14
6 3 12 3 12 3 12
7 2 8 2 8 2 8
8 3 12 3 12 3 12
9 2 8 2 8 2 8
10 3 12 3 12 3 12
11 4 16 4 16 4 15*
12 5* 20* 4 14 4 14
13 1 4 1 4 1 4

14 3* 12* 2 10 2 10
(*represents a different verse or pāda number comparing with other two versions)


37 All three versions of Group B have the same number of verse pādas.

Group A
set number of verse Chinese Sanskrit Tibetan English Translation

1 4 pādas / 1 verse è 惡 Pāpa sdig pa evil
2 16 pādas / 4 verses shuō 說 Ākhyeya brjod bya what can be expressed
3 20 pādas / 5 verses tān 貪 Rāga 'dod chags lust
4 12 pādas / 3 verses liú 流 Ogha chu bo flood
5 14 pādas / 3 verses bù 怖 Utrasta 'jigs being frightened
6 12 pādas / 3 verses lèi 類 abhivarṇṇatā rigs caste
7 8 pādas / 2 verses yù 譽 Yaśasvin grags ldan obtaining fame
8 12 pādas / 3 verses chí 池 Sara mtsho current
938 8 pādas / 2 verses liú 流 Oghau chu bo gnyis two floods
10 12 pādas / 3 verses
11 16 pādas / 4 verses tān 貪 rāgadveṣa 'dod chags zhe sdang lust and hatred
12 20 pādas / 5 verses zuò 作 Kāryatā bya ba what ought to be done
13 4 pādas / 1 verse39 qúláo 劬勞 Āyūhana 'dzin pa struggle
14 12 pādas / 3 verses déyì 得義 Arthaprāpti don thob attainment of the aim

38 The ninth and tenth verse sets share the same title.
39 This verse has not been mentioned in the Śarīrārthagāthā. See §3.1.13 for details.
 

Group B

set number of verse Tibetan title English Translation40 Chinese Translation from Tibetan41
1 4 pādas / 1 verse mi rtag pa impermanence wúcháng 無常
2 4 pādas / 1 verse bag yod carefulness wúyì 無逸
3 4 pādas / 1 verse 'dod chags shas che strong lust měnglìtān 猛利貪
4 4 pādas / 1 verse chos la gnas dwelling in the dharma zhùfǎ 住法
5 4 pādas / 1 verse nges rtag contemplating the truth dìsī 諦思
6 4 pādas / 1 verse legs par smras pa good speech shànyǔ 善語
7 4 pādas / 1 verse dad pa faith jìngxìn 淨信
8 4 pādas / 1 verse thos pa learning duōwén 多聞
9 4 pādas / 1 verse nam mkha' mtshungs being like the sky rúxūkōng 如虛空
10 20 pādas / 5 verses gzugs physical appearance sè 色
11 4 pādas / 1 verse rgyal king wáng 王
12 4 pādas / 1 verse rus pa'i ra ba city of bone gǔwéiqiáng 骨為墻
13 4 pādas / 1 verse rus sbal turtle guī 龜


40 Translated by myself according to the Tibetan text.

41 Translated by Hán (1947) from the Tibetan text.
 

14 4 pādas / 1 verse mtshungs equality děng 等
15 4 pādas / 1 verse 'dam mud yūní 淤泥
16 4 pādas / 1 verse gnas gnyis two dwellings èrchù 二處
17 4 pādas / 1 verse
18 4 pādas / 1 verse bsal gyur having cleaned up xūnchú 熏除
19 4 pādas / 1 verse byin pa giving huìshī 惠施
20 4 pādas / 1 verse sdig pa kun all evils zhūè 諸惡
21 4 pādas / 1 verse ltung ba falling down lúnzhuì 淪墜
22 4 pādas / 1 verse mkhas skilled shànzhī 善知
23 4 pādas / 1 verse bzo craft gōngqiǎo 工巧
24 4 pādas / 1 verse ring du 'gro travelling far away xīnyuǎnxíng 心遠行
25 40 pādas / 10 verses bsgribs obscuring néngfù 能覆
26 24 pādas / 6 verses 'dod sensual pleasures zhūyù 諸欲
27 6 pādas / 1 verse bzang po auspicious xiánshàn 賢善
 

1.1.3. The meaning of the title Śarīrārthagāthā

As mentioned above, there are three gāthā sections in the Cintāmayībhūmi. However, the title Śarīrārthagāthā is more confusing than the other two—the Paramārthagāthā and Ābhiprāyikārthagāthā—whose meanings can be found in the commentary.42 The title Śarīrārthagāthā is neither connected with the content of the verses nor the commentary, nor are there any clues in the Yogacārabhūmi. Therefore, scholars have made various interpretations of the title Śarīrārthagāthā. In this section, the various interpretations will first be discussed, and then I will provide a few possibile solutions to address this.


The difficult part of the term Śarīrārthagāthā is śarīra-artha. There are various interpretations of this compound. Based on a survey of the Tibetan version, Wayman (1994: 97) regarded śarīra as “relics” which, in this context, could refer to the Buddha’s speech.43 According to this interpretation, śarīra-artha is a genitive tatpuruṣa compound, literally meaning “the meaning of relics.” However, Hán (1947: 36)

translated the title as “having gathered meanings,44 and took “'dus pa” in Tibetan to mean “gathering.” Kragh (2013: 96) translated the Sanskrit term śarīra-artha as “the meaning of the [[[Wikipedia:canonical|canonical]]] corpus” and took śarīra in Sanskrit to mean “corpus.”45

According to this interpretation, the compound is a genetive tatpuruṣa. In Chinese, two scholars—Huìjǐng 惠景 and Kuījī
窺基—mistake “tǐ 體” as bhāva rather than śarīra due to limited information from the
Chinese translation. Their interpretations thus are more likely to be “character”/“nature” (tǐxìng 體性).46 Moreover, Ui (1958: 321) suggested that tǐ 體 (śarīra) and yì 義 (artha) mean “verse” and “commentary,” while Hán (1983: 617, 661) took 體 (śarīra) as “the
 
42 The Paramārthagāthā literally means “verse on the supreme meaning.” “Supreme meaning” involves selfless and empty concepts. This section has been studied by Wayman (1961) and Schmithausen (1987a). The Ābhiprāyikārthagāthā literally means “verse on the intended meaning.” “Intended meaning” discusses

in detail the three learnings/trainings, namely, the higher moral conduct (zēngshàng jiè 增 上 戒 , lhag pa'i tshul khrims), the higher mind (zēngshàng dìng 增上定, lhag pa'i sems), and the higher wisdom (zēngshàng huì 增上慧, lhag pa'i shes rab).
43 In an earlier work, Wayman (1961: 163) translated śarīra as “corpus.” Thus, he seems to have changed his interpretation of this term.

44 Hán (1947: 36) defined it as “jíyì gātā 集義” (having gathered meanings) from the Tibetan term. He mentioned that each verse set includes several meanings (zǒngshè zhòngyì 總攝眾義). However, I think 'dus pa'i don is more likely to be a genitive tatpuruṣa compound.
45 Kragh (2013: 48) mentioned that the equivalent terms do not mean actual literal translations of the original Sanskrit terms. The English equivalent term is rather a short explanatory phrase.

46 The following is from a description in a Chinese text: 體義伽陀者 景云謂明諸法體性差別 基云謂明伽陀中體性義理 (T1828, 42.416b5–6). Huìjǐng 惠景 translated it as “understand the difference of the dharma characters” and Kuījī 窺基 translated it as “understand the characters and doctrine in verses.”
 
points of dharma” (fǎyào 法要) referring to the first fourteen verse sets and took yì 義(artha) as “the meanings of doctrine” referring to the remaining verse sets. Lǚ (1924) suggested that the term may be translated as “the meaning of the verses.” In other words, tǐ 體 (śarīra) means “verse” itself. Yìnshùn (1969: 509) said “from the verses, tǐyì 體義of dharma can be understood” (依頌文而明法的體義).47 This explanation is ambiguous in that the meaning of tǐ 體 (śarīra) and yì 義 (artha) are not clear. From the Chinese interpretations, the compound can be seen as a genitive tatpuruṣa or dvandva. The following table shows all of the interpretations:


śarīra artha compound

Wayman relics meaning genitive tatpuruṣa

Hán gathering meaning accusative tatpuruṣa

Kragh corpus meaning genitive tatpuruṣa

Huìjǐng character/nature X X

Kuījī character/nature doctrine dvandva

Ui verse commentary dvandva

Hán the points of

dharma the meanings of

doctrine dvandva

Lǚ verse meaning genitive tatpuruṣa

Yìnshùn [[[Wikipedia:ambiguous|ambiguous]]] [[[Wikipedia:ambiguous|ambiguous]]] X


Most of the Chinese and Japanese scholars do not utilise the Sanskrit and Tibetan versions for reference, so they interpret śarīra-artha in various ways. In my opinion, the compound śarīra-artha is a genitive tatpuruṣa compound referring to the other two gāthā sections, the Paramārthagāthā and Ābhiprāyikārthagāthā. Each gāthā section focuses on one specific meaning (parama, ābhiprāyika, and śarīra) of the verses.

From the above discussion, it is clear that śarīra is particularly important in understanding the meaning of the title. According to the Sanskrit-English Dictionary by


47 From next sentence of the same paragraph, Yìnshùn (1969: 509) seems to have taken tǐ 體 as character/nature (tǐ xìng 體性), based on the Chinese translation of the Yogacārabhūmi. The assumption that Yìnshùn intended to say is “The meaning of dharma character will be understood by verses” (依頌文而明法的體義).
 
Monier-Williams, the term śarīra is related to one of two roots, namely, √śṝ (to hurt, injure), and √śri (to lean on, rest on, lay on, etc.). The literal meaning is “body, bodily frame, solid parts of the body.”48 The definition from √śṝ emphasizes that the body is “easily destroyed or dissolved,” but the definition from √śri stresses the body as a support or supporter. In the latter, the body is like a collection

of parts. This meaning is also reflected in the Tibetan translation “'dus pa” (aggregation, collection) 49 for the title Śarīrārthagāthā.50
Based on these two definitions, there are three ways to explain the term śarīra in the Śarīrārthagāthā. The first definition, the body (śarīra from √śṝ) metaphorically means the Buddha’s teaching, developed from the meaning “relics,” just like the interpretation by Wayman (1994: 97).51 This meaning is consistent with the fact that the verses in the Śarīrārthagāthā are seemingly left from the Buddha’s time and are all found in early Buddhist texts (see chapter 2–4). In other words, these verses represent the heritage of the Buddha, another aspect of

“relics,” besides dead body or bodily ashes. Here, Śarīrārthagāthā means “verses on the meaning of relics.” From the second definition, the body (śarīra from √śri) is more like “corpus,” or “collection” in the Śarīrārthagāthā. This can be explained in two ways. The first is that the Śarīrārthagāthā is a corpus or collection of early verses. Therefore, the meaning of Śarīrārthagāthā is “verses on the meaning of the corpus/collection.” The

second connects with the function of geya, “to gather.” In chapter two, it will be shown that the verses in the Śarīrārthagāthā are all geya verses, although the Group B verses are also attributed to the vyākaraṇa aṅga.52 “Gathering” [the prose or the prose’s meaning] is the early function of the geya verse.

That is, each verse represents a collection which gathers together the prose or the prose’s meaning. Thus, the term śarīra may mean “collection.” The Śarīrārthagāthā can be defined as “verses on the meaning of collection.”
In brief, the Śarīrārthagāthā can be regarded as “verse on the meaning of ‘relics’, or ‘corpus’ or ‘collection’.” The key word śarīra (literal meaning “body”) has been explained as relics (the Buddha’s teaching), corpus (collecting the verses), and collection


48 This is the meaning of the singular form. In the plural form, it can mean “bones.” See MW s.v. śarīra.
49 See JH-Eng and RY s.v. 'dus pa.
50 The Tibetan title is 'dus pa'i don gyi tshigs su bcad pa.
51 Wayman (1994: 97) regarded śarīra as relics which, in this context, could mean “the Buddha’s speech.”
52 See chapter two for further details.
 
(the geya verse). Importantly, the meaning of śarīra from √śri is also valued by Tibetan translator(s).53

1.2. Literature review

To date, the Śarīrārthagāthā has not attracted a comprehensive investigation. Each scholar has investigated one or a few aspects of the Śarīrārthagāthā. In this section, scholarly studies concerning the Śarīrārthagāthā will be summarised and discussed.

1.2.1. Relevant information regarding the Śarīrārthagāthā

There are at present three components of relevant information concerning the Śarīrārthagāthā: background, title and topic, and commentary. These three aspects can portray the initiatory scope of the Śarīrārthagāthā.

1.2.1.1. Background

The background information refers to the school to which the Śarīrārthagāthā might be attributed, and the origination of some of its verses. Four articles are as follows:

1. Schmithausen (1970) suggested that the Udānavarga of Bernhard’s (1965) edition belongs to the Sarvāstivādins but the verses from the Udānavarga that the Śarīrārthagāthā cited belong to the Mūlasarvāstivādins. The difference between these two versions is the wording.

2. Aramaki (1976) argued that the verses are older than the prose. The evidence he presented is that some verses in the Suttanipāta and the Yogacārabhūmi also exist in the Mahābhārata and an old Jain sūtra called Isibhāsiyāin.

3. Schmithausen (1987), who edited, translated and studied the eighth set of the Group A verses, stated that the canonical citations in the Yogacārabhūmi could be ascribed to the Mūlasarvāstivādins.

53 This is why the Tibetan translator(s) chose “'dus pa” (aggregation, collection) as the translation.
 
4. Enomoto (1989), after analyzing the words of some verses, deduced that both the Śarīrārthagāthā and the Eastern Turkestan versions were later than the Chinese Saṃyuktāgama and Arthavargīyasūtra, and the Pāli Suttanipāta. He suggested that this is a later recension of the Mūlasarvāstivādins.
In sum, although some verses might originate from very old texts, According to Enomoto (1989), the verses in the Śarīrārthagāthā seem to represent a later recension belonging to the Mūlasarvāstivādins.

1.2.1.2. Title and topic

The first five following articles are related to studies on the title, while the other two articles are about the topic:

1. Regarding the term tǐyì 體 義 (śarīrārtha), Lǚ (1924) suggested that, according to two other gāthā sections, the Paramārthagāthā and the Ābhiprāyikārthagāthā, “tǐ 體(existence, nature) is yì 義 (meaning).” He stated that the meaning is based on the existence of the verses.
2. Hán (1947) explored the Tibetan Śarīrārthagāthā, and suggested that 'dus pa (śarīra) means “gathering” which means “having gathered meanings from the verses” (zǒngshè zhòngyì 總攝眾義). He also pointed out that the number of verses between the Chinese and Tibetan versions are different. He also stated that the meaning of upadeśa (a section including twenty six verse divisions) is the selection between mundane and supramundane dharmas.

3. Ui (1958) suggested that tǐ 體 (śarīra) and yì 義 (artha) mean “verse” and “commentary,” respectively.

4. Yìnshùn (1969) said “based on the verses, tǐyì 體義 of dharma can be understood” (依頌文而明法的體義). This explanation is ambiguous in that the meaning of tǐ 體 (śarīra) and yì 義 (artha) is not clear.

5. Hán (1983) took 體 (śarīra) to mean “the points of dharma” (fǎyào 法要), referring to the first fourteen verse sets, and yì 義 (artha) to mean “the meanings of doctrine,” referring to the rest of the verse sets.
 
6. Wayman (1984)54 suggested that the Śarīrārthagāthā is relevant to mind training (zēngshàng xīnxué 增上心學, lhag pa'i sems).

7. Mayeda (1992) stated that the Śarīrārthagāthā is relevant to “three learnings/trainings”
(sānxué 三學, ṭrīṇi śikṣāṇi) of moral conduct, meditation, and wisdom. He mentioned that the Śarīrārthagāthā is about “wisdom training” in the study of Wayman (1984). However, this is a misunderstanding. Wayman (1984: 354) stated “His Paramārtha- gāthā emphasizes the instruction of insight, and his following extended section in the Yogacārabhūmi—the Śarīrārtha-gāthā —deals much with Mind Training.”

In brief, there are various explanations of the title “Śarīrārthagāthā.” The topic of the
Śarīrārthagāthā might relate to mind training.

1.2.1.3. Commentary

There are two articles which discuss the features of the Śarīrārthagāthā
commentary:


1. Ui (1958: 341–342) mentioned that the commentary on the verses in the Śarīrārthagāthā is comprehensively developed with its own ideas. This is a common circumstance in the development of Buddhist philosophy.

2. Azami (2007) compared one verse in the Śarīrārthagāthā with its paralles in other texts. He ascertained that the commentary on this verse had its own interpretation because the explanation is only found in the Śarīrārthagāthā.

The commentary on the Śarīrārthagāthā verses seems to be unique, which raises the possibility that it was composed by the author(s) of the Yogacārabhūmi.


54 This study of Wayman (1984) is a translation of the Paramārthagāthā and Ābhiprāyikārthagāthā.
 
1.2.2. Sources of the Śarīrārthagāthā verses

This is a popular topic regarding the study of the Śarīrārthagāthā. There are seven articles which investigate the sources of the Śarīrārthagāthā verses, summarised as follows:

1. Lǚ (1924) hypothesized that the mātr̥kā (categorizing lists)55 of the Eight Assemblies section (Bāzhòng sòng 八眾誦) in the Saṃyuktāgama is the Śarīrārthagāthā (Group A in particular). But Lǚ still found differences and expressed uncertainty about his ideas.

2. Lǚ (19??)56 hypothesized that the mātr̥kā of the Dharmapada/Udānavarga is the Śarīrārthagāthā (Group B verses in particular); however, his evidence is not comprehensive.

3. Ui (1958: 320–342) attempted to identify the sources of the Śarīrārthagāthā verses based on insufficient resources, which resulted in an incomplete conclusion. There also seem to be several discrepancies between the more accepted recensions of the text and the sources upon which he relied.

4. Schmithausen (1970) provided evidence that a section of the verses of Group B was cited from the Mūlasarvāstivāda version of the Udānavarga.

5. Enomoto (1989) edited the verses of the Śarīrārthagāthā based on the Śrāvakabhūmi, Yogacārabhūmi manuscripts and Ui’s research (1958: 320–342); identified the source of the verses; and indicated parallels in other texts in the Pāli, Prakrit, Sanskrit, Chinese and Tibetan versions. However, in his findings, information about the parallels of the Śarīrārthagāthā verses relevant to the verses in T210 (Fǎjù jīng 法句經); T211 (Fǎjù
pìyù jīng 法句譬喻經); T212 (Chūyào jīng 出曜經); T213 (Fǎjíyàosòng jīng 法集要頌經)
is not insufficient.

6. Zōngzhèng (2006) was unaware of the study of Enomoto (1989) and Schmithausen (1970); thus, his findings are less comprehensive by comparison.

7. Chén (2006; 2007) did not draw upon Enomoto (1989) and Schmithausen (1970). Her findings are similarly less comprehensive in comparison with Enomoto’s.



55 See Cox 1995: 18–19 note 29 and Anālayo 2014: 222 note 26.
56 The actual date of this article cannot be found, but it should be earlier than 1953.
 
Amongst these finding, Enomoto (1989) undertook a more complete and accurate investigation. Azami (2007) drew upon Enomoto’s achievements and used them as the basis for new research. The breakthrough for Enomoto came when he obtained copies of Sanskrit manuscripts in Germany with the assistance of experts, such as Lambert Schmithausen.

1.2.3. Textual formation

Three scholars have studied the stratification of the Yogacārabhūmi as follows:


1. Schmithausen (1987a) suggested that there are three main layers of the Yogacārabhūmi. The first layer makes no reference to ālayavijñāna (“storehouse consciousness”), the second sporadically mentions ālayavijñāna but has no reference to the Saṃdhinirmocanasūtra, and the third contains detailed content on ālayavijñāna and information about the Saṃdhinirmocanasūtra. Also he mentioned that the Yogacārabhūmi was not based on a preparatory plan but was gradually developed from complex and separate textual units and materials.

2. Based on the study of Schmithausen (1987a), Deleanu (2006) suggested six phases of the formation of the Yogacārabhūmi, including the text of the Saṃdhinirmocanasūtra: 1. the Śrāvakabhūmi; 2. the Bodhisattvabhūmi; 3. the rest of the Maulyo bhūmayaḥ; the Vastusaṃgrahaṇī; the Vyākhyāsaṃgrahaṇī and the Paryāyasaṃgrahaṇī; 4. the Saṃdhinirmocanasūtra, closely related to the Yogacārabhūmi; 5. the early parts of the Viniścayasaṃgrahaṇī; 6. the citation of the Saṃdhinirmocanasūtra in the Viniścayasaṃgrahaṇī, and the final redaction of the entire Yogacārabhūmi.

3. Aramaki (2013) discovered a principle behind the compilation of the Maulyo bhūmayaḥ from statements in the Sacittakābhūmi. That is, for sake of propounding the mundane truth (saṃvr̥tisatya), seven bhūmis57 are the first part of the Maulyo bhūmayaḥ, so that the deeper meditative concentrations can be achieved and the subtler subconscious defilements can be found and then be removed. Moreover, he also asserted that the Śrāvakabhūmi and the Bodhisattvabhūmi are no doubt the two oldest independent

57 Seven bhūmis are the Pañcavijñānakāyasaṃprayuktabhūmi, Manobhūmi, Savitarkāsavicārābhūmi, Avitarkavicāramātrābhūmi, part of the Avitarkāvicārābhūmi, Samāhitābhūmi, and Asamāhitābhūmi. See Aramaki 2013: 426–427.
 
texts. These oldest texts were developed together with three bhūmis (Śrutamayī-, Cintāmayī-, and Bhāvanāmayī-), “playing the role of being practical manuals on the level of the mundane truth (saṃvr̥tisatya) to foster liberation from the deepest pursuing (ādāna) or cumulative subconsciousness (ālayavijñāna)” (Aramaki 2013: 427). Thus, these bhūmis are compiled with the seven bhūmis (as mentioned above).

The suggestions of Deleanu (2006) and Aramaki (2013) are for the most part based on the study of Schmithausen (1987a). Deleanu (2006) and Aramaki (2013) developed different details, but these do not conflict with each other.

1.2.4. Materials pertaining to aṅga classification

The following information about aṅga classification is relevant to the
Śarīrārthagāthā:


1. Three important studies regarding the comparison of the two aṅga classification systems (navāṅga and dvādaśāṅga) are Hirakawa 1960, Mayeda 1964 and Yìnshùn 1971. In these studies, Hirakawa analyzed aṅga classification based on various traditional Vinayas; Mayeda comprehensively investigated various aspects of aṅga classification based on early Buddhist texts (Nikāya, Āgama, Vinaya, etc.); and Yìnshùn deduced the functional development of aṅga classifications from wide-ranging sources, particularly Abhidharma texts. While drawing on different sources, these three studies come to the same conclusion that the nine-aṅga classification system (navāṅga) is the earliest.

2. Mayeda (1964) studied the relationship between two aṅgas, vyākaraṇa and upadeśa, which can help us understand the section called Upadeśa in the Śarīrārthagāthā.

3. Yìnshùn (1971) identified three elements to the Saṃyuktāgama structure, namely, the three aṅgas of sūtra, geya and vyākaraṇa. By applying Yìnshùn’s deduction concerning the development of geya and vyākaraṇa, I can situate the Śarīrārthagāthā verses in the formation of Buddhist texts in this study. Additionally, based on one of the vyākaraṇa meanings, the reason why the section Upadeśa in the Śarīrārthagāthā includes verses from various texts can be discerned.
 
4. Choong (2000) compared the sūtra aṅga portion between the Pāli Saṃyuttanikāya and Chinese Saṃyuktāgama (SĀ-G).

These studies on aṅga classification help us understand the relationship between early Buddhist texts and the Śarīrārthagāthā.

1.2.5. Other relevant studies

1. Mukai’s (1985) study is similar to those of Yìnshùn (1971, 1983). Importantly, he discovered the verse section in the Saṃyuktāgama called “Saṃgīta” and criticized Xuánzàng’s 玄奘 translation jiéjí 結集 as being incorrect.58

2. Enomoto (1986) investigated the formation of the four Chinese Āgamas. The
structures and schools of the texts he analysed can help us understand the texts from which the Śarīrārthagāthā is cited.

3. Kragh (2013) collected many significant studies about the Yogacārabhūmi. Regarding the Śarīrārthagāthā, he identified the structure of the verses and relevant studies of other scholars.


1.3. Aim and significance of this study


1.3.1. Aim

There are five directions for the instigation in this study:


Firstly, the sources of the verses, the place of the Śarīrārthagāthā verses in the formation of Buddhist texts (especially aṅga classification), and the reason why early verses were collected in the Śarīrārthagāthā will be investigated. As mentioned previously, many researchers have found the sources of these verses, and Enomoto’s study is the most complete. In this thesis, I will carry out further analysis based on the results found by these researchers, in particular the work of Yìnshùn. He was the first person who investigated the development and relationship between aṅga and Āgamas from the texts

58 I propose a different idea that Xuánzàng’s 玄奘 translation is traditional. See §2.2.1.
 
dating from early Buddhism to Mahāyāna. His study provides significant information for the investigation of the above issues.

Secondly, the verses in the Śarīrārthagāthā were highly regarded. This is evidenced by their appearance in many different texts, including sūtras and śāstras. In this study, background information found in early parallels will be explored. Also the distinctive characteristics of the Śarīrārthagāthā verses will be investigated by comparing them with their parallels, with the aim of assessing the school affiliation of the texts from which these verses were taken.

Thirdly, due to the stratification of the Yogacārabhūmi, the development of the Śarīrārthagāthā needs to be examined. Although the Śarīrārthagāthā does not appears to have been transmitted as an individual text, as a textual unit the stage of its development still can be investigated. This investigation is based on three scholars’ studies (Schmithausen 1987a, Aramaki 2000, Deleanu 2006), which provide the theory of textual development.

Fourthly, according to Enomoto (1989), these verses of the Śarīrārthagāthā were selected from several texts rather than from only one text. It is confusing that only particular verses from each text were cited in the Śarīrārthagāthā. Therefore, the principle of selecting the verses from early Buddhist texts will be explored.

Schmithausen (1987: 377–380) indicated that the verses of the Śarīrārthagāthā might belong to the Mūlasarvāstivāda school. Clues from the texts attributed to the Mūlasarvāstivādins will be explored.

Finally, there is a difference in the uddāna that structures the Śarīrārthagāthā verses between the Sanskrit/Chinese and Tibetan versions. The uddāna in the Tibetan translation does not have the last title, Upadeśa, as mentioned above. Instead, it is substituted with 26 other titles for the verses of Group B. Therefore, the possible reasons for this issue will be probed.

The aims of this research are as follows:
 

1. To further verify the source of the verses, the place of the Śarīrārthagāthā verses in the formation of Buddhist texts, and the reason why early verses were collected in the Śarīrārthagāthā.
2. To deduce which Buddhist school the verses are close to.
3. To shed light on the distinctive features of the Śarīrārthagāthā verses by comparing them with other parallels.
4. To explore the principles that lay behind the selection of the verses from canonical texts.
5. To understand the formation of the Śarīrārthagāthā and its place within the textual development of the Yogacārabhūmi.
6. To investigate the possible reasons why the Tibetan uddāna is different from the Chinese/Sanskrit versions.


1.3.2. Significance


Based on the literature review and the study’s aims, the significance of this study is as follows:

1. A study of its Śarīrārthagāthā section, which quotes canonical Śrāvakayāna verses, will contribute towards our understanding of the relationship between Mahāyāna and Śrāvakayāna Buddhism.

2. The relationship between texts belonging to different schools during the early phase of Buddhism and the navāṅga classification system will be clarified. This is particularly important in light of the fact that some scholars59 that think the navāṅga classification system is just a classification of literatures, not a classification of literary style.

3. The authorship of the Yogacārabhūmi is a complicated issue. However, an analysis of the wordings in the Śarīrārthagāthā verses (with the commentary) between other versions and parallels will reveal the author’s and translator’s interpretations.
4. Light will be shed on the relationship of the Yogacārabhūmi to the texts of the school(s) whose texts also quote the same or similar verses.



59 See Ui 1925: 152–153, Shiio 1935: 21, and Hayashiya 1936: 663–664.
 
5. This study will provide an example of the kind of developments which took place from early Buddhism to Mahāyāna Buddhism.

6. The distinctive features of the Sanskrit, Chinese and Tibetan versions of the
Śarīrārthagāthā will be identified.

7. The possible explanation(s) for the difference in the Tibetan Śarīrārthagāthā uddāna might also provide answers to the questions of why there are different structures and name(s) in the uddānas, and different content in the uddānas of the Tibetan Yogacārabhūmi.

1.4. Methodology

The general methodologies I have used in my thesis are comparative philology, textual criticism from philology, 60 as well as philosophical analysis. 61 In this thesis, comparative philology involves the comparison of wording between Chinese, Sanskrit, Tibetan, and, partly, Pāli. By this method, light is shed upon the differences of the concepts and/or the relationship of the versions/recensions. Textual criticism involves the examination and discussion of textual variants between Chinese, Sanskrit, and Tibetan versions (and occasionally Pāli texts). Philosophical analyses have been used to resolve some issues, and have involved the analysis of definitions and/or concepts from different texts or schools.

In this section, I will simply explain the methods (under the above two methodologies, philology and philosophical analysis) employed to deal with the following five aspects of this study:

1. aṅga classification and Śarīrārthagāthā verses (chapter two);

2. sources of the Śarīrārthagāthā verses (chapter two);

3. analysis of the Śarīrārthagāthā verses (chapter three and four);

4. textual development and stratification (chapter five); and

5. principles of selecting verses (chapter five).


60 This is a study combining several areas of literary criticism, history, and linguistics. The details of philology can be found in Watkins 1990. Comparative philology and textual criticism are branches of philology.
61 Philosophical analysis is a study method in which complex concepts can be analysed into elements. The relationships of the elements can thereby be brought into focus. See Dictionary of Philosophy and Psychology by Baldwin (1925), s.v. “analytical philosophy.”
 

1.4.1. Aṅga classification and Śarīrārthagāthā verses
¨

In order to indentify the place of Śarīrārthagāthā verses in Buddhist textual history, it is important to survey the development of early Buddhist texts. Aṅga classification, which is the earliest system of cataloging texts, appears to be a key approach to the identification of Śarīrārthagāthā verses and the reason why these particular early Buddhist verses were collected in a later Mahāyāna text, the Yogacārabhūmi.

In this study, at first the oral transmission in the formation of canonical texts will be briefly summarised from modern studies. Two important issues are involved. The first is the relationship between the navāṅga (nine-division) and dvādaśāṅga (twelve-division) systems. Three important studies—Hirakawa 1960, Mayeda 1964 and Yìnshùn 1971— are particularly discussed because the sources they cover not only include Pāli texts but also Abhidharma and Mahāyāna materials. Hirakawa analyzed aṅga classification based on various traditional Vinayas, Mayeda comprehensively investigated various aspects of aṅga classification based on early Buddhist texts (Nikāya, Āgama and Vinaya etc.), and Yìnshùn deduced the functional development of aṅga classifications from wide-ranging sources, particularly Abhidharma texts. The second is to analyse the development phases of aṅga classification. This is based on the studies of Mayeda (1964: 478–479) and Yìnshùn (1971: 621). Most notably, Yìnshùn (1971) pointed out the importance of the first three aṅgas (sūtra, geya and vyākaraṇa).

How and why are the first three aṅgas more important than the other aṅgas? Yìnshùn (1971, 1983) proposed that the structure of the Saṃyuktāgama was determined by the first three aṅgas. The methodology Yìnshùn applied was to analyse descriptions in the Yogacārabhūmi and the Mūlasarvāstivāda Vinaya. This also helped Yìnshùn correctly re- organize the Chinese Saṃyuktāgama, which was disarranged with regard to its fifty fascicles. This publication of Yìnshùn was a scholarly breakthrough at that time.62

This three aṅgas theory importantly sheds light on the identification of the Śarīrārthagāthā verses. According to the studies of Yìnshùn (1971, 1983), the three aṅgas are valued by the author(s) of the Yogacārabhūmi, particularly the sūtra-aṅga, which assisted Yìnshùn in discovering the relationship between the Vastusaṃgrahaṇī of

62 The contribution of Yìnshùn's work has been highly regarded and accepted by many scholars, including Akira (1985), Mizuno (1988), Nagasaki and Kaji (2004), Sujato (2005), Glass (2006), Bucknell (2007), and Bingenheimer et al. (2011).
 
the Yogacārabhūmi and the sūtra-aṅga section of the Saṃyuktāgama.63 The other two aṅgas, geya and vyākaraṇa, which are “verse” categories, have not yet be found to connect with parts in the Yogacārabhūmi. Therefore, the development of these two aṅgas of geya and vyākaraṇa will be presented. The methodology is to explore the meanings of the two aṅgas in various texts. Through this investigation, the Śarīrārthagāthā verses will be shown to connect with one of their meanings. The relationship between these two aṅgas, geya and vyākaraṇa, and the Śarīrārthagāthā verses will be discussed. The results reinforce the importance of the three aṅgas theory and reveal the reason why early Buddhist verses appear in the Śarīrārthagāthā.


1.4.2. Sources of the Śarīrārthagāthā verses

Before locating the Śarīrārthagāthā verses in Buddhist textual history, there is a need to identify the texts from which these verses were drawn. Four scholars—Ui (1958), Enomoto (1989), Zōngzhèng (2005), and Chén (2006)—have mostly found the specific sūtras to which the Śarīrārthagāthā verses correspond. I will compare and examine their studies, and then integrate the information.

1.4.3. Analysis of the Śarīrārthagāthā verses


The Śarīrārthagāthā verses can be examined through internal and external comparisons. Internal comparisons study the three versions (Chinese, Sanskrit, and Tibetan) of the Śarīrārthagāthā against each other. In the Sanskrit version, the verses have been edited from two manuscripts by Enomoto (1989).64 The Chinese version is taken from the Taishō collection.65 In the Tibetan version, the Yogacārabhūmi is found in the Tanjur. The Buddhist text collections (four classical xylograph editions) of the Tanjur can be roughly divided into two lineages: Narthang, Peking against Derge and



63 The sūtra-aṅga section of the Saṃyuktāgama is cited in the Vastusaṃgrahaṇī of the Yogacārabhūmi, which was identified by Yìnshùn 1983. Also, due to the sequence of each sūtra, Lǚ (1924) and Yìnshùn (1983) argued that the Vastusaṃgrahaṇī is the mātṛka of the sūtra-aṅga section of the Saṃyuktāgama.

64 The Śarīrārthagāthā is found in Śrāvakabhūmi and Yogacārabhūmi manuscripts.
65 This is the definitive edition of the Chinese Buddhist text collection. The Śarīrārthagāthā is located at T1579, 30.370a10–388b1.
 
Cone.66 Therefore, I will take the Peking and Derge editions for sake of comparison. Before comparing them, the English translation of each verse from the three versions will be presented. This will help readers understand the differences and similarities of the three versions. External comparisons study the Śarīrārthagāthā verses with reference to parallel verses in other texts. The wordings, structure and content of each verse will be surveyed. In this comparison, I will also explore the background narrative of each verse if it can be found.


1.4.4. Textual development and stratification


The Yogacārabhūmi contains a wide range of Buddhist concepts from Śrāvakayāna to Mahāyāna Buddhism. However, Schmithausen (1969b: 817; 1987a: 13–14) convincingly suggested that the Yogacārabhūmi involves heterogenic material in both content and structure, which was not arranged in a perfectly balanced whole in the final redaction. He believed that the Yogacārabhūmi was not based on a preparatory plan but was gradually developed from complex and separate textual units and materials (Schmithausen 1987a: 267–269). Three scholars— Schmithausen (1987a), Aramaki (2000), and Deleanu (2006)—identified the stratification of the Yogacārabhūmi.67

Based on the stratification theory from the above scholars, the Śarīrārthagāthā will be examined in terms of its formation and the phase of textual development. Also, the stratification theory can help us understand the difference in the summary verse (uddāna) between the Tibetan and Sanskrit/Chinese versions.


1.4.5. Principles of selecting verses


There are two directions of investigation into the potential principles which may have been applied in Śarīrārthagāthā verse selection. Firstly, the content and classifications

66 See Vogel 1965: 25–30, Eimer 1978: 77, Pasadika 1987: 45, Schoening 1995: 123–124, 132–134, 141–
143, Skilling 1991, Mathes 1996: 37–40, Erb 1997: 261–268, Dietz 2000: 176, Mochizuki 2004, Deleanu 2006: 78. Further, Deleanu (2006: 89) provided a detailed diagram and discussion of the relationships between twenty collections.
67 Basically, the suggestions by Aramaki (2000) and Deleanu (2006) are based on the idea by Schmithausen (1987a). However, this does mean that it is a perfect chronological or linear development from one phase to another one. The phases might have partially overlapped.
 
of verses in the Saṃyuktāgama and early Buddhist texts will be examined to find the connection with Śarīrārthagāthā verses. Secondly, I will discuss relevant descriptions found in Buddhist literature (especially (Mūlasarvāstivādin texts) which are concerned with selecting verses.68


1.5. Structure of this study


In order to arrive at the aims described above, the following structure will be followed:


Chapter two focuses on the relationship between aṅga classification (especially the first three aṅgas—sūtra, geya, vyākaraṇa) and the Yogacārabhūmi. The verses in the Śarīrārthagāthā can be divided into two groups (Group A and Group B). Importantly, Group A and Group B verses are related to the geya and vyākaraṇa aṅgas.

Chapters three and four engage in the analysis of the Śarīrārthagāthā verses (of Group A and Group B separately). There are two ways by which this will be investigated, namely, internal and external comparisons. Internal comparisons study the three versions (Chinese, Sanskrit, and Tibetan) of the Śarīrārthagāthā against each other. External comparisons study the verses of the Śarīrārthagāthā with reference to parallel verses in other texts. By this analysis, the distinctive characteristics of the Śarīrārthagāthā verses will be highlighted.

Chapter five explores three textual issues: the formation of the Śarīrārthagāthā, the differences in the summary verse (uddāna) between the Tibetan and Sanskrit/Chinese versions, and the potential principles of selecting verses.

Chapter six provides the conclusion to the study. The findings of each chapter will be reviewed and summarised. From this summary, and from a comparison with relevant studies of the Śarīrārthagāthā, the contribution this study has made to the topic will be clarified. Avenues for future research on the Śarīrārthagāthā will also be provided.

The main points of each chapter are arranged in the following chart, which will help the reader understand the overall structure of this study:


68 The relationship between the Mūlasarvāstivāda and Sarvāstivāda schools is still a complicated issue. Many texts cannot easily be attributed to the Mūlasarvāstivādins or Sarvāstivādins. In such cases, I will write “(Mūla)sarvāstivādins” as shorhand for “Mūlasarvāstivādins and/or Sarvāstivādins”.
 

 
1.6. Sanskrit manuscripts, Chinese and Tibetan translations and exegetical literature

In order to comprehensively understand the content of the Śarīrārthagāthā, one should of necessity compare its different versions. To this end, the background and condition of two Sanskrit manuscripts (the Śrāvakabhūmi69 and Yogacārabhūmi), the first two translations (the Chinese and Tibetan versions),70 and the commentaries involving the Śarīrārthagāthā in Sanskrit, Chinese and Tibetan, should be explored.
The Sanskrit Śarīrārthagāthā can be found in the Śrāvakabhūmi and Yogacārabhūmi manuscripts. These two manuscripts were found and photographed at Zha-lu in Tibet in 1938 (Sāṅkṛityāyana 1938: 137–142) and Sakya monasteries in 1937 (Sāṅkṛityāyana 1937: 6–24)71 respectively, in Tibet by Rāhula Sāṅkṛityāyana.72 Negatives of the photographs were brought to India and preserved by the Bihar Research Society in Patna (Bandurski 1994: 13–15). Copies of the negatives were also stored in the Seminar für Indologie und Buddhismuskunde of Georg-August University of Göttingen, Germany (Matsunami 1992, Bandurski 1994: 13).'

In 1994, Bandurski analyzed and catalogued this collection of copies, which has made it easier for researchers to access.73 Another scholar, Giuseppe Tucci, also visited the same locations in Tibet shortly after Sāṅkṛityāyana found the manuscripts and took photographs of the manuscripts.74 According to Delhey (2005: 505 note 23), Tucci’s collection includes the Śrāvakabhūmi and Yogacārabhūmi manuscripts. However, unfortunately the materials of Tucci’s collection are still currently unavailable to the public.



69 The Śrāvakabhūmi manuscript contains not only the Śrāvakabhūmi section, but also some other sections, including the Asamāhitābhūmi, Sacittikābhūmi, Acittikābhūmi, Śrutamayībhūmi, Cintāmayibhumi and Pratyekabuddhabhūmi.

70 The Chinese and Tibetan Yogacārabhūmi were translated from a Sanskrit manuscript, but there are several secondary translations from Chinese or Tibetan versions, such as the Mongolian and Japanese Yogacārabhūmi.

71 See Sāṅkr̥tyāyana 1938: 137–142; 144 and Śrāvakabhūmi study group 1994: 3 note 3. Delhey (2013: 505) mentions that these two manuscripts were found in “Zha lu.” This is not correct; Sakya monastery is located at Sakya (town) in Shigatse city in Tibet.

72 For details of the expeditions, see Sāṅkr̥tyāyana 1935, 1937, 1938, 1984. For Sāṅkṛityāyana’s discoveries
of the Sanskrit manuscripts, see Chowdhury 1994: 16–22, 40–42, 106–113. Cf. Deleanu 2006: 63 n. 1.

73 For a detailed analysis and catalogue, see Bandurski 1994.
74 The details of Tucci’s expeditions in Tibet and the Himalayas are described by Sferra (2009) and Nalesini (2009).
 
After 1938, the situation regarding the two manuscripts was unclear due to global, regional, political and social changes.75 The Śrāvakabhūmi manuscript was discovered at the Ethnic Library of China (Zhōnguó mínzú túshūguǎn 中國民族圖書館) in Beijing around 1990 (Matsunami 1992, Bandurski 1994: 27 n. 76). A few years later, the Śrāvakabhūmi manuscript, along with other manuscripts, was possibly taken back to Nor-bu-gling-kha Palace in Lhasa, Tibet.76 However, there is little information available
about where the Yogacārabhūmi manuscript is now preserved. Additionally, there is little information about the current location of a transcript of the Yogacārabhūmi codex handwritten by Sāṅkṛityāyana under an extremely tough situation in Tibet.77

According to Steinkellner (2004: 20–21), the manuscripts from Zha-lu were sent to the Potala Palace in 1959, and in 1961 they were taken to the Library of the Palace of National minorities, the Ethnic Library of China, on loan from the Tibetan Government.78 So, this is why the Śrāvakabhūmi manuscript, along with other manuscripts, was returned to Tibet in 1993.79 Steinkellner added that since 1962, most of the manuscripts kept by the monasteries around Lhasa and Sakya were gathered together in Lhasa. Then some were brought to the Potala Palace, a few to Norbulingka, and some to the TAR (Tibet Autonomous Region) Archive. Steinkellner (2004: 21) stated that those
manuscripts had been salvaged from the ruins left by the Red Guard. Steinkellner (2004: 23) further stated that in 2002 he was told that the manuscripts were still in the same places, i.e., in “the Potala, Norbulingka, Drepung, Sakya, Tsethang, and possibly in the Lhasa Archives.” Thus, logic suggests that the Yogacārabhūmi manuscript should be preserved in one of the above places. It is noteworthy that due to Luo Zhao's effort,80 the Yogacārabhūmi manuscript has not been listed in the catalogues of Sanskrit manuscripts preserved in TAR. According to Steinkellner (personal communication, 15 June 2012), it


75 Cf. Steinkellner 2004: 19–23.
76 For details, see von Hinüber 2006: 286 and Steinkellner 2004: 19–23. However, Gyurme (2009: 303) mentioned that they are kept in the Tibet museum.

77 See Sugawara 1990:322, Delhey 2013:509 note 37 for details.
78 In the meantime, microfilms were made for further research in Beijing. See von Hinüber 2006 for the details.
79 Steinkellner (2004: 30) stated, “In 1993, under Hu Jintao...the mss [i.e. manuscripts] collection in the Palace of National Minorities in Beijing was returned to Tibet except for three mss kept in the Palace library as specimens.”
80 The catalogue has never been published, but some description of its history and character can be found in Steinkellner 2007: ix-xii n. 5.
 
may never have been moved to the Lhasa depositories and may still be at the Sakya.81 However, the catalogues by Luo Zhao do not list the complete collections of manuscripts. So, the above is only conjecture until all of the manuscripts in TAR can be accessed.

Around 1990, the Japanese scholar Yasuo Matsunami and his colleagues at Taishō University learned that the Śrāvakabhūmi manuscript was preserved at the Ethnic Library of China in Beijing, the Institute for Comprehensive Study of Buddhism at Taishō University and the Ethnic Library of China started a joint project aiming to produce a facsimile edition of the manuscript.82 The end result was published in 1994.83 This publication, which is a high quality facsimile reproduction, reduced the possibility of mistakes arising from the use of negatives of the photographs of the Śrāvakabhūmi manuscript. Photographs taken more than half a century ago may have become flawed with the passage of time. For example, many letters are obscured by drawing-pins or overlapping palm-leaves. These two editions of Śrāvakabhūmi manuscript will be utilized in this research.

Based on the above information, the two Sanskrit manuscripts can be summarized as follows:84


Śrāvakabhūmi Yogacārabhūmi
Current location Lhasa Sakya
Material Palm leaf Palm leaf
Possible date 11th century85 1000 CE86
Reproductions 1. Photographs by Sāṅkṛityāyana 1938: 144f (No. 350 and 351) 2.Negatives of the photographs by Sāṅkṛityāyana87 (at Bihar
Research Society in Patna) 1. Photographs by Sāṅkṛityāyana 1937: 24 (No. 199)
2. Negatives of the photographs by Sāṅkṛityāyana88 (at Bihar
Research Society in Patna)

81 I am indebted to Professor Steinkellner for this and other valuable information provided in his email.
82 See Matsunami 1992. Before 1990, this group only accessed the copies of Sāṅkṛityāyana’s photographs.
83 The publication is The Facsimile Edition of the “Śrāvakabhūmi” Sanskrit Palm-leaf Manuscript 瑜 伽 師 地論聲聞地梵文原文影印本. Beijing: Minzu chubanshe.

84 Cf. Delhey 2013: 508–509.
85 See Deleanu 2006: 54–58.
86 See Delhey 2009: 78–80.
87 No. 26 [Xc 14/27], See Bandurski 1994: 61f.
88 No. 27 [Xc 14/28], See Bandurski 1994: 64f.
 
3. Copies of the photograph negatives by Sāṅkṛityāyana (in the Seminar für Indologie und Buddhismuskunde of Georg-August University of Göttingen in Germany)

4. Photographs from Tucci’s collection

(in the Oriental Department of the library of the Istituto Italiano per l'Africa el'Oriente in Rome)
5. A microfilm copy in Beijing

6. Facsimile edition (published in 1994 by Taishō University and the Ethnic Library of China) 3. Copies of the photograph negatives by Sāṅkṛityāyana

(in the Seminar für Indologie und Buddhismuskunde of Georg- August University of Göttingen in Germany)
4. Photographs from Tucci’s collection

(in the Oriental Department of the library of the Istituto Italiano per l'Africa el'Oriente in Rome)

Regarding the edition of the Śarīrārthagāthā from the two manuscripts, only the verse part has been published by Enomoto (1989).89 The commentary is still unavailable. According to Enomoto (personal communication, 10 July 2010), a tentative edition of the whole commentary from the two manuscripts has been made, but he felt hesitant to publish it because many parts of the Yogacārabhūmi manuscript are not readable.
Unfortunately, while it appears that no Indian Sanskrit commentaries survived, one Chinese and one Tibetan translation survived. From these two translations, it may be seen that there were five Indian commentaries. However, four of them were relevant to the Bodhisattvabhūmi and were translated only into Tibetan.90 The fifth


89 Enomoto (1989) also explored where these verses were cited from.

90 They are the Bodhisattvabhūmivr̥tti (Byang chub sems dpa'i sa'i 'grel pa), Bodhisattvaśīlaparivartabhāṣya (Byang chub sems dpa'i tshul khrims kyi le'u bshad pa), Bodhisattvaśīlaparivartaṭīkā (Byang chub sems dpa'i tshul khrims kyi le'u'i rgya cher 'grel pa), and Yogācārabhūmau Bodhisattvabhūmivyākhyā (rNal 'byor spyod pa'i sa la byang chub sems dpa'i sa'i rnam par bshad pa). See Ui et al. 1934: 615–616.
 
(Yogacārabhūmivyākhyā) is not a complete translation of either the Chinese or the Tibetan and does not cover the Śarīrārthagāthā or Cintāmayībhūmi.91

With regard to the Chinese translation of the Yogacārabhūmi, the only definitive and complete version was produced by Xuánzàng 玄 奘 (602–664).92 The title of the Chinese translation is suggested to be Yogacārabhūmiśāstra (Yújiāshīdì lùn 瑜伽師地論). There are ten other partial translations of the Yogacārabhūmi, but most are from the Bodhisattvabhūmi.93 As well, there are two ancient Chinese commentaries which involve the Śarīrārthagāthā—A Summary Concise Compendium of the Yogacārabhūmi (Yújiāshī dì lùn luèzuǎn 瑜伽師地論略纂) (T. No. 1829) and Notes on the Yogacārabhūmi (Yújiā lùn jì 瑜伽論記) (T. No. 1828).94 Yújiāshī dì lùn luèzuǎn (16 fascicles), compiled by Kuījī 窺 基 (632–682), a famous and outstanding disciple of Xuánzàng, briefly explains the first 66 fascicles of the Yogacārabhūmi that cover only the Maulyo bhūmayaḥ and some of the Viniścayasaṃgrahaṇī. Yújiā lùn jì (48 fascicles) was produced by Dàolún 道倫(known as Dùnlún 遁倫),95 a Korean Buddhist monk; but there is little extant information about Dàolún's background, including his age. Yang (1984: 292–293) deduced that he lived between ca. 650 and 730 C.E. Yújiā lùn jì comments on all 100 fascicles of the Yogacārabhūmi; it not only refers to the Yújiāshī dì lùn luèzuǎn, but also cites many

91 There are 70 folios of the Tibetan translation of this text in the Derge Tripiṭaka, which covers up to the Ātmavāda section of the Savitarkādi Bhūmiḥ, but only one fascicle (juàn 卷 ) of the Chinese. It seems more reasonable that Jinaputra (with other translators) is the author of this text, shown in the Chinese translation by Xuánzàng 玄奘. See Yūki [1962] 1985: 75, Mukai 1979: 33–42 and Deleanu 2006: 248–
250.
92 This date of Xuánzàng's life is the most widely accepted (Kamata 1999: 260–261).
93 These are the ten translations: 1. Búsàdì chí jīng 菩 薩 地 持 經 (Bodhisattvabhūmi-sūtra; 10 fasc.; T. No. 1581; Dharmakṣema 曇無讖; 414–426 C.E.); 2. Búsàjiè běn 菩薩戒本 (Bodhisattvaprātimokṣa; 1 fasc.; T.

No. 1500; Dharmakṣema 曇 無 讖 ; 414–426 C.E.); 3. Búsàshànjiè jīng 菩 薩 善 戒 經 (Bodhisattvabhūmi; 9 fasc.; T. No. 1582, 1583; Guṇabhadra 求那跋陀羅; 431 C.E.); 4. Yōupósāi wǔjiè wēiyí jīng 優婆塞五戒威儀經 (Bodhisattvaprātimokṣa; 1 fasc.; T. No. 1503; Guṇabhadra 求那跋陀羅; 431 C.E.); 5. Shíqī dì lùn 十七 地 論 (Saptadaśa bhūmayaḥ; 5 fasc.; not extant; Paramārtha 真 諦 ; 550 C.E.); 6. Juédìng càng lùn 決 定 藏 論 (Vinirṇītapiṭakaṡāstra; 3 fasc.; T. No. 1584; Paramārtha 真 諦 ; 557–569 C.E.); 7. Búsàjiè běn 菩 薩 戒本 (Bodhisattvaprātimokṣa; 1 fasc.; T. No. 1501; Xuánzàng 玄 奘 ; 649 C.E.); 8. Púsàjiè jiémó wén 菩 薩 戒羯磨文 (Bodhisattvavinaya; 1 fasc.; T. No. 1499; Xuánzàng 玄奘; 649 C.E.); 9. Wángfǎ zhènglǐ lùn 王法正理 論 (Savitarkāsavicārābhūmi, Avitarkāvicāramātrābhūmi and Avitarkāvicārābhūmi in the Viniścayasaṃgrahaṇī; 1 fasc.; T. No. 1615; Xuánzàng 玄 奘 ; 649 C.E.); 10. Fó wéi yōutiánwáng shuō wángfǎ zhènglùn jīng 佛為優填王說王法政論經 (Savitarkāsavicārābhūmi, Avitarkāvicāramātrābhūmi and Avitarkāvicārābhūmi in the Viniścayasaṃgrahaṇī; 1 fasc.; T. No. 524; Amoghavajra 不 空 ; 720–771 C.E.). Cf. Katō 1930.
94 These two English translations were undertaken by Deleanu (2006: 251).
95 Yūki (1931: 150–155; [1962] 1985: 264–267) and Yang (1984: 292–293) claimed that Dùnlún 遁倫 is not his actual name and that it should be Dàolún 道倫 (from the pronunciation of his Korean name).
 
masters’ ideas96 at that time. In 1983, the modern scholar Hán Qīngjìng published a valuable work, titled The Collection of the Detailed Catalogue and Comprehensive Exegesis on Yogacārabhūmi (Yújiāshī dì lùn kējù pīxún jì huìbiān 瑜伽師地論科句披尋 記 彙 編 ). This publication, which consists of four volumes, not only developed a new detailed catalogue and wide-ranging commentary on the Yogacārabhūmi, but also referred to the related explanatory treatises and commentaries. These three commentaries will help the reader comprehend the context of the Śarīrārthagāthā.
The Tibetan translation of the Yogacārabhūmi was undertaken by a group of experts at the beginning of the ninth century.97 Ye-she-sde, the chief-editor and translator, translated, collated and revised this work.98 The Tibetan Yogacārabhūmi is divided into six main parts, unlike the five parts in the Chinese version.99 Each part was translated by different translators; for example, the Śrāvakabhūmi by Jinamitra and Ye-shes-sde, the Bodhisattvabhūmi by Prajñāvarma and Ye-shes-sde and the Maulyo bhūmayaḥ (containing the Cintāmayībhūmi, within which is found the Śarīrārthagāthā) by Ye-shes- sde and others.100 The Tibetan Tripiṭaka does not contain a complete commentary on the Yogacārabhūmi. Rather, all we find are four Indian commentaries translated into Tibetan which focus on the Bodhisattvabhūmi.

According to Deleanu (2006: 55–57, 76, 106–107), the date of the Chinese version of the Yogacārabhūmi (first half of the seventh century) is earlier than that of the Tibetan version (end of the eighth century) and the earliest manuscript witness of the Sanskrit versions.101 This can help us understand the differences occurring between the various versions when analysing the content or wording of the Śarīrārthagāthā verses.

96 The number of masters is around 50. Due to Katō's effort, the most frequently quoted people are Huìjǐng 惠景 (1271 times), Shéntài 神泰 (627 times), Yuáncè 圓測 (269 times), Wénbèi 文備 (187 times), and Huìdá 惠達 ( 97 times). See Katō Kokuyaku issaikyō 國譯一切經, Wa-Kan senjutsubu 和漢撰述部, Ronsho-bu 論疏部, Vol. 9, Introduction, p.3.
97 Cf. Bacot 1954, Simonsson 1957: 210–237, and Zimmermann 2002: 213.
98 Cf. Bacot 1954: 329, Skilling 1997: 112, Seyfort Ruegg 1981: 211, Zimmermann 2002: 368, and Deleanu
2006: 73.

99 The five main parts of the Chinese version are closer to the original structure. They are Maulyo bhūmayaḥ (Běn dì fèn 本地分), Viniścayasaṃgrahaṇī (Shè juézé fèn 攝決擇分;), Vyākhyāsaṃgrahaṇī (Shè shì fèn 攝釋分), Paryāyasaṃgrahaṇī (Shè yìmén fèn 攝異門分), Vastusaṃgrahaṇī (Shè shì fèn 攝事分). In the Tibetan version, the sixth part is the Vinayasaṃgrahaṇī, which was separated from the Vastusaṃgrahaṇī. See Hakamaya 1985: 220, Suguro 1989: 280–283 and Deleanu 2006: 46–47.

100 Ui et al. (1934: 614–615) did not obtain much information about the translators of the Maulyo bhūmayaḥ. However, it has been investigated by Deleanu (2006: 75).
101 The dates of the two Sanskrit manuscripts have been investigated by different scholars. The
 
Chapter two: Aṅga classification and the Śarīrārthagāthā verses

At the outset of the Introduction to this dissertation, the verses of the Śarīrārthagāthā were described as being quite diverse. This diversity of verses can also be seen in early Buddhist texts. In aṅga classification of Buddhist literature, several aṅgas particularly focus on the genre of verse. This chapter will analyze the relationship between aṅga classification and the Śarīrārthagāthā verses. The first section will explore many aspects of aṅga classification (especially the first four aṅgas), while the second and third sections will particularly focus on the geya and vyākaraṇa aṅgas and then connect them with the Śarīrārthagāthā verses.

2.1. Aṅga classification and the Yogācārabhūmi

This section will discuss several topics concerning aṅga classification of early Buddhist texts. First, the development of aṅga classification will be examined in order to understand its importance in the formation of early Buddhist texts. Second, the structure of the Saṃyuktāgama in terms of the first three aṅgas, discovered and described by Yìnshùn, will be introduced because of their relevance to the verses of the Śarīrārthagāthā. Third, an examination will be undertaken of why the verses of the Śarīrārthagāthā are considered gāthā (according to the title Śarīrārthagāthā) and what the definition of gāthā would be in this context. All of these topics pertain directly to the background of the Śarīrārthagāthā.

2.1.1. Aṅga classification and the formation of early Buddhist texts

Aṅga classification is a complex issue that has been discussed and debated for a very long time. In Buddhist texts, the “nine-divisions” (navāṅga)102 and “twelve-divisions”



Śrāvakabhūmi manuscript was dated to around the first half of the eleventh century (Deleanu 2006: 55– 57) and the Yogacārabhūmi manuscript was dated to around 1000 CE (Delhey 2009: 78–80).
102 In the Pāli tradition, the nine aṅgas are sutta, geya, veyyākaraṇa, gāthā, udāna, itivuttaka, jātaka, abbhutadhamma and vedalla. In Sanskrit and Chinese texts, the nine aṅgas are sūtra, geya, vyākaraṇa, gāthā, udāṇa, ityuktaka or itivṛttaka, jātaka, vaipulya and adbhūtadharma.
 
(dvādaśāṅga)103 are the two main aṅga classification systems. However, differences are encountered in different texts in the meanings and definitions of each aṅga, as well as their sequential ordering. Other common issues are whether aṅga classification represented collections of texts or literary styles, which of the two classification systems was earlier, and whether aṅga classification existed prior to the formation of the five Nikāyas or four Āgamas.
Aṅga classification was an early system of dividing and cataloguing canonical Buddhist texts. The formation of early Buddhist texts occurred within a wholly oral context. In a traditional account, one of the Buddha’s principal disciples, Mahākāśyapa, called for the first council (saṅgīti or saṅgāyanā) at Rājagṛha so that the Buddha’s teachings could be recited.104 The monk Upāli presided over the recitation of the monastic discipline (vinaya), while Ānanda rehearsed the Buddha’s teachings (dharma). Afterwards, all the monks in the council convened to examine, rearrange and confirm the resulting textual corpus.105 A second council occurred one hundred years later at Vaiśālī to dispute ten points of monastic discipline. At that time, the Vinaya and Dharma were also recited, expanded and processed.106 The early Buddhist texts were subsequently transmitted orally for several centuries. After the second council, there were no councils that were accepted generally by all schools, although councils still occurred.107

Due to differences in dates, places and the numbers of participants between the texts of different schools, modern scholars are skeptical of the historicity of the first two councils.108 For example, Oldenburg (1927) depicted the first council as pure fiction and

103 The above nine aṅgas plus the following three: nidāna, avadāna and upadeśa.

104 The relevant studies are Miayeff 1984, La Vallée Poussin 1910; 1976, Przyluski 1926–1928, Hofinger
1946, Demiéville 1951, Renou and Filliozat 1953, Bareau 1955a; 1955b; 1966, Warder 1970: 195–212,
Williams 1970: 157, Lamotte 1976: 136–154; 297–300, Bechert 1978, Dutt 1978: 377, Norman 1983: 13–
14, Gombrich 1988, Hirakawa 1990: 76–94, Hallisey 1992; 2005, Frauwallner 1994, Tilakaratne 2000,
Goyal 2003: 1–11, Witanachchi 2006, Skilling 2009.
105 The corpus of texts at the council is known as the “root recitation” (Sanskrit mūla-saṃgīti) or the “root compilation” (Pāli mūla-saṃgīti). See Skilling 2009.
106 Records of the first two councils were also found in the Vinaya texts of different schools. Lamotte (1976: 153–154) indicated that the account of the councils is merely an appendix. Skilling (2009) says that the Vinaya masters of each school “retrospectively shaped the records.” Therefore, it is difficult to trace back to the original situations.

107 There are a few Buddhist texts recording the other councils, like the texts of the Mahāvihāravāsa Theravādins of Sri Lanka, the Sāṃmitīyas of northern India, and Sarvāstivādin tradition. See Skilling 2009.

108 See Franke 1908, Oldenburg 1927: 9, Lamotte 1958/1976: 15–18, Prebish 1974, Hallisey 1992, and Skilling 2009. Some of this debate was described by de Jong (1976: 30–31; 67).
 
stated that councils should be taken as “ideas” rather than real events. This is a somewhat extreme position. Following Hallisey’s (1992: 136) conclusion,109 it can be said that the history of phenomena is changing and human movement is complex in the processes of constitution and reconstitution. Therefore, it is difficult to judge whether the councils were real or not. Two other studies have drawn meaningful conclusions. Although Lamotte (1958) described councils as “legends,” he also stated that, whether monks or specialists assembled in the first council or not, they had developed “a coherent dharma and prātimokṣa” which was accepted by early monastic specialists and became the heritage of the schools. Skilling (2009) mentioned the accounts of councils should be taken as “validation of tradition rather than historical or historicist reality.” It was an inevitable consequence of the collective memory of a tradition that the events were taken as authentic. In brief, the Buddha’s teachings should still be valued whether the Buddhist councils were carried out in reality.

In the phase of oral transmission, the first categorization of early Buddhist texts was either navāṅga (nine-division) or dvādaśāṅga (twelve-division).110 Scholars still disagree on which aṅga classification emerged first, due to different interpretations of the relevant information. Those favoring Pāli sources usually take navāṅga to be the earlier of the two,111 while those using Sanskrit and/or Chinese sources take dvādaśāṅga to be the earlier.112 Three important studies worth mentioning here are Hirakawa 1960, Mayeda 1964 and Yìnshùn 1971. In these studies, Hirakawa analyzed aṅga classification based on various traditional Vinayas, Mayeda comprehensively investigated various aspects of aṅga classification based on early Buddhist texts (Nikāya, Āgama and Vinaya, etc.), and Yìnshùn deduced the functional development of aṅga classifications from wide-ranging sources, particularly the Abhidharma texts. While drawing on different sources, these three studies have come to the same conclusion that nine-aṅga classification (navāṅga) is the earliest.113

109 Hallisey said that taking councils as “events” is informed by information preserved in Theravāda texts. Hallisey (1992: 136) stated that the standpoints “are geographically and historically very particular. This particularity, however, may be obscured for Buddhist individuals and groups by the phenomenon highlighted by a perspective which is considerably loftier and less determinately located.”
110 See Ui 1925 and Williams 1970: 158.

111 Ui 1925: 152, Shiio 1935: 21, Ogiwara 1869–1937: 399, Akanuma 1939: 165, Watanabe 1948: 87,
Masunaga 1948: 25, Nishi 1953: 53.

112 Mino 1926: 39, Hayashiya 1928.

113 Nattier (2004) summarised the results of several studies on aṅga classification (with the exception of
 
According to Yìnshùn (1971: 493–497), each of the aṅgas was created separately and at different times before being compiled into the nine-aṅga classification. He identified three stages in the development of each aṅga classification system (Yìnshùn 1971: 497). In the first phase, the terms were simply common terms used by people in India. For example, the term sūtra could refer to a thread, string, line, or wire, and was also applied to some types of texts (see MW s.v. sūtra). At this point, Buddhists would also have used the term; however, most likely with a meaning more specific to the Buddhist context.

Secondly, due to the increased popularity of chanting the Buddha’s teachings and the increased number of texts recited, these terms developed more specific meanings. It appears that at this stage the terms defined different styles of texts and that these texts were each grouped together to form different collections. In the third stage, these various collections then appear to have been combined into the nine-aṅga classification system. However, even while this was happening, the definitions of each aṅga were still being modified.114 Due to the development of early Buddhist texts, the nine-aṅga classification was felt to be no longer comprehensive enough, so three aṅgas were added, resulting in the twelve-aṅga classification system.115 Importantly, Yìnshùn (1971: 497) mentioned that although the twelve-aṅga classification was created later, the three additional aṅgas could have already been in existence.

Based on surviving Buddhist texts, it is extremely difficult to identify the specific group of texts to which each aṅga refers. Therefore, most tend to oppose the claim that each aṅga represents an actual collection or group of early Buddhist texts.116 Rather, they believe that aṅga classification was a description of different literary styles. However, the fifth century Buddhist scholar Buddhaghosa said that aṅga could be regarded as referring to the content of early Buddhist texts, thus determining collections, or groups, of texts. Here, Buddhaghosa’s interpretation of the meaning of aṅga is not solely based

Yìnshùn's studies). Nattier firstly noted that the nine-aṅga classification existed earlier and that the twelve-aṅga classification demonstrates “a subsequent elaboration.” Secondly, she noted that the twelve- aṅga classification appears much more frequently than the nine-aṅga classification, particularly in Mahāyāna texts.

114 With the development of new texts and text styles, the definitions of each aṅga appears to have been modified. In particular, the uses of the terms seen in the various Abhidharma or Mahāyāna texts do not always match the earlier meanings. This situation can be regarded as a natural adjustment during the development of Buddhist texts. See Yìnshùn 1971: 622.

115 The idea of that the twelve-aṅga was an expansion of the nine-aṅga is still in dispute. Mayeda (1964) and Yìnshùn (1971) support this idea, while Hirakata (1964) does not.
116 See Kanakura 1939: 372, Dutt 1957: 89, Kalupahana 1965: 616, Lamotte 1980, and Norman 1983.
 
on literary style,117 an idea that is supported by Mayeda and Yìnshùn. Mayeda (1964: 227–475) attempted to identify the actual sections of early Buddhist texts that each aṅga represents. Yìnshùn (1971: 500–620) postulated the specific group of texts that each aṅga may correspond to by using definitions found in Abhidharma and Mahāyāna texts.

Although drawing from similar sources, Mayeda and Yìnshùn come to slightly different conclusions about the development of aṅga classification. The development of aṅga classification can be divided into a few separate stages. By tracing the evolution of each aṅga from early Buddhist texts, Mayeda (1964: 478–479) concluded that there were three stages in the development of the twelve-aṅga classification. According to Mayeda (1964: 479), “the first stage is sutta, geya, vyākaraṇa, gāthā, udāna, the second itivuttaka, jātaka, vedalla, abbhutadhamma, and the third nidāna, avadāna, upadeśa.”118 Yìnshùn (1971: 621) generally accepts Mayeda’s idea, but went on to propose another possibility. Yìnshùn (1971: 500–521) provided evidence that the first two aṅgas (sūtra and geya) represent the earliest stage of aṅga development. Therefore, he suggested that there are four stages of the twelve-aṅga classification, achieved by splitting Mayeda’s first stage into two, the first being sūtra and geya, the second vyākaraṇa, gāthā and udāna.119

Yìnshùn (1971: 501–506) hypothesized that the content and styles of the first two aṅgas already existed at the first council, but that the Buddha’s teaching at this stage was generally called dharma or sūtra.120 In fact, the term sūtra did not represent any specific genre or specific group of texts. At this early stage, the definition of sūtra is a “string” of the Buddha’s teaching, just like a thread.121 In the initial period, the most obvious

117 See Papañcasūdanī II 106, Samantapāsādikā I 28, and Williams 1970. In addition, Lamotte (1958: 141) mentioned that the canonical texts at the first council are certainly not the same as the texts we currently access. Therefore, it is difficult to know whether the texts that Buddhaghosa referred to and the ones that aṅga correspond to are the same or different.
118 In this account, Mayeda mostly used the Pāli terms, but the last term upadeśa was in Sanskrit. This might be a typo.
119 Yìnshùn (1971: 623) also mentioned that the aṅgas in the first stage shed light on the style rather than the content of Buddhist collections. Others emphasize not only the style but also the content of the Buddhist collection.

120 Yìnshùn 1971: 502. This idea is the same as Hirakawa's (1990: 69). However, Yìnshùn (1971: 505–506) also mentions that the defined range of sūtra is broader than dharma. The term sūtra involves dharma and vinaya.

121 The definitions of sūtra are various in Buddhist texts, so are the interpretations of sūtra by modern scholars. Yìnshùn (1971: 501–506) based this statement on the texts of the Sarvāstivāda and Mahāyāna Yogācāra traditions. He also believed that this is the oldest meaning of sūtra, which suits the literal meaning of the word “sūtra.” Besides, Hirakawa (1990: 70) noted, “the term sūtra originally referred to the warp in weaving, and was used to suggest that rich meanings were included in short sentences.”
 
classification was into prose and verse, with sūtra referring to prose and geya referring to verse. These two styles of texts gradually formed collections, and then gathered with others as the nine-aṅga classification. Yìnshùn (1971: 509) also used the description in the Yogācārabhūmi as evidence that division into sūtra and geya was the first stage in the development of aṅga classification. It reads as follows:

云何思擇諸法。此復二種應知。一思擇素呾纜義。二思擇伽他義。122 (T1579, 30.363a11–12.)
What is the investigation of the dharma? There are two which should be known: first is the investigation on the meaning of sūtra (sūtra-artha),123 and second is the investigation on the meaning of the verses (gāthā-artha).

Both sūtra-artha and gāthā-artha represent the dharma,124 which followers of the Yogācāra tradition should examine. Therefore, Yìnshùn (1971: 509) said that this description indicates that there appears to be a stage of two aṅgas in the formation of the Buddha’s teaching which should be considered as two of the earliest categories.
The early development of aṅga classification can be found in the Saṃyuktāgama and will be explained in detail in the next section. From the examples found in the Saṃyuktāgama, it is obvious that the third aṅga, vyākaraṇa, was added to sūtra and geya. However, the meanings of the sūtra and geya aṅgas had changed during the time that


However, some scholars have different ideas about the term sūtra. Gombrich (1988) mentioned that sūtra was sanskritized from sutta, which could corresponds to sūkta in Sanskrit. Gombrich believed that Buddhist poems in the early stage were called sūkta “well spoken.” Norman (1997a) said that the genre of sūtra texts is normally considered as prose, but in fact there are different characters of Buddhist sūtra texts. The reason for the difference is that sūtra is the Sanskritization of the Middle Indo-Aryan term sutta, which is possibly derived from sūkta in Sanskrit. Sūkta is also a synonym for subhāṣita which was used by King Aśoka. Aśoka described “All that was spoken by the Lord Buddha was well-spoken.” (Norman 1997a: 104). Also see Walleser 1914: 4 note 1, Mayrhofer 1976: 492, and von Hinüber 1994:
132 note 28.

122 The corresponding account of the Sanskrit version is: dharmapravicayataḥ katamaḥ / sa dvidho draṣṭavyaḥ / sūtrārtha-pravicayataḥ / gāthārtha-pravicayataś ca. This was edited by Wayman (1961: 163) from the manuscript.

123 Sūtras here are Vastusaṃgrahaṇi and bodhisattvapiṭakāvavāda. (tatra sūtrārthapravicayaṃ vastusaṃgrahaṇyām bodhisatvapiṭakāvavāde ca vistareṇa vakṣyāmi). (Śrāvakabhūmi manuscript 25b1M)

124 In his book Yìnshùn (1971: 509) described dharma as the Śrāvakayāna piṭaka. However, the Śrāvakayāna piṭaka is a collection which was became fully formed at a fairly late stage. From the context, it appears that Yìnshùn did not intend to mean “the piṭaka,” but rather “early Buddhist texts” (as opposed to Mahāyāna texts).
 
they were organized in the Saṃyuktāgama. The evolution of aṅgas will be fleshed out in greater detail later in the chapter as well.

2.1.2. The first three aṅgas and the Yogācārabhūmi

The importance of the first three aṅgas (sūtra, geya and vyākaraṇa) was evident in the way in which it was applied to the structure of the Saṃyuktāgama. This discovery of the aṅga basis to the structure of the Saṃyuktāgama was made by Yìnshùn (1983) in the process of reconstructing the order of the Chinese Saṃyuktāgama, which is in disarray in its canonical presentation. These three aṅgas helped Yìnshùn identify the chronological composition of the Saṃyuktāgama. Moreover, Yìnshùn (1971: 62) used this as evidence to show that aṅga classification and the grouping of texts into Āgama/Nikāya were developed in parallel, contrary to the view held by some scholars that either the aṅga classification came before the Āgama/Nikāya grouping, or vice versa.125

The idea of the first three aṅgas was derived from the Vastusaṃgrahaṇī of the Yogācārabhūmi, basically a mātr̥kā on part of the Saṃyuktāgama.126 The account in the Vastusaṃgrahaṇī indicates that sections of the Saṃyuktāgama can be divided into three categories: 1. those who speak (C.: néngshuō 能 說 , T.: su ston pa); 2. what is spoken (C.: suǒshuō 所 說 , T.: ci ston pa); and 3. for whom it is spoken (C.: suǒwéishuō 所 為 說 , T.: gang la ston pa).127 Yìnshùn (1983: b8) connected the “what is spoken” with sūtra-aṅga, “for whom it is spoken” with geya-aṅga, and “those who speak” with vyākaraṇa-aṅga.

These connections are drawn from the comparison of three descriptions (the Maulyo bhūmayaḥ, Vastusaṃgrahaṇī of the Yogācārabhūmi and Mūlasarvāstivāda Vinaya) which indicate the content of the Saṃyuktāgama. The three descriptions are as follows:128






125 Ui (1925) and Masunaga (1948: 25) had different opinions on whether Āgama/Nikāya appeared earlier than aṅga classification.
126 The correspondence between the Saṃyuktāgama and the Vastusaṃgrahaṇī has been studied by Lǚ (1923); (1992), Yìnshùn (1983), and Mukai (1985).
127 T1579, 30.72c17–18. The English titles of the three categories listed above were translated by Bucknell (2007).
128 Yìnshùn1971: 622. Cf. Bucknell 2007, Choong 2010.
 
Maulyo bhūmayaḥ129 Vastusaṃgrahaṇī130 Mūlasarvāstivāda
Vinaya131

1. Five Aggregates
(五取蘊) 3. Aggregates
(蘊 skandha) 1. Five Aggregates
(五蘊)

2. Twelve Sense Spheres
(十二處) 5. Sense Spheres
(處 āyatana) 2. Six Sense Spheres
(六處)

3. Twelve Causal Conditions
(十二緣起) 6. Causal Condition
(緣起 pratītya-samutpāda) 4. Causal Condition
(緣起)

4. Four Nutriments
(四食) 7. Nutriments
(食 āhāra)

5. Four Truths
(四聖諦) 8. Truths
(諦 satya) 5. Noble Truths
(聖諦)

6. Innumerable Dhātus
(無量界) 4. Dhātu
(界 dhātu) 3. Eighteen Dhātu
(十八界)

7. Buddha and his Disciples (佛及弟子) 1. Spoken by the Tathāgata (如來所說)
2. Spoken by Disciples
(弟子所說) 7. Spoken by the Buddha (佛所說)
6. Spoken by Śrāvakas
(聲聞所說)

8. Four Stations of Mindfulness, etc. of the bodhipakṣyadharmāḥ
(四念住等菩提分法) 9. Stations of Mindfulness, etc.
(念住正斷神足根力覺支道
支入出息念學證淨等相應) smr̥ti-upasthāna, saṃyak- prahāṇa, r̥ddhi-pada, indriya, bala, bodhyaṅga, mārga, ānāpāna-smr̥ti, śikṣā,
avetya-prasāda, etc.) 8. Stations of Mindfulness, etc. in the Noble Path section
(念處正勤神足根力覺道分相
應者於聖道品處而為建立)

9. Eight Assemblies
(八眾) 10. Eight Assemblies
(八眾 aṣṭau pariśadaḥ)132 9. Discourses connected with
gāthās (經與伽他相應)



129 T1579, 30.294a21–b2.
130 T1579, 30.772c11–15.
131 T1451, 24.407b20–27.

132 The eight assemblies referred to in the title of this section are Kṣatriya-parṣada (assembly of warriors), Brāhmaṇa-parṣada (assembly of priests), Gṛhapati-parṣada (assembly of householders), Śramaṇa- parṣada (assembly of renunciants), Cāturmahārājika-parṣada (assembly of the gods of the Four Great Kings), Trāyastriṃśa-parṣada (assembly of the Thirty-three gods), Māra-parṣada (assembly of tempter gods), Brahmā-parṣada (assembly of higher gods). See Bucknell 2007.
 
“Discourses connected with gāthās” refers to the Eight Assemblies section. The genre of the Eight Assemblies section appears to be “prose with verses.”133 The sequences of the sections are different between the three, a situation which seems to appear often in Buddhist texts due to the different times in which the content was composed or compiled. Overall, the sections of the three Saṃyuktāgama accounts are similar. These Saṃyuktāgama sections are linked with the three categories in the following chart:


1. those who speak 1. Spoken by the Tathāgata and his disciples
2. what is spoken 2. Aggregates, Sense Spheres, Causal Condition, Nutriments, Truths, Dhātu.
2. Stations of Mindfulness, etc. (Elements of
the Path 道品, Mārga-varga)
3. for whom it is spoken 3. Eight Assemblies or Discourses connected
with gāthās

Moreover, these three categories can be identified as sūtra, geya and vyākaraṇa
based on two descriptions. The first is found in the Chinese Maulyo bhūmayaḥ:134
云何契經。謂薄伽梵於彼彼方所。為彼彼所化有情。依彼彼所化諸行差別。宣說無量蘊相應語。處相應語。緣起相應語。食相應語。諦相應語。界相應語。聲聞乘相應語。獨覺乘相應語。如來乘相應語。念住正斷神足根力覺支道支等相應語。不淨息念諸學證淨等相應語。
What is sūtra? For the sake of living beings in different places [and] due to the various capacities of different beings, the Tathāgata spoke [the following]: “[[[Sūtra]] means] discourses relevant to the aggregates, discourses relevant to sense spheres, discourses relevant to causal conditions, discourses relevant to nutriments, discourses relevant to truths, discourses relevant to dhātus, and discourses relevant to the Śrāvakayāna, Pratyekabuddhayāna and Tathāgatayāna;135 as well as discourses relevant to mindfulness, right efforts, basis of supernormal power,

133 This will be further explained later.

134 T1579, 30.418b23–c1.

135 In the Saṃyuktāgama, there are only the sections spoken by the Tathāgata and his disciples, but here the disciples in the Saṃyuktāgama are divided into the śrāvakas, pratyekabuddhas in the Maulyo bhūmayaḥ of the Yogācārabhūmi. Yìnshùn explained that tri-yānas (Śrāvakayāna, Pratyekabuddhayāna and Tathāgatayāna) was a trend in later Buddhism so the content of the descriptions changed to reflect the culture.
 
faculties, powers, enlightenment factors, noble eightfold path, etc., discourses relevant to uncleanliness, mindfulness of breathing, all kinds of learning [and] obtaining purity.

This account describes the content of sūtra. Compared with the above-mentioned contents of the Saṃyuktāgama, only the Eight Assemblies section is not included. The Eight Assemblies are “discourses connected with gāthās” as the section is filled with “verses.”136 The sections in this account are supposedly the content of the sūtra-aṅga. Further, Yìnshùn (1983: b10) believed this description emphasizes the distinction between the sūtra “prose” and verse.

Another account is slightly different from the above description. Yìnshùn (1983: b10) pointed out that the sūtra-mātr̥kā in the Vastusaṃgrahaṇī of the Yogācārabhūmi excludes not only the Eight Assemblies, but also the “sections spoken by the Tathāgata and disciples.”137 Therefore, the following sections remain:


1. Discourses relevant to the Aggregates,
2. Discourses relevant to the Sense Spheres,
3. Discourses relevant to the Causal Conditions,
4. Discourses relevant to the Nutriments,
5. Discourses relevant to the Truths,
6. Discourses relevant to the Dhātus,
7. Discourses relevant to the Stations of Mindfulness, etc. of the bodhipākṣikā dharmāḥ.


Yìnshùn (1983: b10–b11) said that these seven sections correspond to the sūtra-aṅga or saṃyukta-sūtra (相應修多羅), which is the fundamental part of the Saṃyuktāgama, otherwise known as the Saṃyukta-kathā. In other words, at the beginning of collecting the Buddha’s teachings, these seven sections were supposedly the basic and earliest parts of the teachings.
How do the other two categories (speaker and audience) connect with geya and vyākaraṇa? The Eight Assemblies section (bāzhòng 八 眾 aṣṭau pariśadaḥ) of the Saṃyuktāgama has a counterpart in the Sagāthavagga of the Saṃyuttanikāya.


136 Due to “verse” style, the Eight Assemblies section is attributed to geya-aṅga, which will be discussed later.
137 The sūtra-mātr̥kā is contained in scrolls (juàn 卷 ) 85 to 98 of the Yogācārabhūmi.
 
Buddhaghosa regarded the Sagāthavagga as geya, meaning that the Eight Assemblies section of the Saṃyuktāgama should also be considered as geya. 138 Therefore, there seems to have been a tradition of regarding verses as geya in the early time of collecting the Saṃyuttanikāya or the Saṃyuktāgama. There is additional evidence to support this idea (Yìnshùn 1983: b19). The final line of a verse set in the Chinese Śarīrārthagāthā of the Yogācārabhūmi indicates that the verse preceding it is a geya verse, i.e. “cǐzhōng qíyànsòng yīngzhī 此中祇焰頌應知.”139 The verse being commented on can be traced back to the Eight Assemblies of the Saṃyuktāgama.140 Therefore, based on this evidence, the Eight Assemblies should also be designated as geya.

One of the vyākaraṇa meanings in the Abhidharma Mahāvibhāṣāśastra was defined as a conversation (answer to a question).141 These conversations are between the Tathāgata, śrāvakas (disciples), or gods by four ways of talking (catvāri praśna- vyākaraṇāni).142 The conversation, designated as vyākaraṇa, focuses on determining where a saint will appear, or a person will be reborn in the next life. In the Saṃyuktāgama, similar descriptions appear only in the “sections spoken by the Tathāgata and his disciples.”143 For example, in SĀ-G 853, Pūraṇa and his brother R̥ṣidatta went to Tuṣita Heaven after death and attained the level of the sakṛd-āgāmin (once-returner).144 This is how the “sections spoken by the Tathāgata and his disciples” connect with vyākaraṇa.

138 Sabbam pi sagāthakaṃ suttaṃ geyyan ti veditabbaṃ, visesena saṃyuttake sakalo pi sagāthavaggo (All the discourses including verses should be considered as geya, especially the whole Sagāthavagga in the Saṃyutta[[[nikāya]]]). See Papañcasūdanī II 106; Samantapāsādikā I 28.
139 This passage from the Yogācārabhūmi is in T1579, 30.378a10–11. However, according to Enomoto (1989), the counterpart of the description in the Sanskrit text he edited does not mention geya verse. The Sanskrit part of this text can be checked in the Yogācārabhūmi manuscript 122b5f and the Śrāvakabhūmi manuscript 4b8. Detailed information can be seen in chapter three (the thirteenth verse set).
140 The verse is found in T99, 2.348b8–14. See Yìnshùn 1983: b25.

141 T1545, 27.659c28–660a2. The development of vyākaraṇa will be analyzed in § 2.3.
142 Compared with the Saṃyuktāgama, the “speakers” or “talkers” section here includes one more category: gods. Yìnshùn (1981: b28) explained that Buddhism adapted to the need of Indians' belief in gods. The style of the content is normally verse, so it was later collected into the Eight Assemblies section (bāzhòng 八眾 aṣṭau pariśadaḥ) in the Saṃyuktāgama. The four ways of talking were 1. a confirmatory reply to a
question (yīxiàngjì 一向記, ekāṃśavyākaraṇaṃ), 2. a discriminating answer (fènbiéjì 分別記,

vibhajyavyākaraṇaṃ), 3. a questioning in return (fǎnjiéjì 反詰記, paripṛcchāvyākaraṇaṃ), 4. silence as an answer (shězhìjì 捨置記, sthāpanīyavyākaraṇaṃ).
143 See the table above describing the three-part structure of the Saṃyuktāgama.

144 T99, 2.240b5–10. Another story in Saṃyuktāgama 936 (T99, 2.240b5–10) is as follows: Sarakāni gave up drinking wine and obeyed the precepts. As a result, he attained the level of srota-āpanna after his life's end.
 
The above discussion briefly described how Yìnshùn identified the three-aṅga structure in the Saṃyuktāgama. Yìnshùn also highlighted similar supporting evidence in the Mahāśūnyatā Sūtra of the Madhyamāgama (and Majjhimanikāya). The passage reads as follows: “Ānanda! A disciple should follow the Tathāgata for the sake of sūtra, geya, vyākaraṇa.”145 Yìnshùn believed that this is clear evidence of a stage or time in which the Buddha’s teachings were divided according to the three-aṅgas.146

The contribution of Yìnshùn’s work has been highly regarded and accepted by many scholars, including Akira (1985), Mizuno (1988), Nagasaki and Kaji (2004), Sujato (2005)147, Glass (2006), Bucknell (2007), Bingenheimer (2011), et al. However, Anālayo (2011: 697 note 69) holds an opposing view. Anālayo adopts a different angle in interpreting the passages in the Vastusaṃgrahaṇī which Yìnshùn used.

I am going to attempt to clarify a few points. Firstly, while the Vastusaṃgrahaṇī offers vital information, the hypothesis of the three aṅgas of the Saṃyuktāgama is based not only on the descriptions in the Vastusaṃgrahaṇī, but also information in the Maulyo bhūmayaḥ and the Mūlasarvāstivāda Vinaya, in addition to the passage by Buddhaghosa. Secondly, Lévi’s assumption from the description of the Mūlasarvāstivāda Vinaya is one possibility;148 however, the existence of the first council still remains an issue, and as such the formation of the early Buddhist texts is not well determined. Therefore, the interpretation of Yìnshùn remains feasible. Also, Lévi’s research, published in 1916, may no longer be considered particularly accurate in that he wasn’t able to compare his assumptions with more recent studies and sources, such as modern studies of the Buddhist councils. Yìnshùn connected the various texts (the Mūlasarvāstivāda Vinaya,


145 T26, 1.739c4–5 and MN III 115. Cf. Anālayo 2011: 696–697. It is noteworthy that Anālayo (2011: 697) mentioned that, based on the different Pāli editions of the Majjhima-nikaya parallel, the passage could include two or three out of these aṅgas. Also, the counterpart of the Tibetan version lists the whole set of twelve aṅga (see Skilling 1994a: 242).

146 Yìnshùn 1983: a1–2. Anālayo (2011: 697 note 69) stated, “Yìnshùn regarded this as the earliest organizing principle in the evolution of the early Buddhist canon.” To be exact, Yìnshùn mentioned that these three aṅgas could be considered as the foundation of the Saṃyuktāgama (sūtra appeared earlier than the other two). Yìnshùn did not describe them as the “earliest” organizing principle.

147 Sujato provided several cases to support the importance of the three aṅgas, including passages in which they are listed in declined forms while the remaining aṅgas are combined into a long compound. He argued that it appears that this long compound is a latter addition to the initial list of three aṅgas. Anālayo (2011: 698 note 69), however, argued that these declined forms are sim
ply the result of a standard process in the transmission of Buddhist texts and not evidence of their temporal priority.
148 The assumption is that the Saṃyuktāgama is not a temporal priority of the formation of canons, but a temporal priority of its recitation by Ānanda at the first council. See Anālayo 2011: 697 note 69.
 
Maulyo bhūmayaḥ, Vastusaṃgrahaṇī, Madhyamāgama, and Saṃyuktāgama) to investigate the development of the aṅga classification and the Saṃyuktāgama. These texts used by Yìnshùn have a relationship which is relevant to their (Mūla)sarvāstivādin school affiliation.149 Yìnshùn (1983: b9) believed that he discovered these details by means of the knowledge heritage in the (Mūla)sarvāstivāda tradition. Therefore, accounts from the texts of other schools proposed by Anālayo are not effective against Yìnshùn’s hypothesis.

Moreover, Anālayo mentioned that such an early three aṅga system would have been described directly in the texts, instead of needing to be deduced from several indirect descriptions. In fact, extant Buddhist texts are not totally complete and the transmission of Buddhist collections has been complicated. Therefore, it is impossible to know whether the three aṅga system has been directly recorded anywhere. Even though the early three aṅga system is not described directly in early extant Buddhist literature, this does not mean that this system definitely never existed. Another relevant passage given in the Mahāśūnyatā Sūtra of the Madhyamāgama,150 which also belongs to the (Mūla)sarvāstivādins, was considered by Anālayo to be the result of a transmission error or accident. The reason Anālayo mentioned this is that the two instances (in the Madhyamāgama and Majjhimanikāya) of the three aṅgas are rare.151 I agree with the assertion that with a greater amount of evidence comes a higher degree of persuasion.

However, the reason that Yìnshùn seemed to be confident in his discovery is that he not only found the instance in the Mahāśūnyatā Sūtra of the Madhyamāgama (which is also considered a text of the (Mūlasarvāstivādins), but also connected it with related information from the Mūlasarvāstivāda Vinaya, Maulyo bhūmayaḥ, Vastusaṃgrahaṇī. These descriptions seem to indicate the core message and knowledge contained in the (Mūla)sarvāstivāda tradition. From this case, it is obvious that Yìnshùn has considered the related accounts as a whole in order to determine the message left behind. Anālayo does not appear to have noticed this methodology used by Yìnshùn.

149 The school affiliation of the texts has been discussed by Schmithausen (1987: 377–380). Enomoto (2000) believed that Sarvāstivāda and Mūlasarvāstivāda were the same school and that different names were simply applied to this school at certain times and in certain regions; however, this remains controversial.

Therefore “(Mūlasarvāstivāda” is taken as a shorthand for “Mūlasarvāstivāda” and/or “Sarvāstivāda.”
150 T26, 1.739c4–5.
151 See Anālayo 2011: 699. There are different interpretations of the Pāli editions of the Mahāsuññata-sutta. The passages of the Pāli editions could be rendered as either two or three aṅgas. See Anālayo 2011: 697 note 67.
 
2.1.3. The fourth aṅga, gāthā, and the Śarīrārthagāthā verses

As mentioned in the first chapter, the dharma described in the Cintāmayībhūmi has two forms: sūtra and gāthā (discourse and verse).152 The Śarīrārthagāthā is part of the gāthā group. Yìnshùn (1971: 502–503, 621) regarded this passage describing the two forms of the dharma as evidence of the first stage of the aṅga classification system using the categories sūtra and geya. Furthermore, Buddhaghosa mentioned that the Sagāthavagga should be considered as geya.153 Thus, gāthā here appears to be geya, despite the fact that gāthā is normally regarded as the fourth aṅga of the nine aṅga classification system. This raises some important questions, namely, how can gāthā be geya and what is the relationship and difference between gāthā and geya?

The meanings of gāthā and geya appear to overlap. Both words are derived from the verbal root √gai (to sing, speak or recite; MW s.v. gai). This is the reason that “verse” is related to both gāthā and geya. Highlighting the inadequacy of the definition of the term geya, Jayawickrama (1959: 12) accepted that geya seems to have become a “technical term meaning gāthā interspersed with prose.” However, he felt that the original definition of geya might have been lost and expressed doubt that the early meaning of geya was “prose with verse.” From Jayawickrama’s study, based on a brief examination of Buddhaghosa’s description, the definition of geya still remains mysterious.
Similarly, most western scholars have accepted Buddhaghosa’s description, namely, “Dhammapada, Theragāthā, Therīgāthā, and the sections entirely in verse in the Suttannipāta which are not designated as suttas should be known as gāthā.”154 Nevertheless, this also raises two arguments. Firstly, some scholars have mentioned that Dhammapada, Theragāthā, Therīgāthā were compiled later than the existence of the nine-aṅga classification system, so these texts should not be considered gāthā (Ui 1925:

152 Dharmapravicaya entails examination of the dharma, which takes two forms, namely, the meaning of the discourses (sūtra-artha) [of the Buddha] and the meaning of the verses (gāthā-artha). See T1579, 30.361c2–3.

153 Sabbam pi sagāthakaṃ suttaṃ geyyan ti veditabbaṃ, visesena saṃyuttake sakalo pi sagāthāvaggo (All the discourses including verses should be considered as geya, especially the whole Sagāthavagga in the Saṃyutta[[[nikāya]]]). See Papañcasūdanī Ⅱ 106; Samantapāsādikā Ⅰ 28.
154 This translation is by Norman (1983: 15). dhammapada-theragāthā-therīgāthā suttanipāte
nosuttanāmikā suddhikagāthā ca gāthāti veditabbā. Samantapāsādikā Ⅰ 28. However, Jayawickrama (1959) said that it is difficult to demarcate between sutta and gāthā in the Suttanipāta, except that prose is excluded from gāthā.
 
160 and Shiio 1935: 26). However, some mention that although new verses were added to the Dhammapada, Theragāthā, and Therīgāthā by later generations, the old contents existed much earlier (Ogiwara 1869–1937: 405 and Hayashiya 1936: 716). Thus, these texts could still be regarded as gāthā. Secondly, Ui (1925: 160) found that the Pārāyanavagga (Suttanipāta V) suits the definition of gāthā, indicating that there are other texts not involved in Buddhaghosa’s description. Likewise, Ogiwara (1869–1937: 405) indicated that a number of sūtras from the passage of the

Abhidharma mahāvibhāsā-śāstra also suit the defintion of gāthā.155 But Ogiwara regarded the Pārāyanavagga as veyyākaraṇa, not gāthā. Mayeda (1964: 324) disagreed with Ogiwara’s opinions because Buddhaghosa’s commentaries and the Abhidharma mahāvibhāsāśāstra come from different traditions. Akanuma (1981: 170) rejected Buddhaghosa’s definition of gāthā in aṅga classification, since the texts indicated by
Buddhaghosa as representative of gāthā were all collected after the period in which aṅga classification was in full usage. He also disagreed with Ui’s idea about the Pārāyanavagga. Akanuma pointed out that the Pārāyanavagga is a late text by examining one verse as an example. Therefore, it would seem that Buddhaghosa’s definition of gāthā is not decisive.

It is extremely difficult to know which texts of the Dhammapada, Theragāthā, Therīgāthā and Suttanipāta may be classified as gāthā (Mayeda 1964: 326). Mayeda (1964: 327–28) thus examined the whole Āgama and Nikāya collections. He defined the gāthā aṅga by a process of exclusion. Examining the Khuddaka-nikāya first, he excluded content that belongs to other aṅgas, e.g. geya, vyakaraṇa, etc., with the verses remaining being designated by him as gāthā. He extended this investigation to the other four Āgamas and Nikāyas, to refine his definition. However, a definite understanding of the terms geya and gāthā still remains elusive, due to the unclear relationship between the terms geya and gāthā and the overlapping descriptions. Yìnshùn’s deduction of the development of gāthā might solve some of these issues.
Yìnshùn (1971: 540) stated that all verses intended for chanting the dharma are considered gāthā. This idea was derived from the Āpídámó dàpípóshā lùn 阿毗達磨大毗


155 These sūtras are the Khaggavisāna Sutta (Suttanipāta I 3), Uraga Sutta (Suttanipāta I 1), Metta Sutta
(Suttanipāta I 8), Hiri Sutta (Suttanipāta II 3), Nāvā Sutta (Suttanipāta II 8).
 
婆 沙 論 (Abhidharma Mahāvibhāṣā-śastra)156 and Yogacārabhūmi.157 Given such a broad description, it seems that there is an overlap between the definitions of gāthā compared with udāna and other types of verse. Take udāna for example. These verses can be described as being created in response to moods and emotions,158 such as the verses in the Dhammapada of the Pāli tradition, or the Udānavarga of the (Mūla)sarvāstivāda tradition.159 Therefore, in a general sense all verses might be considered as gāthā.

However, it is still possible to observe verses with variable functions within that category.

The Shùnzhènglǐ lùn 順正理論 (Abhidharma Nyāyānusāra-śāstra, T1652)

distinguishes between geya and gāthā by providing different definitions of each. It defines gāthā as verses which praise the Buddha’s teaching by means of delicate words, but which are not connected with the preceding prose.160 In contrast, this text defines geya as verses which praise the dharma and which are connected to the preceding prose. Geya verses are normally made in order to summarize or repeat the preceding prose,161 whereas gāthā verses are created independently (gūqǐ sòng 孤 起 頌 ) without the preceding prose, that is, the verse is based on the thoughts of its composer.
Yìnshùn (1971: 543) also discussed the relationship between geya and gāthā by looking at two specific passages:

一切偈名祇夜，六句、三句、五句，句多少不定，亦名祇夜，亦名伽陀。All verses are geya and may contain six or three or five stanzas, [i.e., it is] not fixed. They are also named “gāthā.” 162

祇夜名偈。偈有二種，一名伽陀，二名路伽。路伽有二種，一順煩惱，二

156 T1545, 27.660a2–3.
157 T1579, 30.418c9–11. (云何諷頌？謂非直說，是結句說。或作二句，或作三句，或作四句，或作五句， 或作六句等，是名諷頌。)
158 See Udānaṭṭhakathā p. 2: kim idaṃ udānaṃ nāma pītivegasamuṭṭhāpito udāhāro (What is it that is named udāna? A saying which has elicited joy and shock). Also see Āpídámó dàpípóshā lùn 阿毗達磨大毗婆沙論(Abhidharma Mahāvibhāṣā-śastra) T1545, 27. 660a6:自說云何。謂諸經中因憂喜事世尊自說。(What is udāna? It is in the sūtras; the Buddha talking due to something sorrowful or pleasing).
159 See T1545, 27.660a6. The Dhammapada is a parallel text to the Udānavarga.
160 言諷頌者。謂以勝妙緝句言詞。非隨述前而為讚詠。(T1562, 29.595a9–10).
161 That is also the reason why geya is normally defined as being all suttas with verses. See Mayeda 1964: 267–268. Cf. Lamotte 1980 and Norman 1983.
162 T1509 25.307a18–20. (Dàzhìdù lùn 大智度論, Mahāprajñāpāramitāśāstra)
 
不順煩惱。不順煩惱者，祇夜中說，是名伽陀。
The kind of verse is geya. There are two types of verses: gāthā and loka. The two kinds of loka verses are verses connected to afflictions (kleśa) and verses not connected to afflictions. Those that are not connected with afflictions are geya.163

According to the first passage, not only can all verses be gāthā, but they can also be geya. This means that geya was also a general name for verses at the time. According to the second passage, verses (geya) can be divided into two kinds: gāthā and loka. The definition of loka here is a verse relevant to ordinary life in the secular world. In contrast, gāthā is a verse related to supra-mundane life, i.e. the Buddha’s teaching. In brief, gāthā is a Buddhist verse and loka is not. The verse of loka can also be separated into two types: those verses connected to afflictions (kleśa) and those which are not. The verses not connected to afflictions are also called geya. The second passage can be summarised in the following chart:

gāthā
geya connected to afflictions
loka
not connected to
afflictions
geya


Why can verses be geya or gāthā in a broad sense? Yìnshùn said that the word geya was commonly used for verse in India. But for a special usage, geya verses were those about stories in society but not connected to afflictions (kleśa). Therefore, at the beginning of collecting the Buddha’s teachings, verse was generally known as geya.164 The geya verses are like the section of the Eight Assemblies in the Saṃyuktāgama, or Udānavarga (優陀那), Pārāyaṇa (波羅延那), and Arthavargīya (義品).165 After a number of years, Buddhist verses had proliferated to become voluminous and were then considered gāthā, and others loka. However, while verses had already been catalogued into the piṭaka or the aṅga classification system, the early verses were, nevertheless, traditionally still called geya. In brief, only in Buddhist collections were the representation of all verses

163 T1646 32.245a9–12. (Chéngshí lùn 成實論, Satyasiddhiśāstra)
164 The details about the evolution of geya will be discussed in a later section of this chapter.
165 These verses existed in early times, but they were collected later. See Yìnshùn 1971: 541.
 
considered geya or gāthā. Geya is a very old traditional term for verse in India, and
gāthā is particularly reserved for Buddhist verse.
In the Cintāmayībhūmi, there are three parts with verses (gāthā), namely, the Paramārthagāthā, Ābhiprāyikārthagāthā and Śarīrārthagāthā. The origins of the Paramārthagāthā and Ābhiprāyikārthagāthā have not as yet been identified with certainty, but may have been derived from Śrāvakayāna canonical texts.166 The Śarīrārthagāthā verses are all quoted from Śrāvakayāna canonical texts.167 The functions and definitions of the Śarīrārthagāthā verses seem diverse, which will be analyzed in the next section. From the above information we may conclude that the three gāthās in the Cintāmayībhūmi generally refer to “Buddhist verse” and should not be attributed to the gāthā aṅga.

2.1.4. Summary

This section focused on several aspects of aṅga classification. By better appreciating the formation of aṅga classification, its role and function in Buddhist texts can be better understood. Understanding the early development of aṅga classification is particularly assisted by the study of the Saṃyuktāgama, in which the three aṅgas can be identified as essential to its organisation. The three aṅga classification is important in the case of the Śarīrārthagāthā, because all of its verses can be assigned to either the geya or vyākaraṇa aṅga. This will be discussed further in the following sections. In addition, according to the meaning of gāthā, in the title Śarīrārtha-gāthā it is thought to be referring to either verse in general, or more narrowly, Buddhist verse.









166 Ui (1958: 316–319) mentioned that the verses in the Paramārthagāthā and Ābhiprāyikārthagāthā are canonical, though he is unable to locate them in the surviving canons. Wayman (1961: 164) conjectured that the composer of these verses utilized components of canonical verses to create his own verses.

Schmithausen (1987: 223) suggested that the verses in the Paramārthagāthā, which are not entirely original compositions, may have been quoted from para-canonical sources.

167 From the Saṃyuktāgama, Udānavarga, Pārāyaṇa, Arthavargīya and Madhyamāgama. See Enomoto 1989 and Chpater three and four.
 

2.2. The second aṅga, geya, and the Group A verses of the Śarīrārthagāthā


As mentioned in chapter one, the verses in Group A of the Śarīrārthagāthā are all quotations from the Saṃyuktāgama. Additionally, it has been found that these verses are all from the section identified as geya-aṅga in the Saṃyuktāgama.168 Is it also similar to the relationship between the Vastusaṃgrahaṇī of the Yogācārabhūmi and the sūtra-aṅga section of the Saṃyuktāgama?169 At first, this section is going to investigate the definition of geya. The definition of each aṅga is not evident in Śrāvakayāna canonical texts; rather, all information relating to aṅga was first described in later texts such as Buddhaghosa’s commentary, and Abhidharma or Mahāyāna texts. The definitions of aṅga in texts, nevertheless, appear quite divergent. Yìnshùn (1971: 493–494) stated that this is because each aṅga’s definition kept changing with time. Therefore, while exploring the meanings of geya in different texts, the development of the term geya will be deduced. The
meaning of geya in the Saṃyuktāgama will be outlined as well. Second, the specific sūtras of the Saṃyuktāgama to which Group A of the Śarīrārthagāthā verses correspond, will be examined to ascertain whether all verses are actually from the section of geya- aṅga in the Saṃyuktāgama. Finally, the relationship between the Saṃyuktāgama and the Group A of the Śarīrārthagāthā will be discussed.


2.2.1. The development of geya


The meaning of the word geya (Pāli: geyya) is “a song” or “singing” (a gerundive from √gai, to sing).170 There are three general ways in which the term geya is rendered in


168 This issue is discussed further in § 2.2.2.

169 The sūtra-aṅga section of the Saṃyuktāgama is cited in the Vastusaṃgrahaṇī of the Yogācārabhūmi, which was identified by Yìnshùn (1983). Also, due to the sequence of each sūtra, Lǚ (1924) and Yìnshùn (1983) mentioned the Vastusaṃgrahaṇī is the mātṛka of the sūtra-aṅga section of the Saṃyuktāgama.

170 See MW s.v. geya. Besides, Mayeda (1964: 267) stated that from the original meaning “to be sung”, “being sung,” geya should correspond to the Chinese term 應 頌 . He seems to regard the Chinese term as “should be sung” (yīng sòng 應頌). The first character yīng 應 of the Chinese term, which is the first tone, is an auxiliary verb (“should” or “should be”). However, the actual meaning of 應頌 should be taken as
“the verse of responding to or corresponding to [the previous sūtra].” 應 (yìng) is the fourth tone in Chinese, in which case it is a gerundive or adjective (“corresponding”). See T1562, 29.595a4–6. (言應頌者，謂以勝妙緝句言詞，隨述贊前契經所說。)
 
Chinese. The first is qíyè 祇夜, qíyè jīng 祇夜經, qíyè 岐夜, which are transcriptions based on the sounds of the word geya. Second, the commonly used terms are yìngsòng 應頌, chóngsòng 重頌, and jiésòng 結頌, which, in terms of the relevant preceding prose (sūtra), mean a verse of “responding” or “corresponding” (yìng 應 ), of “repeating” (chóng 重 ), of ‘“summarizing” or “gathering together” (jié 結 ). Examples of the third kind are sòngjīng 頌經, gē 歌, měiyīn jīng 美音經, shī 詩, sòng 誦, etc. Briefly, the

meanings of these terms are “song”, “verse”, or “poem.” Based on these three kinds of Chinese terms, the word geya obviously indicates a verse or song. The specific functions could be responding, repeating, summarizing or gathering together the previous prose.
These are literal interpretations of the Chinese terms. However, according to the Pāli tradition (or Buddhaghosa’s explanation), geyya was defined as “mixed prose and verses.”171 In other words, the term geya should not be considered as only “verse.” This definition is different from that seen in the Chinese tradition. Mino (1926: 69–70) and Mayeda (1964: 267) have stated that the term geyya is not often treated as a noun and not frequently seen in Pāli texts applied to aṅga classification, suggesting the development of geyya might not have been complete at the time. This might be the reason why we can see various interpretations.

Scholars have generally based their understanding of the oldest traditional meaning of geya upon two passages. First, the Āpídámó dàpípóshā lùn 阿毗達磨大毗婆沙論(Abhidharma Mahāvibhāṣā-śastra) says:

應頌云何? 謂諸經中，依前散說契經文句，後結為頌而諷誦之，即結集文結集品等。 (T1545, 27.659c21–23.)
What is the corresponding verse (geya)? In all sūtras, those verses were made in order to be sung corresponding to the preceding prose (sūtra),172 such as the “gathering sayings” (jiéjí wén 結集文) and [those in] the “gathering section” (jiéjí pǐn 結集品) and so on.


171 See Papañcasūdanī Ⅱ 106, Samantapāsādikā Ⅰ 28. Also see Lamotte 1980: 144, Norman 1983: 15.
172 Here sūtra refers to the genre of literature, i.e. prose (sànshuō qìjīng wénjù 散說契經文句), not to an individual unit of literature.
 
From this definition, geya were regarded as verses and were composed to correspond to the preceding prose. Therefore, the function of a geya verse is not definite, and might be either responding, repeating, summarizing or “gathering” (jiéjí 結 集 ) the previous prose. In other words, a geya verse should correspond to the relevant prose no matter what the specific function of the verse is. After this description, the Āpídámó dàpípóshā lùn 阿 毗達磨大毗婆沙論 gives an example of a verse which has a repeating function; however,173 this should be considered as only one function of a geya verse. Therefore, it might be said that chóngsòng 重 頌 (repeating verse) is a subordinate term to yìngsòng 應 頌(corresponding verse). However, if the intention was to define geya only as chóngsòng 重頌 (repeating verse), the text above should have specified “zhòngsòng yún hé 重頌云何 ” (what is a repeating verse), instead of speaking in terms of yìngsòng yún hé 應 頌 云何 (what is a corresponding verse). Similar accounts can also be seen in other texts, like the Yogācārabhūmi.174 These definitions also emphasize the difference between geya and gāthā, in that gāthā verses were created independently (gūqǐ sòng 孤 起 頌 ) and not in response to the preceding prose. The geya definition sheds light on the connection between prose (sūtra) and verse. Mayeda (1964: 267–268), making use of the definition given above, stated that geya is the integration or combination (jiéhétǐ 結合體) of prose (sūtra) and gāthā (verse);175 that is, gāthā (verse) should appear with prose in the definition of geya. Mayeda’s interpretation appears to closely follow the definition of geya by Buddhaghosa, that is, “suttas with verse.”



173 See T1579, 30.659c23–27.

174 T1579, 30.753a13. See also Chéngshí lùn 成 實 論 (Satyasiddhi-śāstra), T1646, 32.244c16; Shùnzhènglĭ lùn 順正理論 (Abhidharma Nyāyānusāra-śāstra), T1562, 29.595a4–6; Dàhéng āpídámó zájí lùn 大乘阿毘達磨雜集論 (Mahāyānābhidharma Samuccaya-vyākhyā), T1606, 31.743c4; Xiănyáng shèngjiāo lùn 顯揚聖教論 (Prakaranāryavāca-śāstra), T1602, 31.538b23.

175 As for the position, the Xiănyáng shèngjiāo lùn 顯揚聖教論, the outline of the Yogācārabhūmi, points

out that geya verses could be located after or between sūtras, a situation also observed in the Eight Assemblies of the Saṃyuktāgama. Mayeda (1964: 268) also mentioned that gāthā (verse) cannot only repeat the previous sūtra, but also provide some additional information according to an account in the Chéngshí lùn 成實論.
 
Buddhaghosa’s description, another traditional meaning of geya, is given as follows:176

Sabbam pi sagāthakaṃ suttaṃ geyyan ti veditabbaṃ, visesena saṃyuttake sakalo pi sagāthavaggo.
All discourses with verses (gāthā) should be considered as geya, especially the whole Sagāthavagga in the Saṃyutta[[[nikāya]]].

This definition emphasizes the style of geya—sūtra with verses (gāthā, especially).177 However, this geya definition is too broad to understand and causes many arguments. For example, von Hinüber (1996: 47) and Bucknell (2007) stated that the style (sūtra with verses) and some gāthās in the Sagāthavagga can also be seen elsewhere in the Suttapiṭaka.178 Therefore, it is difficult to know the distinct feature of geya in the aṅga classification system according to Buddhaghosa’s definition. Scholars also question whether the example of the Sagāthavagga suits the definition of geya. In the Sagāthavagga, the verses do not correspond to the previous prose. Von Hinüber (1996: 46) stated that it is more like prose introductions combining with existing verses.179 Jayawickrama (1959: 12) argued that geya seems to refer to the ākhyāna-type, “containing stanzas punctuated with narrative prose.” In early times, in the ākhyāna-type only the verses had a fixed style, while the narrative prose was “given by the reciter in his own words.” But as time went by the prose became fixed and in some cases even



176 Papañcasūdanī Ⅱ 106; Samantapāsādikā Ⅰ 28.


177 Some scholars, like Norman (1983: 17), have taken sūtra here to mean the individual unit of literature, while others, e.g. Kalupahana (1965: 617), explained sūtra as a genre of literature—prose. Bucknell (2007: 7–8) explained “sūtra with gāthā” as “verses embedded in a prose framework,” which I tend to use.

178 According to Bucknell's investigation, the intention of the compilers of the Sagāthavagga and the Eight Assemblies section was not only to collect the Buddha's teaching (dharma) that appear to include verses (gāthā), but also to constitute “a coherent collection with a definite underlying structure.” That is, the twelve topics (saṃyuttas) of the Sagāthavagga are connected closely with the twelve classes of beings in the Assemblies section (Kosala, Brāhmaṇa, Vaṅgīsa, Bhikkhu, Bhikkhunī, Devatā, Devaputta, Yakkha, Vana, Sakka, Māra, and Brahmā). Therefore, many sūtras with the sagātha form (sūtra with gāthā) could not be added into the Sagāthavagga. See Bucknell 2007.

179 Regarding the verses of the Sagāthavagga, Bodhi (2000: 70) stated, “Doubtlessly some of the verses were not original to the suttas in our collection but belonged to the vast, free floating mass of Buddhist didactic verse which the compilers of the texts pinned down to specific contexts by providing them with narrative settings such as those found in the Sagāthavagga.”
 
versified. 180 Mino (1926: 94) said that this is an excessive interpretation and Buddhaghosa’s definition can barely be trusted. Mayeda (1964: 270), based on an examination of the whole Suttapiṭaka, also stated that Buddhaghosa’s comment on the Sagāthavagga is not really an orthodox or standard geya definition. Both of them value more the definition from the Āpídámó dàpípóshā lùn 阿毗達磨大毗婆沙論. In contrast, some scholars still accept the Sagāthavagga as an example of geya in the broad sense. Ui (1925: 159) and Hayashiya (1936: 717) regarded the Sagāthavagga as geya, but Ogiwara (1972: 404) and Akanuma (1939: 170) took not only the Sagāthavagga but also other sūtras (like Suttanipāta III. 3 Subhāsita Sutta, III. 10 Kokāliya Sutta, III. 12 Dvayatānupassanā Sutta) as geya.

Differently, Yìnshùn (1971) discovered that both descriptions above (from the Āpídámó dàpípóshā lùn 阿毗達磨大毗婆沙論 and Buddhaghosa) share similar ideas, especially the same examples of geya. At first, Yìnshùn (1971: 512) pointed out in the case where geya is defined as “sūtras with gāthā,” there is a difficulty in that geya always ranks first or second and gāthā comes after (fourth mostly) in the aṅga classifications. Yìnshùn (1971: 512–519) deduced the development of geya from the important information which is located at the end of the geya definition of the Āpídámó dàpípóshā lùn 阿毗達磨大毗婆沙論 and which had been ignored by previous scholars. The account of the Āpídámó dàpípóshā lùn 阿毗達磨大毗婆沙論 mentions that examples of geya are jiéjí wén 結 集 文 (saying or literary work of gathering) and jiéjí pǐn 結 集 品(the gathering section).181 The term jiéjí pǐn 結 集 品 can also be found in the Chinese Yogācārabhūmi.182 The description in the Chinese Yogācārabhūmi detailing the content of the Saṃyuktāgama indicates that the Eight Assemblies section (bāzhòng 八眾) is also called “jiéjí pǐn 結 集 品 .” Enomoto (1986: 29 note 50) found that the Eight Assemblies section (bāzhòng 八 眾 ), according to a Sanskrit fragment of the Saṃyuktāgama, was designated also by the title Saṃgītavarga. 183 This roughly corresponds to the Sagāthavagga of the Saṃyuttanikāya.184 Therefore, the jiéjí pǐn 結集品 is considered to

180 This can be seen in the Pabbajjā and Padhāna Suttas of the Suttanipāta. See Jayawickrama 1959: 12.
181 T1545 27.659c22–23.

182 若諸苾芻天魔等眾。是所為說。如結集品。(T1579 30.772c22).
183 Enomoto (1994: xiii) changed the title to saṃgītanipāta. He did not mention the reason for doing so, but maybe it was to make it correspond to other titles in the Saṃyuktāgama.

184 Details of sūtra correspondence between the Saṃyuktāgama and Saṃyuttanikāya can be found in
 
be the Saṃgītavarga. The meaning of Xuánzàng’s 玄奘 translated term jiéjí pǐn 結集品 is regarded as “the section (varga) of gathering/collecting” which at first seems essentially different to the Sanskrit term Saṃgītavarga (the section of song/verse).185 But they are in fact actually synonymous. Referring to a similar case, the term saṃgīti (singing together; MW s.v. saṃgīti), can prove this idea. The term saṃgīti is usually rendered as “Buddhist Council,” which entailed the collecting and joint recitation of the Buddhist sūtras at a time when the Buddha’s teachings were orally transmitted.186 In Chinese texts, saṃgīti was generally translated into jiéjí 結 集 (gathering or assembly). The Chinese translation obviously emphasizes the purpose or function of the term. The Chinese translation jiéjí pǐn 結 集 品 is supposed to be “Saṃgītavarga.”187 The term Saṃgītavarga also explains the counterpart of the Tibetan Yogācārabhūmi that was translated into “the section of verses” (brjod pa'i sde tshan).188 The term jiéjí pǐn 結集品 in the Āpídámó dàpípóshā lùn 阿毗達磨大毗婆沙論 has taken to be equivalent to the Eight Assemblies section (aṣṭaupariṣadaḥ) of the Saṃyuktāgama.189 Similarly, Buddhaghosa’s definition shows that an example of geya is the Sagāthavagga of the Saṃyuttanikāya, which is the counterpart of the Eight Assemblies section. From the preceding information, we may surmise that geya is jiéjí pǐn 結 集 品 (Saṃgītavarga), and also the Eight Assemblies section (aṣṭaupariṣadaḥ) of the Saṃyuktāgama.
Anesaki 1908: 34–37, Akanuma 1929, and Bucknell 2007.

185 Mukai (1980: 181) stated that saṃgīta is usually known as “jìsòng 偈 頌 (verse)” from the Tibetan version. But, Mukai (1985: 14) argued that the Chinese translation by Xuánzàng (玄奘), “jiéjí pǐn 結集品,” of the title Saṃgītavarga is not correct. However, Xuánzàng does not seem to have been translating literally when providing the translation jiéjí pǐn 結集品.

186 See Ogiwara 1869–1937: 67–99, Cousins 1983, Nakamura 1987: 25, and Hirakawa 1990: 38. However, Norman (1997b) argued that “council” is an inappropriate translation which has led to a misunderstanding of the meaning of saṃgīti. He translated this term as “joint recitation.” “Communal recitation” and “collective recitation” are two other common translations (see Prebish 1987, Gombrich 1988, Collins 1990, Gethin 2005). These translations stress the “oral” transmission of texts and better suit

the original meaning of the term. As time went on, saṃgīti came to refer to the assembly, called “council,” whose purpose was not only to gather and organize the Buddhist literary texts, but also for combating the forces of decay within the Saṅgha since the Third Council (for more detail see Bechert 1970: 763).

Saṃgāyanā is another common term for “council” (see Gombrich 1988).
187 Another piece of evidence that can support this suggestion is provided by Enomoto (1984: 29 note 50), who discovered the term Saṃgīta Sūtra (or Saṃgīti Sūtra) in a Turfan Sanskrit fragment. This could correspond to Saṃgīti Suttanta of the Dīghanikāya. The name of this sūtra, the Saṃgīta Sūtra, was translated in Chinese as sēngqítuó jīng 僧 祇 陀 經 (T1428, 22.968b17 and T1421, 22.191a20) which is

explained as zhòngjí 眾集 (the collection from many) (T2130, 54.984a21). This meaning is also similar to
jiéjí 結集, as discussed above with reference to the term jiéjí pǐn 結集品.
188 The Peking edition, No. 5540, Hi 1436. Also see Mukai 1985: 11.
189 See § 2.1.2.
 
The difference between the two descriptions (of the Āpídámó dàpípóshā lùn 阿 毗 達磨 大 毗 婆 沙 論 and Buddhaghosa) is whether the verse should correspond to the prose. As the content of the Sagāthavagga has been analyzed by many scholars, the verses do not always reflect the meaning of the former prose (sūtra). Therefore, it accords with Buddhaghosa’s interpretation, i.e. prose (sūtra) “with” verses.190 If the prose (sūtra) and verses are brought together, and the verses do not always reflect the prose (sūtra), how do we regard the relationship between the prose (sūtra) and verses? The term jiéjí wén 結集文 (the saying or literary work of gathering) is a key. Yìnshùn (1971: 517) said that the meaning of the term jiéjí pǐn 結 集 品 (the Eight Assemblies section (aṣṭaupariṣadaḥ) of the Saṃyuktāgama) is related to the term jiéjí wén 結 集 文 . An account in the Yogācārabhūmi, located after another description about the content of the Saṃyuktāgama, shows that for sake of extending the existence of the holy doctrine, those (saṃgītikāra) who collected the Buddha’s teachings made the uddāna gāthās (wūtuōnán sòng 嗢拕南頌),191 which were sequentially placed in correspondence to the prose (sūtra).192 Yìnshùn (1971: 517) stated that uddāna gāthās (the verses of mnemonic summary) are taken to be jiéjí wén 結集文 (the sayings or literary works of “gathering”), so the function implied in the term jiéjí wén 結 集 文 is the gathering [of sūtras].193 That is, at the first Buddhist Council, the uddāna gāthās were composed to reflect the organization of the saṃyukta- sūtras (mentioned in § 2.1.2)194 in case the sūtras were lost or disarrayed. The saṃyukta-

190 The content and style of the Eight Assemblies (jiéjí pǐn 結集品) in the Saṃyuktāgama to date has not been examined.
191 The term wūtuōnán sòng 嗢拕南頌 is generally regarded as uddāna. However, the Chinese translation
wūtuōnán sòng 嗢 拕 南 頌 seems to indicate that the original Sanskrit term was a compound. Therefore, uddāna gāthās suits this description. This term uddāna gāthās can also be seen in the titles of two other texts, attested in the Schøyen Collection related to the Mūlasarvāstivādin Vinaya, “Nidāna and Muktaka Uddāna Gāthās” (根本説一切有部毘奈耶尼陀那目得迦攝頌, T. 1456) and “Vinayakśudrakavastu Uddāna Gāthās” (根本說一切有部略毘奈耶雜事攝頌, T1457). “Uddāna Gāthās” also corresponds to the Chinese translation shèsòng 攝 頌 (cf. Clarke 2002). Moreover, as Yìnshùn stated, “uddāna verses” is a case in

which the verse corresponds to the previous prose, which suits the geya definition, “sūtra with gāthā.” Uddāna verses are gāthās. The compound “uddāna-gāthā” from the titles of two other texts seems to be another piece of evidence to support Yìnshùn’s idea.
192 後結集者為令聖教久住。結嗢拕南頌。隨其所應。次第安布 (T1579, 30.772c15–16).

193 Uddāna is like an index which indicates 10–12 sūtras or vastus in order. Each uddāna is filled with keywords indicating each sūtra or vastu. See Panglung 1979, Clarke 2002 and Bingenheimer 2011: 16.

The detailed content is


1. Discourses relevant to the aggregates;

2. Discourses relevant to sense spheres;

3. Discourses relevant to causal conditions;

4. Discourses relevant to nutriments;

5. Discourses relevant to truths;

6. Discourses relevant to dhātus; and

7. Discourses relevant to the stations of mindfulness, etc. of the bodhipākṣikā dharmāḥ.
 

sūtras and the uddāna gāthās were the first two parts to be collected at the time.195 This is the reason sūtra and geya are the first two aṅgas in the aṅga classification. Furthermore, Yìnshùn (1971: 518) postulated that “gathering” is the earlier meaning of the term geya in Buddhist aṅga classification. This also matches the definition of geya in the Āpídámó dàpípóshā lùn 阿毗達磨大毗婆沙論. 196 Yìnshùn (1983: b28) said that the Eight Assemblies section (or Saṃgītavarga) was collected after the jiéjí wén 結集文 (uddāna gāthās), and so was designated as “jiéjí pǐn 結集品.”197 The function of the word jiéjí 結集 in this term is therefore taken to literally mean “gathering.”198 However, does the function also imply “gathering sūtras”? This needs further discussion by means of an investigation of the Eight Assemblies section (jiéjí pǐn 結集品).
The Eight Assemblies section can be examined in two versions of the Chinese

Saṃyuktāgama, as shown in Table 2.1.199 This table demonstrates the correspondence between SĀ-U and SĀ-G, the functions of the verses in each sūtra, and the titles of their counterparts in the Saṃyuttanikāya.200 Several functions of verses appear in the Eight Assemblies section. First, a verse may “respond” to the prose (sūtra), in that it replies to a conversation or situation. For example, in SĀ-U 023, the Buddha knows that the demon king Māra has come to confuse and disturb, so he speaks a verse about this situation. Second, a verse may “summarize” or “repeat” the prose (sūtra).201 In SĀ-U 001, the Buddha tells his disciples a story about Sujāta. After that, he versifies the story to encourage his disciples to practice harder. Third, a verse may neither respond to nor

195 Yìnshùn (1983: b23) mentioned that the uddāna gāthās are eventually gathered as a collection or group, such as Vinayakśudrakavastu Uddāna Gāthās" (根本說一切有部略毘奈耶雜事攝頌, T1457), so geya still indicates a specific kind of text, not a style of literature.
196 That is, “in all sūtras, those verses were composed to be sung according to the preceding prose (謂諸經中。依前散說契經文句。後結為頌而諷誦之).” T1545, 27.659c21–22.

197 Yìnshùn (1971: 518) stated that the combination of the jiéjí wén 結 集 文 and the Eight Assemblies section of the Saṃyuktāgama was entitled “jiéjí pǐn 結集品.” But later Yìnshùn (1983: b23 and b28) did not describe these two parts as “jiéjí pǐn 結 集 品 ” anymore. It seems that he corrected his previous description and noticed in the Dàpípóshā lùn 大毗婆沙論 the jiéjí pǐn 結集品 is clearly regarded as corresponding to the Eight Assemblies section.

198 See Yìnshùn 1971: 518.
199 These two versions in the Taishō edition were no. 99 Zá āhánjīng 雜阿含經 (henceforth abbreviated “SĀ-G”) and no. 100 Biéyì zá āhánjīng 別譯雜阿含經 (henceforth SĀ-U). Another text, no. 101, is also entitled Biéyì zá āhánjīng 別譯雜阿含經; however, as it is comprised of only 27 sūtras, it will not be included in this study.

200 This table was based on Bingenheimer 2011 and Bucknell 2011.
201 The verse functions to condense the meaning of the sūtra, so the verbs “summarize” and “repeat” both suit this definition.
 
summarize the prose. For instance, SĀ-U084 contains a prose line and two kinds of verses (question and answer). The verses are not actually related to the prose. The first line of prose acts as introduction about a Brahman approaching the Buddha, while the two following verses were about the Brahman asking questions and the Buddha’s answer. Fourth, there is no verse to be found in the sūtras of SĀ-U and SĀ-G. Examples of this situation can be seen in SĀ-U 032 together with SĀ-G 1093, SĀ-U 035 together with SĀ-G 1106, and SĀ-U 329 together with SĀ-G 1130.

It is difficult to understand why these four different types of verses are found in the Eight Assemblies section. The first and second conform to the definition of geya in the Āpídámó dàpípóshā lùn 阿毗達磨大毗婆沙論 (Abhidharma Mahāvibhāṣā-śastra) in that they correspond to the preceding prose. Amongst the four different types of verses, the first two have the largest proportion in the Eight Assemblies section. Also, the Chinese Saṃyuktāgama and Āpídámó dàpípóshā lùn 阿毗達磨大毗婆沙論 are both considered to belong to the (Mūla)sarvāstivāda school.202 Therefore, as the geya definition might be derived from the Eight Assemblies section, so also the Āpídámó dàpípóshā lùn 阿 毗 達 磨大 毗 婆 沙 論 indicates that gaya verses should correspond to the preceding prose, just as Buddhaghosa considered geya to be “sūtra with verse” for the Sagāthavagga. The literary style of the third type of verse is very common in the Sagāthavagga of the Saṃyuttanikāya. If the sūtras are organized by group as per Table 2.2, it is apparent that this style occurs most commonly in a part of the Eight Assemblies section named Zhūtiān 諸 天 (the gods), which corresponds to the Devatā and Devaputta saṃyuttas of Saṃyuttanikāya.203 A possible explanation for the absence of the verse (the fourth type) is


202 Bingenheimer (2011) and Bucknell (2011) indicate that SĀ-U and SĀ-G are from the (Mūla-)Sarvāstivāda school.

203 The reason for this situation is still in question. The correspondence can be seen as the following table. (Cf. Bucknell 2007)

Assemblies from SĀ Saṃyuttas from SN

1. Khattiya: Chàlì 剎利 Kosala
2. Brāhmaṇa: Póluómén 婆羅門 Brāhmaṇa
3. Gahapati: Póqíshě 婆耆舍 Vaṅgīsa
4. Samaṇa: Bǐqiū 比丘, Bǐqiūní 比丘尼 Bhikkhu, Bhikkhunī
5. Cātummahārājika: Zhūtiān 諸天, lín 林, yèchā 夜叉 Devatā, Devaputta, Vana, Yakkha
6. Tāvatiṃsa: Dìshì 帝釋 Sakka
7. Māra: Mó 魔 Māra
8. Brahmā: Fàntiān 梵天 Brahmā
 

the imperfect nature of the transmission of Buddhist collections from generation to generation.204 By looking at the relative proportions, the function of verses in the jiéjí pǐn 結 集 品 (the Eight Assemblies section) can overall be regarded as “responding and summarizing.” This function can also connect with the literal meaning of the term jiéjí pǐn 結 集 品 . In my opinion, “responding and summarizing” could be regarded as an instance of

“gathering the meaning” of the previous prose. Therefore, the function described by jiéjí 結集, with respect to the terms jiéjí wén 結集文 and jiéjí pǐn 結集品might literally be considered as “gathering.” Nevertheless, the practical functions of the jiéjí wén 結集文 and jiéjí pǐn 結集品 are different. The verses of the jiéjí wén 結集文function to “gather” (i.e. “collect”) sūtras together [as uddānas] while the verses of the jiéjí pǐn 結集品 “gather” (i.e. “summarize”) the meanings of sūtras.

“Gathering” was considered to be the initial function of geya in aṅga classification;
however, Yìnshùn (1971: 518) said that this function was gradually ignored as time went by.205 Instead, “repeating” or “summarizing” become the widespread meaning of geya, as is evidenced by the fact that it is found in most texts. The geya definition of the (Mūla)sarvāstivādins seems to have the idea that verses should be in correspondence with the prose (sūtra) or have a connection with the prose.206 This is also the significant difference between geya and gāthā. Gāthā can independently appear in the text, while geya always remain connected with section(s) of prose.

2.2.2. The source of the Group A verses in the Śarīrārthagāthā

As mentioned in chapter one, the verses in Group A were all cited from the Saṃyuktāgama. This section will discuss from which sūtras of the Saṃyuktāgama the Śarīrārthagāthā verses were quoted. There will also be a further analysis of the

204 Cf. Bucknell 2007.
205 Maybe this is why Jayawickrama (1959: 12) said, “it is quite possible that the true significance of this term (geya) too has been lost.” Besides, Yìnshùn (1971: 518) stated that these geya verses are close to those known in general society, but not connected with afflictions (kleśa), according to the geya definition of the Chéngshí lùn 成 實 論 (Satyasiddhiśāstra), as mentioned in § 2.1.4. Therefore, this is not exclusive.

206 The ancient meanings of geya in India can still be seen in the Āpídámó dàpípóshā lùn 阿毗達磨大毗婆

沙論, Yúgāshīdì lùn 瑜伽師地論 (Yogācārabhūmi), Chéngshí lùn 成實論 (Satyasiddhi-śāstra), and so on, particularly the idea that geya was considered as gāthā and the special usage of geya. These texts are affiliated with the (Mūla-)Sarvāstivāda school.
 
relationship between the Śarīrārthagāthā and Saṃyuktāgama, and the reason why the author(s) of the Yogācārabhūmi chose to quote these verses.
There are four scholars—Ui (1958), Enomoto (1989), Zōngzhèng (2005), and Chén 2006—who have found the specific sūtras to which the Śarīrārthagāthā verses correspond. The detail of their findings can be seen in Table 2.3. The verses of Group A are divided into fourteen sets. The differences between these scholars’ findings are as follows:


1. the first set (è 惡, pāpa): Ui mentioned only SĀ-G 1270
2. the sixth set (lèi 類 , abhivarṇṇatā): Zōngzhèng mentioned SĀ-G 597+SĀ-G 1315
3. the eighth set (chí 池, sara): Ui mentioned SĀ-G 595, not SĀ-G 601
4. the ninth set (liú 流 , ogha): Ui mentioned only SĀ-G 603, and Chén combined the ninth and tenth set and did not mention SĀ-G 603
5. the tenth set (liú 流 , ogha): all are different, all scholars mentioned SĀ-G 1269, SĀ-G 1326, and SĀ-G 1329.
6. the fourteenth set (dé yì 得義, arthaprāpti): Ui mentioned SĀ-G 549.


These differences need to be examined with respect to the verses’ content. Ui’s study was done in 1958, so the materials and tools he could use were limited, thereby possibly restricting the accuracy of his study. The first, ninth, and tenth sets of Ui’s study represent the content of only one sūtra each, but in the case of the eighth and thirteenth sets, the corresponding sūtras are not correct because the contents of SĀ-G 959 and SĀ- G 549 do not totally match the

verses. The sixth set of Zōngzhèng’s study includes the extra sūtra SĀ-G 1315. This is because SĀ-G 597 in the sixth set lacks half a verse — 住
此於此學，不懼後世死 (atra sthitaḥ atra ca śikṣamāṇo, nāyaṃ martyaḥ paralokādbibheti), so Zōngzhèng added the similar account (住彼學彼法，則無他世畏) from SĀ- G 1315 into SĀ-G 597. In the content of this verse of the sixth set, the objects of learning are jiè 戒 (morality, śīla), dìng 定 (meditation, samādhi), and huì 慧 (wisdom, prajñā), but SĀ-G 1315 indicates xìn 信 (faith), huì 惠 (wisdom), cái 財 (wealth), and fǎshī 法 施(gift of [[[teaching]]] dharma). They are different, so it is not necessary to add SĀ-G 1315 after SĀ-G 597. The ninth and tenth are from the same topic (liú 流 , ogha), so Chén combined them as a group. However, the contents of these two sets are distinct. It is
 
better to separate them to find the corresponding sūtras, which was done in the studies of Enomoto and Zōngzhèng. As for the tenth set, Enomoto does not mention SĀ-G 1316, but it is indeed related. Overall, the studies of Enomoto and Zōngzhèng are more accurate and complete. The precise correspondence of the verse sets with sūtras can be seen in the following tabulation:


Title corresponding sūtra
1 è 惡 (pāpa) SĀ-G 1270–1274
2 shuō 說 (ākhyeya) SĀ-G 1078
3 tān 貪 (rāga) SĀ-G 1214
4 liú 流 (ogha) SĀ-G 1092
5 bù 怖 (utrasta) SĀ-G 596
6 lèi 類 (abhivarṇṇatā) SĀ-G 597
7 yù 譽 (yaśasvin) SĀ-G 1282
8 chí 池 (sara) SĀ-G 601
SĀ-G 1329

9 liú 流 (ogha) SĀ-G 603
SĀ-G 1326
SĀ-G 1329


10 liú 流 (ogha) SĀ-G 1269
SĀ-G 1316
SĀ-G 1326
SĀ-G 1329
11 tān 貪 (rāgadveṣa) SĀ-G 1314
SĀ-G 1324
12 zuò 作 (kāryatā) SĀ-G 1311
13 qú láo 劬勞 (āyūhana) SĀ-G 1267
14 dé yì 得義 (arthaprāpti) SĀ-G 1092

Some sets correspond to several sūtras in the Saṃyuktāgama, because the verses of these sūtras are identical. Moreover, these corresponding sūtras all appear in the Eight Assemblies section. The Eight Assemblies section has been identified as an example of
geya. Therefore, the verses of Group A are taken to be geya verses.

If these corresponding sūtras were compared as they appear in the Biéyì zá āhánjīng 別譯雜阿含經 (SĀ-U), Eight Assemblies section of the Saṃyuktāgama, and Saṃyuttanikāya, more information might be found. SĀ-U (T. 100) is usually been taken to have the correct order of sūtras in SĀ-G, especially those in the geya parts.207 The Eight Assemblies section and Saṃyuttanikāya could provide information about the specific group to which the verses belong. Table 2.4 shows three facts:

1. the correspondence between the Śarīrārthagāthā verses and SĀ-G or SĀ-U sūtras reflects that the sequential arrangement has not been maintained.
2. the verses of Group A in the Śarīrārthagāthā do not include all the verses (geya part) of SĀ-G or SĀ-U.
3. large parts of the Śarīrārthagāthā verses (10 of 14 sets) particularly come from the assembly of the Cātummahārājika, or the Saṃyuttas of Devatā and Yakkha.

From the first two points, one thing which can be certain is that the relationship between the Group A of the Śarīrārthagāthā and Saṃyuktāgama (geya part) is not like that of the Vastusaṃgrahaṇī and Saṃyuktāgama (sūtra part).208 Group A of the Śarīrārthagāthā is not a mātṛka which sequentially comments on the geya part of Saṃyuktāgama.

Why are the verses of Group A mostly from the Cāturmahārājika assembly or the Devatā and Yakkha Saṃyuttas? The Eight Assemblies represent those groups for whom the dharma is spoken.209 In other words, these are the eight classes of listeners taught by the Buddha. Therefore, a possible answer is that the author(s) of the Yogācārabhūmi intended to emphasize the fact that the Buddha has educated the deity (Devatā) and yakṣa (Yakkha).210


207 See Mayeda 1964: 654–655, Yìnshùn 1983: b23–24, Bingenheimer 2011 and Bucknell 2011.
208 Lǚ (1924) and Yìnshùn (1983) gave evidence that the Vastusaṃgrahaṇī is the mātṛka of the sūtra-aṅga part of the Saṃyuktāgama. Both of them use the Vastusaṃgrahaṇī to correct the order of sūtras in the Zá āhánjīng 雜阿含經 (T.99).

209 The eight classes are Kṣatriya (chàlì 剎利), Brāhmaṇa (póluómén 婆羅門), Gṛhapati, (póqíshě 婆耆舍),
Śramaṇa (bǐqiū 比丘, bǐqiūní 比丘尼), Cāturmahārājika (zhūtiān 諸天, lín 林, yèchā 夜叉), Trāyastriṃśa
(dìshì 帝釋), Māra (Mó 魔), and Brahmā (fàntiān 梵天).
210 It is possible that belief in deities and yakṣas had become a trend at the time in India. The author may have wanted to use Buddhist ideas to change the cultural perception of such beings and make Buddhism more popular. Cf. Yìnshùn 1971: 732.
 
In brief, the verses of Group A have two special features. Firstly, due to the fact that verses are all quoted from the Eight Assemblies of the Saṃyuktāgama, the verses are considered as geya. Secondly, most verses of Group A were chosen from the saṃyuttas of the Devatā and Yakṣa in the Saṃyuktāgama.

2.2.3. Summary

It has been shown that the verses of Group A are all quoted from the Eight Assemblies section of the Saṃyuktāgama, which is also known as jiéjí pǐn 結 集 品 . Therefore, due to the function of geya, the attribute of the verses of Group A is “gathering [the meanings of the prose (sūtra)].” In addition, Group A of the Śarīrārthagāthā is not like a mātr̥kā on the Eight Assemblies section of the Saṃyuktāgama. The connection found between them is that the verses of both texts are geya. Maybe this is the reason why the author(s) of the Yogācārabhūmi particularly chose the verses from the Eight Assemblies of the Saṃyuktāgama. The verses of Group A are mostly from the saṃyuttas of the Devatā and Yakkha.

2.3. The third aṅga, vyākaraṇa, and the verses of Group B of the
Śarīrārthagāthā

This section will investigate the meanings of vyākaraṇa in various texts. At the same time, the development of vyākaraṇa will also be discussed. Second, the specific sūtras of the Udānavarga, Pārāyaṇa, Arthavargīya and Madhyamāgama, which Group B of the Śarīrārthagāthā verses correspond to, will be ascertained.
Finally, the relationship between the meaning of vyākaraṇa and Group B of the Śarīrārthagāthā will be analyzed. The reason why the verses of Group B were chosen in the Śarīrārthagāthā will additionally be discussed.
 
2.3.1. The development of vyākaraṇa

The third aṅga in the nine-aṅga classification is vyākaraṇa (separation, distinction, explanation, etc.).211 The word vyākaraṇa is employed in Sanskrit and Pāli.212 However, in the Pāli tradition, the third aṅga is veyyākaraṇa. Mayeda (1964: 281) mentioned that veyyākaraṇa is close to the Buddhist Hybrid Sanskrit term vai(y)yākaraṇa.213 Both terms vyākaraṇa and veyyākaraṇa derive from the verb vyākaroti (the root √kr̥ with the two prefixes vi and ā), meaning “to undo, sever, divide, separate from,”214 so these two nouns generally share the similar meanings, such as explanation and exposition. However, Yìnshùn (1971: 528) said that, strictly speaking, the applications of the two terms are different. If vyākaraṇa already existed within Pāli,215 why did the Pāli tradition instead favour veyyākaraṇa216 for aṅga classification? One possibility is that vyākaraṇa is an early term which shares the general meanings (explanation, answer, or distinction) and was broadly used at that early time, while the term veyyākaraṇa, which is applied in later Pāli, emphasizes the Buddhist concept. This concept concerns the achievement of the saints or rebirth of the disciples. It is a special usage which turned out to be more significant in later Buddhism. The Pāli tradition, therefore, may have adopted the term veyyākaraṇa instead of vyākaraṇa for aṅga classification (Yìnshùn 1971: 528).
In this case, veyyākaraṇa could be considered as “revelation” which involves the
achievement of a person in the present or future as well as the karma of the past (Yìnshùn 1971: 528). This means that someone could reveal the mysterious, occult and unmanifest occurrence by distinct explanations. The term veyyākaraṇa practically extends the usage of vyākaraṇa. Thus, in Pāli texts, the term veyyākaraṇa has a broader usage.
The meanings of vyākaraṇa in Buddhist texts are various. The definition of the third aṅga vyākaraṇa (veyyākaraṇa) by Buddhaghosa is most difficult to understand. The description states:


211 See MW s.v. vyākaraṇa.
212 See PED s.v. vyākaraṇa; veyyākaraṇa. Cf. Mayeda 1964: 281.
213 See BHS s.v. vaiyākaraṇa. PED (s.v. veyy°) states that veyy° is a diaeretic form of vy°.
214 See MW s.v. vyākr̥; vyākaroti.
215 Sn 19. 8, Sn 10. 1. 7.
216 The prefix veyy° is usually used for abstract nouns. See PED s.v. veyy°.
 
sakalam pi abhidhammapiṭakaṃ, niggāthakaṃ suttañ ca, yañ ca aññam pi aṭṭhahi aṅgehi asaṅgahitaṃ buddhavacanaṃ, taṃ veyyākaraṇaṃ ti veditabbaṃ.
The whole of the Abhidhamma piṭaka, suttas without verses and any other sayings of the Buddha which is not included in the other eight aṅga, should be regarded as the veyyākaraṇa.

The range of veyyākaraṇa from Buddhaghosa’s definition seems quite broad. Jayawickrama (1959) mentioned that from his description, Buddhaghosa was not really familiar with the content of veyyākaraṇa.217 Mayeda (1964: 294) said that this definition is not precise enough to understand what specific texts belong to veyyākaraṇa. Therefore, even though some scholars have attempted to find the Nikāya texts of veyyākaraṇa with this definition, the results are still controversial. For example, Hayashiya (1928) took “explanation” as the meaning of veyyākaraṇa, and then indicated that the texts in the Saṃyuttanikāya are those not included the Sagāthavaggga and the Nidānasaṃyutta.218 However, Mayeda (1964: 296) pointed out that because scholars have surveyed the texts mentioned by Buddhaghosa using a meaning of “vyākaraṇa” from other traditions,219 the results of these studies cannot be reliable.
Mayeda (1964: 296–306) identified the meaning of veyyākaraṇa by using the basic meanings of the verb vyākaroti/byākaroti. There are two distinct meanings in Nikāya texts, namely, an answer to a question and a declaration (jì bié 記 別 )220. The first, an answer to a question, is most common and important meaning of veyyākaraṇa. This meaning in the broad sense can also be regarded as “explanation.” The second, “declaration”, may be divided into four types, as follows:221

217 The phrase “not included in the other eight aṅgas” is not a characteristic thing for Buddhaghosa to state. See Jayawickrama 1959.
218 Another scholar, Ui (1925: 159), also used the meaning “explanation” to examine the texts of vyākaraṇa. He roughly indicated that the first eight sections of Nidānsaṃyutta of the Saṃyuttanikāya belong to vyākaraṇa.
219 Some scholars, such as Mino (1926: 24), Akanuma (1939: 168), Ogiwara (1869–1937: 405), and Hikata (1935–1941), have taken “answer to a question” as the meaning of vyākaraṇa. But, the results of their studies are divergent.
220 “Declaration” to Jì bié 記別 is the preferred definition of veyyākaraṇa by Analāyo (2009). Basically, jì bié 記別 is considered as “prediction or prophecy” on the context of most later Buddhist texts. However, according to Mayeda’s opinion on the definition of Jì bié 記 別 , “declaration” for Jì bié 記 別 is more suitable which could involve a broader meaning.
221 The first two emphasize the difference between other people and self; and the last two lay stress on the difference of time between the present and future.
 
1. Declaration of others (particularly the Buddha and his disciples): An example can be seen in the Janavasabhasutta (DN 18) which mentions that the Buddha indicated where a disciple is going to when he passes away. The Mahāparinibbānasutta (DN 16) mentions the disciples’ future circumstances after dying.
2. Declaration of self: a person can announce his own future. An example can also be seen in the Mahāparinibbānasutta (DN 16)222, which shows that the Buddha told Ānanda that a disciple can declare his own achievement.
3. Declaration of the future: Due to personal enlightenment, one can predict where one is going to be after passing away. An example is the Naḷakapānasutta (MN 68), which contains a story in which the Buddha indicated where a disciple is going to be reborn after death. However, Mayeda (1964: 299) said that the previous life of a person could also be indicated sometimes.
4. Declaration of a present achievement: Mayeda (1964: 303) mentioned that in early Buddhism, the highest ideal of practice was to reach liberation in the present life, rather than the achievement in a future life. A good example is a common statement found in Pāli texts, “destroyed is birth, the holy life has been lived, what had to be done has been done, there is no more for this state of being.”223
Differently, Anālayo (2009) took the words veyyākaraṇa and vyākaraṇa to examine its practical meaning in Nikāya texts. He concluded that the meanings were replies, expositions, explanations and declarations. “Replies” appear in response to question(s),
while “expositions and explanations” need not be prompted by a question. “Declarations” are about the Buddha’s future circumstances or disciples’ attainment; or by a retired
ruler to his son about how to manage the country; or by a god revealing his identity.
The traditional descriptions of vyākaraṇa in other texts could be divided into three types (Mayeda 1964: 282). The first is “answer to a question.” The account of the





222 anacchariyaṃ kho pan' etaṃ ānanda yaṃ manussabhūto kālaṃ kareyya, tasmiṃ tasmiṃ ce kālakate tathāgataṃ upasaṃkamitvā etam atthaṃ pucchissatha, vihesā v' esā ānanda tathāgatassa. tasmāt ih' ānanda dhammādāsaṃ nāma dhammapariyāyaṃ desessāmi yena samannāgato ariyasāvako ākaṅkhamāno attanā va attānaṃ vyākareyya : khīṇanirayo 'mhi khīṇatiracchānayoni khīṇapettivisayo khīṇāpāyaduggativinipāto, sotāpanno 'ham asmi avinipātadhammo niyato sambodhiparāyaṇo ’ti (DN II 93.11–20).
223 khīṇā jāti vusitaṃ brahmacariyaṃ, kataṃ karaṇīyaṃ nāparaṃ itthattāyā ti abbhaññāsi (SN 6.1.3 at SN II 140.27–30).
 
Āpídámó dàpípóshā lùn 阿毗達磨大毗婆沙論 (Abhidharma Mahāvibhāṣāśastra) describes:224

「記說云何？謂諸經中，諸弟子問，如來記說；或如來問，弟子記說；或弟子問，弟子記說；化諸天等，問記亦然。若諸經中，四種問記；若記所證所生處等。
What is vyākaraṇa? It is those sūtras in which the Tathāgata answers the disciples' question(s), or the disciples answer the Tathāgata’s question(s); the Tathāgata answers the question(s) of gods, or gods answer the Tathāgata’s question(s). These are the four types of “answer to a question” in sūtras. The reply of “answer to a question” is where to be achieved or reborn.

This passage shows three aspects of the meaning “answer to a question” of vyākaraṇa— the persons who questions and answers, the methods, and content of “answer to a question.” The persons are the Tathāgata, disciples, and gods. The four types of answer to a question (catvāri praśna-vyākaraṇāni) are:225 a confirmatory reply to a question (一
向記, ekāṃśavyākaraṇaṃ), discriminating answer (分別記, vibhajyavyākaraṇaṃ),
questioning in return ( 反 詰 記 , paripṛcchāvyākaraṇaṃ), silence for answer ( 捨 置 記 , sthāpanīyavyākaraṇaṃ). The content of conversation is where a saint will gain realisation or where a person will be reborn in the next life. From this description of vyākaraṇa, Mayeda (1964: 282) also took the content of the answer to the question(s) as another meaning of vyākaraṇa, namely, “declaration” or “prediction” (shòu jì 授 記 ), which I will discuss later.
Another definition of vyākaraṇa can be seen in the Yúqié shīdì lùn 瑜伽師地論
(Yogācārabhūmi). There are two accounts:


記別者？謂廣分別略所摽義，及記命過弟子生處。 (T1579, 30.753a15–16)





224 T1545, 27.659c28–660a2. Also see Chéngshí lùn 成 實 論 (Satyasiddhi-śāstra), T1646, 32.244c29;
Shùnzhènglĭ lùn 順正理論 (Abhidharma Nyāyānusāra-śāstra), T1562, 29.595a8;
225 These four ways are commonly seen in early texts, such as MĀ 119 and AN I 197.
 
What is vyākaraṇa? This is a detailed explanation (guǎng fènbié 廣 分 別 ) for a brief connotation, and the account about where disciples who passed away will be reborn.

云何記別？謂於是中，記別弟子命過已後當生等事，或復宣說已了義經。
(T1579, 30.418c7–9)
What is vyākaraṇa? This is an account about where disciples will be reborn after dying, or [an account] is taken as a fully explained discourse (leyì jīng 了義經).

From the above descriptions,226 we see that there are two meanings of vyākaraṇa. One is the detailed explanation or fully explained discourse.227 This meaning contrasts with the definition of geya in which the verses of geya are unexplained discourse.228 The meaning of geya is the opposite to that of vyākaraṇa; geya verses need to be explained. Therefore, the function of vyākaraṇa is to elucidate the content of geya verses which are regarded as succinct sayings (luèsuǒbiāo yì 略 所 摽 義 ). This is why vyākaraṇa stands for detailed explanation. Evidence may be found in the Āgamas.229 For example, there are four kinds of detailed explanations in the Saṃyuktāgama. The first is related to the Bōluóyánnà 波
羅延那 (Pārāyaṇa). The specific texts are SĀ-G 345,230 SĀ-G 982,231 SĀ-G 983232 and
SĀ-G 1164.233 The second is related to the Yì pǐn 義 品 (Aṭṭhakavaggika). The specific text is SĀ-G 551.234 The third is related to the Yōutuónà 優 陀 那 (Udāna), of which the texts are SĀ-G 64235 and SĀ-G 566.236 The fourth is the Bāzhòng sòng 八 眾 誦(Saṃgītavarga) which can be found in SĀ-G 549. In the Chinese Madhyamāgama



226 A similar account is in the Xiănyáng shèngjiāo lùn 顯 揚 聖 教 論 (Prakaranāryavācaśāstra), T1602, 31.509a1–2.
227 Also see Chéngshí lùn 成實論 (Satyasiddhiśāstra), T1646, 32.244c27–28.
228 應頌者。謂長行後宣說伽他。又略摽所說不了義經。 See Yúqiéshīdì lùn 瑜伽師地論(Yogācārabhūmi), T1579, 753a13–14.
229 Cf. Yìnshùn 1971: 521.
230 The counterpart in the Nikāyas is SN V. 2 (Ajitapañha).
231 The counterpart in the Nikāyas is SN V. 4 (Puṇṇakapañha).
232 The counterpart in the Nikāyas is SN V. 14 (Udāya).
233 The counterpart in the Nikāyas is SN V. 3 (Metteyyapañha). 234 The counterpart in the Nikāyas is SN 884 (Māgandiyapañha). 235 The counterpart in the Nikāyas is SN 22. 55 (Udānasutta).
236 The counterpart in the Nikāyas is SN 41. 5 (Paṭhamakāmabhūsutta).
 
(Zhōng āhánjīng 中阿含經), the texts are sūtras 165–167.237 In the above texts, the verses (geya) are interpreted or explaned. In other words, the verses condense ideas or feelings in limited stanzas, in such a way that they cannot always be fully understood by the reader. Hence, the function of vyākaraṇa is to make the geya verses comprehensible.
Another meaning of vyākaraṇa is an account or record of where disciples will be reborn in the next life. This is similar to the content of “the answer to the question(s)” in the Āpídámó dàpípóshā lùn 阿毗達磨大毗婆沙論. The account of the Yúqié shīdì lùn 瑜伽 師 地 論 (Yogācārabhūmi) mentions only the rebirth of disciples, while the Āpídámó dàpípóshā lùn 阿毗達磨大毗婆沙論 mentions the achievements of saints and the rebirth of disciples. In brief, these accounts are about the future of people after death. In a broad sense, this is another kind of explanation or answer to a question. The Buddha explains where the disciples are going to in the future.
Overall, the above three meanings of vyākaraṇa suits the literal meaning of vyākaraṇa (vi+ ā+ √kr̥), which is to make clear the teachings of the Buddha, or the future situation of his disciples. Yìnshùn (1971: 525) further indicated that the first two meanings emphasize the mode of vyākaraṇa, which is the general meaning; and the meaning specially focuses on the content of vyākaraṇa. The relationship of the three meanings is shown in the following chart:


answer to question(s)
mode (general)
 
vyākaraṇa detailed explanation
content (special) achievements of saints and rebirth of
disciples


As mentioned in § 2.1.2, there is a collection of vyākaraṇa texts spoken by the Tathāgata and his disciples in the Saṃyuktāgama. The three meanings of vyākaraṇa can also be seen in this section of the Saṃyuktāgama, although the definition of the Āpídámó dàpípóshā lùn 阿毗達磨大毗婆沙論 lays particular stress on the meaning “answer to a question” and the achievements of the saints and rebirth of the disciples, while the


237 The counterpart in the Nikāyas is MN 133, 134, 132.
 
account of the Yúqié shīdì lùn 瑜伽師地論 (Yogācārabhūmi) highlights the meaning of the detailed explanation and rebirth of the disciples.
Yìnshùn (1971: 525) commented that the three meanings of vyākaraṇa have changed over time. The contents of the first two meanings of vyākaraṇa have further developed and become longer in the Madhyamāgama/Majjhimanikāya and Dīrghāgama/Dīghanikāya, such as MN 109 (Mahāpuṇṇamasuttaṃ) and DN 1, 21 (Brahmājala Sutta and Sakkapabha Sutta).238 Yìnshùn (1971: 525) additionally commented that with the passage of time, the first two became the aṅga “vaipulya” in the twelve-aṅga classification (dvādaśa-anga).
The third meaning, stressing the content of vyākaraṇa, has also evolved over time. “Prediction” or “prophecy” became its dominant meaning in later Buddhism. For instance, Gautama Buddha’s prediction that Ajita will become Maitreya Buddha (Mílè 彌勒) in his next life is described in the Dà bōnièpán jīng 大般涅槃經(Mahāparinirvānasūtra) as vyākaraṇa.239 Mayeda (1964: 303) also commented that the meaning “prediction” focuses on future circumstances and does not relate to the present
achievements of a person.240 The reason why “prediction” could suit the literal meaning of vyākaraṇa is that “prediction” explains or reveals mysterious and subtle meanings.
This is also described in the Dàhéng āpídámó zájí lùn 大乘阿毘達磨雜集論
(Mahāyānābhidharma Samuccaya-vyākhyā).241
In sum, all three meanings of vyākaraṇa are often found in the texts. The section spoken by the Tathāgata and his disciples in the Saṃyuktāgama is considered to be vyākaraṇa. This section is found after of the sections of sūtra and geya (Yìnshùn 1971: 527). Furthermore, Yìnshùn (1971: 527) stated that although the verses in the Eight Assemblies section of the Saṃyuktāgama are regarded as geya, the verses which were not incorporated into texts at the first Buddhist Council are also named geya in a broad sense. The function of vyākaraṇa is to elucidate the content of these verses, which existed at the time but were not incorporated into Buddhist collections. Also, the modes

238 See Yìnshùn 1971: 524–525. Cf. Shiio 1935: 27.
239何等名為授記經。如有經律如來說時。為諸天人授佛記別汝阿逸多。未來有王名曰蠰佉。當於是世而
成佛道。號曰彌勒。(T357, 12.693c6–9) Another example of a prediction being labelled vyākaraṇa is in the Dàzhìdù lùn 大 智 度 論 (Mahāprajñāpāramitāśāstra) (T1509, 25.306c25–307a18). Cf. Mayeda 1964: 283–284.
240 The example is the Pāthikasutta (DN 24.3).
241 記別開示深密意故 (T1606, 31.743c6–8).
 
of vyākaraṇa (the first two meanings) are “the answer to question(s)” and “the detailed explanation.” The special application of vyākaraṇa (the last meaning), which can be considered as “declaration,” concerns where a saint will gain realisation or where a person will be reborn to in the next life. Due to the fact that the special application has gradually become more highly valued, the meaning of vyākaraṇa has gradually changed into “prediction”, which reveals mysterious and subtle meanings.

2.3.2. The source of Group B verses in the Śarīrārthagāthā

As mentioned in chapter one, the verses in Group B were cited from the Udānavarga,
Pārāyaṇa, Arthavargīya and Madhyamāgama. This section will show from which specific sūtras the Śarīrārthagāthā verses were quoted. Further analysis of the relationship between the Śarīrārthagāthā and these texts will be undertaken.
As mentioned in § 2.2.2, there are four scholars who have already found the specific sūtras to which the Śarīrārthagāthā verses correspond. The detail of their findings can be seen in Table 2.5.242 The verses of Group B are divided into twenty seven sets with their titles, which were translated from the Tibetan version.243 As is evident from Table 2.5, the results of these four scholars are largely divergent. The reason being that, unlike Enomoto (1989), Ui (1958), Zōngzhèng (2005), and Chén (2006) did not examine the Sanskrit version of the Udānavarga (edited by Schmithausen (1970)).
The verses in Group B can also be separated into two types. The first type involves the verses (the first twenty four sets) all from the Udānavarga (Uv-S),244 according to Enomoto (1989). The verse content of the Śarīrārthagāthā and Udānavarga are exactly the same. However, the findings of Ui (1958), Zōngzhèng (2005), and Chén (2006) are generally based on texts relevant to the Udānavarga (Dharmapada, Fǎjù jīng 法句經,



242 Dharmapada, Suttanipāta, Udānavarga, Fǎjù jīng 法句經 (T210), Fǎjù pìyù jīng 法句譬喻經 (T211), Chūyào jīng 出 曜 經 (T212), Fǎjí yàosòng jīng 法 集 要 頌 經 (T213) are abbreviated as “Dhp,” “Sn,” “Uv,” “FJ,” “FJP,” “CY,” and “FJY.”
243 As mentioned in chapter one, the titles of the 27 sets of Group B appear only in the Tibetan
Yogācārabhūmi. The title of Group B is called upadeśa.
244 This version was found and edited by Schmithausen (1970); it is not the version edited by Bernhard (1965). There are two kinds of Sanskrit Udānavarga, namely, the versions of the Mūlasarvāstivādins and Sarvāstivādins. The one edited by Schmithausen (1970) is from the Mūlasarvāstivāda school.
 
Fǎjù pìyù jīng 法句譬喻經, Chūyào jīng 出曜經, Fǎjí yàosòng jīng 法集要頌經).245 Therefore, the study of Enomoto (1989) is more accurate than the studies of the other three scholars. The following table shows the specific sūtras of the Udānavarga from which the Śarīrārthagāthā verses are quoted.


title of verses sūtra
1 impermanence Uv-S 1.3
2 carefulness Uv-S 4.1
3 strong lust Uv-S 3.1
4 dwelling in the dharma Uv-S 5.24
5 contemplating the truth Uv-S 9.6
6 good speech Uv-S 8.11
7 faith Uv-S 10.1
8 learning Uv-S 12.6
9 being like the sky Uv-S 17.12
10 physical appearance Uv-S 22.12–16
11 king Uv-S 16.22
12 city of bone Uv-S 16.23
13 turtle Uv-S 26.1
14 equality Uv-S 26.30
15 mud Uv-S 29.37
16 two dwellings Uv-S 29.38
17 Uv-S 29.51
18 having cleaning up Uv-S 29.56
19 giving Uv-S 28.2
20 all evils Uv-S 28.1
21 falling down Uv-S 31.1
22 skilled Uv-S 31.51
23 craft Uv-S 32.5
24 travelling far away Uv-S 33.55



245 The Fǎjù jīng 法 句 經 and Fǎjù pìyù jīng 法 句 譬 喻 經 are close to the version of the Pāli Dhammapada, while the Chūyào jīng 出 曜 經 and Fǎjí yàosòng jīng 法 集 要 頌 經 are differet recensions of the Udānavarga.
 
From the above table, we see that the sequence of Śarīrārthagāthā verses are overall in accordance with the order of chapters in the Udānavarga, although a few sets have a different order.
Did the author(s) of the Yogācārabhūmi choose verses from each chapter of the Udānavarga? The distribution of verses in the Udānavarga is displayed in the following table:246


Chapter of
Udānavarga
English translation247 Number of verse sets in the
Śarīrārthagāthā
1 Anityavarga Impermanence 1
2 Kāmavarga Desire
3 Tṛṣṇāvarga Lust 3
4 Apramādavarga Purity 2
5 Priyavarga Agreeable Things 4
6 Śīlavarga Morality
7 Sucaritavarga Virtuous Conduct
8 Vācavarga Speech 6
9 Karmavarga Deeds 5
10 Śraddhāvarga Faith 7
11 Śramaṇavarga The Śramaṇa
12 Mārgavarga The Path
13 Satkāravarga Honor
14 Drohavarga Mischief
15 Smṛtivarga Reflection
16 Prakirṇakavarga Miscellaneous 11, 12
17 Udakavarga Water 9
18 Puṣpavarga The Flower
19 Aśvavarga The Horse
20 Krodhavarga Anger
21 Tathāgatavarga The Tathāgata
22 Śrutavarga The Hearer 8, 10
23 Ātmavarga Self
24 Peyālavarga Numbers

246 The titles of chapters are in accordance with Bernhard 1965. The version edited by Bernhard (1965) is considered to belong to the Sarvāstivādins. However, according to Schmithausen (1970), the chapter titles should be the same as those in the version of the Mūlasarvāstivādins.
247 Cf. Rockhill 1975.
 
25 Mitravarga Friendship
26 Nirvāṇavarga Nirvāṇa 13, 14
27 Paśyavarga Sight
28 Pāpavarga Sin 19, 20
29 Yugavarga Pair 15, 16, 17, 18
30 Sukhavarga Happiness
31 Cittavarga The Mind 21, 22
32 Bhikṣuvarga The Bhikṣu 23
33 Brāhmaṇavarga The Brāhmaṇa 24

We may conclude two main points from this table. First, the twenty ninth chapter, Yugavarga, has been quoted the most (four times) in the Śarīrārthagāthā. Second, not every chapter of the Udānavarga has been quoted in the Śarīrārthagāthā. Therefore, Group B of the Śarīrārthagāthā does not fully comment on the Udānavarga and the Śarīrārthagāthā is not the mātr̥kā of the Udānavarga.
As previously mentioned, the verses in Group B can be separated into two types. The first type has been discussed above. The second type constitutes the last three sets. The findings of the four scholars are similar but the situation is complicated, so these three sets will be analysed separately. The twenty fifth set is related to the Ajitapraśna of Pārāyaṇa (Bōluóyánnà 波羅延那). The results of the four scholars’ studies is as follows:

1 Ui Sn 1032–1039



2 Enomoto Sn 1032–1039, 1110f.; SNII, 47, 49; JA IV, 266; Nett10–17, 70f.; Peṭ13, 17,84f.; SHT VI, 198f.; SĀ TI 2, 95b 13f.; TI
25, 82c14f (Dàzhìdù lùn 大智度論/TI no.1509); Vi 1, TI 27, 230bf., 379b 11–15; Vi 2, TI 28, 285b13–17; Vi 3, TI 28,
454c27–455a2; Ibh 29.1.
3 Zōngzhèng Sn 1032–1039, 1110–1111
4 Chén Sn 1032–1039

Ui (1958), Zōngzhèng (2005), and Chén (2006) indicated that the verses correspond to
Suttanipāta 1032–1039,248 but the texts Enomoto (1989) listed are not corresponding


248 The verses of the Suttanipāta 1032–1039 are as follows:
kenassu nivuto loko, iccāyasmā Ajito kenassu nappakāsati | ki 'ssābhilepanaṃ brūsi, kiṃ su tassa mahabbhayaṃ ||
 
sūtras but parallel sūtras. This is because the verses of the Suttanipāta and the Śarīrārthagāthā are from different traditions.249 Moreover, if the verses of the Śarīrārthagāthā are examined with SĀ-G 245 (Ajitapraśna), the contents are different. Therefore, the study of Enomoto (1989) is correct that the corresponding sūtra has not yet been found.
The situation of the twenty sixth verse set is similar to the twenty fifth set. The results of the four scholars’ studies are as follows:

1 Ui Sn 766–771




2 Enomoto Sn 766–770; Dhp 250ab, 336cd; Th 401cd, 457; JA IV, 172; Nett 5f., 69; Peṭ 9, 45f.; Vism 576; Uv III. 10cd, X. 13ab; PSVy 619;
Abhidh-k-bh 9; Abhidh-k-T Tu 19b5–20a1; Chinese Arthavargīyasūtra, TI 4, 175c17–23; Vi 1, TI 27, 176a29–b3, 288b8f., 980c4f.; Vi 2, TI 28, 133c7f., 223a12f.; TI 29, 337b19f.
(Shùnzhènglǐ lùn 順正理論/TI no.1562)
3 Zōngzhèng Sn 766–771; Chinese Arthavargīyasūtra (T198)
4 Chén Sn 766–771; Chinese Arthavargīyasūtra (T198)

As for the twenty sixth verse set, Ui (1958), Zōngzhèng (2005), and Chén (2006) indicated that the verses correspond to Suttanipāta 766–771, while Enomoto (1989) instead listed Suttanipāta 766–770. Suttanipāta 766–771 is the entire Kāmasutta of the Pāli Suttanipāta.250 However, the last verse of the twenty sixth set is different from verse

avijjāya nivuto loko, Ajitā ti bhagavā vevicchā pamādā nappakāsati | jappābhilepanaṃ brūmi, dukkham assa mahabbhayaṃ ||
savanti sabbadhī sotā, iccāyasmā Ajito sotānaṃ kiṃ nivāraṇaṃ | sotānaṃ saṃvaraṃ brūhi, kena sotā pithiyyare ||
yāni sotāni lokasmiṃ, Ajitā ti bhagavā sati tesaṃ nivāraṇaṃ | sotānaṃ saṃvaraṃ brūmi, paññāy’ ete pithiyyare ||
paññā c' eva satī ca, iccāyasmā Ajito nāmarūpañ ca mārisa | etaṃ me puṭṭho pabrūhi, katth' etaṃ uparujjhati ||
yam etaṃ pañhaṃ apucchi, Ajita taṃ vadāmi te | yattha nāmañ ca rūpañ ca, asesaṃ uparujjhati | viññāṇassa nirodhena, etth' etaṃ uparujjhati ||
ye ca saṃkhātadhammāse, ye ca sekhā puthū idha | tesaṃ me nipako iriyaṃ, puṭṭho pabrūhi mārisa ||
kāmesu nābhigijjheyya, manasānāvilo siyā | kusalo sabbadhammānaṃ, sato bhikkhu paribbaje ti ||
249 Schmithausen (1970) and Enomoto (1989) indicated that the verses of the Śarīrārthagāthā belong to the Mūlasarvāstivāda school.
250 Sn 766–770: kāmaṃ kāmayamānassa, tassa ce taṃ samijjhati | addhā pītimano hoti, laddhā macco yad icchati ||
tassa ce kāmayānassa, chandajātassa jantuno | te kāmā parihāyanti, sallaviddho va ruppati ||
 
771 of the Suttanipāta.251 Therefore, Enomoto’s listing is more accurate. Besides, due to the fact that the Chinese Arthavargīyasūtra (T198) and Kāmasutta of Pāli Suttanipāta does not belong to Mūlasarvāstivāda school (Mizuno 1952 and Yìnshùn 1969: 101), the exact corresponding sūtra of the twenty sixth set has not yet been found.
The comparison of findings between the four scholars for the last set of Group B is displayed in the following table:


1 Ui SĀ-G 955
2 Enomoto MĀ 165–167
3 Zōngzhèng SĀ-G 955; close to SĀ-G 1072
4 Chén SĀ-G 955

Again, the results of Ui (1958), Zōngzhèng (2005), and Chén (2006) are different from that of Enomoto (1989). Ui (1958), Zōngzhèng (2005), and Chén (2006) pointed out that the corresponding sūtra is SĀ-G 995, whereas Enomoto (1989) instead listed MĀ 165–
167. The commentary on this verse set in the Śarīrārthagāthā mentions that this verse is known as Bhadraikarāgāthā (Zàoxiánshàn sòng 造賢善頌), which is MĀ 165.252 Perhaps Ui (1958), Zōngzhèng (2005), and Chén (2006) did not notice this statement in the commentary. Also, if the content of SĀ-G 995 is examined with the twenty seventh verse set, the meaning of SĀ-G 995 is not precisely the same as the twenty seventh verse set. Therefore, SĀ-G 995 should be regarded as the parallel sūtra.

yo kāme parivajjeti, sappasseva padā siro | so imaṃ visattikaṃ loke, sato samativattati ||
khettaṃ vatthuṃ hiraññaṃ vā, gavāssaṃ dāsaporisaṃ | thiyo bandhū puthu kāme, yo naro anugijjhati || abalā va naṃ balīyanti, maddante naṃ parissayā | tato naṃ dukkham anveti, nāvaṃ bhinnam ivodakaṃ || 771: tasmā jantu sadā sato, kāmāni parivajjaye | te pahāya tare oghaṃ, nāvaṃ siñcitvā pāragū ti ||
251 The verses in the twenty sixth set are (See Enomoto 1989):
kāman kāmāyamānasya tasya cet tat samr̥dhyati /
addhā prītamanā bhavati labdhvā martyo yad īpsitaṃ // 1
tasya cet kāmāyamānasya chandajātasya jaṃtunaḥ / te kāmāh parihīyaṃte śalyaviddha iva rūpyate // 2 yaḥ kāmāṃ parivarjayati sarppasyeva śirāt padaṃ / sa imāṃ viṣaktikāṃ loke smr̥taḥ samativarttate // 3 kśetravastuhiraṃnyaṃ ca gavāśvamaṇikuṇḍalaṃ /
striyo dāsān pr̥thakkāmān yo naro hy abhigr̥dhyati // 4
abalaṃ vā balīyāṃso mr̥dnaṃty enaṃ parisravāḥ /
tata enaṃ duḥkham anveti bhinnāṃ nāvaṃ ivodakaṃ // 5 yasya tv etat samucchinnaṃ tālamastakavad dhataṃ / śokās tasya nivarttante udabindur iva puśkarāt // 6
252 In chapter four, the corresponding sūtra is yet to be found, based on the comparison of verse content.
 
In sum, the verses of Group B are cited from different texts. All verses in the first type (24 sets) of Group B are cited from the Udānavarga. The corresponding sūtras of the twenty fifth, twenty sixth and twenty seventh verse set have not been found to date but they are related to the Ajitapraśna of Pārāyaṇa, the Kāmasūtra of the Arthavargīya, and MĀ 165. Therefore, the second type of Group B may simply quote from the Pārāyaṇa, Arthavargīya and Madhyamāgama.

2.3.3. Summary

The verses of Group B in the Śarīrārthagāthā can be found in the Udānavarga, Pārāyaṇa, Arthavargīya and Madhyamāgama. However, why did the author(s) choose the verses from a number of different texts? As mentioned in § 2.3.1, there are three meanings of vyākaraṇa. The texts which refer to the meaning “detailed explanation” are SĀ-G 345,253 SĀ-G 982,254 SĀ-G 983,255 SĀ-G 1164256 of the Bōluóyánnà 波羅延那
(Pārāyaṇa), SĀ-G 551257 of the Yì pǐn 義 品 (Arthavargiya), SĀ-G 64258 and SĀ-G 566259
of Yōutuónà 優陀那 (Udāna), and the Chinese Madhyamāgama 165–167260 (Zhōng āhánjīng 中阿含經). The specific correspondence between the corresponding texts of Group B of the Śarīrārthagāthā and the early texts is displayed in the following table.


Early texts referring to the meaning “detailed
explanation” The corresponding texts of Group B
Yōutuónà 優陀那 (Udāna) of the Saṃyuktāgama The first 24 sets from the Udānavarga
SĀ-G 345 of the Bōluóyánnà 波羅延那
(Pārāyaṇa) The 25th set related to the Ajitapraśna of
Pārāyaṇa
SĀ-G 551 of the Yì pǐn 義品 (Arthavargiya) The 26th set related to the Chinese
Arthavargīyasūtra (T198)
MĀ 165-167 The 27th set related to the MĀ 165


253 The counterpart of the Nikāya is SN V. 2 (Ajitapañha).
254 The counterpart of the Nikāya is SN V. 4 (Puṇṇakapañha).
255 The counterpart of the Nikāya is SN V. 14 (Udāya).
256 The counterpart of the Nikāya is SN V. 3 (Metteyyapañha). 257 The counterpart of the Nikāya is SN 884 (Māgandiyapañha). 258 The counterpart of the Nikāya is SN 22. 55 (Udānasutta).
259 The counterpart of the Nikāya is SN 41. 5 (Paṭhamakāmabhūsutta).
260 The counterpart of the Nikāya is MN 133, 134, 132.
 
The first twenty four sets of Group B were cited from the “Udānavarga.” Some verses of the Udāna may have existed very early before being collected into the Udānavarga text. Therefore, the verses may also have been seen in the Saṃyuktāgama and been explained. The other texts referring to the meaning “detailed explanation” in the table are SĀ-G 345 of the Pārāyaṇa, SĀ-G 551 of the Arthavargiya, and MĀ 165–167. These texts are also quite relevant to the corresponding sūtras of the twenty fifth, twenty sixth and twenty seventh verse sets, which are the Ajitapraśna of Pārāyaṇa, the Chinese Arthavargīyasūtra, and MĀ 165. Therefore, the verses of Group B are attributed to the vyākaraṇa aṅga. Maybe this is the reason the author(s) of the Yogācārabhūmi choose this kind of verse in the Śarīrārthagāthā.
From the definition of vyākaraṇa in the Yogācārabhūmi, we may conclude that the function of vyākaraṇa is to elucidate the content of geya verses. This is why vyākaraṇa stands for detailed explanation. Hence, the verses of Group B are also geya verses in a sense, but different from those verses in the Eight Assemblies section of the Saṃyuktāgama. As mentioned in chapter one, the title of Group B in the Śarīrārthagāthā uddana is Upadeśa. The literal meaning of upadeśa is “specification” (MW s.v. upadeśa ). From this, it is evident that the Group B verses of the Śarīrārthagāthā are quite close to the meaning of vyākaraṇa, the detailed explanation.

2.4. Conclusion

The first section of this chapter demonstrated that the first three aṅgas (sūtra, geya, and veyākaraṇa) constitute the Saṃyuktāgama. The verses in the Śarīrārthagāthā can be divided into two groups (Group A and B). The second section showed that the Group A verses are considered to be geya verse. Due to the function of geya, the attribute of the verses of Group A is possibly “gathering [the meanings of the prose (sūtra)].” The third section of this chapter demonstrated that the Group B verses belong to vyākaraṇa. The particular meaning of vyākaraṇa is “detailed explanation”, that is, the function of vyākaraṇa is to elucidate the content of geya verses. In other words, the Group A and B of the Śarīrārthagāthā verses are geya, but different types.
The division of the Śarīrārthagāthā verses in the Tibetan version is different from that of the Sanskrit and Chinese versions. In the Sanskrit and Chinese versions, twenty
 
seven verse sets in Group B have the title upadeśa (lùnyì 論 議 ), but in Tibetan version the verses are not in a single division called upadeśa. In the Tibetan version, the twenty seven verse sets are labelled with twenty six individual titles.261 My assumption is that in the Tibetan version the compiler(s) might have intended to emphasize the common feature of all Śarīrārthagāthā verses—geya.262
If the Śarīrārthagāthā verses are of the geya type, why does the title (Śarīrārtha- gāthā) indicate the verses are gāthā? In § 2.1.3, it was shown that gāthā is a general term for verse or Buddhist verse. Therefore, it is possible to indicate further the specific type of gāthā verses.
In brief, in the early time, there were two types of geya verses. One was like the verses in the Eight Assemblies section of the Saṃyuktāgama. Another was sung amongst ordinary people, and then later collected in texts such as the Udānavarga, Pārāyaṇa, and Arthavargīya. The second type of verses is often explained, so this type was taken into the vyākaraṇa aṅga. This suits one of the meanings of vyākaraṇa, the detailed explanation, and the literal meaning of the Group B title, upadeśa (specification). These two types of verses are equally important in the Śarīrārthagāthā, due to the fact that the proportion of the two types of verses is nearly half in the Sanskrit and Chinese versions.
From the above research, we can also find out that the author(s) of the Yogācārabhūmi valued the first three aṅgas, which are the foundation of the Saṃyuktāgama. According to the works of Yìnshùn (1983) and Mukai (1985), the Vastusaṃgrahaṇī (the last chapter of Yogācārabhūmi) not only is the mātr̥kā of some sections of the Saṃyuktāgama but also belongs to the sūtra-aṅga. In addition, in this chapter, the Śarīrārthagāthā verses may be attributed to geya-aṅga and veyākaraṇa-aṅga. In other words, the first three aṅgas are adopted in the Yogācārabhūmi. The author(s) or complier(s) knew the importance of these three aṅgas, which at a later time seems gradually to have been forgotten or developed with different definitions. This outcome also seems to support Yìnshùn’s idea that the first three aṅgas can be considered as an early part of the evolution of early Buddhist texts.



261 The sixteenth and seventeenth sets of Group B share the title “two dwellings” (gnas gnyis).
262 Although this assumption is dubious, another possibility is that the Tibetan version just follows the structure of the Śarīrārthagāthā verses. In chapter five, I will give evidence that the Tibetan uddāna might have been modified somehow in Tibet. This is the reason why the Tibetan uddāna is different.
 
Chapter three: The Group A verses

This chapter is divided into two main sections. The first presents the three versions of the Group A verses with the corresponding translations, along with background information on each set. These translations will indicate my own interpretation of the source material and also help readers understand the content of three versions in case they are not proficient in each language. The second section provides an analysis of the Group A verses. These verses can be examined through internal and external comparisons. Internal comparisons study the three versions of the Śarīrārthagāthā against each other. External comparisons study the Group A verses of the Śarīrārthagāthā with reference to parallel verses in other texts, especially those in corresponding sūtras (i.e. those which are attributed to the same school, the (Mūla)sarvāstivāda). Other issues relevant to the Group A verses will also be discussed, namely, the relationship between the two Chinese Saṃyuktāgamas and select technical terms.

3.1. The three versions of the Group A verses with corresponding translations, and the narratives of corresponding sūtras and parellels

In this section, fourteen verse sets will be studied in two ways. Firstly, three versions (Chinese, Sanskrit and Tibetan) with their translations will be presented to facilitate comparisons between their content and structure. Secondly, background information on each set of verses will be supplied. The Śarīrārthagāthā does not provide this information; while some verses are evidently conversations, it is often hard to know the identities of the speakers. Therefore, the background narratives of the verses from corresponding sūtras (in the Saṃyuktāgama) and parallels (mostly from the Saṃyuttanikāya) will be presented. Different parallels of these verses may have different background narratives. The effective way to analyze this is to extract the elements of the narrative and then compare them, namely, speaker/replier, hearer/one asking and the setting of the verse. From this information, we may discern the differences between the Śarīrārthagāthā, and the Saṃyuktāgama and parallels.
 
I have attempted to provide a fairly literal translation, except in instances where doing so may mislead the reader as to the meaning of the text. The number of verse pāda is quite divergent between the different translators and is influenced by the skill and knowledge of the translators (Zürcher 1991: 284–285). Therefore, pāda number will not be used to critically compare different versions or texts, aside from the basic additional information they provide.

3.1.1. The first set: evil (è 惡, pāpa, sdig pa); 4 pādas / 1 verse

1. Three versions with their translations263
(1) Chinese version:
於身語意諸所有 一切世間惡莫作 由念正知離諸欲 勿親能引無義苦 (T1579, 30.370a11–12)
[Cūḷakokanadā264 said:]
Do not commit evils (è 惡) in the whold world,
in terms of all that is given forth by one’s body, speech and mind; staying away from all desires with mindfulness and right understanding, do not personally approach meaningless suffering.

(2) Sanskrit version:265
pāpaṃ na kuryān manasā na vācā, kāyena vā kiñcana sarvaloke |
riktaḥ kāmaiḥ smṛtimān saṃprajānan, duḥkhaṃ na seveta anarthasaṃhitam || (Enomoto 1989: 24 [YBh 112b1f; ŚrBh 3b6.1f])
[Cūḷakokanadā said:]
One should not do any evil in the whole world, not by mind, speech or body;
free from desires, mindful [and] clearly comprehending,
one should not cultivate suffering connected with what is unbeneficial.


263 The three versions are mostly the same. The Chinese term wúyì 無義, “meaningless,” corresponds to anartha in Sanskrit. The Tibetan version instead uses the term “harm” (gnod pa), a common traslation of “unbenefical” (anartha).
264 The Sanskrit counterpart of this female deity’s name is not attested, therefore I use the Pāli name instead.
265 The following Sanskrit verses from the Śarīrārthagāthā were edited by Enomoto (1989).
 

(3) Tibetan version
/ 'jig rten kun du yid dang ngag rnams dang /
/ lus kyi sdig pa ci yang mi bya ste /
/ 'dod rnams bral zhing dran ldan shes bzhin can /
/ gnod par ldan pa'i sdug bsngal bsten mi bya / (Peking 5536.253b4–5; Derge 4035.219b2–3)
[Cūḷakokanadā said:]
In all the world, do not commit any misdeeds whatsoever of body, speech, or mind.
Freed from desires, and with mindfulness and right understanding, one should not engage in any suffering connected with harm.

2. The narratives of verse(s) from corresponding sūtras and parallels


According to Enomoto (1989), the source texts of this verse are Zá Āhánjīng 雜阿含經 (Chinese Saṃyuktāgama), SĀ-G (Guṇabhadra’s Chinese translation of the Saṃyuktāgama, Zá āhánjīng 雜阿含經, Taishō edition no. 99) and four other parallels as follows:266
(1) SĀ-G
1270–1274: These sūtras share the identical verse containing six pādas.267 The speaker is mentioned as a female deity called Jūjiāní tiānnǚ 拘 迦 尼 天 女 (in 1270–1271) or Jūjiānàsuō tiānnǚ 拘迦那娑天女268 (in 1272–1274). But in SĀ-G 1271, it also mentions another speaker Ānán 阿難. 269 The Buddha approved what the female deity or Ānán said. The setting is Wángshèchéng shāngǔjīngshè 王舍城山谷精舍, a monastery in the valley



266 As mentioned by Bingenheimer (2011) and Bucknell (2012), Biéyì Zá Āhánjīng 別譯雜阿含經 (SĀ-U) can be treated as the text of (Mūla)Sarvāstivādin provenance, but this is speculative. I will examine the similarity between SĀ-G and SĀ-U by anaylizing its narratives and verses.
267 The identical verse is 其心不為惡 及身口世間 五欲悉虛空 正智正繫念 不習近眾苦 非義和合者. The meaning of the verse is the same as the one in the Chinese Yogacārabhūmi.
268 This should be the same as jūjiāní tiānnǚ 拘迦尼天女, Cūḷakokanadā in Pāli.
269 Ānán 阿難 spoke it to some other monks. The Buddha approved it and mentioned to a Brahman that it was originally uttered by Jūjiāní tiānnǚ 拘迦尼天女.
 
of Rājagṛha (in 1270–1273), or Píshèlí Míhóuchícè zhònggéjiǎngtáng 毘舍離獼猴池側重閣講堂, a double-storeyed hall next to Monkey River in Vaiśālī (in 1274).

(2) SĀ-U
269–272: These sūtras share two identical verses containing eight pādas.270 The speaker of the verse is a female deity called Qiújiānísuō 求 迦 尼 娑 . The setting is a mountain called Qíní 耆尼 in Rājagṛha (in 269–271) or Píshèlí běimíhóubǐàn jīngshè 毘舍離北獼猴彼岸精舍, a monastery located near the north of Monkey river in Vaiśālī (in 272). But in SĀ-U 270, it also mentions another speaker Ānán 阿 難 and the setting is Shèwèiguó qíshùjǐgūdúyuán 舍衛國祇樹給孤獨園, Jeta’s Grove in Śrāvastī. The Buddha approved what the female deity or Ānán said.

(3) SN
The following two sūtras share the same verse containing four pādas.271
1.2.10272: The speaker is an unknown female deity. The monk Samiddhi helped her ask the Buddha. The setting is Rājagaha in the Hot Springs Park (Tapodārāma).
1.4.10273: The speaker is a female deity called Cūḷakokanadā. The hearer is the Buddha. The setting is Vaiśālī in the Great Wood (Mahāvana).

(4) Gēnběnshuōyīqièyǒubù pínàiyē 根本說一切有部毗奈耶 (the Chinese
Mūlasarvāstivāda-vinaya)
Only one occurrence of the verse, containing four pādas.274
T1442: This verse was spoken by Dàlù 大路 (Mahāpanthaka) to his younger brother Yúlù 愚路 (Cūḍapanthaka). The setting is Shìluófá Chéng 室羅伐城, a town called Śrāvastī.


270 The verse is 口意宜修善 不應作諸惡 身不以小惡 加害於世間 觀欲空無實 修於念覺意 若自不樂苦 莫作損減業. The difference in content is that here the suffering is connected with “negative karma.”
271 pāpaṃ na kayirā vacasā manasā, kāyena vā kiñcana sabbaloke | kāme pahāya satimā sampajāno (SNN
satisampajāno), dukkhaṃ na sevetha anatthasaṃhitan ti ||
272 SN I 12.19–22 = SNN I 26.1–4.
273 SN I 31.5–8 = SNN I 63.17–64.3.
274 身語意業不造惡 不惱世間諸有情 正念觀知欲境空 無益之苦當遠離 (T1442, 23.796b9–10).
 

(5) Āpídámó dàpípóshā lùn 阿毘達磨大毘婆沙論 (the Chinese Abhidharmamahā- vibhāṣā)
Only one occurrence of the verse, containing four pādas.275
T1545: This verse was spoken by Dàlù 大路 (Mahāpanthaka) to his younger brother Xiǎolù 小路 (Cūḍapanthaka). The setting is Shìluófá 室羅筏 (Śrāvastī).

The following Table 3.1 is an overview of the information from these five texts:



Speaker Hearer Location
SĀ-G 1270 Jūjiāní 拘迦尼 The Buddha and
his disciples A monastery in the valley
of Rājagṛha
SĀ-G 1271 Ānán 阿難 and Jūjiāní
拘迦尼 The Buddha and his disciples A monastery in the valley of Rājagṛha
SĀ-G 1272 Jūjiānàsuō 拘迦那娑 The Buddha and
his disciples A monastery in the valley
of Rājagṛha
SĀ-G 1273 Jūjiānàsuō 拘迦那娑 The Buddha and
his disciples A monastery in the valley
of Rājagṛha
SĀ-G 1274 Jūjiānàsuō 拘迦那娑 The Buddha and his disciples A double-storeyed hall
next to the Monkey River in Vaiśālī.
SĀ-U 269 Qiújiānísuō 求迦尼娑 The Buddha A mountain called Qíní 耆
尼 in Rājagṛha
SĀ-U 270 Ānán 阿 難 and Qiújiānísuō 求迦尼娑 The Buddha and his disciples Jeta’s Grove in Śrāvastī and a mountain called Qíní
耆尼 in Rājagṛha
SĀ-U 271 Qiújiānísuō 求迦尼娑 The Buddha A mountain called Qíní 耆
尼 in Rājagṛha
SĀ-U 272 Qiújiānísuō 求迦尼娑 The Buddha A monastery located near the north of Monkey River
in Vaiśālī.


275 身語意莫作 一切世間惡 離欲念正知 不受苦無義 (T1545, 27.902b3–4).
 
SN 2.10 A female deity The Buddha Hot Springs Park at
Rājagaha
SN 4.10 Cūḷakokanadā The Buddha Great Wood (Mahāvana)
at Vaiśālī
T1422 Dàlù 大路
(Mahāpanthaka) Yúlù 愚路
(Cūḍapanthaka) Śrāvastī
T1545 Dàlù 大路
(Mahāpanthaka) Xiǎolù 小路
(Cūḍapanthaka) Śrāvastī

Apparently, in the part of Saṃyuttanikāya cited in Table 3.1, this verse is composed and spoken by a female deity, Cūḷakokanadā in Pāli276 although Ānán 阿難 is once mentioned. The Buddha, who is the key person, agreed with what Cūḷakokanadā said. However, the location seems to be either Rājagṛha or Vaiśālī, yet Rājagṛha is mentioned more frequently than Vaiśālī (8 times compared with 2 times). Two texts mention that the
older brother, Mahāpanthaka, spoke the verse to the younger brother, Cūḍapanthaka, in Śrāvastī.

3.1.2. The second set: what can be expressed277 (shuō 說, ākhyeya, brjod bya); 16 pādas
/ 4 verses

1. Three versions with their translations278
(1) Chinese version:

應說想眾生 依應說安住 不了知應說 而招集生死
若了知應說 於說者無慮 由無有此故 他不應譏論
若計等勝劣 彼遂興諍論 於三種無動 等勝劣皆無



276 The Sanskrit counterpart of this female deity’s name is not attested. The names in Chinese transcription, Jūjiāní 拘迦尼, Jūjiānàsuō 拘迦那娑, Qiújiānísuō 求迦尼娑, obviously point to the same person.
277 This title is based on a translation by Bodhi (2000: 99) of a counterpart in the Saṃyuttanikāya.
278 These three versions are essentially the same. However, it is difficult to understand the meaning of this verse set. The main issue is the technical term yìngshuō 應 說 , ākhyeya, brjod bya, because it cannot be understood by the literal interpretation. This term will be discussed in § 3.2.3.2. In Chinese, “in tandem with birth and death” (zhāojí shēngsǐ 招 集 生 死 ) means “will cause reincarnation.” This seems to be the interpretation of Xuánzàng 玄 奘 because the Sanskrit “come under the yoke of death,” the Tibetan “come to meet with death” and other corresponding sūtras (see below) connect it only with “death” which does not
mention rebirth.
 
斷名色愛慢 無著煙寂靜 無惱悕不見 此彼天人世 (T1579, 30.370b29–c7) [The Buddha said:]
What can be expressed is the perception of sentient beings, [who] dwell by what is to be expressed;
not understanding what can be expressed, is in tandem with birth and death.
If fully understanding what can be expressed, One does not worry about what others express; because he does not have this [[[concern]]],
others should not censure [him].
If one thinks “the equal, the superior and the inferior,” one would then engage in disputes [on this account]; because one is not shaken in the three [discriminations],
there is no [notion of] equality, superiority or inferiority at all. One who cuts off name and form, craving, conceit,
has no attachment [like] the stillness of smoke; absence of vexation and expectation,
this one cannot be found in the heavens or any of the abodes.


(2) Sanskrit version:
ākhyeyasaṃjñinaḥ sattvā, ākhyeye 'smin pratiṣṭhitāḥ | ākhyeyam aparijñāya, yogam āyānti mṛtyunaḥ || ākhyeyaṃ tu parijñāya, ākhyātāraṃ na manyate |
tad vai na vidyate tasya, vadeyur yena taṃ pare ||
samo viśeṣa uta vāpi hīno, yo manyate sa vivadeta tena |
vidhātraye 'smin na vikampate yaḥ, samo viśiṣṭaś ca na tasya bhavati || ācchidya tṛṣṇām iha nāmarūpe, prahāya mānaṃ ca na saṅgam eti | taṃ śāntadhūmam anighaṃ nirāśaṃ, nādrākṣus te devamanuṣyaloke | iha bāhirataś ca || (Enomoto 1989: 24 [YBh 113a3–5; ŚrBh 3a7.4f]) [The Buddha said:]
Sentient beings conscious of what can be expressed, fixed in what can be expressed;
 
not fully understanding what can be expressed, come under the yoke of death.
But having fully understood what can be expressed, one does not think about the one who expresses; that does not exist for him
by which others might blame him.
One who thinks “equal, superior or inferior,” might thereby engage in dispute.
One who is not shaken in these three kinds [of discrimination] does not have the [[[thought]]] “equal or better [than others].” Cutting off craving in this life in regard to name and form, and abandoning conceit, one does not develop attachment.
They did not see him who is like abated smoke, free from troubles, without desire in the world of gods and humans,
in this life and beyond.


(3) Tibetan version:
/ sems can brjod bya'i 'du shes can /
/ brjod bya la ni gnas pa rnams /
/ brjod bya yongs su mi shes pas /
/ 'chi ba dang ni phrad par 'gyur /
/ brjod bya yongs su shes na ni /
/ gang gis de la gzhan dag gis /
/ brjod pa de ni de la med /
/ brjod pa por ni sems pa med /
/ mtshungs sam khyad zhugs 'on te dma'o zhes /
/ gang 'dzin de ni des na rtsod par 'gyur /
/ gang zhig rnam pa gsum la mi g.yo ba /
/ de la mtshungs dang khyad par zhugs pa med /
/ 'dir ni ming dang gzugs la sred bcad cing /
/ nga rgyal spangs nas chags pa med pa de /
 
/ dud pas279 sangs shing 'joms med re ba med /
/ lha mi'i 'jig rten dag na de mi snang /
(Peking 5536.255a4–7; Derge 4035.220b5–221a1)
[The Buddha said:]
Sentient beings conscious of what can be expressed, who abide in what can be expressed;
not fully understanding what can be expressed, come to meet with death.
[As to one] fully understanding what can be expressed, what is expressed about him by others
does not exist for him,
[because] he has given no attention to the one who expresses. Fixating on that which is called “equal, superior or inferior,” one will lead to dispute.
But for one who is unmoving in the three types [of discrimination], there is no equality or superiority.
Cutting off craving in this life in regard to name and form, having abandoned conceit, there is no attachment
like vanished smoke, it cannot be destroyed or hoped for; one cannot be seen in the worlds of gods and humans.

2. The narratives of verse(s) from corresponding sūtras with parallels


In the Śarīrārthagāthā, these four verses seem to be connected. However, in the following texts they are combined answers to a deity’s questions.

(1) SĀ-G
1078: There are four verses containing eighteen pādas.280 The verses are four answers given by the Buddha. The one asking is an unknown deity. This happened in

279 Derge: pa.
280 眾生隨愛想 以愛想而住 以不知愛故 則為死方便 (T99, 2.282a3–4); 若知所愛者 不於彼生愛 彼此無所有 他人莫能說 (T99, 2.282a8–9); 見等勝劣者 則有言論生 三事不傾動 則無軟中上 (T99, 2.282a12– 13); 斷愛及名色 除慢無所繫 寂滅息瞋恚 離結絕悕望 不見於人天 此世及他世 (T99, 2.282a16–18).
 
Wángshèchéng Jiālántuózhúyuán 王舍城迦蘭陀竹園, Karaṇḍa bamboo grove in Rājagṛha.

(2) SĀ-U
17281: There are five verses containing twenty two pādas. The verses are the three answers made by the Buddha in Wángshèchéng Jiālántuózhúlín 王舍城迦蘭陀竹林, Karaṇḍa bamboo grove in Rājagṛha. The one asking is an unknown deity.

(3) SN
1.2.10282: There are four verses containing eighteen pādas. The verses are three answers made by the Buddha. The question is from an unknown deity. The setting is Rājagaha283.

From these three texts, the verses are clearly spoken by Buddha, the one asking is an unknown deity and the location is Rājagaha (Wángshèchéng 王 舍 城 ). The number of answer verses seems different between SĀ-G and SĀ-U/SN.

3.1.3. The third set: lust (tān 貪, rāga, 'dod chags); 20 pādas / 5 verses

1. Three versions with their translations284
(1) Chinese version:
欲貪所摧蔽 我心遍燒然 惟大仙哀愍 為說令寂靜

281名色中生相 謂為真實有 當知如斯人 是名屬死徑 若識於名色 本空無有性 是名尊敬佛 永離於諸趣(T100, 2.379b15–18); 勝慢及等慢 并及不如慢 有此三慢者 是可有諍論 滅除此三慢 是名不動想 (T100, 2.379b21–23); 斷愛及名色 滅除三種慢 不觸於諸欲 滅除於瞋恚 拔除諸毒根 諸想願欲盡 若能如是者 得度生死海 (T100, 2.379b26–29).
282 akkheyyasaññino sattā, akkheyyasmiṃ patiṭṭhitā | akkheyyam apariññāya, yogam āyanti maccuno || akkheyyañ ca pariññāya, akkhātāraṃ na maññati | tañ hi tassa na hotī ti, yena naṃ vajjā na tassa atthi || (SN I 11.22–25 = SNN I 23.16–24.6); samo visesī udavā nihīno, yo maññatī so vivadetha tena | tīsu vidhāsu avikampamāno, samo visesīti na tassa hoti || (SN I 12.1–4 = SNN I 24.13–16);
pahāsi saṅkhaṃ na vimānam ajjhagā (SNN āgā), acchecchi taṇhaṃ idha nāmarūpe | taṃ chinnaganthaṃ anighaṃ nirāsaṃ, pariyesamānā nājjhagamuṃ| devā manussā idha vā huraṃ vā, saggesu vā sabbanivesanesu || (SN I 12.10–15 = SNN I 25.6–11).
283 It is located in the Hot Springs Park (Tapodārāma) of Rājagaha.
284 These three versions are basically the same. Gautama and dà xiān 大 仙 (great sage) in the Chinese version should not necessarily be taken as “the Buddha” but a member from the Gautama family. From the Āgama and Nikāya texts (see below), Ānanda (from the Gautama family) replied to the questions.
 
由汝想顛倒 令心遍燒燃 是故常遠離 引貪淨妙相
汝當修不淨 常定於一境 為貪火速滅 數數應澆灌
觀非妙諸行 為苦為無我 亦繫念於身 多修習厭離
修習於無相 壞慢及隨眠 由於慢現觀 當證苦邊際 (T1579, 30.372a24–b4)

[[[Vaṅgīsa]] said:]
“Ravaged by sensual lust,
my mind is pervasively burning;
[I] wish the great sage will take pity on me, [and] have a talk to calm me down.” [[[Ānanda]] said:]
“Due to your inversion of perception, your mind is pervasively burning; therefore, always keep far away
from an object of great beauty which provokes lust. You should practice meditation on impurity,
and always concentrate on one object;
in order to rapidly extinguish the fire of lust, [you] should irrigate [the mind] frequently.
Observe all unpleasant conditionings as suffering and “not-self,”
also, be mindful of the body [and] practice more on revulsion.
Practice on the formless
[that] discards conceit and latent tendency; due to insight into conceit,
one reaches the end of suffering.”


(2) Sanskrit version:
kāmarāgābhibhūtatvāc, cittaṃ me paridahyate | aṅga me gautama brūhi, śāntiṃ tvaṃ anukampayā || viparyāsena saṃjñānāṃ, cittaṃ te paridahyate |
nimittaṃ varjyatāṃ tasmāc, chubhaṃ rāgopasaṃhitam ||
 
aśubhāṃ bhāvaya sadā tvam, ekāgraḥ susamāhitaḥ | nirvāpayāśu rāgāgniṃ, dahyase mā punaḥ punaḥ || saṃskārāvaratāṃ paśya, duḥkhato 'nātmatas tathā | smṛtiṃ kāyagatāṃ kṛtvā, nirvedabahulo bhava || bhāvyatām animittaṃ ca, mānānuśayanāśanam | tato mānābhisamayād, duḥkhasyāntaṃ kariṣyasi || (Enomoto 1989: 25 [YBh 115a4–6; ŚrBh 4a1.3f]) [[[Vaṅgīsa]] said:]
“Because of the state of being overwhelmed by sensual lust, my mind is tormented.
Gautama, please speak to me
in regard to calmness, out of compassion.” [[[Ānanda]] said:]
“By an inversion of perceptions, your mind is tormented;
therefore, beautiful object[s] connected with lust should be abandoned.
Always meditate upon the impure, one-pointed and well-concentrated.
Having quickly extinguished the fire of lust, you should not be tormented again and again.
Contemplate the inferiority of conditioned things as suffering and also not-self;
having established mindfulness on the body, you should cultivate much revulsion.
You should cultivate the formless
and the destruction of latent tendency (anuśaya) to conceit; then by the comprehension of conceit,
you will put an end to suffering.”
 
(3) Tibetan version:
/ 'dod pa'i chags pas zil mnan pas /
/ bdag gi sems ni yongs su bsregs /
/ kye kye go'u ta ma thugs brtse ba /
/ khyod kyis zhi 'gyur bdag la gsung285 /
/ 'du shes phyin ci log pa yis /
/ khyod kyi sems ni yongs su bsregs /
/ de phyir 'dod chags ldan pa yi /
/ sdug pa'i mtshan ma spang bar gyis /
/ khyod kyis rtse gcig mnyam bzhag286 pas /
/ mi sdug rtag tu bsgom par gyis /
/ 'dod chags me ni myur du sod /
/ phyi phyir zhing ni bsregs sa re /
/ sdug bsngal de bzhin bdag med pas /
/ 'du byed thams cad nyid la ltos /
/ lus la dran pa gnas byos la /
/ shas cher yang ni skyo bar gyis /
/ nga rgyal bag nyal287 sel ba yi /
/ mtshan ma med pa'ang bsgom par gyis /
/ des na nga rgyal mngon rtogs pas /
/ sdug bsngal mthar ni 'byin par 'gyur /
(Peking 5536.255a4–7; Derge 4035.224b4–224b6)
[[[Vaṅgīsa]] said:]
“Overwhelmed by sensual lust, my mind is thoroughly burnt.
O Gautama! [With your] loving-kindness, speak to me of that which pacifies.” [[[Ānanda]] said:]
“By distorted perception,
your mind is thoroughly burnt.

285 Derge: gsungs.
286 Derge: gzhag.
287 Peking: la.
 
Therefore, connected with lust, attractive objects should be abandoned. As you constantly meditate on impurity, with one-pointed concentration,
quickly extinguish the fire of lust.
The field [of lust] will be burned up again and again.
Observe the essence of all conditioned things as suffering and also not-self;
establish mindfulness on the body, and cultivate much revulsion.
Also practice intent on signlessness which dispels latent conceit.
Therefore, by the comprehension of conceit, suffering will become exhausted.”

2. The narratives of verse(s) from corresponding sūtras with parallels


(1) SĀ-G
1214: There are five verses containing twenty two pādas.288 The setting is Shěwèiguó qíshù gěigūdú yuán 舍衛國祇樹給孤獨園, Anāthapiṇḍika’s Park of Jeta’s Grove in Śrāvastī. Ānàn 阿難 answers the question of a disciple, Póqíshě 婆耆舍, Vāgīśa in Sanskrit.

(2) SĀ-U
230: There are five verses containing twenty pādas.289 The setting is Shěwèiguó qíshù gěigūdú yuán 舍衛國祇樹給孤獨園, Anāthapiṇḍika’s Park of Jeta’s


288 貪欲所覆故 熾然燒我心 今尊者阿難 為我滅貪火 慈心哀愍故 方便為我說 (T99, 2.331a25–27); 以彼顛倒想 熾然燒其心 遠離於淨想 長養貪欲者 當修不淨觀 常一心正受 速滅貪欲火 莫令燒其心 諦觀察諸行 苦空非有我 繫念正觀身 多修習厭離 修習於無相 滅除憍慢使 得慢無間等 究竟於苦邊(T99, 2.331a29–b7).
289 為欲結所勝 燋然於我心 唯願為我說 除欲善方便 (T100, 2.458b5–6); 起於顛倒想 能燒然其心 淨想能生欲 應修不淨觀 獨處而坐禪 速滅於貪欲 莫數受燒然 當觀察諸行 無常無有樂 并及無我法 安心念此身 多厭惡生死 修習正智慧 除七慢結使 若知斷慢已 苦則有邊際 (T100, 2.458b8–15).
 
Grove in Śrāvastī. Ānàn 阿難 answers the question from a disciple, Póqíshě 婆耆舍, Vāgīśa in Sanskrit.

(3) EĀ-S
9: There are two verses containing eight pādas.290 The setting is Luóyuèchéng jiālánzhúyuánsuǒ 羅閱城迦蘭竹園所, Karaṇḍa bamboo grove in Rājagṛha. Ānàn 阿難 answers the question from a disciple, Póqíshě 婆耆舍, Vāgīśa in Sanskrit.

(4) SN
8.4: There are five verses containing twenty pādas.291 Ānanda answers the question from Vaṅgīsa in Anāthapiṇḍika’s Park of Jeta’s Grove in Sāvatthī (Śrāvastī in Sanskrit).

These texts all show that the one asking is Vaṅgīsa and the replier is Ānanda (Ānàn 阿 難 ). The setting is apparently Anāthapiṇḍika’s Park of Jeta’s Grove in Sāvatthī, although the Chinese Ekottarikāgama differs on this.

3.1.4. The fourth set: flood (liú 流, ogha, chu bo); 12 pādas / 3 verses

1. Three versions with their translations292
(1) Chinese version:
云何苾芻多所住 越五暴流當度六 云何定者能度廣 欲愛而未得腰舟身輕安心善解脫 無作繫念不傾動 了法修習無尋定 憤愛惛沈過解脫


290 欲火之所燒 心意極熾然 願說滅此義 多有所饒益 (T125, 2.701a18–19); 知欲顛倒法 心意極熾然 當除想像念 欲意便自休 (T125, 2.701a21–22).
291 kāmarāgena ḍayhāmi, cittaṃ me pariḍayhati| sādhu nibbāpanaṃ brūhi, anukampāya gotamā ti || saññāya vipariyesā, cittaṃ te pariḍayhati | nimittaṃ parivajjehi, subhaṃ rāgūpasaṃhitaṃ || saṅkhāre parato passa, dukkhato mā ca attato |
nibbāpehi mahārāgaṃ, mā ḍayhittho punappunaṃ || asubhāya cittaṃ bhāvehi, ekaggaṃ susamāhitaṃ | sati kāyagatā ty atthu, nibbidābahulo bhava || animittañ ca bhāvehi, mānānusayam ujjaha |
tato mānābhisamayā, upasanto carissasī ti || (SN I 188.15–24 = SNN I 405.7–406.6).
292 These three versions are essentially the same.
 
如是苾芻多所住 越五暴流當度六 如是定者能度廣 欲愛而未得腰舟 (T1579, 30.372c29–373a5)
[One293 asked:]
“How does a monk dwell much,
after crossing the five violent floods, thereupon cross the sixth?
How does a meditator cross vast sensual lust and not obtain a waist boat (lifebuoy)294?” [The Buddha replied:]
“With a tranquil body and a well liberated mind,
one does not generate conditioned thoughts, is mindful and does not waver.
Having mastered the dharma and having practiced the meditation of non-discursive thought,
one is liberated from the faults of anger, craving and lethargy.
In this way, the monk dwells much
where he has crossed the five violent floods and is traversing the sixth.
In this way, the meditator can cross vast sensual craving who has not attained the waist boat.”

(2) Sanskrit version:
kathaṃvihārabahulo bhikṣuḥ, pañcaughatīrṇas taratīha ṣaṣṭham | kathaṃdhyāyī vipulāṃ kāmatṛṣṇāṃ, tīrṇo bhavaty apratilabdhayoktraḥ || praśrabdhakāyaḥ suvimuktacitto, hy asaṃskurvan smṛtimān akopyaḥ | ājñāya dharmam avitarkadhyāyī, kopaspṛhāstyānadoṣaiḥ viyuktaḥ || evaṃvihārabahulo bhikṣuḥ, pañcaughatīrṇas taratīha ṣaṣṭham | evaṃdhyāyī vipulāṃ kāmatṛṣṇāṃ, tīrṇo bhavaty apratilabdhayoktraḥ || (Enomoto 1989: 25 [YBh 116a3f; ŚrBh 4b1.8f])
[One asked:]
“How does a monk who often dwells much

293 See the discussion below. This is likely to be one of Māra's three daughters.
294 Yāozhōu 腰 舟 , which literally translates as “waist boat,” refers to a big gourd tied to the waist that prevents one from sinking while crossing the river. It thus refers to something similar to a lifebuoy. The indigenous term yāozhōu is probably a translation of yoktra, which probably refers to something not found in China. The commentary glosses this term as “the mind.” For further details on the interpretation of this
term, see § 3.2.3.2.
 
[and] has crossed the five streams here cross the sixth [[[Wikipedia:stream|stream]]]? How does a meditator who has not obtained the tie of the yoke, become [one who] has crossed the vast sensual craving.”
[The Buddha replied:]
“Having a tranquil body, a well-liberated mind, not creating, mindful, steadfast,
understanding the dharma, meditating without discursive thought, one is detached from the faults of anger, eager desire and lethargy. A monk who thus dwells much,
has crossed the five streams, crosses the sixth here. meditating thus, having not obtained the tie of the yoke, becomes [one who] has crossed the vast sensual craving.

(3) Tibetan version:
/ ji ltar dge slong phal cher gnas gyur na /
/ chu bo lnga rgal 'di la drug pa brgal295 /
/ ji ltar gnya' dag ma thob bsam gtan pa /
/ 'dod pa'i sred pa chen po rgal296 bar 'gyur /
/ tshogs sbyangs sems ni shin tu rnam par grol /
/ 'du byed med cing dran ldan 'khrugs297 pa med /
/ chos kun shes shing rtog med bsam gtan pa /
/ 'khrug pa chags dang rmugs pa'i nyes pa med /
/ de ltar dge slong phal cher gnas gyur na /
/ chu bo lnga brgal298 'di la drug pa brgal299 /
/ de ltar gnya' dag ma thob bsam gtan pa/
/ 'dod pa'i sred pa chen po brgal300 bar 'gyur /
(Peking 5536.261b8–262a3; Derge 4035.226a7–226b1)
[One asked:]


295 Derge: rgal.
296 Derge: rgal.
297 Derge: 'khrug.
298 Derge: rgal.
299 Derge: rgal.
300 Derge: rgal.
 
“How does a monk who dwells much,
after crossing the five rivers, thereupon cross the sixth?
How does a meditator who does not obtain the tie of the yoke301 cross the great sensual craving?”
[The Buddha replied:]
“Endowed with a trained mind [which is] completely liberated, unconditioned, mindful and undisturbed;
the thought-free meditator who understands all dharma, is without the faults of anger, craving and lethargy.
In that way does a monk who dwells much,
having crossed the five rivers, thereupon cross the sixth.
In that way does the meditator who has not obtained the tie of the yoke, cross the great sensual craving.”

2. The narratives of verse(s) from corresponding sūtras with parallels


(1) SĀ-G
1092: There are six verses containing twenty four pādas.302 The setting is on the bank of the river Níliánchán 尼連禪 (Nairañjanā) at Yùbìluó 欝鞞羅 (Uruvilvā). Àiniàn 愛 念 , the female deity Aratī (the second daughter of Māra) questioned and the Buddha replied.

(2) SĀ-U
31: There are five verses containing eighteen pādas.303 The setting is on the bank of the river Níliánchán 尼連禪 (Nairañjanā) at Yōulóubǐluó 優樓比螺 (Uruvilvā). Shìyì 適意, the third daughter of Māra, Ratī, questioned and the Buddha replied.

301 Yokoyama and Hirosawa (1996: 1076) indicated that yāozhōu 腰 舟 can be seen as an equivalent of “gnya' dag” but do not provide any Sanskrit counterpart, so I assume that gnya' dag might be a synonym of gnya' shing, which can correspond to yoktra in Sanskrit.
302 多修何妙禪 而度五欲流 復以何方便 度於第六海 云何修妙禪 於諸深廣欲 得度於彼岸 不為愛所持(T99, 2.287a29–b3); 身得止息樂 心得善解脫 無為無所作 正念不傾動 了知一切法 不起諸亂覺 愛恚睡眠覆 斯等皆已離 如是多修習 得度於五欲 亦於第六海 悉得度彼岸 如是修習禪 於諸深廣欲 悉得度彼岸 不為彼所持 (T99, 2.287b5–b12).
303比丘住何處 能度五駛流 六駛流亦過 入何禪定中 得度大欲岸 永離有攝縛 (T100, 2.383c29); 身獲柔軟
 

(3) SN
4.3.5: There are three verses containing twelve pādas.304 The setting is on the bank of the river Nerañjarā at Uruvelā. Māra’s daughter Aratī asked the question and the Buddha replied.

(4) Mahāvastu Avadāna
There are three verses containing twelve pādas.305 Māra’s daughter Tantrī asked the question and the Buddha replied. The setting is not explicitly stated; however, it mentions the place is where the Buddha became awakened under the Bodhi tree.
Therefore, it must be Uruvilvā.


From the above texts, we may surmise that the setting is Uruvilvā, but the identity of the questioner is unclear because all of Māra’s three daughters are mentioned. SĀ-G states that the questioner is Aratī, the second daughter of Māra, and perhaps we may also assume that this is the speaker for the Śarīrārthagāthā since both these texts are attributed to same school. A difference needs to be noted between these texts and the Yogacārabhūmi. They do not mention something related to a “yoke” or “waist boat” which only can be seen in the fourth verse set of the Śarīrārthagāthā.







樂 心得善解脫 心離於諸業 意不復退轉 得斷覺觀法 得離瞋愛掉 得住此處住 能度五駛流 并度第六者 作如是坐禪 能度大欲結 并離有攝流 (T100, 2.384a1–6).
304 kathaṃ vihārībahulo dha (SNN 'dha) bhikkhu, pañcoghatiṇṇo atarīdha chaṭṭhaṃ |
kathaṃ jhāyiṃ (SNN kathaṃjhāyiṃ) bahulaṃ kāmasaññā, paribāhirā honti aladdha yo tan ti ||
passaddhakāyo suvimuttacitto, asaṅkharāno satimā anoko |
aññāya dhammaṃ avitakkajhāyī, na kuppati na sarati na thino (SNN na kuppatī na saratī na thīno) ||
evaṃvihārībahulo dha (SNN 'dha) bhikkhu, pañcoghatiṇṇo atarīdha chaṭṭhaṃ |
evaṃ jhāyiṃ (SNN evaṃjhāyiṃ) bahulaṃ kāmasaññā, paribāhirā honti aladdha yo tan ti || (SN I 126.21–22
= SNN 277.1–8).
305 kathaṃvihārī bahulīha bhikṣu, paṃcoghatīrṇo taratīha ṣaṣṭaṃ / kathaṃdhyāyī bahulī kāmasaṃjñā, paribāhito bhavati alabdhagāḍhā // aññāya dharmaṃ avitarkadhyāyī, no rajyate no saratīha thīnaṃ / evaṃvihārī bahulīha bhikṣu, paṃcoghatīrṇo taratīha ṣaṣṭaṃ // praśrabdhakāyo suvimuktacitto, smṛtimānakopyo apanītamāno /
avaṃdhyāyī bahulī kāmasaṃjñā, paribāhito bhavati alabdhagāḍhā // (Mvu III 283.17–284.10).
 
3.1.5. The fifth set: being frightened (bù 怖, utrasta, ‘jigs); 14 pādas / 3 verses

1. Three versions with their translations306
(1) Chinese version:
常有怖世間 眾生恒所厭 於未生眾苦 或復已生中 若有少無怖 今請為我說天我觀解脫 不離智精進 不離攝諸根 不離一切捨
我觀極久遠 梵志般涅槃 已過諸恐怖 超世間貪著 (T1579, 30.374a1–7) [The deity said:]
“In the world with perpetual fear,
sentient beings perennially dislike the sufferings that have not arisen
or even those that have already arisen in which [they are caught]. If there is a way to reduce or do away with fear,
please now preach it for me.” [The Buddha said:]
“O Deity! I contemplate that liberation
which is not separate from wisdom and diligence, not separate from restraint of the sense faculties, not separate from complete relinquishing.
I have contemplated for a very long time that having achieved parinirvāṇa,
the Brahman has overcome fear
and transcended worldly attachment of desires.”


(2) Sanskrit version:
nityotrasto hy ayaṃ loko, nityodvignā iyaṃ prajā | anutpanneṣu duḥkheṣu, samutpanneṣu vā punaḥ | yadi kiṃcid anutrastaṃ, pṛṣṭa ācakṣva tan mama ||

306 These three versions are mostly the same. Only one term in the Chinese version is different from the others, namely, Jīngjìn 精進, “diligence,” which is parallel to tapas in Sanskrit and dka' thub in Tibetan. The term tapas, “austerity,” would normally be kǔxíng 苦行 in Chinese, but “diligence” (Jīngjìn 精進) would be a more positive attitude which better fits this context. This might be the reason the Chinese translators commonly use Jīngjìn 精進 insead of kǔxíng 苦行.
 
nānyatra jñānatapaso, nānyatrendriyanigrahāt | nānyatra sarvasaṃtyāgān, mokṣaṃ paśyāmi devate || cirasya bata paśyāmi, brāhmaṇaṃ parinirvṛtaṃ | sarvavairabhayātītaṃ, tīrṇaṃ loke viṣaktikām || (Enomoto 1989: 26 [YBh 117b2; ŚrBh 4a3.4f]) [The deity said:]
“For this world is constantly frightened, [and] humankind is constantly anxious, about suffering which is unarisen
and also arisen.
If there is anything which is unfrightened, may you, being asked, tell me.”
[The Buddha said:]
“Not apart from wisdom and austerity,
not apart from restraint of the sense faculties, not apart from relinquishing all,
do I see liberation, O deity.
Ah! after a long time I see
a Brahman who is fully quenched,
who has left behind all enmity and fear,
who has crossed over attachment to the world.”


(3) Tibetan version:
/ sdug bsngal dag ni ma skyes dang /
/ yang na skyes par gyur pa na /
/ 'jig rten 'di dag rtag tu 'jigs /
/ skye dgu 'di dag rtag tu skyo /
/ gal te 'jigs med 'ga' mchis na /
/ zhus pa'i lan307 ni bdag la gsungs /
/ shes dang dka' thub ma gtogs dang /
/ dbang po bcil ba ma gtogs dang /

307 Derge: len.
 
/ lha mo kun spangs ma gtogs par /
/ thar par 'gyur ba ma mthong ngo /
/ bram ze mya ngan yongs 'das pa /
/ mkhon308 dang 'jigs pa kun 'das shing /
/ 'jig rten dag na sred rgal ba /
/ kye ma yun rings gdod mthong ngo /
(Peking 5536.265a3–265a6; Derge 4035.229a1–229a3)
[The deity said:]
“With suffering that has not yet arisen, or else has already arisen,
the worlds are constantly frightened;
[and] those sentient beings are constantly saddened. If there is any that exist without fear,
tell me the answer to my query.” [The Buddha said:]
“Apart from knowledge and austerity, and restraint of the sense faculties,
and apart from total renunciation, O Goddess, liberation is never seen.
A Brahman who has fully transcended affliction, transcended all enmity and fear,
and has crossed over attachment to the worlds, ah! after for a long time [I] see [him].”

2. The narratives of verse(s) from corresponding sūtras with parallels


(1) SĀ-G
596: There are three verses containing fourteen pādas.309 The setting is Shěwèiguó qíshù gěigūdú yuán 舍衛國祇樹給孤獨園, Anāthapiṇḍika’s Park of Jeta’s Grove in Śrāvastī. An unkown deity talked with the Buddha.

308 Derge: khon.
309此世多恐怖 眾生常惱亂 已起者亦苦 未起亦當苦 頗有離恐處 唯願慧眼說 (T99, 2.159c24–26); 無有異
 
(2) SĀ-U
181: There are six verses containing twenty six pādas.310 The setting is Shěwèigu ó qíshù gěigūdú yuán 舍衛國祇樹給孤獨園, Anāthapiṇḍika’s Park of Jeta’s Grove in Śrāvastī. An unknown deity talked with the Buddha.

(3) SN
2.2.7+ SN 1.1.1311: There are three verses containing fourteen pādas.312 The deity Subrahmā talked with the Buddha. The location is not mentioned.313

From the above information, we may surmise that a deity talked with the Buddha, probably in Śrāvastī. But it is hard to know whether the deity should be referred to as “Subrahmā,” as the Saṃyuttanikāya indicates. Another difference is that the subject of the verses in Chinese is “beings” (zhòngshēng 眾生, prajā) but in Pāli, it is “the mind” (citta).

3.1.6. The sixth set: the caste system314 (lèi 類, abhivarṇṇatā, rigs); 12 pādas / 3 verses

1. Three versions with their translations315
(1) Chinese version:
誰獎勝類生 及開出離道 於何住何學 不懼後世死

苦行 無異伏諸根 無異一切捨 而得見解脫 (T99, 2.159c28–29); 久見婆羅門 逮得般涅槃 一切怖已過 永超世恩愛 (T99, 2.159c28–29).
310世間常驚懼 眾生恒憂惱 未得財封利 及已得之者 於得不得中 能無喜懼心 如斯之等事 唯願為我說(T100, 2.439a7–10); 若有智慧者 苦行攝諸根 棄捨一切務 除如此等人 更無出生死 若不捨諸務 常處於生死 驚畏而怖迮 憂愁等諸患 苦惱所纏逼 若捨於一切 能除上諸患 則離於生死 憂怖等諸惡 (T100, 2.439a12–18); 往昔已曾見 婆羅門涅槃 嫌怖久棄捨 能度世間愛 (T100, 2.439a20–21).
311 The verse is supposed to be a final verse which can be seen in many sūtras, such as SN 2.2.8 (SN I 54.27–29 = SNN 125.13–16). The third pāda would normally be a bit changed to better fit the context. 312 niccaṃ utrastam idaṃ cittaṃ, niccaṃ ubbiggam idaṃ mano |
anuppannesu kicchesu, atho uppatitesu ca |
sace atthi anutrastaṃ, taṃ me akkhāhi pucchito ti ||
nāññatra bojjhaṅgatapasā, nāññatr (SNN nāññatr') indriyasaṃvarā |
nāññatra sabbanissaggā, sotthiṃ passāmi pāṇinan ti || (SN I 53.32–54.5 = SNN 123.3–12).
cirassaṃ vata passāmi, brāhmaṇaṃ parinibbutaṃ |
appatiṭṭhaṃ anāyūhaṃ, tiṇṇaṃ loke visattikan ti || (SN I 1.21–22 = SNN I 2.10–13).
313 Although this sūtra omits the location, it is under a chapter where in each sūtra, one deity addresses. The location, Rājagaha, is shown in the first sūtra of the chapter.
314 This rendering follows Ulrich 2013: 96.
315 These three versions are essentially the same.
 
戒慧自薰修 具定念正直 斷諸愁熾燃 正念心解脫
能獎勝類生 及開出離道 住此於此學 不懼後世死 (T1579, 30.374c16–21) [The deity asked:]
“Who instructs superior kinds of human beings and guides the path to liberation?
By what [should] one abide, and what [should] one learn in order to not fear death to the next life?”
[The Buddha said:]
“The one who personally cultivates moral conduct and wisdom, who has concentration, mindfulness and uprightness;
he cuts off all burning worries,
[achieves] right mindfulness and liberation of the mind.
[Such a person] could instruct/guide superior kinds of human beings and explain the path of liberation.
One abides by this and learns this,
and does not fear the death of the next life.”


(2) Sanskrit version:
kenābhivarṇā janatā praṇītā, mārgaś ca nairyāṇikataḥ prayuktaḥ | kutra sthitaḥ kutra ca śikṣamāṇo, nāyaṃ martyaḥ paralokād bibheti || yaḥ śīlavān jñānavān bhāvitātmā, samāhitaḥ smṛtimān ṛjugataś ca |
sarve 'sya śokajvarathāḥ prahīṇāḥ, samyaksmṛto yasya cittaṃ vimuktam ||
tenābhivarṇā janatā przaṇītā, mārgaś ca nairyāṇikataḥ prayuktaḥ | atra sthitaḥ atra ca śikṣamāṇo, nāyaṃ martyaḥ paralokād bibheti || (Enomoto 1989: 26 [YBh 118b1f; ŚrBh 4a4.5f])
[The deity asked:]
“By whom are good people guided
and the path undertaken from that which is conducive to liberation? Standing where and learning where,
does this mortal being not fear the next world?” [The Buddha said:]
“He who is virtuous, wise, cultivated,
 
concentrated, mindful and upright,
all of whose afflictions of sorrow have been abandoned, who has right mindfulness, whose heart is liberated— by him are the good people guided
and the path undertaken from that which is conducive to liberation. [Whether] standing or learning,
here a mortal being does not fear the next world.”


(3) Tibetan version:
/ gang gis rigs lhag316 skye bo bstan gyur cing /
/ lam dang nges par 'byung la btsud pa lags /
/ gang la gnas shing gang la bslabs bgyis na /
/ mi 'di 'jig rten pha rol 'jigs med 'gyur /
/ gang zhig khrims ldan317 shes ldan bdag nyid bsgoms /
/ mnyam bzhag318 dran ldan drang por song ba dang /
/ gang gis mya ngan rims nad thams cad spangs /
/ yang dag dran ldan sems ni rnam grol gang /
/ de yis rig319 lhags skye bo bstan gyur cing /
/ lam dang nges par 'byung la'ang btsud pa yin /
/ de la gnas shing de la bslabs pa na /
/ mi 'di 'jig rten pha rol 'jigs med 'gyur /
(Peking 5536.267a4–267a7; Derge 4035.230b4–230b7)
[The deity asked:]
“Who teaches beings of superior lineage,
and sets them on the path of certain liberation? Where does one abide and where does one learn to become fearless in this and the next world?” [The Buddha replied:]
“The one who is completely liberated, virtuous, wise,


316 Derge: lhags.
317 khrims ldan is an abbreviation for tshul khrims dang ldan pa.
318 Derge: gzhag.
319 Derge: rigs.
 
concentrated, mindful, upright
and who has removed all the contagious illness of sorrow, with a mind that is perfectly mindful and liberated.
He who teaches beings of superior lineage, and sets them on the path to certain liberation, whoever abides and learns with such a person,
will become fearless in this and the next world.”


2. The narratives of verse(s) from corresponding sūtras with parallels


(1) SĀ-G
597: There are five verses containing twenty pādas.320 The setting is Shěwèiguó qíshù gěigūdú yuán 舍衛國祇樹給孤獨園, Anāthapiṇḍika’s Park of Jeta’s Grove in Śrāvastī. An unknown deity asked and the Buddha replied.
(2) SĀ-U
182: There are four verses containing eighteen pādas.321 The setting is Shěwèiguó qíshù gěigūdú yuán 舍衛國祇樹給孤獨園, Anāthapiṇḍika’s Park of Jeta’s Grove in Śrāvastī. An unknown deity asked and the Buddha replied.

(3) SN
2.2.4 : There are three verses containing fourteen pādas.322 The deity Nandana talked to the Buddha. The setting is not mentioned.323



320云何諸眾生 受身得妙色 云何修方便 而得乘出道 眾生住何法 為何所修習 為何等眾生 諸天所供養 (T99,
2.160a10–13); 持戒明智慧 自修習正受 正直心繫念 熾然憂悉滅 得平等智慧 其心善解脫 斯等因緣故 受身得妙色 成就乘出道 心住於中學 如是德備者 為諸天供養 (T99, 2.160a15–20).
321誰得色最勝 誰乘和合逝 當於其處住 習學何事業 是何等種類 而能供養天 (T100, 2.439a27–29); 持戒有智慧 善能修己者 念禪不放逸 除去四熱惱 正法意解脫 如此得上色 美妙獲最勝 和合斯乘道 應形彼處住習學於善法 若有如是人 名知供養天 (T100, 2.439b2–7).
322 kathaṃvidhaṃ sīlavantaṃ vadanti, kathaṃvidhaṃ paññavantaṃ vadanti |
kathaṃvidho dukkham aticca iriyati, kathaṃvidhaṃ devatā pūjayantī ti ||
‘yo sīlavā paññavā bhāvitatto, samāhito jhānarato satīmā |
sabb (SNN sabb') assa sokā vigatā pahīnā, khīṇāsavo antimadehadhārī ||
tathāvidhaṃ sīlavantaṃ vadanti, tathāvidhaṃ paññavantaṃ vadanti |
tathāvidho dukkham aticca iriyati, tathāvidhaṃ devatā pūjayantī ti || (SN I 53.1–53.12 = SNN 120.8–121.3). 323 Although this sūtra omits the location, it is under a chapter where in each sūtra, one deity addresses. The location, Rājagaha, is shown in the first sūtra of the chapter.
 
The Chinese texts do not name the deity; however, the Saṃyuttanikāya identifies him as Nandana. This deity talked to the Buddha, possibly in Śrāvastī. Comparing the content, what should be cultivated is much the same across these three texts and the Śarīrārthagāthā, but the results of this cultivativation are quite divergent. For example, one is completely liberated, virtuous, wise, concentrated, mindful, upright; however,
(1) in SĀ-G, such a person obtains a marvelous form and is worshiped by deities;
(2) in SĀ-U, such a person obtains a superior form and their conduct is the way to worship the deities;
(3) in SN, this person is worshiped by deities;
(4) in the Śarīrārthagāthā, such a person can guide superior kinds of human beings and not fear the next world.
Therefore, because of these divergences, this sixth verse set in the Śarīrārthagāthā might not really correspond to the verses in the above sūtras. The corresponding sūtra might not exist in the texts we witness. Alternatively, they do correspond, yet over time the content has changed.

3.1.7. The seventh set: obtaining fame (yù 譽324, yaśasvin, grags ldan); 8 pādas / 2 verses

1. Three versions with their translations325
(1) Chinese version:
云何擅名譽 云何具珍財 云何獲美稱 云何攝親友
持戒擅名譽 布施具珍財 諦實獲美稱 惠捨攝親友 (T1579, 30.375b5–8) [The deity asked:]
“How does one obtain name and fame? How does one have treasures?
How does one gain a compliment?
How does one gather close ones and friends?” [The Buddha replied:]

324 The origial word in the summary verse (uddāna) is yǔ 與 but it should be regarded as a scribal error. Yù 譽 is consistent with the Sanskrit and Tibetan versions and is also found in another Chinese Buddhist text colletion called Jīnlíng kè jīng 金陵刻經.
325 These three versions are essentially the same.
 
“By moral conduct one obtains name and fame. By donation one possesses treasures.
By truth one gains a compliment.
By giving one gathers close ones and friends.”


(2) Sanskrit version:
kathaṃ yaśasvī bhavati, kathaṃ bhavati bhogavān | kathaṃ kīrtim avāpnoti, kathaṃ mitrāṇi vindati || śīlād yaśasvī bhavati, dānād bhavati bhogavān | satyena kīrtim āpnoti, dadan mitrāṇi vindati || (Enomoto 1989: 26 [YBh 119a5f; ŚrBh 4a5.2f]) [The deity asked:]
“How does one become famous? How does one become wealthy? How does one obtain praise?
How does one acquire companions?” [The Buddha replied:]
“From moral conduct one becomes famous. From liberality one becomes wealthy.
By truth one obtains praise. Giving, one acquires companions.”

(3) Tibetan version:
/ ji ltar grags dang ldan par 'gyur /
/ ji ltar longs spyod can du 'gyur /
/ ji ltar snyan pa thob par 'gyur /
/ ji ltar mdza' bo rnyed par 'gyur /
/ tshul khrims kyis ni grags ldan 'gyur /
/ sbyin las326 longs spyod can du 'gyur /
/ bden pas snyan pa thob par 'gyur /
/ byin na mdza' bo rnyed par 'gyur /

326 Derge: pas.
 
(Peking 5536.268b5–6; Derge 4035.232a1–2)
[The deity asked:]
“How does one become famous? How does one become wealthy? How does one gain renown?
How does one find friends?” [The Buddha replied:]
“By moral conduct one becomes famous. By giving one becomes wealthy.
By truthfulness one becomes renowned.
If you are generous, you will find friends.”


2. The narratives of verse(s) from corresponding sūtras with parallels


(1) SĀ-G
1282: There are two verses containing eight pādas.327 The setting is Shěwèiguó qíshù gěigūdú yuán 舍衛國祇樹給孤獨園, Anāthapiṇḍika’s Park of Jeta’s Grove in Śrāvastī.
An unknown deity asked and the Buddha replied.


(2) SĀ-U
280: There are two verses containing eight pādas.328 The setting is Shěwèiguó qíshù gěigūdú yuán 舍衛國祇樹給孤獨園, Anāthapiṇḍika’s Park of Jeta’s Grove in Śrāvastī. An unknown deity asked and the Buddha replied.

These two texts share the same information; therefore, we may surmise that an unknown deity asked and the Buddha replied, and that the setting is Śrāvastī. In Pāli texts, there are similar verses in SN 10.12329 and Sn 185–188, but the content330 and the narratives of

327 云何得名稱 云何得大財 云何德流聞 云何得善友 (T99, 2.353a11–12); 持戒得名稱 布施得大財 真實德流聞 恩惠得善友 (T99, 2.353a14–15).
328 云何得名稱 云何得財業 云何得稱譽 云何得親友 (T100, 2.471b19–20); 持戒得名稱 布施得財寶 實語得稱譽 普施眾皆親 (T100, 2.471b22–23).
329 SN I 214.28–215.5 = SNN 462.21–463.18.
330 In SĀ-G and SĀ-U there are four questions with answers, but in SN and Sn there are five questions with answers, only three of which match the verses of SĀ-G and SĀ-U.
 
these verses are different. For example, in both SĀ-G and SĀ-U the one asking the Buddha is a diety and the setting is Śrāvastī, but in SN and Sn the one asking is a yakkha named Āḷavaka and the setting is Āḷavī (a city in Magadha). I am in agreement with Enomoto (1989: 26) that these two Pāli passages should not be regarded as direct parallels.

3.1.8. The eighth set: current331 (chí 池, sara, mtsho); 12 pādas / 3 verses

1. Three versions with their translations332
(1) Chinese version:
齊何泉止息 於何逕不通 世間諸苦樂 何處無餘滅若於是處所 眼耳及與鼻 舌身意名色 永滅盡無餘
齊此泉止息 於斯逕不通 世間諸苦樂 是處無餘滅 (T1579, 30.375b29–c5) [The deity asked:]
“From where does the spring cease? Where does the path not go on?
Where do all of the sufferings and happiness in the world, end without remainder?”
[The Buddha replied:] “If this is the place
where eye, ear, nose, tongue, body, mind, and name and form cease forever without remainder, this is where the spring ceases, where the path does not go on.
All of the sufferings and happiness in the world, this is the place where all ends without remainder. From this place the spring ceases,
this path does not go on.
All of the sufferings and happiness in the world,

331 Bodhi (2000: 103) referred to this set as “stream.” Ulrich (2013: 96) calls it “the analogy of wandering.”
332 These three versions are mostly the same in content.
 
it is this place where all ends without any remainder.”


(2) Sanskrit version:
kutaḥ sarā nivartante, kutra vartma na vartate |
kutra duḥkhasukhaṃ loke, niḥśeṣam uparudhyate || cakṣuḥ śrotraṃ tathā ghrāṇaṃ, jihvā kāyo manas tathā | yatra nāma ca rūpaṃ ca, niḥśeṣam uparudhyate ||
tataḥ sarā nivartante, tatra vartma na vartate | tatra duḥkhasukhaṃ loke, niḥśeṣam uparudhyate || (Enomoto 1989: 26 [YBh 118b4f; ŚrBh 4b5.1f]) [The deity asked:]
From where do the currents turn back? Where does the path not go on?
Where does suffering and happiness towards the world cease without remainder?
[The Buddha replied:] “Where eye, ear, nose, tongue, body, mind, name and form
cease without remainder,
from there it is that the currents turn back, there the path does not go on,
there the suffering and happiness towards the world ceases without remainder.”

(3) Tibetan version:
/ mtsho ni gang333 las ldog par 'gyur /
/ lam ni gang las ldog pa lags /
/ gang du 'jig rten sdug bsngal dang /
/ bde ba ma lus 'gag par 'gyur /
/ mig dang rna ba de bzhin sna /

333 Peking: gapa.
 
/ lce dang lus dang de bzhin yid /
/ ming dang gzugs ni gang334 zhig tu /
/ ma lus par ni 'gag 'gyur ba /
/ de las mtsho ni ldog 'gyur zhing /
/ lam ni de las ldog pa yin /
/ 'dir ni 'jig rten sdug bsngal dang /
/ bde ba ma lus 'gag par 'gyur /
(Peking 5536.269b3–5; Derge 4035.232b3–5)
[The deity asked:]
“From where do the currents turn back? Where does the path turn back?
Where do the suffering and happiness of the world cease without remainder?”
[The Buddha replied:] “Where eye, ear, and nose, tongue, body and mind name and form,
cease without remainder, There the currents turn back, there the path turns back,
there the suffering and happiness of the world cease without remainder.”

2. The narratives of verse(s) from corresponding sūtras with parallels


(1) SĀ-G
601: There are two verses containing fourteen pādas.335 The setting is Shěwèiguó qíshù gěigūdú yuán 舍衛國祇樹給孤獨園, Anāthapiṇḍika’s Park of Jeta’s Grove in Śrāvastī. An unknown deity asked and the Buddha replied.

334 Peking: gapa.
335 薩羅小流注 當於何反流 生死之徑路 於何而不轉 世間諸苦樂 何由滅無餘 (T99, 2.160c20–22); 眼耳鼻舌身 及彼意入處 名色滅無餘 薩羅小還流 生死道不轉 苦樂滅無餘 (T99, 2.160c24–26); 久見婆羅門 逮得般涅槃 一切怖已過 永超世恩愛 (T99, 2.160c28–29).
 
(2) SĀ-U
176: There are three verses containing twelve pādas.336 The setting is Shěwèiguó qíshù gěigūdú yuán 舍衛國祇樹給孤獨園, Anāthapiṇḍika’s Park of Jeta’s Grove in Śrāvastī. An unknown deity asked and the Buddha replied.

(3) SN
3.7: There are three verses containing ten pādas.337 No background information about the verses is supplied. The location is not mentioned.338

(4) Vibhāsā
The following texts all share the same information: A Brahman asked a question and the Buddha replied. The setting is not given.
T1545 (Āpídámó dàpípóshā lùn 阿毘達磨大毘婆沙, the Chinese
Abhidharmamahāvibhāsā): There are three verses containing twelve pādas.339
T1546 (Āpítán pípóshā lùn 阿毘曇毘婆沙論, the Chinese Abhidharmavibhāsā): There are two verses containing eight pādas.340
T1547 (Pípóshā lùn 鞞婆沙論, the Chinese Vibhāsā): There are two verses containing ten pādas.341

There are four divergences between these texts. Firstly, the two Saṃyuktāgama texts mention that the one asking is a deity but the vibhāsā texts all indicate the he is a


336 池水云何竭 有何流還返 世間之苦樂 何處都消盡 (T100, 2.438a8–9); 眼耳與鼻舌 并及於身意 名色都消盡 如是池枯竭 盡於諸結業 世間之苦樂 於斯盡無餘 亦無有還返 (T100, 2.438a11–14); 往昔已曾見 婆羅門涅槃 嫌怖久棄捨 能度世間愛 (T100, 2.438a16–17).
337 kuto sarā nivattanti, kattha vaṭṭaṃ na vattati |
kattha nāmañ ca rūpañ ca, asesaṃ uparujjhatī ti ||
yattha āpo ca pathavī (SNN paṭhavī), tejo vāyo na gādhati |
ato sarā nivattanti, ettha vaṭṭaṃ na vattati ||
ettha nāmañ ca rūpañ ca, asesaṃ uparujjhatī ti || (SN I 15.14–18 = SNN I 33.6–15).
338 Although this sūtra omits the location, it is under a chapter where in each sūtra, one deity addresses. The location, Rājagaha, is shown in the first sūtra of the chapter.
339 水從何池出 何處道不通 何處攝世間 苦樂等皆盡 (T1545, 27.379b3–4); 眼耳鼻舌身 意及諸餘處 此攝名及色 能令無有餘 水從此池出 此處道不通 此處攝世間 苦樂等皆盡 (T1545, 27.379b6–9).
340 何處泉水生 何處道不通 世間諸苦樂 何處得滅盡 (T1546, 28.284b6–7); 眼耳及與鼻 舌身及與意 此處盡名色 能令無有餘 (T1546, 28.284b9–10).
341泉從何轉 何轉不轉 何所苦樂 無餘滅盡 (T1547, 28.454a20–21); 眼耳及鼻 舌身并意 泉從是轉 此轉不轉此苦及樂 無餘滅盡 (T1547, 28.454a23–24).
 
Brahman. Secondly, in the two Chinese Saṃyuktāgama texts there is a final verse which may be translated “after a long time at last I see a Brahman who is fully quenched, who has surpassed all fears, has crossed over attachment to the world.” This final verse is not particularly relevant to the previous verses and is not found in other parallel texts.
Thirdly, the elements listed in the Saṃyuttanikāya (water, earth, fire and air) are different to those listed in the other parallel texts. Lastly, in the Pípóshālùn 鞞婆沙論 (the Chinese Vibhāsā), míngsè 名 色 (name and form) does not appear to be mentioned in the verses, unlike the other parallel texts.

3.1.9. The ninth set: (two) floods342 (liú 流, ogha, chu bo gnyis); 8 pādas / 2 verses

1. Three versions with their translations343
(1) Chinese version:
誰能越暴流 誰能超大海 誰能捨眾苦 誰能得清淨
正信越暴流 無逸超大海 精進捨眾苦 智慧得清淨 (T1579, 30.375c27–376a1) [The yakṣa asked:]
“Who can cross the violent flood? Who can traverse the great ocean? Who can discard all sufferings?
Who can obtain purification?” [The Buddha replied:]
“One with correct faith crosses the violent flood. One with carefulness344 traverses the great ocean. One with diligence discards all sufferings.
One with wisdom obtains purification.”



342 Ulrich (2013: 96) takes the eighth and ninth sets as one set and refers to them as “the analogy of the two floods.”
343 These three versions are essentially the same.
344 In Chinese, fàngyì 放 逸 means “indulgence.” But in Buddhist literature, it normally corresponds to pramāda in Sanskrit and bag-med-pa in Tibetan, so fàngyì 放逸 should be interpreted as “carelessness” or “negligence.” “Indulgence” is the situation in which the Buddhist practitioner does not pay attention to the mind, like cultivating goodness and repelling evil (See T1579, 30.379b–3). Therefore wúyì 無 逸 , “non- indulgence,” is the meaning of “vigilance” and “carefulness.” I will translate “carelessness” for fàngyì 放逸, and “carefulness” for wúyì 無逸.
 
(2) Sanskrit version:
kena svid oghaṃ tarati, kenottarati cārṇavam | duḥkhaṃ tyajati kena svit, kena svit pariśudhyati || śraddhayā tarati hy ogham, apramādena cārṇavam | vīryeṇa duḥkhaṃ tyajati, prajñayā pariśudhyati || (Enomoto 1989: 27 [YBh 120a3; ŚrBh 4a6.2])
[The yakṣa asked:]
“By what does one cross over the flood and by what does one cross the ocean? By what does one abandon suffering? By what does one purify [oneself]?” [The Buddha replied:]
“By faith one crosses the flood
and by vigilance [one crosses] the ocean. By energy one abandons suffering.
By wisdom one purifies [oneself].”


(3) Tibetan version:
/ gang gis chu bo brgal345 bar 'gyur /
/ gang gis rgya mtsho gal bar 'gyur /
/ gang gis sdug bsngal spong bar 'gyur /
/ gang gis yongs su dag par 'gyur /
/ dad pas chu bo rgal bar 'gyur /
/ bag yod pas ni rgya mtsho rgal /
/ brtson 'grus kyis ni sdug bsngal spong /
/ shes rab kyis ni yongs su 'dag346 /
(Peking 5536.270a8–b2; Derge 4035.233a5–6) [The yakṣa asked:]
“By what does one cross the river? By what does one cross the ocean?

345 Derge: rgal.
346 Derge: dag.
 
By what does one eliminate suffering? By what is one wholly purified?” [The Buddha replied:]
“By faith one crosses the river.
By carefulness one crosses the ocean. By diligence one eliminates suffering. By wisdom one is wholly purified.”

2. The narratives of verse(s) from corresponding sūtras with parallels


(1) SĀ-G
603: There are two verses containing eight pādas.347 The setting is Shěwèiguó qíshù gěigūdú yuán 舍衛國祇樹給孤獨園, Anāthapiṇḍika’s Park of Jeta’s Grove in Śrāvastī. An unknown deity asked and the Buddha replied.
1326: There are three verses contaning twelve pādas.348 The setting is Magadha. The
yakṣa Ālà 阿臈 (Āṭavakā) asked and the Buddha replied.
1329: There are two verses containing eight pādas.349 The setting is Wángshěchéng jiālántuó zhúyuán 王舍城迦蘭陀竹園, Karaṇḍa’s garden in Rājagṛha. The two yakṣas Suōduōqílì 娑多耆利 and Xīmóbōdī 醯魔波低 (Sātāgiri and Haimavata) asked and then the Buddha replied.

(2) SĀ-U
325: There are two verses containing eight pādas.350 The setting is Magadha. A yakṣa
Kuàngyě 曠野 (Āṭavakā) asked and then the Buddha replied.






347 云何度諸流 云何度大海 云何能捨苦 云何得清淨 (T99, 2.161a26–27); 信能度諸流 不放逸度海 精進能除苦 智慧得清淨 (T99, 2.161a29–b1).
348 以何法度流 以何度大海 以何捨離苦 以何得清淨 (T99, 2.365a9–10); 以信度河流 不放逸度海 精進能除苦 以慧得清淨 汝當更問餘 沙門梵志法 其法無有過 真諦施調伏 (T99, 2.365a12–15);
349 何法度諸流 以何度大海 云何捨離苦 云何得清淨 (T99, 2.367a9–11); 正信度河流 不放逸度海 精進能斷苦 智慧得清淨 (T99, 2.367a13–14).
350 誰渡於駛流 誰渡於大海 誰能捨離苦 誰得於清淨 (T100, 2.483a7–8); 信能渡駛流 不放逸渡海 精進能離苦 智慧能清淨 (T100, 2.483a10–11).
 
328: There are three verses containing fourteen pādas.351 The setting is Wángshěchéng jiālántuó zhúlín 王舍城迦蘭陀竹林, Karaṇḍa’s garden in Rājagṛha. A yakṣa Xuěshān 雪山 (Haimavata) asked and the Buddha replied.

(3) SN
10.12: There are two verses containing eight pādas.352 The setting is Āḷavī (a city in Magadha).The yakkha Āḷavaka asked and the Buddha replied.

(4) Gēnběnshuōyīqiēyǒubù pínàiyē 根本說一切有部毘奈耶 (the Chinese
Mālasarvāstivāda-vinaya)
T1442: There are three verses containing twelve pādas.353 The setting is the Wángshěchéng jiālántuó zhúlínyuán 王舍城迦蘭陀竹林園, Karaṇḍa’s garden in Rājagṛha. An unkown yakṣa asked and the Buddha replied.

These texts have divergent information; however, SĀ-G, SĀ-U, and SN all mention that a yakṣa called Āṭavakā asked questions and the Buddha replied; and that the setting is Magadha.

3.1.10. The tenth set: (two) floods (liú 流, ogha, chu bo gnyis); 12 pādas / 3 verses

1. Three versions with their translations354
(1) Chinese version:

誰超越暴流
圓滿眾尸羅 晝夜無惛昧
具慧善安定 於無攀無住
內思惟繫念 甚深無減劣
能度極難度
諸欲想離染 亦超色界結 彼無攀無住 甚深無減劣 (T1579, 30.376b5–10)

351 誰度於駛流 孰能越大海 誰能捨於苦 云何得清淨 (T100, 2.485a4–6); 信能度駛流 不放逸越海 精進能捨苦 智慧能使淨 汝詣諸沙門 及諸婆羅門 各各種別問 誰有知法者 誰能說實捨 離我誰能說 (T100, 2.485a8–12).
352 kathaṃ su tarati oghaṃ, kathaṃ su tarati aṇṇavaṃ | kathaṃ su dukkham acceti, kathaṃ su parisujjhatī ti || saddhāya tarati oghaṃ, appamādena aṇṇavaṃ |
vīriyena dukkham acceti, paññāya parisujjhatī ti || (SN I 214.25–28 = SNN 462.11–19).
353 誰能渡瀑流 誰能越大海 誰能離諸苦 誰得心清淨 (T1442, 23.884c16–18); 信能渡瀑流 謹慎越大海 精勤離諸苦 有慧心清淨 汝今咸可問 沙門婆羅門 離實語布施 更有勝法不 (T1442, 23.884c20–23).
354 These three versions are essentially the same.
 
[The deity asked:]
“Who can cross over the violent flood without confusion day and night?
In being without grasping and dwelling,
[who can achieve the] deep [[[state]]] without degeneration?” [The Buddha replied:]
“One who perfects moral conduct,
possesses wisdom, is skilled in concentration, is inwardly contemplative and mindful;
he can traverse that which is extremely difficult to traverse. One who detaches from the impure in all sensual thoughts, also transcends the fetters of the realm of form
without grasping and dwelling,
he [achieves the] deep [[[state]]] without degeneration.”


(2) Sanskrit version:
ka etam oghaṃ tarati, rātriṃdivam atandritaḥ | anālambe 'pratiṣṭhe ca, ko gambhīre na sīdati || sarvataḥ śīlasaṃpannaḥ, prajñāvān susamāhitaḥ | adhyātmacintī smṛtimāṃs, taratīmaṃ sudustaram || viraktaḥ kāmasaṃjñābhyo, rūpasaṃyojanātigaḥ | anālambe 'pratiṣṭhe ca, sa gambhīre na sīdati || (Enomoto 1989: 27 [YBh 120b4f; ŚrBh 4b6.1f]) [The deity asked:]
“Who crosses over this flood, unwearied day and night?
Who does not sink in the deep,
that is bereft of support and without a [safe] abode.” [The Buddha replied:]
“The one thoroughly endowed with moral conduct, [and] wise, well concentrated,
who inclines the mind towards inner contemplation and is mindful,
 
crosses this [flood] which is very difficult to cross. One who is unattached to sensual perceptions,
has surpassed the fetters of form. He does not sink in the deep,
that is bereft of support and without a [safe] abode.”


(3) Tibetan version:
/ su zhig gdugs mtshan mi g.yol355 bar /
/ chu bo 'di las rgal bar bgyid356 /
/ mi dmigs mi gnas zab mo la /
/ su zhig sgyid lug mi 'gyur lags /
/ rnam pa kun tu357 tshul khrims ldan /
/ shes rab ldan zhig mnyam bzhag358 la /
/ nang du sems shing dran ldan pa /
/ shin tu brgal359 dka' 'di las brgal360 /
/ 'dod pa'i 'du shes chags pa med /
/ gzugs gyi kun du sbyor las 'das /
/ mi dmigs mi gnas zab mo la /
/ de ni sgyid lug mi 'gyur ro /
(Peking 5536.271b6–8; Derge 4035.234a6–b1) [The deity asked:]
“Who, unflagging day and night, crosses over this flood?
Who does not become disheartened in the deep, which is bereft of support, lacking a [safe] abode?” [The Buddha replied:]
“One who is endowed with moral conduct in all circumstances, has wisdom and is well concentrated,

355 Derge: g.yel.
356 Derge: gyis.
357 Derge: dang.
358 Derge: gzhag.
359 Derge: rgal.
360 Derge: rgal.
 
inwardly contemplates and has mindfulness.
[Such a one] crosses that which is so difficult to cross, without attachment to sensual perceptions,
gone beyond fetters of form.
Such a one does not become disheartened in the deep, which is bereft of support, lacking a [safe] abode.”

2. The narratives of verse(s) from corresponding sūtras with parallels


(1) SĀ-G
1269: There are three verses containing twelve pādas.361 The setting is Shěwèiguó qíshù gěigūdú yuán 舍衛國祇樹給孤獨園, Anāthapiṇḍika’s Park of Jeta’s Grove in Śrāvastī. An unknown deity asked and the Buddha replied.
1326: There are three verses containing twelve pādas.362 The setting is Magadha. The
yakṣa Ālà 阿臈 (Āṭavakā) asked and the Buddha replied.
1329: There are three verses containing twelve pādas.363 The setting is Wángshěchéng jiālántuó zhúyuán 王舍城迦蘭陀竹園, Karaṇḍa’s garden in Rājagṛha. The two yakṣas Suōduōqílì 娑多耆利 and Xīmóbōdī 醯魔波低 (Sātāgiri and Haimavata) asked and then the Buddha replied.
1316: There are three verses containing twelve pādas.364 The setting is Shěwèiguó qíshù gěigūdú yuán 舍衛國祇樹給孤獨園, Anāthapiṇḍika’s Park of Jeta’s Grove in Śrāvastī. An unknown deity asked and the Buddha replied.









361 誰度於諸流 晝夜勤精進 不攀亦不住 何染而不著 (T99, 2.348c13–14); 一切戒具足 智慧善正受 內思惟繫念 度難度諸流 不樂於欲想 超越於色結 不繫亦不住 於染亦不著 (T99, 2.348c16–19).
362 誰能度諸流 晝夜勤方便 無攀無住處 孰能不沈沒 (T99, 2.364c29–365a2); 一切戒具足 智慧善正受 正念內思惟 能度難度流 不樂於五欲 亦超度色愛 無攀無住處 是能不沒溺 (T99, 2.365a4–7).
363 云何度諸流 日夜勤方便 無攀無住處 而不溺深淵 (T99, 2.367a1–3); 一切戒具足 智慧善正受 如思惟繫念 是能度深淵 不樂諸欲想 亦超色諸結 無攀無住處 不溺於深淵 (T99, 2.367a5–8).
364 誰度於諸流 晝夜勤不懈 不攀無住處 云何不沒溺 (T99, 2.361a27–28); 一切戒具足 智慧善正受 內思惟正念 能度[[[離]]>難]度流 不染此欲想 超度彼色愛 貪喜悉已盡 不入於難測 (T99, 2.361c1–4).
 
(2) SĀ-U
178: There are three verses containing sixteen pādas.365 The setting is Shěwèiguó qíshù gěigūdú yuán 舍衛國祇樹給孤獨園, Anāthapiṇḍika’s Park of Jeta’s Grove in Śrāvastī. An unknown deity asked and the Buddha replied.
315: There are four verses containing sixteen pādas.366 The setting is Shěwèiguó qíshù gěigūdú yuán 舍衛國祇樹給孤獨園, Anāthapiṇḍika’s Park of Jeta’s Grove in Śrāvastī. A deity called Zhāntán 栴檀 (Candana) asked and the Buddha replied.
328: There are three verses containing sixteen pādas.367 The setting is the Wángshěchéng jiālántuó zhúlín 王舍城迦蘭陀竹林, Karaṇḍa’s garden in Rājagṛha.The yakṣa Xuěshān 雪山 (Haimavata) asked and the Buddha replied.

(3) SN
2.2.5 : There are three verses containing twelve pādas.368 The setting is Sāvatthī. The deity Candana talked to the Buddha.

(4) Suttanipāta
173–175: There are three verses containing twelve pādas.369 The yakkha Hemavata asked questions and the Buddha replied. The setting is not provided.


365 都無所緣攀 又無安足處 甚深洪流中 誰能不沈沒 誰有勤精進 能度瀑駛流 (T100, 2.438b12–14); 淨持於禁戒 修智及禪定 觀察內身念 難度而得度 得離於欲結 出過色有使 盡於歡喜有 如是能履深 而不為沒溺能度瀑駛流 (T100, 2.438b16–20).
366 云何度駛流 晝夜恒精進 如此駛流中 濤波甚暴急 無有攀挽處 亦無安足地 誰能處深流 而不為漂沒(T100, 2.479c20–23); 一切戒完具 定慧充其心 思惟內心念 此能度難度 除去於欲想 度有結使流 盡於喜愛有 處深不沈沒 (T100, 2.479c25–28).
367 云何修善法 晝夜不懈怠 云何度駛流 無有安足處 亦無所攀緣 處深不沈沒 (T100, 2.484c24–26); 一切戒無犯 智慧具禪定 思惟眾過患 具足於念力 此能度難度 遠離欲和合 捨諸有結使 盡於歡喜有 如是人名為處深不沈沒 (T100, 2.484c28–485a3).
368 ko sū 'dha tarati oghaṃ, rattindivam atandito | appatiṭṭhe anālambe, ko gambhīre na sīdatī ti || sabbadā sīlasampanno, paññavā susamāhito | āraddhaviriyo pahitatto, oghaṃ tarati duttaraṃ || virato kāmasaññāya, rūpasaṃyojanātigo |
nandīrāgaparikkhīṇo (SNN nandirāgaparikkhīṇo), so gambhīre na sīdatī ti || (SN I 53.16–21 = SNN 121.8– 122.2).
369 ko sū 'dha tarati oghaṃ, ko 'dha tarati aṇṇavaṃ | appatiṭṭhe anālambe, ko gambhīre na sīdati || sabbadā sīlasampanno, paññavā susamāhito | ajjhattacintī satimā, oghaṃ tarati duttaraṃ ||
virato kāmasaññāya, sabbasaṃyojanātigo |
 
The information among these texts seems divergent. However, SĀ-G, SĀ-U and SN agree that a deity (perhaps called Candana) asked questions and the Buddha replied; and that the setting is Śrāvastī.

3.1.11. The eleventh set: lust and hatred370 (tān 貪, rāgadveṣa, ‘dod chags zhe sdang); 16
pādas / 4 verses

1. Three versions with their translations371
(1) Chinese version:
貪恚何因緣 由何故欣慼 毛豎意尋思 如孩依乳母潤所生自生 如諾瞿陀樹 別縛於諸欲 猶摩迦處林是貪恚因緣 由斯故欣慼 毛豎意尋思 如孩依乳母
知彼彼因緣 生已尋除滅 超昔未超海 暴流無後有 (T1579, 30.376c11–18) [The yakṣa Sūciroma asked:]
“What are the causes and conditions of lust and anger? For what reasons does one become happy or sad,
have one’s hair stand on end, or have a discursive mind, like a child relying on the wet nurse (rǔmǔ 乳 母 )?” [The Buddha replied:]
“That which moistens the produced and the self-produced are like the Nyagrodha tree;372
they separately cling to all sensual pleasures, like Mālutā plants in the forest.
These are the causes and conditions of lust and anger, due to them one becomes happy or sad,
has one’s hair stand on end or has a discursive mind, like a child relying on the wet nurse.
Having known their causes and conditions,

nandībhavaparikkhīṇo, so gambhīre na sīdati ||
370 In the Chinese uddāna of the Śarīrārthagāthā the title of this verse set is “lust” (tān 貪) without “hatred.” 371 There is no mention of a yakṣa in the Chinese root text or commentary. Besides this, the three versions are similar.
372 These trees send down aerial roots, creating multiple trunks. See MW s.v. nyagrodha.
 
those produced are instantly eliminated.
One crosses the ocean not crossed in the past [and] the violent flood for no renewed existence.”

(2) Sanskrit version:
rāgadveṣau bhagavan kiṃnidānāv, aratiratī romaharṣaḥ kuto 'yam | kutaḥsamutthāś ca mano vitarkāḥ, kumārakā dhātrīm ivāśrayante || snehajā ātmasaṃbhūtā, nyagrodhaskandhakā yathā | pṛthagviṣaktāḥ kāmeṣu, mālutā vā latā vane ||
rāgaś ca dveṣaś ca itonidānāv, aratiratī romaharṣaḥ ito 'yaṃ | itaḥsamutthāś ca mano vitarkāḥ, kumārakā dhātrīm ivāśrayante || ye tān prajānanti yatonidānāṃs, te tāñ janā yakṣa vinodayanti |
te arṇavaṃ saṃpratarantīhaugham, atīrṇapūrvam apunarbhavāya ||
(Enomoto 1989: 27 [YBh 121a5f; ŚrBh 4a7.1f]) [The yakṣa Sūciroma asked:]
Blessed one, what are the causes of lust and hatred?
From what [[[cause]]] is there non-delight and delight, and does hair stand on end? From what [[[cause]]] do conjectures rise to the mind,
as boys resort to their mother? [The Buddha replied:]
“[They are] born by moisture [and] generated by oneself, like the trunks of the Nyagrodha.
[They] separately cling to sensual pleasures, like Mālutā vines in the forest.
Lust and hatred are the causes here.
From this [[[cause]]] is there non-delight and delight and hair stands on end. From this [[[cause]]] conjectures rise to the mind,
as boys resort to their mother.
They who understand from where these are the causes, those people dispel them, O Yakṣa.
In this life they cross over the ocean, the flood, previously uncrossed, for the non-renewal of being.”
 
(3) Tibetan version:
/ 'dod chags zhe sdang gnyis kyi gzhi ni gang zhig lags /
/ dga' dang mi dga' spu zing bgyid pa gang las byung /
/ yid kyi rtog pa rnams kyang gang las byung ba lags /
/ nya gro dha yi sdong po ltar /
/ rlan las skyes shing bdag las byung /
/ nags la ma lus373 'i khri374 shing ltar /
/ 'dod pa rnams la so sor 'phro /
/ byis pa dag ni ma ma dag la brten pa ltar /
/ 'dod chags zhe sdang gnyis kyi gzhi ni de las byung /
/ dga' dang mi dga' spu zing byed pa de las byung /
/ yid kyi rtog pa rnams kyang de las byung ba yin /
/ gnod sbyin de dag gang las byung ba'i gzhi shes pa'i /
/ skye bo gang yin de rnams de dag spang gyur zhing /
/ de dag 'dir ni yang srid med par bya ba'i phyir /
/ sngan375 chad ma brgal376 rgya mtsho chu bo brgal377 bar 'gyur
(Peking 5536.273a5–8; Derge 4035.235a7–b2) [The yakṣa Sūciroma asked:]
“What indeed is the basis of both lust and hatred?
What is it that makes joy and sadness, [as well as] the hairs stand [on end]? From where do concepts of the mind also arise?”
[The Buddha replied:]
“Like the trunk of a Nyagrodha tree,
it is born from moisture and arises from itself. Just like creeping vines in the forest,
they pervasively spread out to the sense objects as children rely upon their mothers,
lust and hatred both arise from this root.


373 Peking: lu.
374 Derge: 'khri not 'i khri.
375 Derge: sngon.
376 Derge: rgal.
377 Derge: rgal.
 
Hairs stand [on end as well as] joy and sadness [[[exist]]] because of this. And the concepts of the mind also arise from this.
O Yakṣa, knowing the basis from which these arise, a sentient being is able to abandon them all.
In order to have no continued existence,
cross the ocean [and] flood not been crossed before.”


2. The narratives of verse(s) from corresponding sūtras with parallels


(1) SĀ-G
1314: There are three verses containing fourteen pādas.378 The setting is Shěwèiguó qíshù gěigūdú yuán 舍衛國祇樹給孤獨園, Anāthapiṇḍika’s Park of Jeta’s Grove in Śrāvastī. A deity called Jiāmó 迦摩 (Kāma) asked and the Buddha replied.
1324: There are four verses containing sixteen pādas.379 The setting is Magadha. The
yakṣa Zhēnmáo 針毛 (Sūciroma) asked and the Buddha replied.


(2) SĀ-U
313: There are four verses containing eighteen pādas.380 The setting is Shěwèiguó qíshù gěigūdú yuán 舍衛國祇樹給孤獨園, Anāthapiṇḍika’s Park of Jeta’s Grove in Śrāvastī. A deity called Jiāmò 迦默 (Kāma) asked and the Buddha replied.
323: There are seven verses containing twenty two pādas.381 The setting is Magadha.
The yakṣa Jiànmáo 箭毛 (Sūciroma) asked and the Buddha replied.





378 貪恚何所因 不樂身毛竪 恐怖從何起 覺想由何生 猶如鳩摩羅 依倚於乳母 (T99, 2.361a27–b1); 愛生自身長 如尼拘律樹 處處隨所著 如榛綿叢林 若知彼因者 發悟令開覺 度生死海流 不復更受有 (T99, 2.361b3–6).
379 一切貪恚心 以何為其因 不樂身毛竪 恐怖從何起 意念諸覺想 為從何所起 猶如新生兒 依倚於乳母(T99, 2.363c18–22); 愛生自身長 如尼拘律樹 展轉相拘引 如籐綿叢林 若知彼所因 當令鬼覺悟 度生死海流 不復重增有 (T99, 2.363c24–27).
380 貪欲及瞋恚 以何為根本 樂不樂恐怖 為是而誰耶 孾孩捉母乳 (T100, 2.479b21–23); 意覺從何生 從愛至我心 如尼拘陀樹 根鬚從土生 然後入于地 各各於異處 愛著生於欲 亦如摩樓多 纏縛覆林樹 若知其根本夜叉當捨離 能渡生死海 度更不受有 (T100, 2.479b25–c1).
381貪欲瞋恚 以何為本 樂及不樂 怖畏毛竪 為是何耶 彼意覺者 住在何處 孾孩小兒 云何生便知捉於乳(T100, 2.482a3–6); 愛從以我生 如尼拘陀樹 愛隨所著 亦如摩樓多 纏縛尼拘樹 夜叉應當知 若知其根本必能捨棄離 [如>知]彼根本者 能度生死海 度於有駛流 更不受後有 (T100, 2.482a8–13).
 
(3) SN
10.3: There are four verses containing sixteen pādas.382 The setting is Gayā. The
yakkha Sūciloma asked, and the Buddha replied.


(4) Suttanipāta
270–273:383 There are four verses containing sixteen pādas.384 The setting is Gayā. A
yakkha Sūciloma asked and the Buddha replied.


There are two versions in SĀ-G and SĀ-U, but in the Yogacārabhūmi it clearly mentions the word yakṣa, so for this eleventh verse set, we may assume that the background information is that the yakṣa Sūciroma asked and the Buddha replied in Magadha. But this is different from the Pāli texts, which mention that the location is Gayā.
The Pāli texts are also different in another respect. In the fourth pāda of the third verse, they state “as boys toss up a crow,” while this eleventh set in the Yogacārabhūmi and the Saṃyuktāgama state “as boys rely on their mother.”










382 rāgo ca doso ca kuto nidānā, aratī ratī lomahaṃso kutojā |
kuto samuṭṭhāya manovitakkā, kumārakā dhaṅkam iv' ossajantī ti ||
rāgo ca doso ca ito nidānā, aratī ratī lomahaṃso itojā |
ito samuṭṭhāya manovitakkā, kumārakā dhaṅkam iv' ossajanti ||
snehajā attasambhūtā, nigrodhasseva khandhajā |
puthū visattā kāmesu, māluvāva vitatā vane ||
ye naṃ pajānanti yato nidānaṃ, te naṃ vinodenti suṇohi yakkha |
te duttaraṃ ogham imaṃ taranti, atiṇṇapubbaṃ apunabbhavāyā ti || (SN I 207.25–208.4 = SNN 446.11– 447.8).
383 The Nettipakaraṇa cites these verses (Nett 147.12–27).
384 rāgo ca doso ca kuto nidānā, aratī ratī lomahaṃso kutojā |
kuto samuṭṭhāya manovitakkā, kumārakā dhaṅkam iv' ossajantī ti ||
rāgo ca doso ca ito nidānā, aratī ratī lomahaṃso itojā |
ito samuṭṭhāya manovitakkā, kumārakā dhaṅkam iv' ossajanti ||
snehajā attasambhūtā, nigrodhasseva khandhajā |
puthū visattā kāmesu, māluvāva vitatā vane ||
ye naṃ pajānanti yato nidānaṃ, te naṃ vinodenti suṇohi yakkha |
te duttaraṃ ogham imaṃ taranti, atiṇṇapubbaṃ apunabbhavāyā ti ||
 
3.1.12. The twelfth set: what ought to be done385 (zuò 作, kāryatā, bya ba); 20 pādas / 5
verses 386

1. Three versions with their translations387
(1) Chinese version:
應作婆羅門 謂斷無縱逸 求棄捨諸欲 不希望此有若更有所作 非真婆羅門 當知婆羅門 於所作已辦諸身分劬勞 未極底未度 已得度住陸 無勤到彼岸天汝今當知 此喻真梵志 謂永盡諸漏 得常委靜慮
彼永斷一切 愁憂及熾燃 恒住於正念 亦常心解脫 (T1579, 30.377b16–25) [The deity Dāmani said:]
“This should be done by the Brahman: to strive without slackness,
one seeks to discard all desires
and does not yearn for this existence. [The Buddha said:]
“If one has more to be done, he is not a real Brahman;
one should know that a Brahman has done what is to be done.
All the body parts are struggling together; without reaching the bottom, he has not crossed. Having crossed, he stands on the ground

385 Enomoto (1989: 21) and Ulrich (2013: 96) both referred to this set as “what ought to be done.”
386 In the Sanskrit and Tibetan versions, the number of pādas is fourteen, and Enomoto (1989) arranged these pādas into four verses but he does not explain the reason for doing so.
387 The Chinese and Tibetan versions each refer to a deity, but do not offer the name. In the Sanskrit version, the name Dāmani is provided but it is not clear whether this is the name of the diety. Another difference is that the number of verse pāda in Chinese (20) is more than that in Sanskrit and Tibetan (14). The structures of the Sanskrit and Tibetan versions are similar as well. The possible reason to explain the difference
between the Chinese and Sanskrit/Tibetan is that the manuscript that Xuánzàng 玄 奘 used was originally different. More importantly, this verse set seems to be an answer to the sixth verse set. The sixth set asks “Who can guide superior kinds of human beings?” The description of the one who can guide is the same as
that described in the twelfth set, namely, wise, has acquired meditation, all the afflictions of sorrow have been abandoned, has right mindfulness and whose mind is liberated. The twelfth set labels such a person as a “Brahman.”
 
and reaches the other shore without effort.
Deity! You should now know
that this is the simile of the real Brahman who permanently eliminates all taints
and attained constant comprehension and meditation. He cuts off all worries and burnings,
dwells forever in right mindfulness and is also always mentally liberated.”

(2) Sanskrit version:
kāryam etad brāhmaṇena, prahāṇam akilāsinā | kāmānāṃ viprahāṇārthaṃ, na hi kāṃkṣaty asau bhavam || na kāryaṃ brāhmaṇasyāsti, kṛtārtho brāhmaṇaḥ smṛtaḥ ||
yāvan na gādhaṃ labhate na tīram, āyūhate sarvagātraiḥ sa tāvat | tīraṃ tu labdhveha saṃtiṣṭhati sthale, nāyūhate pāragato nirucyate || eṣopamā dāmane brāhmaṇasya, kṣīṇāsravo yo nipako dhyānalābhī |
sarve 'sya śokajvarathāḥ prahīṇāḥ, samyaksmṛto yasya cittaṃ vimuktam ||
(Enomoto 1989: 28 [YBh 122a4f; ŚrBh 4b7.7–4a8.1]) [The deity Dāmani said:]
“This should be done by a Brahman:
exertion by one who is unwearied for sake of abandoning desires,
for he does not desire this existence.” [The Buddha said:]
“A Brahman does not have a [further task] to be accomplished,
a Brahman is known as the one whose purpose has been achieved. As long as he has not found the ford, nor the bank,
he struggles with all his limbs.
But having found the bank, he stands here on the ground.
He does not struggle,388 he is said to be one who has gone beyond.


388 Nāyūhate can also mean “not strive” which is close to the word’s meaning in Chinese and Tibetan. I choose “struggle” by the context because it is mentioned previously.
 
Dāmani, this is a simile of the Brahman whose taints are destroyed, who is wise, has acquired meditation.
All the afflictions of sorrow have been abandoned
for he who has right mindfulness [and] whose mind is liberated.”


(3) Tibetan version:
/ de ni srid pa mi 'tshal bas /
/ 'dod pa rnams ni spang389 slad du /
/ bram ze ngal ba ma mchis pas /
/ spang ba de ni bgyi ba lags /
/ don byas bram ze 'dod pa'i phyir /
/ bram ze la ni bya ba med /
/ ji srid gting dang 'gram du ma phyin pa /
/ de srid de ni lus rnams kun gyis rtsol /
/ brgal390 nas de ni 'di na thang la gnas /
/ rtsol ba med pa pha rol phyin ces bya /
/ gang zhig zag zad rtag 'grus bsam gtan thob /
/ de yis rims nad mya ngan kun spangs shing /
/ gang zhig yang dag dran sems grol pa yi /
/ bram ze'i dper bya ba ni 'di yin lha /
(Peking 5536.275a7–b2; Derge 4035.237a2–4) [The deity Dāmani said:]
“Not looking for further existence, by eliminating all desires
the Brahman, without weariness, accomplishes abandonment indeed.” [The Buddha said:]
“The purpose has been achieved by the Brahman, on account of [having obtained] what he wished for;
for a Brahman there are no more tasks to be done.


389 Derge: spangs.
390 Derge: rgal.
 
Until he has touched the ford and reached the shore, he keeps striving with each and every part of his body. Having crossed over, he will stand here on the ground, for he is said to have gone beyond without effort.
Deity! This is the simile of a Brahman
whose taints are destroyed, who is wise and acquired meditation, who has removed all the contagious illness of sorrow,
who has right mindfulness [and] whose mind is liberated.”


2. The narratives of verse(s) from corresponding sūtras with parallels


(1) SĀ-G
1311: There are four verses containing sixteen pādas.391 The setting is Shěwèiguó qíshù gěigūdú yuán 舍衛國祇樹給孤獨園, Anāthapiṇḍika’s Park of Jeta’s Grove in Śrāvastī. A deity called Tuómóní 陀摩尼 (Dāmani) spoke a verse and the Buddha replied.

(2) SĀ-U
310: There are five verses containing twenty pādas.392 The setting is Shěwèiguó qíshù gěigūdú yuán 舍衛國祇樹給孤獨園, Anāthapiṇḍika’s Park of Jeta’s Grove in Śrāvastī. A deity called Tánmóshī 曇摩尸 (Dāmani) spoke a verse and the Buddha replied.

(3) SN
2.1.5: There are three verses containing fourteen pādas.393 The setting is Sāvatthī. A deity Dāmali recited a verse and then the Buddha replied.

391 為婆羅門事 學斷莫疲惓 斷除諸愛欲 不求受後身 (T99, 2.360c8–9); 婆羅門無事 所作事已作 乃至不得岸 晝夜常勤跪 已到彼岸住 於岸復何跪 此是婆羅門 專精漏盡禪 一切諸憂惱 熾然永已斷 是則到彼岸 涅槃無所求 (T99, 2.360c11–16).
392 婆羅門今者 斷三有欲結 不願求諸有 竟何為所作 (T100, 2.478c29–479a1); 婆羅門無作 念作已終訖 涉水足盡底 以至于彼岸 若足不盡底 不能到彼岸 手足必運動 是名有所作 以此為方喻 以明無作義 曇摩汝當知 已盡於諸漏 住於最後身 諸有愛欲過 一切悉斷除 超渡生死海 (T100, 2.479a6–a13);
393 karaṇīyam etaṃ brāhmaṇena, padhānaṃ akilāsunā |
kāmānaṃ vippahānena na, tenāsiṃsate (SNN ten' āsiṃsate) bhavan ti || natthi kiccaṃ brāhmaṇassa, dāmalīti bhagavā, katakicco hi brāhmaṇo | yāva na gādhaṃ labhati nadīsu, āyūhati sabbagattehi jantu |
 
These three texts share the same information, namely, in Śrāvastī the deity Dāmani spoke a verse and the Buddha replied. However, there is one difference between the Yogacārabhūmi and the above texts. The last pāda of this verse set of the Śarīrārthagāthā (samyaksmṛto yasya cittaṃ vimuktam, “for one who has right mindfulness [and] whose mind is liberated”), differs from the counterpart in
1. SĀ-G: 是則到彼岸 涅槃無所求 (has reached the other shore, [and] nirvāṇa; has nothing to ask for)
2. SĀ-U: 超渡生死海 (has crossed over the sea of birth and death)
3. SN: pappuyya jātimaraṇassa antaṃ n' āyūhatī pāragato hi so ti (having reached the end of birth and death, he need not strain for he has gone beyond).394

3.1.13. The thirteenth set: struggle395 (qú láo 劬勞, āyūhana, ‘dzin pa); 4 pādas / 1 verses

1. Three versions with their translations396
(1) Chinese version:
苾芻苾芻 已度暴流耶 告言如是天無攀無住 已度暴流耶 告言如是天
苾芻 汝今猶如何等無攀無住 已度暴流
如如我劬勞 如是如是劣 如如我劣已 如是如是住 如如我住已 如是如是漂天 我如如捨劬勞 如是如是無減劣
如是廣說鮮白品 此中祇焰頌應知 (T1579, 30.378a3–11)
[The deity asked:] “Monk! Monk! Have you already crossed the violent flood?” [The Buddha] replied: “Yes, Deity!”



gādhañ ca laddhāna thale ṭhito so, nāyūhati (SNN n'āyūhati) pāragato hi so ti ||
esūpamā dāmali brāhmaṇassa, khīṇāsavassa nipakassa jhāyino |
pappuyya jāti maraṇassa antaṃ, nāyūhati (SNN n'āyūhati) pāragato hi so ti || (SN I 47.30–48.7 = SNN 107.20–108.13).
394 This English translation is by Bodhi (2000: 141).
395 Ulrich (2013: 96) refers to this set as “the analogy of struggle.”
396 As for the content, the three versions are essentially the same. No verse is included in these three versions; however, in the Chinese version it mentions that the qíyàn sòng 祇焰頌 (geya, “verse”) should be known. I will identify that geya verse below.
 
[The deity asked:] “Without grasping and dwelling, have you already crossed the violent flood?”
[The Buddha] replied: “Yes, Deity!”
[The deity asked:] “Monk! How is it that you have already crossed the violent flood without grasping and dwelling?”
[The Buddha replied:] “Just as I have struggled, so have I degenerated; just as I have degenerated, so have I dwelled; just as I have dwelled, so have I drifted.
Deity! Just as I have given up struggling, so I do not degenerate.”
[The detail] is explicited in the pure white section(s). In this [section], the geya verse(s) should be known.397

(2) Sanskrit version:
bhikṣo bhikṣo ogham atārṣīḥ | āma devate | anālambe 'pratiṣṭhe ogham atārṣīḥ | āma devate | yathā kathaṃ tvaṃ bhikṣo anālambe apratiṣṭhe ogham atārṣīḥ | yathā yathāhaṃ devate āyūhāmi tathā tathā saṃsīdāmi | yathā yathā saṃsīdāmi tathā tathā saṃtiṣṭhe | yathā yathā saṃtiṣṭhe tathā tathā uhye | yathā yathāhaṃ devate nāyūhāmi tathā tathā na saṃsīdāmīti vistareṇa śuklapakṣo veditavyaḥ |
(Enomoto 1989: 28 [YBh 122b5f; ŚrBh 4b8.1f])
[The Deity asked:] “Monk, monk, did you cross the flood?” [The Buddha replied:] “Yes, Deity.”
[The Deity asked:] “Did you cross the flood where there is no support, no standing point?”
[The Buddha replied:] “Yes, Deity.”
[The Deity asked:] “Monk, how did you cross the flood where there is no support, no standing point?”
[The Buddha replied:] “Deity, as I struggled, to that extent I sunk. As I sunk, to that extent I stopped. As I stopped, to that extent I was swept away. Deity, as I did not struggle, to that extent I did not sink.”
The positive/bright side should be known in detail.398




397 This appears to be an abbreviation formula.
398 This appears to be an abbreviation formula.
 
(3) Tibetan version:
/ dge slong khyod chu bo las brgal399 tam / lha mo de lta yin no /
/ dmigs pa med pa dang / gnas pa med par chu bo las brgal400 tam / lha mo de lta401 yin no /
/ dge slong khyod ji lta bur dmigs pa med pa dang / gnas pa med pa chu bo las brgal / lha mo nga ni ji lta ji ltar 'dzin pa de lta de ltar 'bying ngo /
/ ji lta ji ltar 'bying ba de lta de ltar 'dug go /
/ ji lta ji ltar 'dug pa de lta de ltar khyer ro /
/ lha mo nga ni ji lta ji ltar 'dzin pa med pa de lta de ltar 'bying ba med do/
/402 zhes bya ba nas / rgyas par dkar po'i phyogs kyang rig par bya'o /
(Peking 5536.276b4–8; Derge 4035.238a2–5)
[The deity asked:] “Monk, have you crossed the river?” [The Buddha replied:] “Yes, deity.”
[The deity asked:] “Have you crossed over the river, which is without basis and support?” [The Buddha replied:] “Yes, deity.”
[The deity asked:] “Monk, how have you crossed over the river, which is without basis and support?”
[The Buddha replied:] “Deity, as I grasped, to that extent I sank. As I sunk, to that extent I stayed on.
As I stayed on, to that extent I was carried away.
O Deity, as I did not grasp, to that extent, I did not sink.” Thus it was said.
The aspect of virtue should be understood in detail.


2. The narratives of verse(s) from corresponding sūtras with parallels


(1) SĀ-G
1267: There is prose and a verse containing four pādas.403 The setting is Shěwèiguó qíshù gěigūdú yuán 舍衛國祇樹給孤獨園, Anāthapiṇḍika’s Park of Jeta’s Grove in Śrāvastī. An unknown deity asked and the Buddha replied.

399 Derge: rgal.
400 Derge: rgal.
401 Derge: de lta bu.
402 The Derge version does not contain two daṇḍas here.
403 時彼天子白佛言世尊比丘比丘度駛流耶 佛言如是天子 天子復問無所攀緣 亦無所住度駛流耶 佛言如是天子 天子復問無所攀緣 亦無所住而度駛流 其義云何 佛言天子我如是如是抱 如是如是直進 則不為水之所㵱 如是如是不抱 如是如是不直進 則為水所㵱 如是天子 名為無所攀緣 亦無所住而度駛流 時彼天子
 
(2) SĀ-U
180: There is prose and a verse containing four pādas.404 The setting is Shěwèiguó qíshù gěigūdú yuán 舍衛國祇樹給孤獨園, Anāthapiṇḍika’s Park of Jeta’s Grove in Śrāvastī. An unknown deity asked and the Buddha replied.

(3) SN
1.1.1: There is prose and a verse containing four pādas.405 The setting is Anāthapiṇḍika’s Park of Jeta’s Grove in Sāvatthī. An unknown deity asked and the Buddha replied.

These three texts share the same information, that is, an unknown deity asked and the Buddha replied in Śrāvastī. These texts additionally provide a final verse which is not found in this thirteenth verse set of the Śarīrārthagāthā. This verse may be translated as:406
After a long time at last I see
A Brahman who is fully quenched407, Who surpassed all fears,408
Has crossed over attachment to the world.
This final verse is supposed to be the “geya” verse which can also be seen in many other sūtras, just like the Chinese description of this verse set in the Śarīrārthagāthā: “This is so widely preached in the pure white section(s).”
復說偈言 久見婆羅門 逮得般涅槃 一切怖已過 永超世恩愛 (T99, 2.348b10–20).
404 問佛言瞿曇汝為能度瀑駛流耶 爾時世尊答言實爾 天言瞿曇如此駛流 深廣無際傍無攀緣 中無安足而能得度 甚為奇特 佛言實爾 天復問曰瞿曇汝今云何於此駛流 無可攀挽無安足處 而能得度 佛答天曰若我懈怠必為沈沒 若為沈沒必為所漂 若我精進必不沈沒 若不沈沒不為所漂 我於如是大洪流中 無可攀挽無安足處 而能得度此大駛流 天即讚言善哉善哉 比丘於此駛流 無所攀挽而能得度 甚為希有 天復以偈讚言我昔已曾見 婆羅門涅槃 久捨於嫌 能度世間愛 (T100, 2.438c15–28).
405 kathaṃ nu tvaṃ mārisa ogham atarī ti | appatiṭṭhaṃ khvāhaṃ, āvuso, anāyūhaṃ ogham atarin ti. yathā kathaṃ pana tvaṃ mārisa, appatiṭṭhaṃ anāyūhaṃ ogham atarī ti. yadā svāhaṃ, āvuso, santiṭṭhāmi tadāssu saṃsīdāmi, yadā svāhaṃ, āvuso, āyūhāmi tadāssu nibbuyhāmi. evaṃ khvāhaṃ, āvuso, appatiṭṭhaṃ anāyūhaṃ ogham atarin ti.
cirassaṃ vata passāmi, brāhmaṇaṃ parinibbutaṃ |
appatiṭṭhaṃ anāyūhaṃ, tiṇṇaṃ loke visattikan ti || (SN I 1.14–22 = SNN I 1.17–2.13).
406 SĀ-G: “久見婆羅門 逮得般涅槃 一切怖已過 永超世恩愛 (T99, 2.348b19–20); SĀ-U: “我昔已曾見 婆羅 門 涅 槃 久 捨 於 嫌 怖 能 度 世 間 愛 (T99, 2.438c27–28); SN: cirassaṃ vata passāmi, brāhmaṇaṃ parinibbutaṃ. appatiṭṭhaṃ anāyūhaṃ, tiṇṇaṃ loke visattikan ti (SN I 1.21–22 = SNN I 2.12–13).
407 This is a synonym for parinirvāṇa.
408 The parallel phrase in the Saṃyuttanikāya is different and states “[who by] not halting, not straining” (appatiṭṭhaṃ anāyūhaṃ).
 
3.1.14. The fourteenth set: attainment of the aim409 (dé yì 得義, arthaprāpti, don thob);
12 pādas / 3 verses410

1. Three versions with their translations411
(1) Chinese version:
獨臻阿練若 靜慮棄珍財 為別有方求 為窺窬封邑 何不與人交 而絕無徒侶
得義心寂靜 摧妙色魔軍 我獨處思惟 受最勝安樂 故不與人交 而絕無徒侶 (T1579, 30.378b16–21)
[[[Māra’s]] daughter, Tantrī, asked:]
Alone, one goes to the wilderness (āliànruò 阿練若, araṇya), meditates and gives up treasures.
Is this for seeking something else, or for coveting lands?
Why do you not interact with people, instead of having no companion at all? [The Buddha replied:]
Having obtained the purpose, the mind is tranquil; one destroys Māra’s army of pleasant forms.
I contemplate alone,
[and] have attained the foremost calm and pleasure.
I therefore do not interact with people and have no companion at all.




409 See Enomoto 1989: 21 and Ulrich 2013: 96.
410 The number of pādas in the Chinese version is 12, but in both the Sanskrit and Tibetan versions it is 10. 411 The meaning of the third and fourth pādas in Chinese, “coveting lands,” is different from the third pāda of other two versions, “committing offense.” These two meanings are also present in the Saṃyuktāgama and Saṃyuttanikāya. The fourth pāda in Chinese, “coveting lands” (kuīyú fēngyì 窺窬封邑), is relevant to
“seeking benefit in the village” (qiú jùluò lì 求聚落利) in the Saṃyuktāgama (SĀ-G). The third pāda in
Sanskrit and Tibetan, “committed offense,” is the same as the Saṃyuttanikāya and Mahāvastu (see below). That indicates that the Chinese version is similar to the Saṃyuktāgama text, while the Sanskrit and Tibetan versions are similar to the Pāli texts. The Chinese Yogacārabhūmi and the Saṃyuktāgama belong to the (Mūla)sarvāstivāda school. The Chinese Yogacārabhūmi is assumed to be earlier than the Sanskrit and Tibetan versions. I conjecture that the manuscripts of the Sanskrit and Tibetan versions have been influenced by other schools’ texts.
 
(2) Sanskrit version:
eko 'raṇye prasṛto dhyāyase tvaṃ, vittād vihīna uta vā prārthayānaḥ |
grāmasya vā kiṃcanāgo nv akārṣīḥ, kasmāj janena na karoṣi sakhyaṃ sakhyaṃ na saṃvidyate kena cit tava ||
arthaprāptyā hṛdayasyeha śāntir, jitveha senāṃ priyaśātarūpām |
eko dhyāyī sukham asmy anvabhotsaṃ, tasmād janena na karomi sakhyaṃ sakhyaṃ na saṃvidyate kena cid mama ||
(Enomoto 1989: 29 [YBh 123a5f; ŚrBh 4b9.2f]) [[[Māra’s]] daughter, Tantrī, asked:]
“Set out to the wilderness alone, you meditate.
Are you destitute of possessions [or] do you desire [them]? Or have you committed some offense to the village?
Why do you not make friends with people?
Why do you have no friendship with anyone?” 412
[The Buddha replied:]
“Because of obtaining the goal, [I have] calmness of the heart here. Having here conquered the army of the pleasurable and agreeable, meditating alone, I am awakened to happiness.
Therefore, I do not make friends with people. I have no friendship with anyone.”

(3) Tibetan version:
/ khyod ni gcig pu dgon par bsam gtan rtsom /
/ nor med gyur tam 'on te sgrub pa yod /
/ grong du nyes pa 'ga'413 zhig byas gyur tam /
/ ci'i phyir skye bo dag dang bshes mi byed /
/ khyod ni su dag dang yang bshes pa med /
/ 'di na dga' bde'i ngo bo'i sde btul nas /
/ don thob pas na 'di na snying zhi gyur /
/ nga ni gcig pu bsam gtan bde ba myong /


412 Literally: “why is friendship with someone not found for you?”
413 Derge: 'ba'.
 
/ de phyir skye bo dag dang bshes mi byed /
/ nga ni su dag dang yang bshes pa med /
(Peking 5536.277b8–278a2; Derge 4035.239a1–3)
[[[Māra’s]] daughter, Tantrī, asked:]
“You cultivate meditation alone in the wilderness.
Have you become destitute of wealth, or have you acquired [it]? Have you committed some faults in the village?
Why do you not make friends with people? Are you not acquainted with anyone at all? [The Buddha replied:]
“Here, overcoming the domain of pleasurable [[[objects]]],
and thereby attaining the goal, my heart has become peaceful herein. I have experienced the bliss of solitary meditation.
Therefore, I do not make friends with people; I am not acquainted with anyone at all.”

2. The narratives of verse(s) from corresponding sūtras with parallels


(1) SĀ-G414
1092: There are four verses containing sixteen pādas.415 The setting is on the bank of the river Níliánchán 尼連禪 (Nairañjanā), in Yùbìluó 欝鞞羅 (Uruvilvā). Mó bōxún 魔波旬 (Māra papīyān) asked and the Buddha replied.
1092: There are four verses containing fourteen pādas.416 The setting is on the bank of the river Níliánchán 尼連禪 (Nairañjanā), in Yùbìluó 欝鞞羅 (Uruvilvā). Àiyù 愛欲(Tantrī), the female deity (the first daughter of Māra) asked the Buddha questions. And the Buddha replied her.



414 The same verses are spoken by Māra first, then by the daughter again.
415 獨入一空處 禪思靜思惟 已捨國財寶 於此復何求 若求聚落利 何不習近人 既不習近人 終竟何所得(T99, 2.286b28–c2); 已得大財利 志足安寂滅 摧伏諸魔軍 不著於色欲 獨一而禪思 服食禪妙樂 是故不與人 周旋相習近 (T99, 2.286c5–c8).
416 獨一禪寂默 捨俗錢財寶 既捨於世利 今復何所求 若求聚落利 何不習近人 竟不習近人 終竟何所得(T99, 2.287a19–23); 已得大財利 志足安寂滅 摧伏諸魔軍 不著於色欲 是故不與人 周旋相習近 (T99, 2.287a25–27).
 
(2) SĀ-U
31417: There are six verses containing twenty two pādas.418 The setting is on the bank of the river Níliánchán 尼連禪 (Nairañjanā), in Yōulóubǐluó 優樓比螺 (Uruvilvā).
Mówáng 魔王 (Māra papīyān) asked and the Buddha replied.
31: There are four verses containing fourteen pādas.419 The setting is on the bank of the river Níliánchán 尼連禪 (Nairañjanā), in Yōulóubǐluó 優樓比螺 (Uruvilvā). Jíài 極愛(Tantrī), the first daughter of Māra, Tantrī, asked the Buddha questions. And the Buddha replied to her.

(3) SN
4.3.4 : There are two verses containing nine pādas.420 The setting is on the bank of the river Nerañjarā in Uruvelā. Māra the Evil one (pāpimā) asked and the Buddha replied.
4.3.5 : There are four verses containing ten pādas.421 The setting is on the bank of the river Nerañjarā in Uruvelā. Māra the Evil one (pāpimā) asked and the Buddha replied.

(4) Mahāvastu Avadāna
There are two verses containing ten pādas.422 Arati, the second daughter of Māra, asked questions and the Buddha replied. It does not explicitly state the setting, but it mentions

417 The same verses are spoken by Māra first, then by the daughter again.
418 汝獨處閑靜 閉默常寂然 光顏顯神體 諸根悉悅豫 譬如失財者 後還獲於財 汝今翫禪寂 歡喜亦如是 既能遺國榮 亦不悕名利 何不與諸人 而共為親友 (T100, 2.383a24–a29); 我久獲禪定 其心常寂然 破壞汝欲軍 得於無上財 我根恒恬豫 心中得寂滅 以壞汝欲軍 修道情歡喜 獨一離憒閙 安用親友為 (T100, 2.383b2–6).
419 端拱樹下坐 閑靜獨思惟 為失於財寶 為欲求大財 城邑聚落中 都無愛著心 何不與眾人 而共作親友(T100, 2.383c15–19); 我以得大財 心中得寂滅 我壞愛欲軍 妙色都不著 獨處而坐禪 最受第一樂 以是因緣故 都不求親友 (T100, 2.383c21–c24).
420 sokāvatiṇṇo nu vanamhi jhāyasi, vittaṃ nu jīno uda patthayāno | āguṃ nu gāmasmim akāsi kiñci, kasmā janena na karosi sakkhiṃ|| sakkhī na sampajjati kenaci te ti ||
sokassa mūlaṃ palikhāya sabbaṃ, anāgu jhāyāmi asocamāno |
chetvāna sabbaṃ bhavalobhajappaṃ, anāsavo jhāyāmi pamattabandhū ti || (SN I 123.1–9 = SNN I 269.14– 270.6).
421 sokāvatiṇṇo nu vanamhi jhāyasi, vittaṃ nu jīno uda patthayāno | āguṃ nu gāmasmim akāsi kiñci, kasmā janena na karosi sakkhiṃ || sakkhī na sampajjati kenaci te ti ||
atthassa pattiṃ hadayassa santiṃ, jetvāna senaṃ piyasātarūpaṃ |
eko 'haṃ jhāyaṃ sukham anubodhiṃ (SNN anvabodhiṃ), tasmā janena na karomi sakkhiṃ ||
sakkhī na sampajjati kenaci me ti || (SN I 126.9–18 = SNN I 276.3–13). 422 sokābhitunno va vanasmiṃ dhyāyasi citāya jīvaṃ abhiprārthayanto | āsādya grāmyā na karoṣi sākhyaṃ kasmājjane na karoṣi sākhyaṃ ||
 
that the place is where the Buddha became awakened under the Bodhi tree. Therefore, it must be Uruvilvā.

(5) Dàzhìdù lùn 大智度論 (the Chinese Mahāprajñāpāramitā)
T1509: There are four verses containing sixteen pādas.423 Rati, the third daughter of Māra asked questions and the Buddha replied under a Nyagrodha tree at an unkown location.

In the Śarīrārthagāthā commentary of this verse set, it clearly mentions that “it is the verse(s) asked by the female deity.”424 The above sources suggest that the first daughter of Māra, Tantrī, asked the Buddha questions in Uruvilvā.
The counterpart “obtaining great fortune (dé dàcái 得 大 財 )” in the Chinese Saṃyuktāgama texts (SĀ-G and SĀ-U) and “obtaining the flavour of nirvāṇa (得涅槃味)” in the Dàzhìdù lùn 大 智 度 論 (the Chinese Mahāprajñāpāramitā), are different from “obtaining the goal” 425 in this fourteenth verse set in the Śarīrārthagāthā.

3.2. The analysis of Group A verses

On the basis of the above materials, the Group A verses will now be analysed in two ways. Firstly, internal comparisons will be made purely based upon the three versions of the Śarīrārthagāthā. Secondly, external comparisons will be made by comparing the Śarīrārthagāthā to parallel verses in other texts. Additionally, in the last section the relationship of the two Chinese Saṃyuktāgama texts and two technical terms will be examined.




śokasya mūlaṃ parikhanya sarvaṃ sarvaṃ prahāya bhavalobhamāpsye |
arthasya prāptiṃ hṛdayasya śāntiṃ dhyānānuyukto sukhasārabuddhiḥ |
tasmājjane na karomi sākhyaṃ sākhyaṃ na saṃvidyate tena asme || Mvu III 284.13–285.4.
423 獨坐林樹間 六根常寂默 有若失重寶 無援愁苦毒 容顏世無比 而常閉目坐 我等心有疑 何求而在此(T1509, 25.181a1–4); 我得涅槃味 不樂處染愛 內外賊已除 汝父亦滅退 我得甘露味 安樂坐林間 恩愛之眾生 為之起慈心 (T1509, 25.181a6–9).
424 天女所問伽他 (T1579, 30.378b22).
425 The commentary defines the goal as the fruit of the śramaṇa (shāmén guǒ 沙門果) and the supernatural power of the sage (shèng shéntōng 聖神通). See T1579, 30.378b22–23.
 
3.2.1. The distinctive characteristics of the Chinese Group A verses

If the three versions of the Śarīrārthagāthā (Chinese, Sanskrit and Tibetan) are compared verse by verse, the Chinese version seems distinct. Three lines of evidence support this conclusion. Firstly, in the fourteenth verse set, the Sanskrit and Tibetan versions both mention “committing some offense [to the village]” (the third pāda of the first verse), but the Chinese instead states “coveting lands.” The corresponding sūtra in the Saṃyuktāgama states, “seeking benefit for the village”426 (qiú jùluò lì 求聚落利) and lacks any information about “committing some offense.” However, in the Saṃyuttanikāya and Mahāvastu, we find the phrase “committing some offense.” It seems that the Chinese version of the Śarīrārthagāthā is close to the Saṃyuktāgama (Mūlasarvāstivāda school) while the Sanskrit and Tibetan versions are close to the Pāli texts (Theravāda school). Secondly, in the twelfth verse set the number of verse pādas in the Chinese version (20) is much greater than that in the Sanskrit and Tibetan versions (14). Xuánzàng 玄 奘 and his team appear to be quite loyal to the pāda structure and content of the exemplar he used for his translation. It is possible that the manuscripts of the Chinese version were different. Thirdly, in the eleventh set the Chinese version (and its commentary) does not provide any information about a yakṣa, but this term is found in the Sanskrit and Tibetan versions.
It is difficult to account for the differences between the Chinese version and the Sanskrit and Tibetan versions. It is possible that the original manuscript of the Chinese represents a different recension from the Sanskrit and Tibetan. However, it is also possible that these differences were introduced by Xuánzàng 玄奘 (and his team).

3.2.2. The features of the Group A verses of the Śarīrārthagāthā

Although the Group A verses in the Śarīrārthagāthā mostly correspond to sūtras in the Saṃyuktāgama, the Group A verses are nonetheless distinct. Firstly, the Group A verses may represent a different version of the verses. Secondly, they omit text found in parallel versions.

426 “Land” can be viewed as “the benefit in the village,” so “coveting lands” can be an example of “seeking benefit in the village.” The Chinese version of the fourteenth set is close to the Saṃyuktāgama.
 
3.2.2.1. Different recension

The following verses show the difference between the Śarīrārthagāthā and their corresponding sūtras with parallels, although the content and structure of the verse sets are mostly similar. Firstly, in the fourth verse set, the fourth pāda mentions a “yoke” (yoktra) or “waist boat” in Chinese, but this item is not mentioned in any of the corresponding sūtras or their parallel texts. Secondly, in the sixth verse set, it mentions “one who instructs superior kinds of human beings... and not fear death of the next life,” which is not found in any of the sūtras.427 Thirdly, the last pāda of the twelfth verse set states, “for one who has right mindfulness [and] whose mind is liberated.”428 The counterparts of the corresponding sūtra and the parallel are different and divergent. For example, SĀ-G includes the phrases “has reached the other shore,” “nirvāṇa” and “has nothing to ask for.”429 Fourthly, in the thirteenth verse set, the content logic of the Buddha’s last reply is struggle→ sink→ stop→ swept away. However, the counterparts
in SĀ-G and SĀ-U are different: struggle (or strive)→ not sinking or not halting→ not
swept away.430 Fifthly, in the fourteenth verse set it mentions “obtaining the goal.” The counterpart in the Chinese Saṃyuktāgama texts is “obtaining great fortune (dé dàcái 得大 財 )” and in the Dàzhìdù lùn 大 智 度 論 (the Chinese Mahāprajñāpāramitā) it is “obtaining the flavour of nirvāṇa (dé nièpán wèi 得 涅 槃 味 ).”431 The commentary in the Śarīrārthagāthā explains it as “obtaining the fruit of the śramaṇa (shāmén guǒ 沙 門 果 )”




427 It states, “this one obtains a marvelous form and is worshiped by deities” in SĀ-G; “this one obtains a superior form and their conduct is is the way to being worshiped by deities” in SĀ-U; “this one is worshiped by deities” in SN.
428 More importantly, this verse set seems to be an answer to the sixth verse set. The sixth set asks “Who can guide superior kinds of human beings?” The description of the one who can guide is the same as that described in the twelfth set, which is as follows: wise, has acquired meditation, all the afflictions of sorrow have been abandoned, has right mindfulness and whose mind is liberated. The twelfth set labels such a person as a “Brahman.” It is hard to know if it is intentional that the twelfth set responds to the sixth set in the Śarīrārthagāthā and if this is the reason that the last pāda of the twelfth set is different from its counterpart in the corresponding sūtra in the Saṃyuktāgama.
429 SĀ-U states, “has crossed over the sea of birth and death,” while SN states, “having reached the end of birth and death, he need not strain for he has gone beyond.”
430 The meaning of the counterpart in the Saṃyuttanikāya has a similar meaning to the Śarīrārthagāthā, but
the wording of the two texts is different. The wording in SN1.1.1 is “that by not halting and by not straining I crossed the flood” (Bodhi 2000: 89).
431 In the Saṃyuttanikāya, it does not contain a corresponding term.
 
and “obtaining the supernatural power of the sage (shèn gshéntōng 聖 神 通 ).”432 Therefore, the Dàzhìdù lùn 大 智 度 論 (the Chinese Mahāprajñāpāramitā) is closer to the Śarīrārthagāthā.
The above verses of the Śarīrārthagāthā do not fully match the counterpart of any sūtra. It is likely that these differences are due to the Śarīrārthagāthā verses belonging to a different recension.433

3.2.2.2. Deficiency of a verse

Two verse sets in the Śarīrārthagāthā have one less verse than the corresponding sūtras in the Saṃyuktāgama. Firstly, a final verse434 is not present in the eight verse set of the Śarīrārthagāthā, but appears in the corresponding sūtras in the two Saṃyuktāgama texts. This final verse is not fully relevant to the previous context and is not present in the other parallel texts. It is possible that this final verse did not exist in an earlier version of the Saṃyuktāgama, but was added somehow at a later time, when compared with all the parallels. Secondly, the same final verse435 is not provided in the thirteenth verse set. However, this verse has been mentioned and is described as a geya verse in the Chinese version.436 This final verse is related to the previous context and is commonly found in many other sūtras of the Saṃyuktāgama and Saṃyuttanikāya. Also, this final verse is important for the thirteenth “verse” set of the Śarīrārthagāthā because if it were not included there would be no verse in this set. It is confusing as to why this final verse is not specifically and directly mentioned. Instead, it appears to be referred to indirectly without any information about a “verse” with the statement, “The positive/bright side should be known in detail” in the Sanskrit and Tibetan versions.


432 See T1579, 30.378b22–23.
433 Enomoto (1989: 22) applied several cases and believds that the Śarīrārthagāthā verses should belong to a later version. However, I propose another possibility in § 6.1.2.
434 SĀ-G: 久見婆羅門 逮得般涅槃 一切怖已過 永超世恩愛 (T99, 2.160c28–29); SĀ-U: 往昔已曾見 婆羅門涅槃 嫌怖久棄捨 能度世間愛 (T99, 2.438a16–17). The verse is translated as follows: “After a long time
at last I see a Brahman who is fully quenched, who has surpassed all fears, has crossed over attachment to the world.”
435 SĀ-G: 久見婆羅門 逮得般涅槃 一切怖已過 永超世恩愛 (T99, 2.348b19–20); SĀ-U: 我昔已曾見 婆羅門涅槃 久捨於嫌怖 能度世間愛 (T99, 2.438c27–28); SN: cirassaṃ vata passāmi, brāhmaṇaṃ parinibbutaṃ. appatiṭṭhaṃ anāyūhaṃ, tiṇṇaṃ loke visattikan ti (SN I 1.21–22 = SNN I 2.12–13).
436 In the Sanskrit and Tibetan versions, information about the “verse” is not mentioned.
 
In these two cases which lack a final verse in the Śarīrārthagāthā, it is important to compare with the corresponding sūtras and the parallels. Therefore, it seems that the final verse is needed in one case and is surplus in another case.

3.2.3.3. Section summary

From the above two points, it can be seen that the verses in the Śarīrārthagāthā do not fully correspond to the sūtras in the Saṃyuktāgama we have currently. Enomoto (1989: 23) believed that the Śarīrārthagāthā represents a later redaction of this material than the Saṃyuktāgama. However, this does not mean that the verses or content in the Saṃyuktāgama are earlier in every single instance, as is evidenced in the above example in which a verse is lacking.

3.2.3. Other issues

In this section, two issuses will be raised. The first is the relationship between two Chinese Saṃyuktāgama collections. The second is the meaning of two technical terms, which will be studied because they cannot be understood by the literal meaning.

3.2.3.1. Comparing SĀ-G with SĀ-U

The relationship between two versions of the Chinese Saṃyuktāgama, SĀ-G and SĀ- U, is an open question.437 Since the Group A verses correspond to the verses in both SĀ- G and SĀ-U, the content of the two texts can also be examinated to check whether these two texts are identical.438
The following list is a comparison of the content between SĀ-G and SĀ-U:




437 The school of SĀ-G is commonly accepted as (Mūla)sarvāstivāda. However, the school of SĀ-U has been viewed as Kāśyapīya, Mahīśāsaka, Dharmaguptaka or (Mūla)sarvāstivāda by scholars. See Akanuma 1939: 48–50; Mizuno 1970; Waldschmidt 1980: 146; Enomoto 1980, 1984; Mayeda 1985: 101; Yinshun
1988: 98; Hiraoka 2000; Bingenheimer 2011: 23–50; Bucknell 2011.
438 Bingenheimer (2011) and Bucknell (2011) examined the relationship of SĀ-G and SĀ-U by the sequence and arrangement of the text structure, but not by the content. I agree with their conclusion that SĀ-U seems to be a (Mūla)sarvāstivādin recension.
 
Verse set Comparison of SĀ-G with SĀ-U
1 Same
2 Second verse (4 pādas) is very different.439
3 The conditions are different.440
4 The person is different.441
5 Same
6 The description is different.442
7 Same
8 Same
9 Same
10 Mostly same443
11 Same
12 Different444
13 Same
14 Different445

Among these fourtheen verse sets, only seven seem to be the same between SĀ-G and SĀ-U. The sixth, tenth sets contain only reasonably minor differences, which might have


439 The term “what is to be expressed” is interpreted as “craving” (ài 愛 ) in SĀ-G, and as “name and form” (míng sè 名 色 ) in SĀ-U. The meaning of the second verse in SĀ-G is the same as the one in the Śarīrārthagāthā, but in SĀ-U, it is different. The second verse in SĀ-U may be translated, “If name and form can be known as empty, without a nature, this is revering the Buddha, and all paths of sentient existence can be gone beyond.” (若識於名色 本空無有性 是名尊敬佛 永離於諸趣 T100, 2.379b17–18). 440 It is said to be “suffering, empty and not-self” (kǔ 苦 , kōng 空 , wúwǒ 無 我 ) in SĀ-G, and “impermanence, suffering and not-self” (wúcháng 無常, wú yǒu lè 無有樂, wúwǒ 無我) in SĀ-U.
441 The one asking is Àiniàn 愛念 (Aratī, the second daughter of Māra) in SĀ-G and Shìyì 適意 (Ratī, the
third daughter of Māra) in SĀ-U. Bingenheimer (2011: 126) suggests that Aratī and Ratī are probably the same name.
442 SĀ-G states “this one is worshiped by deities” (諸天所供養 T99, 2.160a13), while SĀ-U states “this one
can worship deities” (而能供養天 T99, 2.439a29). It seems likely that the translator(s) of SĀ-U might have misunderstood the Sanskrit grammar, that is, the subject and object are taken oppositely.
443 One pāda from one of three corresponding sūtras in SĀ-U is different. SĀ-U reads “what is developing good dharma” (云何修善法 T100, 2.484c24), while the others read “who/how to cross the floods.”
444 The main ideas of the verses in both texts are basically the same, but the structure and content of the verses in SĀ-U are longer and the descriptions of a few pādas are a bit different. For example, SĀ-U states, “has crossed over the sea of birth and death” and SĀ-G states, “has reached the other shore [and] nirvāṇa; has nothing to ask for.”
445 The same verses said by Māra first, then repeated by one of his daughters. However, the verses in SĀ-U spoken by Māra are longer and have different information.
 
arisen from personal interpretations by the translator(s) or errors by scribes, and these can possibly be added to the list above to make nine sets. However, there are still five remaining verse sets in which SĀ-G and SĀ-U differ. The contents of the verses in SĀ-G are mostly close to the verses in the Śarīrārthagāthā, but the contents of SĀ-U are not.
The fact is that there are divergences in content between SĀ-G and SĀ-U and it is difficult to regard them as coming from the same source. However, it is a very common situation to find discrepancies in content between different recensions. The Śarīrārthagāthā is closer to the recension of SĀ-G.

3.2.3.2. Technical terms

Two technical terms need to be further discussed as they cannot be understood by a literal translation. These two are “what can be expressed” (yìngshuō 應說, ākhyeya, brjod bya), and “the tie of yoke” (yāozhōu 腰舟, yoktra, gnya' dag/gnya' shing).
Firstly, the Sanskrit term ākhyeya is a future passive participle from the prefix ā (near, towards) + √khyā (to tell, say), which literally means “to be said or expressed,” but used as a noun meaning “what can be expressed.” In Chinese, literally, yìngshuō 應說 is supposed to be something which is expressed by stating what it is. In Tibetan, brjod bya (a verbal noun) means “what is to be told or said.” Overall, the literal meaning of the term could be rendered as “what can be expressed,” following Bodhi (2000: 99).
What is the meaning of this term? It can be seen in the corresponding sūtras of the two Chinese Saṃyuktāgama texts. The translator’s interpretation is “craving” (ài 愛)446 in SĀ-G and “name and form” (míngsè 名 色 )447 in SĀ-U. However, it is unclear why the term was interpreted as such. The commentary on the Saṃyuttanikāya and the commentary in the Yogacārabhūmi provide information on this term. Buddhaghosa explained akkheyya (Pāli) as the five aggregates.448 As explained by Bodhi (2000: 354 note 35), the five aggregates are “the objective sphere of linguistic reference (not the



446 眾生隨愛想 以愛想而住 以不知愛故 則為死方便 (T99, 2.282a3–4); 若知所愛者 不於彼生愛 彼此無所有 他人莫能說 (T99, 2.282a8–9).
447 名色中生相 謂為真實有 當知如斯人 是名屬死徑 若識於名色 本空無有性 是名尊敬佛 永離於諸趣
(T100, 2.379b15–18).
448 See Sāratthappakāsini I 44.19.
 
terms of expression themselves).”449 This explanation more fully explains the term’s literal meaning “what can be expressed.”
The Yogacārabhūmi does not regard the “five aggregates” as the objective sphere of linguistic reference. Two definitions are supplied in the Yogacārabhūmi. The first is “all formations/conditioned phenomena” (yīqiè yǒuwéifǎ 一切有為法)450 which are attributed to “the expressed phenomena” (yánshì 言 事 )451. The phenomena, which have been expressed, are the objective sphere of linguistic reference. From this definition, “what can be expressed” is simply “everything” a being perceives. Another explanation is “pleasant five sensual pleasures” ( 妙 五 欲 ),452 which specifically refers to the second verse set in the Śarīrārthagāthā. Sensual pleasures are what can be asked for by means of expression, e.g. when the king enjoys sensual pleasures by an order to his servants.453
In conclusion, “what can be expressed” is the objective sphere of linguistic reference. Therefore, it can be everything (all formations) that sentient beings perceive. Specifically, it is regarded as “the five aggregates” by Buddhaghosa, and “five sensual pleasures” in the Yogacārabhūmi.
The second term which requires clarification is yoktra. The meaning in Sanskrit is an instrument for tying or fastening (MW s.v. yoktra), just like a rope or tie. In the context of these verses, this rope is supposed to help a person cross over the flood. In Tibetan, gnya' dag is supposed to be a counterpart of yoktra. In Chinese, Xuánzàng 玄奘 (with his team) does not directly translate this term, but rather refers to an instrument found in China called a waist boat. The waist boat was a common way to cross a river especially in the southern China, but it now only seems to be in use by a tribe in Hainan province.454
What does “the tie of yoke” or “waist boat” imply in the verses? Although this term cannot be found in the corresponding sūtras or the parallel, it has explained in the commentary of the verse in the Yogacārabhūmi. The waist boat means “the mind.”455


449 See Sāratthappakāsinī I 44.22–24.
450 初言應說者謂一切有為法 (T1579, 30.370c8).
451 諸有為法皆三種言事之所攝故 (T1579, 30.370c9). The expressed phenomena can be divided into three kinds, namely, the past, present and future. See T1579, 30.752c6.
452 今此義中說妙五欲以為應說 (T1579, 30.370c9–10).
453 又諸君主於妙五欲從僕使等 以言呼召而受用之 由是因緣亦名應說 (T1579, 30.370c12–13).
454 This information is shown in the government website of Hainan province. See http://www.hainan.gov.cn:1500/hn/zjhn/lsrw/lsyg/200809/t20080911_365167.html (accessed on 01/04/2016) 455 意名腰舟 (T1579, 30.373c26).
 
This mind is a well liberated mind from all fetters.456 In the context of the verses, it is said that the meditator has not achieved a well liberated mind. The meditator has conquered sensual lust, but more subtle forms of lust [of form and the formless] still exist.457 Therefore, he has not yet obtained a well liberated mind.

3.2.4. Summary

In this section, the special characteristics of the Chinese Śarīrārthagāthā version (§ 3.2.1), the verses in the Śarīrārthagāthā (§ 3.2.2), and SĀ-U (§ 3.2.3.1) have been analyzed based on the Group A verses of the Śarīrārthagāthā. The dissimilarities between the different versions might have arisen during the process of translation and manuscript copying, and differing exemplars. Additionally, two technical terms have been discussed. The term “what can be expressed” is the objective sphere of linguistic reference. It can be everything (all formations) that sentient beings perceive. However, it is specifically regarded as “the five aggregates” by Buddhaghosa, and “five sensual pleasures” in the Yogacārabhūmi. Another term “the tie of yoke” is implied as “the mind,” especially “a well liberated mind.”

3.3. Conclusion

In this Chapter, the three versions of the Group A verses (Chinese, Sanskrit, and Tibetan) with their English translations were presented, as well as the narratives of the verses. Following this, based on the previous information, it was argued that the Chinese version of the Group A verses is different from the other two versions and the unique characteristics of the Śarīrārthagāthā verses were highlighted. If SĀ-G and SĀ-U are to be attributed to the the same school, it is the (Mūla)sarvāstivāda school; however, if they are to be regarded as different recensions, the Śarīrārthagāthā is closer to the recension of SĀ-G.





456 此中何等名為腰舟 謂於諸結善解脫心 (T1579, 30.373c28–29).
457 諸有學未得勝意 已離欲貪未離上貪 (T1579, 30.373c25–26).
 
Chapter Four: The Group B verses

The structure of this chapter is similar to that of chapter three. This chapter is divided into two main sections. The first presents the three versions of the Group B verses with corresponding English translations, along with comparisons between the Śarīrārthagāthā verses and other texts. As with chapter three, these English translations will indicate my own interpretation of the source material. An analysis of the Group B verses will be presented in the second section. These verses will be investigated through internal and external comparisons. Internal comparisons study the three versions of the Śarīrārthagāthā against each other. External comparisons study the Group B verses of the Śarīrārthagāthā with reference to parallel verses in other texts. Other issues relevant to the Group B verses will also be examined, namely, the relationship between Uv-B and T212, and the meaning of the term prapañca in the Yogacārabhūmi.

4.1. The three versions of the Group B verses with their English translations, and comparison of the verse with other texts

There are twenty seven verse sets in Group B. These verses will be studied in two ways. Firstly, three versions (in Chinese, Sanskrit and Tibetan) with their English translations will be presented to facilitate comparisons between their content and structure. Secondly, comparisons will be made between the Śarīrārthagāthā verses and other texts. While parallels to the Group B verses are found amongst early Buddhist texts, these parallels unfortunately mostly appear as standalone verses without narrative or background information. According to Schmithausen (1970), the first twenty four verse sets presented in this chapter correspond to verses in the Udānavarga attributed to the Mūlasarvāstivādins (Uv-S). This is different from the Udānavarga of Bernhard’s edition (Uv-B), which is attributed to the Sarvāstivādins (Schmithausen 1970; Brough 2001).
Another three verse sets (twenty fifth to twenty seventh) do not correspond to material of the same school; however, there are some Pāli parallels and one Madhyamāgama parallel (see below).
Most of the comparisons made in this chapter are with texts belonging to the Sūtra Piṭaka, although some of the parallels can also be found in the Vinaya and Abhidharma
 
Piṭakas. The identification of these parallels is largely based upon the three studies of Tanjo (1967), Mizuno (1972) and Enomoto (1989).

4.1.1. The first set: impermanence (mi rtag pa); 4 pādas / 1 verse

1. Three versions with their translations458
(1) Chinese version:
諸行無常 有生滅法 由生滅故 彼寂為樂 (T1579, 30.378c20) All formations are impermanent;
they are with the nature (fǎ 法)459 of arising and vanishing. Because they arise and [then] vanish,
their extinction is blissful.


(2) Sanskrit version:
anityā bata saṃskārā, utpādavyayadharmiṇaḥ | utpadya hi nirudhyante, tesāṃ vyupaśamaḥ sukham || (Enomoto 1989: 29 [YBh 123b5f; ŚrBh 4b9.3])
All formations are impermanent; their nature is to arise and vanish. Having arisen, they cease.
Their extinction is blissful.


(3) Tibetan version:
/ kye ma 'du byed rnams mi rtag / skye dang 'jig pa'i chos can yin/
/ skyes nas 'gag par 'gyur ba ste / / de dag nye bar zhi ba bde / (Peking 5536.278b8–279a1; Derge 4035.239b6–7)
Alas! Formations are impermanent; their nature is to arise and vanish. Having arisen, they come to vanish;


458 The three versions are mostly the same.
459 In the commentary of this verse, fǎ 法 (dharma in Sanskrit) is explained as tǐ 體, which means “body”, “form”, “nature”, or “essence”.
 
that they are fully pacified is bliss.


2. Comparison of the verse with other texts
This is a popular verse which is found in numerous texts. The corresponding Udānavarga verse is Uv-S 1.3.460 Parallel verses may be found at: Uv-B 1.3;461 SN 1.2.1,462 15.20.19;463 SĀ-G 576,464 956,465 1197;466 SĀ-U 110,467 161,468 350;469 DN
16.6.10;470 DĀ T1;471 Th 1159;472 Jā 1.10.5;473 Ap 64.15–16,474 385.15–16;475 GDhp
181;476 Nirvāṇa sūtra: T6 (Bōníhuán jīng 般泥洹經), T7 (Dà bōnièpán jīng 大般涅槃經);477 T189478 (Guòqù xiànzài yīnguǒ jīng 過去現在因果經); T39479 (Dǐngshēngwáng
gùshì jīng 頂生王故事經); T210480 (Fǎjù jīng 法句經); T211481 (Fǎjù pìyù jīng 法句譬喻經); T212482 (Chūyào jīng 出曜經); T374 (Dàbōnipán jīng 大般涅槃經) Chpater 1,483



460 anityā bata saṃskārā, utpādavyayadharmiṇaḥ | utpadya hi nirudhyante, tesāṃ vyupaśamaḥ sukham || (Uv-S 50.15).
461 anityā bata saṃskārā, utpādavyayadharmiṇaḥ | utpadya hi nirudhyante, tesāṃ vyupaśamaḥ sukham || (Uv-B 231.3–4).
462 aniccā sabbasaṅkhārā, uppādavayadhammino | uppajjitvā nirujjhanti, tesaṃ vūpasamo sukho ti || (SN I 6.4–5 = SNN I 11.12–15).
463 aniccā sabbasaṅkhārā, uppādavayadhammino | uppajjitvā nirujjhanti, tesaṃ vūpasamo sukho ti || (SN II 193.13–16).
464 一切行無常 是則生滅法 生者既復滅 俱寂滅為樂 (T99, 2.153c13–14).
465 一切行無常，悉皆生滅法，有生無不盡，唯寂滅為樂 (T99, 2.244a6–7).
466 一切行無常 斯皆生滅法 雖生尋以滅 斯寂滅為樂 (T99, 2.325b17–18).
467 諸行無常 是生滅法 生滅滅已 乃名涅盤 (T100, 2.413c29–414a1).
468 諸行斯無常 是生滅之法 其生滅滅已 寂滅乃為樂 (T100, 2.435a16–17).
469 諸行無常 是生滅法 生滅滅已 寂滅為樂 (T100, 2.489b2–3).
470 aniccā vata saṅkhārā, uppādavayadhammino | uppajjitvā nirujjhanti, tesaṃ vūpasamo sukho ti || (DN II 157.8–9).
471 陰行無有常 但為興衰法 生者無不死 佛滅之為樂 (T1, 1.26c21–22).
472 aniccā vata saṅkhārā, uppādavayadhammino | uppajjitvā nirujjhanti, tesaṃ vūpasamo sukho ||
473 aniccā vata saṅkhārā, uppādavayadhammino | uppajjitvā nirujjhanti, tesaṃ vūpasamo sukho ti || (Jā I 392.19–20).
474 aniccā vata saṅkhārā, uppādavayadhammino | uppajjitvā nirujjhanti, tesaṃ vūpasamo sukho || (Ap 64.15– 16).
475 aniccā vata saṅkhārā, uppādavayadhammino | uppajjitvā nirujjhanti, tesaṃ vūpasamo sukho || (Ap 385.15–16).
476 aṇica vada saghara, upadavayadharmiṇo || upajiti niruj̄adi, teṣa uvaśamo suho ||
477 諸行無常 是生滅法 生滅滅已 寂滅為樂 (T7, 1.204c23–24).
478 諸行無常 是生滅法 生滅滅已 寂滅為樂 (T189, 3.623c21–22).
479 諸法悉無常 生者必壞敗 生生悉歸盡 彼滅第一樂 (T39, 1.824a13–14).
480 所行非常 謂興衰法 夫生輙死 此滅為樂 (T210, 4.559.a11–12).
481 所行非常 謂興衰法 夫生輒死 此滅為樂 (T211, 4.575c4–5).
482 所行非常 謂興衰法 夫生輒死 此滅為樂 (T212, 4.611b10–11).
483 有為之法 其性無常 生已不住 寂滅為樂 (T374, 12.375a26–27).
 
Chapter 7;484 T380485 (Dàbēi jīng 大悲經), T456486(Fóshuō mílè dàchéngfó jīng 佛說彌勒大 成 佛 經 487); Parallels in which the content of the third pāda differs are found in the following two texts: EĀ 31.9,488 32.6,489 34.2,490 41.5,491 42.2,492 52.1;493 T384494 (Púsà
cóngdōushùtiān jiàng shén mǔtāi shuō guǎngpǔ jīng 菩薩從兜術天降神母胎說廣普經). In both texts, the third pāda means “no birth, no death” or “there is no death without birth” (不生必不死/無生則無死/不生則不死/無生亦無死), which seems to respond to the second pāda “that they are born/arisen will die/vanish.” The translator of these two texts are said to be the same person, Zhúfóniàn 竺佛念. Therefore, the difference might have been caused by the translator’s interpretation or skill. These parallels are similar overall.
From the above texts, we find that there are 20 different narratives for this verse.495 It is difficult to decide which information is most strongly related to this verse in the

484 諸行無常 是生滅法 (T374, 12.450a16); 生滅滅已 寂滅為樂 (T374, 12.451a1).
485 諸行無常 是生滅法 生已還滅 滅彼為樂 (T380, 12.951c18–19).
486 諸行無常 是生滅法 生滅滅已 寂滅為樂 (T456, 14.430b24–25).
487 Another version of the sūtra on the Descent of Maitreya.
488 一切行無常 生者必有死 不生必不死 此滅最為樂 (T125, 2.672b14–15).
489 一切行無常 生者必有死 不生必不死 此滅最為樂 (T125, 2.677c18–19).
490 一切行無常 生者必有死 不生必不死 此滅最為樂 (T125, 2.693a8–9).
491 一切行無常 起者必有滅 無生則無死 此滅最為樂 (T125, 2.746c22–23).
492 一切行無常 生者必有死 不生則不死 此滅為最樂 (T125, 2.752c14–15).
493 一切行無常 生者必有盡 不生則不死 此滅為最樂 (T125, 2.823b5–6).
494 一切行無常 生者必有死 無生亦無死 此滅為最樂 (T384, 12.1057c6–7).
495 1. SĀ-G
576: The Buddha spoke this verse to a deity in Shěwèiguó qíshù gěigūdú yuán 舍衛國祇樹給孤獨園, Anāthapiṇḍika’s Park of Jeta’s Grove in Śrāvastī.
956: The Buddha talked to his monk disciples at Vipula Mountain in Rājagṛha.
1197: The deity Śakro devānām indraḥ spoke this verse in the place where the Buddha passed away under a twin Sāla tree in Kuśinagara city of the Mallas.
2. SĀ-U
110: The deity Śakro devānām indraḥ spoke this verse in the place where the Buddha passed away under a twin Sāla tree in Kuśinagara city of the Mallas.
161: The Buddha spoke this verse to a deity in Shěwèiguó qíshù gěigūdú yuán 舍衛國祇樹給孤獨園, Anāthapiṇḍika’s Park of Jeta’s Grove in Śrāvastī.
350: The Buddha talked to his monk disciples at Vipula Mountain in Rājagṛha.
3. SN
1.2.1: While in Sāvatthī, the Buddha mentioned that a deity spoke this verse to a certain deity of the Tāvatiṃsa heaven.
15.20.19: The Buddha spoke this verse to his monk disciples on Vulture Peak in Rājagaha.
4. EĀ
31.9: A deity talked to another deity in Nandana garden.
32.6: The deity Śakro devānām indraḥ spoke this verse to a deity. The setting is not mentioned. 34.2: The Buddha spoke this verse in Śrāvastī.
41.5: The Buddha spoke this verse to Kāśyapa and Ānanda in Shěwèiguó qíshù gěigūdú yuán 舍衛國祇樹給孤獨園, Anāthapiṇḍika’s Park of Jeta’s Grove in Śrāvastī.
 
Śarīrārthagāthā. However, in at least five of the parallels it is stated that the deity496 spoke this verse in the place where the Buddha passed away under a twin Sāla tree in Kuśinagara city of the Mallas.

42.2: The Buddha spoke this verse under a twin Sāla tree in Kuśinagara city of the Mallas.
52.1: The Buddha spoke this verse to the deities and his monk disciples in a lecture hall called Pǔhuì 普會
in Vaiśālī.
5. DĀ
T1: The deity Śakro devānām indraḥ spoke this verse in the place where the Buddha passed away under a twin Sāla tree in Kuśinagara city of the Mallas.
6. DN
16.6.10: Sakka, the king of the deities, spoke this verse just after the Buddha passed away. The setting is not mentioned; however, we may surmise it is at Kuśinagara city of the Mallas.
7. Jā
1.10.5: Bodhisatta Sudassana said this verse to queen Subhaddā. The setting is not provided.
8. Nirvāṇa sūtra
T6 (Bōníhuán jīng 般泥洹經): The deity Śakro devānām indraḥ spoke this verse in the place where the Buddha passed away under a twin Sāla tree in Kuśinagara city of the Mallas.
T7 (Dà bōnièpán jīng 大般涅槃經): The Buddha spoke this verse to his monk disciples under a twin Sāla
tree in Kuśinagara city of the Mallas.
9. Guòqù xiànzài yīnguǒ jīng 過去現在因果經
T189: A deity called Shànhuì 善慧 spoke this verse to another deity in Dōushuài tiān 兜率天 (Tuṣita Heaven).
10. Dǐngshēngwáng gùshì jīng 頂生王故事經
T39: The Buddha spoke this verse to Ānanda in Shěwèiguó qíshù gěigūdú yuán 舍衛國祇樹給孤獨園, Anāthapiṇḍika’s Park of Jeta’s Grove in Śrāvastī.
11. Fǎjù pìyù jīng 法句譬喻經
T211: The Buddha spoke this verse to the deity Śakro devānām indraḥ in a stone room at Vulture Peak.
12. Chūyào jīng 出曜經
T212: One of the Buddha's disciples spoke this to Ānanda. The setting is not mentioned.
13. Dàbōnipán jīng 大般涅槃經
Chapter 1: The Buddha spoke this verse to Chúntuo 純 陀 (Cunda in Sanskrit). The setting is not provided, but from the context, we may surmise it is where the Buddha passed away under a twin Sāla tree in Kuśinagara city of the Mallas.
Chapter 7: The deity Śakro devānām indraḥ spoke this verse to a practitioner of austerities (kǔxíngzhě
苦行者). The setting is not provided.
14. Dàbēi jīng 大悲經
T380: A Tathāgata called Shāngzhǔ 商主 (Sārthavāha in Sanskrit) mentioned to Luóhóuluó 羅睺羅
(Rāhula) that the Buddha spoke this verse. The setting is not provided.
15. Púsà cóngdōushùtiān jiàng shén mǔtāi shuō guǎngpǔ jīng 菩薩從兜術天降神母胎說廣普經
T384: Dàjiāshè 大迦葉 (Kāśyapa) spoke this verse. The setting is not provided, but from the context we may surmise that it is where the Buddha passed away under a twin Sāla tree in Kuśinagara city of the Mallas.
16. Fóshuō mílè dàchéngfó jīng 佛說彌勒大成佛經
T456: Maitreya mentioned that a Buddha of the past spoke this verse. From the context, we may surmise that the setting is the country ruled by King Śaṅkha (Nángqū 蠰佉).
17. Gēnběnshuōyīqiēyǒubù pínàiyē záshì 根本說一切有部毘奈耶雜事
T1451: The deity Śakro devānām indraḥ spoke this verse in the place where the Buddha passed away under a twin Sāla tree in Kuśinagara city of the Mallas.
496 The full title of this deity is Śakro devānam indraḥ.
 
4.1.2. The second set: carefulness (bag yod); 4 pādas / 1 verse

1. Three versions with their translations497
(1) Chinese version:
無逸不死跡 放逸為死跡 無逸者不死 縱逸者常死 (T1579, 30.379a23) Carefulness is the trace of the deathless;
carelessness is the trace of death. One who is careful does not die; one who is careless always dies.

(2) Sanskrit version:
apramādo ’mṛtapadaṃ, pramādo mṛtyunaḥ padaṃ | apramattā na mriyante, pramattās tu sadā mṛtāḥ || (Enomoto 1989: 29 [YBh 124a4; ŚrBh 5a1.3]) Carefulness is the path of the deathless; carelessness is the path of death.
Those who are careful do not die,
but those who are careless are always dead.


(3) Tibetan version:
/ bag yod pa ni 'chi me dag nas / / bag med pa ni 'chi ba'i gnas /
/ bag yod rnams ni 'chi ba med / / bag med rnams ni rtag tu 'chi / (Peking 5536.279b8–280a1; Derge 4035.240b4)
Carefulness is the abode of the deathless; carelessness is the abode of death.
Those who are careful are undying; those who are careless always die.

2. Comparison of the verse with other texts
The corresponding Udānavarga verse is Uv-S 4.1.498 Parallel verses may be found at: Uv-B 4.1;499 Dhp 21;500 Nett 3.A.6;501 Jā 16.10;502 PDhp 14;503 GDhp 115;504 T376

497 The three versions are overall the same.
 
(Fóshuō dàbōníhuán jīng 佛說大般泥洹經);505 T374 (Dà bōnièpán jīng 大般涅槃經);506
T375 (Dà bōnièpán jīng 大般涅槃經);507 T721 (Zhèngfǎ niànchù jīng 正法念處經).508
Parallel verses with some differences in content may be found at: EĀ 10.1,509 35.4;510 T210511 (Fǎjù jīng 法句經), T212512 (Chūyào jīng 出曜經); T213513 (Fǎjíyàosòng jīng 法集要頌經). Except for EĀ 10.1, these texts indicate that “moral conduct” (jiè 戒) rather than “carefulness” is the path of the deathless. Also, “craving” (tān 貪 ) is the key for death. EĀ 10.1 states that “arrogance” (jiāo 憍 ) is the path of death, and those who are conceited are dead.

4.1.3. The third set: strong lust (’dod chags shas che); 4 pādas / 1 verse

1. Three versions with their translations514
(1) Chinese version:
眾生尋思所鑽搖 猛利貪欲隨觀妙 倍增染愛而流轉 便能自為堅固縛 (T1579, 30.379b24) Sentient beings are intensely disturbed by discursive thoughts;
[their] strong lust is caused by looking at beauty.

498 apramādo ’mṛtapadaṃ, pramādo mṛtyunaḥ padaṃ | apramattā na mriyante, pramattās tu sadā mṛtāḥ || (Uv-S 50.16–20).
499 apramādo hi ’mṛtapadaṃ, pramādo mṛtyunaḥ padaṃ | apramattā na mriyante, ye pramattāḥ sadā mṛtāḥ || (Uv-B 126.1–2).
500 appamādo amatapadaṃ, pamādo maccuno padaṃ | appamattā na mīyanti, ye pamattā yathā matā || 501 appamādo amatapadaṃ, pamādo maccuno padaṃ | appamattā na mīyanti, ye pamattā yathā matā ti || (Nett 34.1–2).
502 appamādo amatapadaṃ, pamādo maccuno padaṃ | appamattā na mīyanti, ye pamattā yathāmatā || (Jā V 99.23–24).
503 apramādo amatapadaṃ, pramādo maccuno padaṃ | apramattā na mrīyanti, ye pramattā yathā matā ||
504 apramadu amuda-pada, pramadu mucuṇo pada | apramata na miyadi, ye pramata yadha mudu ||
505 若無放逸 是處不死 若其放逸 是為死徑 無放逸者 得不死處 若放逸者 常處生死 (T 376, 12.889c27–29).
506謹慎無放逸 是處名甘露 放逸不謹慎 是名為死句 若不放逸者 則得不死處 如其放逸者 常趣於死路
(T374, 12.415c2–5).
507謹慎無放逸 是處名甘露 放逸不謹慎 是名為死句 若不放逸者 則得不死處 如其放逸者 常趣於死路
(T375, 12.656b28–c2).
508 謹慎不放逸 是處名甘露 若行放逸者 是名為死句 若不放逸者 常得不死處 若行放逸者 常趣於死路(T721, 17.133a8–12); 不放逸不死 放逸是死處 不放逸不死 放逸常生死 不放逸不死 放逸是死句 (T721, 17.146b6–9).
509 無憍甘露跡 放逸是死徑 無慢則不死 慢者即是死 (T125, 2.563c28–29).
510 戒為甘露道 放逸為死徑 不貪則不死 失道為自喪 (T125, 2.699b17–18).
511 戒為甘露道 放逸為死徑 不貪則不死 失道為自喪 (T210, 4.562b21–23).
512 戒為甘露道 放逸為死徑 不貪則不死 失道為自喪 (T212, 4.636c29– 637a1).
513 戒為甘露道 放逸為死徑 不貪則不死 失道為自喪 (T213, 4.779a1–3).
514 These three versions are basically the same.
 
This multiplies unwholesome craving and causes transmigration, in such a way that they are firmly bonded.

(2) Sanskrit version:
vitarkapramathitasya dehinas, tīvrarāgasya śubhānudarśinaḥ | bhūyas tṛṣṇā vivardhate, sā gāḍhīkurute 'sya bandhanaṃ || (Enomoto 1989: 30 [YBh 124b3; ŚrBh 5a1.9])
For a living creature who is disturbed by discursive thoughts, has strong lust [and] looks at the beautiful,
The craving increases more, it makes his bondage strong.

(3) Tibetan version:
/ rtogs515pas rab tu bcom pa'i skye ba bo / / 'dod chags shas cher sdug par lta ba ni /
/ sred pa phyir zhing rnam par 'phel 'gyur te / / des ni de yi 'ching ba dam par byed / (Peking 5536.280b8; Derge 4035.241b1)
For those who are overwhelmed by thoughts, who looking upon beauty with strong lust; their craving thus manifests increasingly; thus, their bondages are made tight.

2. Comparison of the verse with other texts
The corresponding Udānavarga verse is Uv-S 3.1.516 Parallel verses may be found at: Uv-B 3.1;517 Dhp 349.518 Parallel verses with some differences in content may be found at: T210519 (Fǎjù jīng 法句經); T211520 (Fǎjù pìyù jīng 法句譬喻經); T212521 (Chūyào jīng

515 Derge: rtog.
516 vitarkapramathitasya dehinas, tīvrarāgasya śubhānudarśinaḥ | bhūyas tṛṣṇā vivardhate, sā gāḍhīkurute 'sya bandhanaṃ || (Uv-S 50.21–26).
517 vitarkapramathitasya jantunas, tīvrarāgasya śubhānudarśinaḥ | bhūyas tṛṣṇā pravardhate, gāḍhaṃ hy eṣa karoti bandhanaṃ || (Uv-B 119.1–4).
518 vitakkapamathitassa jantuno, tibbarāgassa subhānupassino | bhiyyo taṇhā pavaḍḍhati esa kho daḷhaṃ karoti bandhanaṃ ||
519 心念放逸者 見婬以為淨 恩愛意盛增 從是造獄牢 (T210, 4.571a23–25).
520 心念放逸者 見婬以為淨 恩愛意盛增 從是造牢獄 (T211, 4.602b27–29).
521 夫人無止觀 多欲觀清淨 倍增生愛著 縛結遂固深 (T212, 4.632b21–22).
 
出曜經); T213522 (Fǎjíyàosòng jīng 法集要頌經). The primary difference is in the first two pādas, which states that one who is careless regards lust as a pure object.

4.1.4. The fourth set: dwelling in the dharma (chos la gnas); 4 pādas / 1 verse

1. Three versions with their translations523
(1) Chinese version:
住法具尸羅 有慚言諦實 能保愛自身 亦令他所愛 (T1579, 30.380b10)
One who abides by the dharma and embodies moral conduct has a sense of shame and tells the truth.
He cherishes himself, and is adored by others.

(2) Sanskrit version:
dharmasthaṃ śīlasaṃpannaṃ, hrīmantaṃ satyavādinaṃ | ātmanaḥ priyakartāraṃ, taṃ janaḥ kurute priyaṃ || (Enomoto 1989: 30 [YBh 125b4f; ŚrBh 5a2.7])
One who dwells in dharma, is endowed with moral conduct, has a sense of moral shame, speaks the truth,
acts dear towards to himself. People treat him dearly.

(3) Tibetan version:
/ chos la gnas shing tshul khrims ldan / / ngo tsha shes ldan bden par smra /
/ bdag nyid kyang ni dga' bar byed / / de la skye bo dga' bar 'dzin / (Peking 5536.283a8; Derge 4035.243b1–2)
One who abides in the dharma, is endowed with moral conduct, has a sense of moral shame and speak truthfully,
also holds himself dear. People treat him as dear.

522 極貪善顯現 有情懷疑慮 若復增貪意 自作堅固縛 (T213. 4.778b16–18).
523 These three versions are mostly the same.
 
2. Comparison of the verse with other texts
The corresponding Udānavarga verse is Uv-S 5.24.524 Parallel verses may be found at: Uv-B 5.24;525 Dhp 217;526 PDhp 294;527 GDhp 322;528 T210529 (Fǎjù jīng 法句經);
T211530 (Fǎjù pìyù jīng 法句譬喻經); T212531 (Chūyào jīng 出曜經); T213532
(Fǎjíyàosòng jīng 法集要頌經). These texts share the same information in content.

4.1.5. The fifth set: contemplating the truth (nges rtag); 4 pādas / 1 verse

1. Three versions with their translations533
(1) Chinese version:
若見他惡業 能審諦思惟 自身終不為 由彼業能縛 (T1579, 30.380b26) If one saw the evil deeds of others,
one could think about it based on the truth; one would never perform them,
because that [[[evil]]] karma can bind oneself.


(2) Sanskrit version:
yat pareṣūpanidhyāyet, karma dṛṣṭveha pāpakam | ātmanā tan na kurvīta, karmabaddho hi pāpakaḥ || (Enomoto 1989: 30 [YBh 126a1; ŚrBh 5b2.2]) Were one to contemplate a bad action in others which one has seen in this world,
one would not perform it himself.

524 dharmasthaṃ śīlasaṃpannaṃ, hrīmantaṃ satyavādinaṃ | ātmanaḥ priyakartāraṃ, taṃ janaḥ kurute priyaṃ || (Uv-S 50.27–28).
525 dharmasthaṃ śīlasaṃpannaṃ, hrīmantaṃ satyavādinaṃ | ātmanaḥ kartāraṃ santaṃ, taṃ janaḥ kurute priyaṃ || (Uv-B 146.1–2).
526 sīladassanasampannaṃ, dhammaṭṭhaṃ saccavedinaṃ | attano kamma kubbānaṃ, taṃ jano kurute piyaṃ
||
527 śilavantaṃ śuciṃ dacchaṃ, dhammaṭṭhaṃ saccavādinaṃ | āttano kārakaṃ śantaṃ, taṃ jano kurute priyaṃ ||
528 śilamadu suyi-drakṣo, dhamaṭho sadhujivaṇo | atvaṇo karako sadu, ta jaṇo kuradi pri’u ||
529 貪法戒成 至誠知慚 行身近道 為眾所愛 (T210, 4.567.c22–23).
530 貪法戒成 至誠知慚 行身近道 為眾所愛 (T211, 4.595c17–18).
531 樂法戒成就 誠信樂而習 能自勅身者 為人所愛敬 (T212, 4.654a16–17).
532 樂法戒成就 成信樂而習 能誡自身者 為人所愛敬 (T213, 4.780b7–9).
533 These three versions are basically the same.
 
For a bad person is bound to his actions.


(3) Tibetan version:
/ ’di na gzhan gyi sdig pa'i las / / mthong nas nges par sems byed pa /
/ las kyi 'ching ba sdig yin te / / de ni bdag gis yong mi bya / (Peking 5536.283b8; Derge 4035.244a1)
One who constantly contemplates
having seen the evil action of others in this world, that there is evil bondage to [such] actions.
He will never perform them himself.


2. Comparison of the verse with other texts
The corresponding Udānavarga verse is Uv-S 9.6.534 Parallel verses may be found at: Uv-B 9.6;535 Th 496536.

4.1.6. The sixth set: good saying (legs par smras pa); 4 pādas / 1 verse

1. Three versions with their translations537
(1) Chinese version:
賢聖常說最善語 愛非不愛語第二 諦非不諦語第三 法非非法語第四 (T1579, 30.381b20) Noble beings always speak the most wholesome words.
Secondly, they speak kind, not unkind, words. Thirdly, they speak truthful, not untruthful, words.
Fourthly, they speak words of the dharma, not of non-dharma.


(2) Sanskrit version:
subhāṣitaṃ hy uttamam āhur āryāḥ, priyaṃ vaden nāpriyaṃ tad dvitīyaṃ |



534 yat pareṣūpanidhyāyet, karma dṛṣṭveha pāpakam | ātmanā tan na kurvīta, karmabaddho hi pāpakaḥ ||
(Uv-S 50.29–30).
535 yat pareṣāṃ vigarheta (v.l. upanidhāti), karma dṛṣṭveha pāpakam | ātmanā tan na kurvīta, karmabaddho hi pāpakaḥ || (Uv-B 171.1–2).
536 na parass’ upanidhāya, kammaṃ maccassa pāpakaṃ | attanā taṃ na seveyya, kammabandhū hi mātiyā ||
537 These three versions are mostly the same.
 
satyaṃ vaden nānṛtaṃ tat tṛtīyaṃ, dharmaṃ vaden nādharmaṃ tat caturtham || (Enomoto 1989: 30 [YBh 127a4f; ŚrBh 5a3.3f])
For first, the noble ones speak what is well said.
Secondly, one should speak that which is kind, not unkind. Thirdly, one should speak that which is true, not untrue.
Fourthly, one should speak that which is dharma, not non-dharma.


(3) Tibetan version:
/ 'phags pa rnams ni legs par smras pa mchog ces gsung /
/ snyan smra mi snyan ma yin de ni gnyis pa yin /
/ bden smra mi bden538 ma yin de ni gsum pa yin /
/ chos smra chos min ma yin539 de ni bzhi pa yin / (Peking 5536.286a4; Derge 4035.245b7)
Firstly, it is said that the noble ones are ones who speak what is well said Secondly, that they speak kindly and not unkindly,
Thirdly, that they speak truthfully and not untruthfully, Fourthly, that they speak dharma and not non-dharma.

2. Comparison of the verse with other texts
The corresponding Udānavarga verse is Uv-S 8.11.540 Parallel verses may be found at: SĀ-G 1218;541 SĀ-U 253.542 The following texts share similar information, but the order of the verses is different from Uv-S: Uv-B 8.11;543 SN 8.5;544 Sn 450.545 These texts place the sentence about “speaking that which is dharma” in the second verse, rather

538 Derge: brten.
539 “ma yin” is omitted in the Peking version.
540 subhāṣitaṃ hy uttamam āhur āryāḥ, priyaṃ vaden nāpriyaṃ tad dvitīyaṃ | satyaṃ vaden nānṛtaṃ tat tṛtīyaṃ, dharmaṃ vaden nādharmaṃ tat caturtham || (Uv-S 50.31–51.6).
541 賢聖善說法 是則為最上 愛說非不愛 是則為第二 諦說非虛妄 是則第三說 法說不異言 是則為第四
(T99, 2.322a11–14).
542 善說最為上 仙聖之所說 愛語非麁語 是名為第二 實語非妄語 是名為第三 說法不非法 是名為第四
(T100, 2.462b23–27).
543 subhāṣitaṃ hy uttamam āhur āryā, dharmaṃ vaden nādharmaṃ tad dvitīyaṃ | priyaṃ vaden nāpriyaṃ tat tṛtīyaṃ, satyaṃ vaden nāsatyaṃ tac caturtham || (Uv-B 166.1–4).
544 subhāsitaṃ uttamam āhu santo, dhammaṃ bhaṇe nādhammaṃ taṃ dutiyaṃ | piyaṃ bhaṇe nāppiyaṃ taṃ tatiyaṃ, saccaṃ bhaṇe nālikaṃ taṃ catutthan ti || (SN I 189.7–10 = SNN I 407.5–9).
545 subhāsitaṃ uttamam āhu santo, dhammaṃ bhaṇe nādhammaṃ taṃ dutiyaṃ | piyaṃ bhaṇe nāppiyaṃ taṃ tatiyaṃ, saccaṃ bhaṇe nālikaṃ taṃ catutthan ti ||
 
than in the fourth. So the sentences concerning speaking that which is kind and true are given after.
In T1545546 (Āpídámó dàpípóshā lùn 阿毘達磨大毘婆沙論, the Chinese
Abhidharmamahā-vibhāṣā) and T1546547 (Āpítán pípóshā lùn 阿毘曇毘婆沙論, the Chinese Abhidharma-vibhāṣā), the order of the verses is the same as this sixth verse set and Uv-S 8.11. These two texts are attributed to the (Mūla)sarvāstivādins. It seems that the order of the verses in Uv-B 8.11, which belongs to the Sarvāstivādins, is uncommon.

4.1.7. The seventh set: faith (dad pa); 4 pādas / 1 verse

1. Three versions with their translations548
(1) Chinese version:
信慚戒施法 善人所稱讚 是名趣天道 能往天世間 (T1579, 30.381c12)
Faith, shame, moral conduct, and giving of the dharmas are lauded by good people.
This is called the path of heaven,
by which one can go to the heavenly realm.


(2) Sanskrit version:
śraddhātha hrīḥ śīlam athāpi dānaṃ, dharmā ime satpuruṣapraśastāḥ | etaṃ hi mārgaṃ divigaṃ vadanti, etena vai gacchati devalokam || (Enomoto 1989: 30 [YBh 127b2; ŚrBh 5b3.6f])
Faith, shame, moral conduct and also giving, these qualities are praised by good people, for they say this is the path leading to heaven.
One goes to the heavenly realm by just this [[[path]]].


(3) Tibetan version:
/ dad dang ngo tsha shes dang tshul khrims sbyin /


546 賢聖法中善言最 二常愛言遠不愛 三常實言離虛誑 四常法言遠非法 (T1545, 27.28c15–16).
547 賢聖法中善言最 二常愛言遠不愛 三常實言離虛妄 四常法言遠非法 (T1546, 28.20c7–8).
548 These three versions are mostly the same.
 
/ chos 'di dag ni skyes bu dam pas bsngags /
/ lam 'di lta yul 'gro ba yin zhes gsungs /
/ 'dis ni lha yi549'jig rten 'gro bar 'gyur / (Peking 5536.286b4; Derge 4035.246a6–7) Faith, shame, moral conduct, generosity, these qualities are praised by good beings.
This is called the path leading to heaven by which one goes to the heavenly realm.

2. Comparison of the verse with other texts
The corresponding Udānavarga verse is Uv-S 10.1.550 Parallel verses may be found at: Uv-B 10.1;551 AN IV 32;552 T210553 (Fǎjù jīng 法句經); T212554 (Chūyào jīng 出曜經); T213555 (Fǎjíyàosòng jīng 法集要頌經). However, AN IV 32 does not mention moral conduct (sīla). Instead, it emphasizes “good action” (kusala). This is also the case in a similar Pāli passage, Kathāvatthu VII.4.556 It seems that the Theravāda school has its own particular reading in this case.

4.1.8. The eighth set: learning (thos pa); 4 pādas / 1 verse

1. Three versions with their translations557
(1) Chinese version:
多聞能知法 多聞能遠惡 多聞捨無義 多聞得涅槃 (T1579, 30.382a9) By learning widely, one can know the dharma.


549 Peking: lhi'.
550 śraddhātha hrīḥ śīlam athāpi dānaṃ, dharmā ime satpuruṣapraśastāḥ | etaṃ hi mārgaṃ divigaṃ vadanti, etena vai gacchati devalokam || (Uv-S 51.7–11).
551 śraddhātha hrīśīlam athāpi dānaṃ, dharmā ime satpuruṣapraśastāḥ | etaṃ hi mārgaṃ divyaṃ vadanti, etenāsau gacchati devalokam || (Uv-B 216.1–4).
552 saddhā hiriyaṃ kusalañ ca dānaṃ, dhammā ete sappurisānuyātā | etaṃ hi maggaṃ diviyaṃ vadanti, etena hi gacchati devalokan ti || (AN IV 236.11–14).
553 信慚戒意財 是法雅士譽 斯道明智說 如是昇天世 (T210 4.560b22–24).
554 信慚戒意財 是法雅士譽 斯道明智說 如是昇天世 (T212 4.672a17–18).
555 信慚戒布施 上士譽此法 斯道明智說 得生於天界 (T213, 4.782a19–21).
556 saddhā hiriyaṃ kusalañ ca dānaṃ, dhammā ete sappurisānuyātā | etaṃ hi maggaṃ diviyaṃ vadanti, etena hi gacchati devalokan ti || (Kv 341.1–4).
557 These three versions are overall the same.
 
By learning widely, one can be distanced from evil. By learning widely, one discards the meaningless. By learning widely, one attains nirvāṇa.

(2) Sanskrit version:
śrutvā dharmaṃ vijānāti, śrutvā pāpān nivartate | śrutvā anarthaṃ tyajati, śrutvā prāpnoti nirvṛtim || (Enomoto 1989: 30 [YBh 128a1f; ŚrBh 5a4.5]) After learning, one knows the dharma.
After learning, one discards evil.
After learning, one abandons what is not beneficial. After learning, one achieves liberation.

(3) Tibetan version:
/ thos pas chos rnams shes par 'gyur /
/ thos pas sdig las ldog par 'gyur /
/ thos pas gnod pa sbong bar 'gyur /
/ thos pas mya ngan 'das pa thob /
(Peking 5536.287b3–4; Derge 4035.247a3) By learning, the dharma is understood.
By learning, evil actions are averted. By learning, harm is abandoned.
By learning, nirvāṇa is attained.


2. Comparison of the verse with other texts
The corresponding Udānavarga verse is Uv-S 22.6.558 A parallel verse may be found at Uv-B 22.6.559 Parallel verses with some differences in content may be found at: T210560 (Fǎjù jīng 法句經); T211561 (Fǎjù pìyù jīng 法句譬喻經); T213562 (Fǎjíyàosòng

558 śrutvā dharmaṃ vijānāti, śrutvā pāpān nivartate | śrutvā anarthaṃ tyajati, śrutvā prāpnoti nirvṛtim || (Uv-S 51.13–16).
559 śrutvā dharmaṃ vijānāti, śrutvā pāpāṃ na sevate | śrutvā hy anarthaṃ varjayate, śrutvā prāpnoti nirvṛtim || (Uv-B 286.3–4).
560 聞為知法律 解疑亦見正 從聞捨非法 行到不死處 (T210, 4.560a16–18).
 
jīng 法集要頌經); T754563 (Fóshuō wèicéngyǒuyīnyuán jīng 佛說未曾有因緣經. The differences lie in the second and third pādas, which state, “one solves the confusion and corrects the knowledge; after learning, one abandons that which is not dharma.” These two pādas seem to respond to the first pāda (“after learning, one knows the dharma”), which means that three pādas concern one topic. But in the eighth verse set, there is one topic per pāda.564 Importantly, this verse can also be found in the vibhāṣā texts attributed to the (Mūla)sarvāstivādins, and the content of these parallels is exactly the same as in the eighth verse set.565

4.1.9. The ninth set: being like the sky (nam mkha' mtshungs); 4 pādas / 1 verse

1. Three versions with their translations566
(1) Chinese version:
智者如空無染污 不動猶如天帝幢 如泛清涼盈滿池 不樂淤泥生死海 (T1579, 30.382b1) The wise one who is like the sky without any defilement,
is immobile like the pillar/banner of the heavenly king,
is like floating across a pond full of fresh and clear water, not taking delight in the muddy ocean of saṃsāra.

(2) Sanskrit version:
ākāśasamo na lipyate, indrakīlapratimo na kaṃpate |
hrada iva samupetakardame, saṃsāre ramate na paṇḍitaḥ || (Enomoto 1989: 31 [YBh 128a6f; ŚrBh 5b4.3])

561 聞為知法律 解疑亦見正 從聞捨非法 行到不死處 (T211, 4.578c13–14).
562 聞為知法律 解疑亦見正 從聞捨非法 行到不死處 (T213, 4.788b4–6).
563 聞為知法律 解疑亦見正 從聞捨非法 行到不死處 (T754, 17.586c29–587a2).
564 The topics are “knows the dharma”, “discards evil”, “abandons what is not beneficial”, and “achieves liberation”.
565 T1545 (Āpídámó dàpípóshā lùn 阿毘達磨大毘婆沙論): 多聞能知法 多聞離不善 多聞捨無義 多聞得涅槃 (T1545, 27.2b8–9; 27.731a15–16); T1546 (Āpítán pípóshā lùn 阿毘曇毘婆沙論): 多聞能知法 多聞能遠惡 多聞離無義 多聞得涅槃 (T1546, 28.2b20–21; 28.273a13–14); T1547 (Pípóshā lùn 鞞婆沙論): 聞法能知 聞惡不作 聞除非義 聞得至滅 (T1547, 28.416c29–417a1; 28.442a23–24).
566 In the third pāda, the Chinese version is different from the other two and is closer to the third pāda in Uv-B 17.12 (see below). In the Sanskrit and Tibetan versions, saṃsāra is likened to a lake with mud; however, in the Chinese version, the simile is positive in that the wise one is likened to a pond with clear water (i.e., without mud). A similar version of the Chinese third pāda can also be found in T210 (淨如水無
垢) and T212 (澄如清泉).
 
Like the sky, he does not get stained.
Like a threshold-stone, he does not tremble. As though it were a lake abounding with mud, the wise one does not delight in saṃsāra.

(3) Tibetan version:
/ nam mkha' 'dra ba gos pa med / / them pa 'dra ba yo567mi 'gyur /
/ mtsho dang568'dam rdzab ldan 'dra ba'i / / 'khor ba la ni mkhas mi dga' / (Peking 5536.288a7; Derge 4035.247b4–5)
Like the sky, he does not get stained. Like a threshold, he does not tremble.
The wise one has no fondness for saṃsāra
which is like a lake or swamp filled with mud.


2. Comparison of the verse with other texts
The corresponding Udānavarga verse is Uv-S 17.12.569 Parallel verses with some differences in content may be found at: Uv-B 17.12;570 Dhp 95;571 T210572 (Fǎjù jīng法句經); T212573 (Chūyào jīng 出曜經); T213574 (Fǎjíyàosòng jīng 法集要頌經). In the first pāda of the ninth verse set in the Śarīrārthagāthā and Uv-S 17.12, the described object is “the sky.” In all other parallels, it is “the earth.” The second pāda of the ninth verse set and Uv-S 17.12 concerns the “threshold-stone” (indrakīlapratima), but it is quite divergent in all other parallels.575 The third pāda in Uv-B576 and Dhp 95577 mentions that

567 Derge: g.yo.
568 Derge: bran.
569 ākāśasamo na lipyate, indrakīlapratimo na kaṃpate | hrada iva samupetakardame, saṃsāre ramate na paṇḍitaḥ || (Uv-S 51.17–24).
570 pr̥thivīsadr̥śo na lipyate, tāyī kīlavad aprakaṃpayaḥ (v.l. apramatto yaḥ) | hrada iva hi vinītakardamo, niṣkaluṣā hi bhavanti na paṇḍitāḥ || (Uv-B 237.3–6).
571 paṭhavīsamo no virujjhati, indakhīlūpamo tādi subbato | rahado va apetakaddamo, saṃsārā na bhavanti tādino ||
572 不怒如地 不動如山 真人無垢 生死世絕 (T210, 4.564b9–10); 受辱心如地 行忍如門閾， 淨如水無垢 生盡無彼受 (T210, 4.573c8–10).
573 忍心如地 不動如安明 澄如清泉 智者無亂 (T212, 4.708b8–9).
574 忍心如大地 不動如虛空 聞法喻金剛 獲味免輪迴 (T213, 4.785c28–29).
575 It is about stake/pillar (kīla) in Uv-B, threshold-stone (indakhīlūpamo) in Dhp, mountain in T210 and T212, bolt in T210, and sky in T213.
576 hrada iva hi vinītakardamo (Uv-B 237.5).
 
the pond is rid of mud, which is a simile for the wise. The Chinese instead contains a positive simile which compares the wise with a clean pond, instead of a pond without mud. A similar description can also be seen T210 (淨如水無垢) and T212 (澄如清泉). However, the Sanskrit and Tibetan versions instead contain a simile which compares a lake abounding with mud to saṃsāra.
Among these texts, the content is overall divergent. However, the content in Uv-B
17.2 is very similar to T212.

4.1.10. The tenth set: physical appearance (gzugs); 20 pādas / 5 verses

1. Three versions with their translations578
(1) Chinese version:
若以色量我 以音聲尋我 欲貪所執持 彼不能知我若於內了知 於外不能見 由內果觀察 彼音聲所引若於內無知 於外而能見 由外果觀察 亦音聲所引若於內無知 於外不能見 彼普障愚夫 亦音聲所引
若於內了知 於外亦能見 英雄出離慧 非音聲所引 (T1579, 30.382b24–c4) If one, who uses [[[physical]]] form to evaluate me,
and uses sound of the voice to find me, is bound by longing and lust,
he cannot know me.
If one, who knows internally and cannot see externally, observes from internal fruit, he is led by sound of the voice.
If one, who does not know internally, and can see externally,
observes from external fruit,
he is also led by sound of the voice. If one who does not know internally
577 rahado va apetakaddamo.
578 These three versions are basically the same.
 
and cannot see externally, is a fool fully obstructed,
he is also led by sound of the voice. If one who knows internally,
and also sees externally,
is a hero with the wisdom of transcendence, he is not led by sound of the voice.

(2) Sanskrit version:
ye rūpeṇa pramiṇvanti, māṃ ghoṣeṇānuyānti ca | cchandarāgavaśopetā, na mā jānanti te janāḥ || adhyātmaṃ ca vijānāti, bahirdhā ca na paśyati | adhyātmaphaladarśī yaḥ, sa vai ghoṣeṇa nīyate || adhyātmaṃ ca na jānāti, bahirdhā ca vipaśyati | bahirdhāphaladarśī yaḥ, so 'pi ghoṣeṇa nīyate || adhyātmaṃ ca na jānāti, bahirdhā ca na paśyati | samantāvaraṇo bālaḥ, so 'pi ghoṣeṇa nīyate || adhyātmaṃ ca vijānāti, bahirdhā ca vipaśyati | dhīro niḥsaraṇaprajño, na sa ghoṣeṇa nīyate || (Enomoto 1989: 31 [YBh 128b5f; ŚrBh 5b5.1f])
Those people who judge me by appearance and follow me by my voice,
who have come under the control of longing and lust, do not know me.
He who knows internally, but does not see externally, observes the fruit internally.
He indeed is led by my voice.
He who does not know internally, but sees externally,
observes the fruit externally, he also is led by my voice.
 
The fool, having obstructions on all sides, does not know internally,
nor sees externally.
He also is led by my voice.
The wise one, having the wisdom of transcendence, knows internally,
and sees externally.
He is not led by my voice.


(3) Tibetan version:
/ gang dag gzugs kyis nges par 'dzin / /nga la sgra yis rjes su 'brang /
/ 'dun pa'i 'dod chags dbang gyur pa / / skye bo de dag nga mi shes /
/ nang ni rnam par shes gyur la / / phyi579rol mthong ba ma yin zhing /
/ nang gi 'bras bu mthong ba gang / / de ni sgra yis bkri ba yin /
/ nang ni shes ba ma yin la / / phyi rol mthong bar gyur pa la /
/ phyi rol 'bras bu mthong ba gang / /de yang sgra yis bkri ba yin/
/ nang yang shes pa ma yin la / /phyi rol mthong ba'ang ma yin zhing /
/ byis pa kun nas bsgribs gyur ba / /de yang sgra yis bkri ba yin/
/ nang yang rnam par shes gyur pa580 / / phyi rol yang ni mthong gyur cing /
/ bstan pa nges par 'byung shes pa / / de ni sgra yis bkri ba yin / (Peking 5536.289a3–7; Derge 4035.248a7–b2)
Those who conceive of me by physical appearance, who follow me by sound of the voice,
and who are controlled by longing and lust, those people do not know me.
Those who clearly understand internally, yet do not see externally;
those who see the internal fruit, are led by sound of the voice.
Those who do not understand internally,


579 Derge: pha.
580 Derge: la.
 
yet see externally,
while seeing the external fruit, are also led by sound of the voice.
Those who neither understand internally, nor see externally,
are the fools fully obscured, also led by sound of the voice.
Those who have clearly understood internally, and have also seen externally,
knowing the teachings of transcendence, are not led by sound of the voice.

2. Comparison of the verse with other texts
The corresponding Udānavarga verse is Uv-S 22.12–16.581 The following parallels have a different verse order:
(1) Uv-B 22.12–16:582 the second and the third verses are switched compared with Uv-S. Also in the fourth and fifth verses, the subjects are not described as “the fool” (bāla) and “the wise” (dhīra), respectively.
(2) AN II 7.65583and Th 469–472:584 these two texts share the same content. However, the third verse in the tenth verse set and Uv-S is missing in these two parallels. Also, the

581 ye rūpeṇa pramiṇvanti, māṃ ghoṣeṇānuyānti ca | cchandarāgavaśopetā, na mā jānanti te janāḥ || adhyātmaṃ ca vijānāti, bahirdhā ca na paśyati | adhyātmaphaladarśī yaḥ, sa vai ghoṣeṇa nīyate || adhyātmaṃ ca na jānāti, bahirdhā ca vipaśyati | bahirdhāphaladarśī yaḥ, so 'pi ghoṣeṇa nīyate || adhyātmaṃ ca na jānāti, bahirdhā ca na paśyati | samantāvaraṇo bālaḥ, so 'pi ghoṣeṇa nīyate || adhyātmaṃ ca vijānāti, bahirdhā ca vipaśyati | dhīro niḥsaraṇaprajño, na sa ghoṣeṇa nīyate || (Uv-S 51.25– 52.17).
582 ye me rūpeṇa minvanti (v.l. ye rūpeṇa pramiṇvanti), ye me ghoṣeṇa ca anvagāḥ (v.l. māṃ ghoṣeṇānuyānti ca) | chandarāgavaśopetā, na māṃ jānanti te janāḥ ||
ādhyātmaṃ ca na jānāti, bahirdhā tu vipaśyati | bahirdhāphaladarśī (v.l. adhyātmaphaladarśī) tu, sa vai ghoṣeṇa nīyate ||
ādhyātmaṃ tu prajānāti, bahirdhā ca na paśyati | ādhyātmaphaladarśī (v.l. bahirdhāphaladarśī) tu, sa vai ghoṣeṇa nīyate ||
ādhyātmaṃ ca na jānānti, bahirdhā ca na paśyati | ubhaye ’phaladarśī tu, sa vai ghoṣeṇa nīyate || ādhyātmaṃ ca prajānāti, bahirdhā ca vipaśyati | sa tu nihsaraṇaprajño, na vai ghoṣeṇa nīyate || (Uv-B 288.1-289.4).
583 ye ca rūpena pāmiṃsu, ye ca ghosena anvagū | chandarāgavasūpetā, na te jānanti taṃ janaṃ || ajjhattañ ca na jānāti, bahiddhā ca na passati | samantāvaraṇo bālo, sa ve ghosena vuyhati || ajjhattañ ca na jānāti, bahiddhā ca vipassati | bahiddhā phaladassāvī, so pi ghosena vuyhati ||
ajjhattañ ca pajānāti, bahiddhā ca vipassati | vinīvaraṇadassāvī, na so ghosena vuyhatī ti || (AN II 71.17–24).
 
order of the content is different. The second and fourth verse in the tenth verse set and Uv-S are the third and second verse in AN and Th. “The fool” (bāla) and “the seer” (dassāvin) are mentioned in the second and fourth verse in both AN and Th.
(3) T212585 (Chūyào jīng 出曜經): The third verse in the tenth verse set and Uv-S is missing in this parellel. The subjects are not described as “the fool” and “the wise.”
(4) T213586 (Fǎjíyàosòng jīng 法集要頌經): the second verse, which corresponds to the eighth verse set, is oddly inserted here. Also, the third verse in the tenth verse set and
Uv-S is missing in this parallel. In T213 the first two pādas of the fourth verse state “one who knows internally but cannot see externally;” however, the corresponding text in the above parallels (1–3) states “one who does not know internally but see externally.” Moreover, the subjects are not described as “the fool” and “the wise.”
From the above texts, it can be seen that the tenth verse set is more similar to AN II
7.65 and Th 469–472, although one verse is missing in these parallels. Importantly, the tenth verse set closely corresponds to the verses587 in T1451 (Gēnběnshuōyīqièyǒubù pínàiyē záshì 根 本 說 一 切 有 部 毘 奈 耶 雜 事 , the Chinese Mūlasarvāstivādavinayakṣudrakavastu), a Mūlasarvāstivādin vinaya text. This verse order and content therefore seem to be specific to this school. Consistent with this argument, another relevant text, Uv-B, which is attributed to the Sarvāstivādins, has a different verse order and content (as described above).









584 ye maṃ rūpena pāmiṃsu, ye ca ghosena anvagū | chandarāgavasūpetā, na maṃ jānanti te janā || ajjhattañca na jānāti, bahiddhā ca na passati | samantāvaraṇo bālo, sa ve ghosena vuyhati || ajjhattañca na jānāti, bahiddhā ca vipassati | bahiddhā phaladassāvī, sopi ghosena vuyhati || ajjhattañca pajānāti, bahiddhā ca vipassati | anāvaraṇadassāvī, na so ghosena vuyhatī ti ||
585 諸有稱己色 有歎說名德 斯皆謂貪欲 然自不覺知 (T212, 4.721c3–4); 內無自知 外無所見 內不見果 便隨聲往 (T212, 4.721c12–13); 內既知之 外無所見 內見果實 便隨聲往 內既不知 外有所見 二果俱成 便隨聲往 內有所知 外有所見 彼有朗智 不隨聲往 (T212, 4.721c21–25).
586 諸有稱己色 有歎說名德 斯皆諸貪欲 然自不覺知 聞為知法律 解疑亦見正 從聞捨非法 行到不死處 內無人自知 外無人所見 內不見其果 便隨聲而住 內既而知之 外無人所見 二果俱已成 便隨聲而住 內有而所知 外有而所見 彼有其明智 不隨聲而住 (T213, 4.788b2–12).
587 若以色見我 以音聲求我 愛染亂彼心 不能當見我 若人但知內 而不見於外 於內而求果 此為聲所迷 若人但知外 而不見於內 於外而求果 此亦聲所迷 若人不知內 亦復不見外 凡夫皆被障 此亦聲所迷 若人善知內 復善見於外 智者當出離 此不為聲迷 (T1451, 24.222b19–28).
 
4.1.11. The eleventh set: king (rgyal); 4 pādas / 1 verse

1. Three versions with their translations588
(1) Chinese version:
第六增上王 染時染自取 於無染不染 染者名愚夫 (T1579, 30.383a1) The sixth is the superior king.
When he is defiled, the defilement is taken by himself. In non-contamination, he is not defiled.
One who is defiled is called a fool.


(2) Sanskrit version:
ṣaṣṭhe adhipatau rājñi, rajyamāne rajasvalaḥ | arajasy arajā bhavati, rakto bālo nirucyate || (Enomoto 1989: 31 [YBh 12916; ŚrBh 5b5.7])
When the sixth ruling king is polluted, one is covered with impurity.
When he is not polluted, one is free from impurity. The polluted one is called a fool.

(3) Tibetan version:
/ drug pa bdag po rgyal po ni / /chags par gyur na chags can yin /
/ chags pa med na chags med 'gyur / / chags pa byis pa zhes bya'o / (Peking 5536.290a4; Derge 4035.249a5)
When the sixth ruling king becomes attached, one is attached.
When without attachment, one is not attached. The attached one is called a fool.








588 The three versions are similar in content.
 
2. Comparison of the verse with other texts
The corresponding Udānavarga verse is Uv-S 16.22.589 Parallel verses may be found at: Uv-B 16.22;590 T190591 (Fóběnxíngjí jīng 佛本行集經); T212592 (Chūyào jīng 出曜經); T213593 (Fǎjíyàosòng jīng法集要頌經). These parallels all share the same content.

4.1.12. The twelfth set: city of bone (rus pa'i ra ba); 4 pādas / 1 verse

1. Three versions with their translations594
(1) Chinese version:
有城骨為墉 筋肉而塗飾 其中有貪恚 慢覆所任持 (T1579, 30.383a18) There is a city that has bones as its pillars,
and is plastered with tendons and flesh. It is supported by lust, hatred,
Conceit, and concealment from within.


(2) Sanskrit version:
nagaram asthiprākāraṃ, snāyumāṃsānulepanaṃ | yatra rāgaś ca dveṣaś ca, māno mrakṣaś ca gāhate || (Enomoto 1989: 31 [YBh 129b3f; ŚrBh 5b5.2])
The town has walls of bone
[and] is plastered with sinews and flesh, into which lust, hatred,
conceit and hypocrisy enter.


(3) Tibetan version:
/ grong khyer rus pa'i ra ba can / / rgyus dang sha yi zhal zhal can /
/ de ni 'dod chags zhe sdang dang / / nga rgyal dang ni 'chab pa 'jug /


589 ṣaṣṭhe adhipatau rājñi, rajyamāne rajasvalaḥ | arajasy arajā bhavati, rakto bālo nirucyate || (Uv-S 52.18– 52.22).
590 ṣaṣṭha adhipatī rājā, rajyamāne rajasvalaḥ | arakte virajā bhavati, rakte bālo nirucyate || (Uv-B 231.1–2).
591 第六自在故 王染名曰染 無染而有染 是故名為癡 (T190, 3.827c3–6).
592 六增上王 染為染首 無染則離 染者謂愚 (T212, 4.706a14–15).
593 六識王為主 愛染為眷屬 無染則離愛 染著是愚癡 (T213, 4.785b24–26).
594 These three versions are basically the same.
 
(Peking 5536.290b4–5; Derge 4035.249b4) A city fenced in by bone,
plastered with tendons and flesh, [into] that [city] lust, hatred, conceit and concealment enter.

2. Comparison of the verse with other texts
The corresponding Udānavarga verse is Uv-S 16.23.595 Parallel verses may be found at: Uv-B 16.23;596 GDhp 284.597 These two parallels share the same content, but only two words are obviously different from Uv-S 16.23. In the second pāda, Uv-S contains the word “sinews” (snāyu), while the corresponding word in both Uv-B and GDhp is “blood” (śoṇita, lohida). In the last pāda, Uv-S contains the word “enter” (gāhate), while the corresponding word in Uv-B is “distressed” (bādhyate) and the corresponding word GDhp is “assembled” (samokadu). Parallel verses in which the last two pādas differ are:
(1) Dhp 150;598 T210599 (Fǎjù jīng法句經): The last two pādas state “old age and death and pride and hypocrisy are deposited.” The elements of “old age” and “death” are different.
(2) T212600 (Chūyào jīng 出曜經); T213601 (Fǎjíyàosòng jīng 法集要頌經): The last two pādas state “the sense facilities all open up, which is like making friends with thieves and being careless.” This content is totally different.
(3) T310602 (Dàbǎojī jīng 大寶積經); T1451603 (Gēnběnshuōyīqièyǒubù pínàiyē záshì 根本說一切有部毘奈耶雜事): The elements are lust, hatred and ignorance, rather than lust, hatred, conceit and hypocrisy.
In the parallels above, “blood” is mentioned rather than “sinews.” It seems that the version of this twelfth verse set and Uv-S is unique.
 
595 nagaram asthiprākāraṃ, snāyumāṃsānulepanaṃ | yatra rāgaś ca dveṣaś ca, māno mrakṣaś ca gāhate || (Uv-S 52.23–52.27).
596 nagaram hy asthiprākāraṃ, māṃsaśoṇitalepanaṃ | yatra rāgaś ca dveṣaś ca, māno mrakṣaś ca bādhyate
(v.l. mrakṣaḥ pragāhati) || (Uv-B 231.3–4).
597 nakara aṭhipakara, matsalohidalevaṇa | yatra rako ya doṣo ya, maṇo makṣo samokadu ||
598 aṭṭhīnaṃ nagaraṃ kataṃ, maṃsalohitalepanaṃ | yattha jarā ca maccu ca, māno makkho ca ohito ||
599 身為如城 骨幹肉塗 生至老死 但藏恚慢 (T210, 4.564c3–4).
600 骨幹以為城 肉血而塗之 根門盡開張 結賊得縱逸 (T212, 4.706b2–3).
601 骨幹以為城 肉血而塗之 根門盡開張 結賊得縱逸 (T213, 4.785b27–28).
602 身城骨牆壁 血肉作塗泥 畫彩貪瞋癡 隨處而莊飾 (T310, 11.335a22–24).
603 身城骨牆壁 血肉作塗泥 畫綵貪瞋癡 隨處而莊飾 (T1451, 24.260b21–23).
 
4.1.13. The thirteenth set: turtle (rus sbal); 4 pādas / 1 verse

1. Three versions with their translations604
(1) Chinese version:
如龜藏支於自殼 苾芻善攝意尋思 無所依止不惱他 證般涅槃無所謗 (T1579, 30.383b8) Just like the turtle which hides its limbs in the shell,
the monk well constrains discursive thoughts of the mind. Without anything to rely upon, he does not bother others.
Having achieved complete extinction (parinirvāṇa), there is no slander.


(2) Sanskrit version:
kūrmaḥ svake 'ṅgāni yathā kapāle, bhikṣur nnidadhyān manaso vitarkāni | aniśrito 'nyān aviheṭhamānaḥ, parinirvṛto nāpavadeta kaṃ cit || (Enomoto 1989: 32 [YBh 130a2; ŚrBh 5b5.6])
As a turtle [may withdraw] its limbs into its own shell,
[so] a monk should restrain discursive thoughts of the mind. Independent, not harassing others,
completely extinguished, he could not revile anyone.


(3) Tibetan version:
/ rus sbal bdag lus khog par sbas pa ltar /
/ dge slong yid kyi rtog rnams sbas nas ni /
/ rten pa med cing gzhan la mi 'tshe na /
/ mya ngan 'das shing gang la 'ang skur pa med / (Peking 5536.291a5–6; Derge 4035.250a4)
As a turtle hides its own body [in its shell],
so a monk holds back all thoughts of his mind, independent and doing no violence to others.
For one who has achieved nirvāṇa there is no slander.





604 These three versions are mostly the same.
 
2. Comparison of the verse with other texts:
The corresponding Udānavarga verse is Uv-S 26.1.605 Parallel verses may be found at: Uv-B 26.1;606 SN 1.2.7, 35.200.8;607 Mil 371;608 SĀ-G 600,609 1167;610 SĀ-U 174;611
T212612 (Chūyào jīng 出曜經); T213613 (Fǎjíyàosòng jīng 法集要頌經). However, in the second pāda of Uv-B, we do not find the terms “monk” (bhikṣu) or “mind” (manas), but rather the term “self” (ātman), which is quite different from other parallels.
The following two texts are different: T210614 (Fǎjù jīng 法句經); T211615 (Fǎjù pìyù jīng 法句譬喻經). The last three pādas state, “protecting the mind, which is like a citadel. [It is like] the fighting between wisdom and evil; there will be no suffering if conqured.” The content is totally different.

4.1.14. The fourteenth set: equality (mtshungs); 4 pādas / 1 verse

1. Three versions with their translations616
(1) Chinese version:
等不等而生 牟尼捨有行 內樂定差別 如俱舍卵生 (T1579, 30.383c1) To be born equal or unequal,
the sage617 discards the formation of existence. Distinguished by internal delight and concentration, [his birth is] like the oviparous birth from a shell.

605 kūrmaḥ svake 'ṅgāni yathā kapāle, bhikṣur nnidadhyān manaso vitarkāni | aniśrito 'nyān aviheṭhamānaḥ, parinirvṛto nāpavadeta kaṃ cit || (Uv-S 52.28–53.6).
606 kūrmo yathāṅgāni svake kapāle, samādadhītātmavitarkitāni | anihśrito hy anyam aheṭhayānaḥ, parinirvṛto nāpavadeta kaṃ cit || (Uv-B 317.1–4).
607 kummo va aṅgāni sake kapāle, samodahaṃ bhikkhu manovitakke | anissito aññam aheṭhayāno, parinibbuto nūpavadeyya kañcī ti || (SN I 7.17–20=SNN I 15.17–16.3; SN IV 179.1–4).
608 kummo va aṅgāni sake kapāle, samodahaṃ bhikkhu manovitakke | anissito aññam aheṭhayāno, parinibbuto na upavadeyya kañcī ti || Mil 371.28–372.2.
609 如龜善方便 以殼自藏六 比丘習禪思 善攝諸覺想 其心無所依 他莫能恐怖 是則自隱密 無能誹謗者
(T99, 2.160c7–10).
610 龜蟲畏野干 藏六於殼內 比丘善攝心 密藏諸覺想 不依不怖彼 覆心勿言說 (T99, 2.311c23–26).
611 比丘覆惡覺 譬如龜藏六 比丘無所依 亦不惱害彼 比丘入涅槃 都無有譏論 (T100, 2.437c11–13).
612 如龜藏其六 比丘攝意想 無猗無害彼 滅度無言說 (T212, 4.730c6–7).
613 如龜藏其六 苾芻攝意想 無倚無害彼 圓寂無言說 (T213, 4.779b16–18).
614 藏六如龜 防意如城 慧與魔戰 勝則無患 (T210, 4.563.a18–19).
615 藏六如龜 防意如城 慧與魔戰 勝則無患 (T211, 4.584b29–c1).
616 These three versions are basically the same.
617 Móuní 牟尼 is muni in Sanskrit, which means “sage,” i.e. the Buddha.
 
(2) Sanskrit version:
tulyam atulyaṃ ca saṃbhavaṃ, bhavasaṃskāram avāsṛjan muniḥ | adhyātmarataḥ samāhita, abhinat kośam ivāṇḍasaṃbhavaḥ || (Enomoto 1989: 32 [YBh 130b1f; ŚrBh 5a6.2])
[With] equal or unequal birth,
The sage abandoned the formation of existence. Internally delighting, concentrated,
like one whose birth is from an egg, he broke the shell.


(3) Tibetan version:
/ mtshungs dang mtshungs pa ma yin 'byung ba dang /
/ nang du dgyes par mdzad cing mnyam bzhag618 nas /
/ thub pas srid pa'i 'du byed spangs mdzad do /
/ sgo ngar skyes pas sbubs ni brtol ba bzhin / (Peking 5536.292a1–2; Derge 4035.250b5–6)
Arising as equal or unequal, delighting internally, concentrated,
the sage has abandoned the formation of existence, like breaking out from the shell of an egg.

2. Comparison of the verse with other texts
The corresponding Udānavarga verse is Uv-S 26.30.619 Parallel verses may be found at: Uv-B 26.30;620 DN 16.3.13;621 AN 70.9;622 Ud 6.1;623 DĀ T1;624 T212625 (Chūyào jīng
出曜經); T213626 (Fǎjíyàosòng jīng 法集要頌經). They all share the same content.


618 Derge: gzhag.
619 tulyam atulyaṃ ca saṃbhavaṃ, bhavasaṃskāram avāsṛjan muniḥ | adhyātmarataḥ samāhita, abhinat kośam ivāṇḍasaṃbhavaḥ || (Uv-S 53.7–10).
620 tulyam atulyaṃ ca sambhavaṃ, bhavasaṃskāram avāsṛjan muniḥ | ādhyātmarataḥ samāhito, hy abhinat kośam ivāṇḍasambhavam || (Uv-B 331.3–6).
621 tulam atulañ ca sambhavaṃ, bhavasaṃkhāram avassaji muni | ajjhattarato samāhito, abhida kavacam iv’ attasambhavan ti || (DN II 107.3–5).
622 tulam atulañ ca sambhavaṃ, bhavasaṅkhāram avassaji muni | ajjhattarato samāhito, abhindi kavacam iv’ attasambhavan ti || (AN IV 312.1–4).
623 tulam atulaṃ ca sambhavaṃ, bhavasaṅkhāram avassajji muni | ajjhattarato samāhito, abhindi kavacam iv’ attasambhavan ti || (Ud 64.29–30).
624 有無二行中 吾今捨有為 內專三昧定 如鳥出於卵 (T1, 1.15c24–25; 1.17a25–26).
 
4.1.15. The fifteenth set: mud (’dam); 4 pādas / 1 verse

1. Three versions with their translations627
(1) Chinese version:
無淤泥等欲 無魑魅等瞋 無羅網等癡 無江河等愛 (T1579, 30.383c20) There is no mud-like desire;
There is no demon-like anger; There is no net-like ignorance; There is no river-like craving.

(2) Sanskrit version:
nāsti kāmasamaḥ paṃko, nāsti dveṣasamo grahaḥ | nāsti mohasamaṃ jālaṃ, nāsti tṛṣṇāsamā nadī || (Enomoto 1989: 32 [YBh 130b5; ŚrBh 5a6.6]) There is no mud like desire;
there is no seizer like hatred; there is no net like ignorance; there is no river like craving.

(3) Tibetan version:
/ 'dod pa lta bu'i 'dam med de / / zhe sdang 'dra ba'i gdon yang med /
/ gti mug 'dra ba'i dra ba med / / sred pa 'dra ba'i chu klung med / (Peking 5536.292b3; Derge 4035.251a5–6)
There is no mud like desire,
nor is there a demon like hatred. There is no net like ignorance; nor is there a river like craving.





625 知節不知節 最勝捨有行 內自思惟行 如卵壞其膜 (T212, 4.753c2–3).
626 知節不知節 最勝捨有行 內自思惟行 如卵壞其膜 (T213, 4.791a27–29).
627 These three versions are basically the same. However, the grammar of the Chinese version is a bit different, which might be due to the interpretation of the translator(s).
 
2. Comparing the verse with other texts
The corresponding Udānavarga verse is Uv-S 29.37.628 Parallels in which the first pāda is different are: Uv-B 29.37;629 Dhp 251;630 T210631 (Fǎjù jīng 法 句 經 ). Uv-B describes desire as a “flood” (ogha), while Dhp and T210 describe it as “fire” (aggi, huǒ 火).

4.1.16. The sixteenth set: two dwellings632 (gnas gnyis); 4 pādas / 1 verse

1. Three versions with their translations633
(1) Chinese version:
虛空無鳥跡 外道無沙門 愚夫樂戲論 如來則無有 (T1579, 30.384a22) There is no bird’s trace in the sky;
there are no monks outside the [[[Buddhist]]] learning path. Fools take delight in idle discourse (xìlùn 戲論),
but the Tathāgata does not.


(2) Sanskrit version:
ākāśe vai padaṃ nāsti, śramaṇo nāsti bāhyakaḥ | prapañcābhiratā bālā, niṣprapañcās tathāgatāḥ || (Enomoto 1989: 32 [YBh 131b1f; ŚrBh 5a7.1]) There is no footprint in the sky;
there is no ascetic external [to the Buddha’s path]; Fools delight in conceptual proliferation;634

628 nāsti kāmasamaḥ paṃko, nāsti dveṣasamo grahaḥ | nāsti mohasamaṃ jālaṃ, nāsti tṛṣṇāsamā nadī || (Uv- S 53.11–13).
629 nāsti kāmasamo hy ogho, nāsti doṣasamo grahaḥ | nāsti mohasamaṃ jālaṃ, nāsti tṛṣṭā samā nadī || (Uv- B 382.7–8).
630 n’ atthi rāgasamo aggi, n’ atthi dosasamo gaho | n’ atthi mohasamaṃ jālaṃ, n’ atthi taṇhāsamā nadī ||
631 火莫熱於婬 捷莫疾於怒 網莫密於癡 愛流駛乎河 (T210, 4.568c12–14).
632 In the Tibetan version the sixteenth and seventeenth sets are taken as one set, named “two dwellings” because there is a dwelling (gnas) in each set.
633 These three versions are basically the same.
634 Prapañca is a noun from the prefix pra (before, forward, in front) + √pañc (to spread out, make clear), which literally means “expansion, manifestation, amplification, prolixity, diffuseness,” etc. In this case, it is more about something expanded in the mind, like conceptual proliferation. The meaning of this term in the Śarīrārthagāthā will be explained in § 4.2.3.2.
 
Tathāgatas are without conceptual proliferation.


(3) Tibetan version:
/ mkha'635 la ni gnas med ltar / / phyi rol pa la dge sbyong med /
/ byis pa rnams ni spros la dga'/ / de bzhin gshegs rnams spros mi mda' / (Peking 5536.293b1–2; Derge 4035.252a2)
Just as there is no dwelling in the sky,
so there is no Buddhist ascetic among those outside [the Buddhist path]; Fools delight in futile discoursing (spros pa),
while Tathāgatas have no futile discoursing.


2. Comparison of the verse with other texts
The corresponding Udānavarga verse is Uv-S 29.38.636 Parallel verses may be found at: Uv-B 29.38637 and Dhp 254.638 Only one word is different in Dhp, which mentions “ordinary people” (pajā) instead of “fools” (bālā) in other texts. The two parallels T210639 (Fǎjù jīng 法句經) and T213640 (Fǎjíyàosòng jīng 法集要頌經) have some differences in content. These two parallels repeat the verse, and become eight pādas of two verses. The last two pādas may be translated “sentient beings like evil, but the Buddha is pure” and “the world is impermanent, and the Buddha is not-self and not- object.”











635 Peking: nmkha'.
636 ākāśe vai padaṃ nāsti, śramaṇo nāsti bāhyakaḥ | prapañcābhiratā bālā, niṣprapañcās tathāgatāḥ || (Uv-S 53.15–17).
637 ākāśe tu padaṃ nāsti, śramaṇo nāsti bāhyakaḥ | prapañcābhiratā bālā, niṣprapañcās tathāgatāḥ || (Uv-B 382.9–10).
638 ākāse ca padaṃ n’ atthi, samaṇo n’ atthi bāhiro | papañcābhiratā pajā, nippapañcā tathāgatā ||
639 虛空無轍迹 沙門無外意 眾人盡樂惡 唯佛淨無穢 虛空無轍迹 沙門無外意 世間皆無常 佛無我所有
(T210, 4.568c14–17).
640虛空無轍迹 沙門無外意 眾人盡樂惡 唯佛淨無穢 虛空無轍迹 沙門無外意 世間皆無常 佛無我所有
(T213, 4.793c11–14).
 
4.1.17. The seventeenth set: two dwellings (gnas gnyis);641 4 pādas / 1 verse

1. Three versions with their translations642
(1) Chinese version:
住戲論皆無 踰墻塹離愛 牟尼遊世間 天人不能識 (T1579, 30.384b16) Neither having a dwelling nor idle discourse at all,
having crossed over walls and moats, [and] discarding craving, the sage travels the realms,
[so that] heavenly beings do not recognize him.


(2) Sanskrit version:
sthitiḥ prapañcāś ca na santi, yasya yaḥ saṃdānaṃ parighaṃ cātivṛttaḥ | taṃ nistṛṣṇaṃ muniṃ carantaṃ, na vijānāti sadevako 'pi lokaḥ || (Enomoto 1989: 32 [YBh 132a1f])
The world, together with the gods, does not recognize that wandering sage without craving,
who has neither dwelling nor conceptual proliferations, who has overcome the fetter [and] obstacle.

(3) Tibetan version:
/ gang la gnas pa dang ni spros med cing /
/ gang zhig 'ching dang 'obs las 'das gyur pa /
/ thub pa sred dang bral ba rgyu bde /
/ lhar bcas 'jig rten dag gis yong643 mi shes / (Peking 5536.294a4–5; Derge 4035.252b3–4)
One has no dwelling or futile discoursing, who has passed beyond the fetter and moat.

641 In the Tibetan version the sixteenth and seventeenth sets are taken as one set, named “two dwellings” because there is a dwelling (gnas) in each set.
642 In the second pāda, one word is different between the three versions. The Chinese and Tibetan versions contain the word “moat” (qiàn 塹, 'obs), while the corresponding word in the Sanskrit version is “obstacle” (parigham).The Sanskrit version and Uv-S are close to the Pāli parallels (see below). The Chinese and Tibetan versions are related to Uv-B.
643 Derge: yongs.
 
The sage who wanders free from craving,
is not comprehended by the worlds and gods.


2. Comparison of the verse with other texts
The corresponding Udānavarga verse is Uv-S 29.51.644 Parallel verses may be found at: Uv-B 29.51;645 Ud 7.7;646 Nett 3.A.647 Only one word need to be discussed—parigha in Uv-S, parikha in Uv-B, and paligha in Ud and Nett. It seems that Uv-S is close to the Pāli parallels while the word “moat” (parikha) in Uv-B is close to the Chinese and Tibetan versions of this verse set in the Śarīrārthagāthā. However, the two following parallels have a different second pāda: T212648 (Chūyào jīng 出曜經); T213649
(Fǎjíyàosòng jīng 法集要頌經). The two texts mention that one plants the seed of
suffering in the body (shēnqiàn 身塹).

4.1.18. The eighteenth set: having cleaning up (bsal gyur); 4 pādas / 1 verse

1. Three versions with their translations650
(1) Chinese version:
若有熏除諸尋思 於內無餘離分別 超過礙著諸色想 四軛蠲除不往生 (T1579, 30.384c16) If one cleans up all discursive thoughts by fumigation,
he has no remainder and is internally freed of distinctions.
He transcends obstructing attachments of the perception of form; with the four yokes eradicated, he is not reborn.

(2) Sanskrit version:
yasya vitarkā vidhūpitā, adhyātmam avikalpitā aśeṣaṃ |

644 sthitiḥ prapañcāś ca na santi, yasya yaḥ saṃdānaṃ parighaṃ cātivṛttaḥ | taṃ nistṛṣṇaṃ muniṃ carantaṃ, na vijānāti sadevako 'pi lokaḥ || Uv-S 53.18–24.
645 yasya ha prapañcitaṃ hi no sat, saṃtānaṃ parikhaṃ ca yo nivṛttaḥ | tṛṣṇāvigataṃ muniṃ carantaṃ, na vijānāti sadevako ’pi lokaḥ || Uv-B 388.1–4.
646 yassa papañcā ṭhiti ca n’ atthi, sandānaṃ palighañ ca vītivatto | na taṃ nittaṇhaṃ muniṃ carantaṃ, n’ āvajānāti sadevako pi loko || Ud 77.17–20.
647 yassa papañcā ṭhiti ca n’atthi, sandānaṃ palighañ ca vītivatto | tan nittaṇhaṃ muniṃ carantaṃ, na vijānāti sadevako pi loko ti || Nett 37.3–6.
648 無垢無有住 身塹種苦子 最勝無有愛 天世人不知 (T212, 4.752b7– 8).
649 無垢無有住 身壍種苦子 最勝無有愛 天世人不知 (T213, 4.794a5–7).
650 These three versions are basically the same.
 
saṃgaṃ so 'tītya rūpasaṃjñāṃ, caturyogāpagato na jātim eti || (Enomoto 1989: 33 [YBh 132b3])
He whose discursive thoughts are destroyed
[and] have been internally freed of indecision without remainder, having overcome attachment which is the perception of form, free from the four attachments, he is not reborn.

(3) Tibetan version:
/ gang gi rtogs651 pa bsal ba dang / / nang gi ma lus rtogs652 med pa /
/ de chags gzugs kyi 'du shes 'das/ / sbyor bzhi bral ba skye mi 'gyur / (Peking 5536.295a2–3; Derge 4035.253a6–7)
One who has cleared away thoughts,
who is free from internal conceptualization without remainder, transcends attachment for perceptions of form,
freed from the four fetters, one will not be [re-]born.


2. Comparison of the verse with other texts
The corresponding Udānavarga verse is Uv-S 29.56.653 A similar parallel verse may be found at Ud 6.7654. The text Uv-B 29.56655 is a bit different in three terms, namely, “turning away” (vinivartitā) in the second pāda, “all perceptions” (sarvasaṃjñāṃ) in the third pāda, and “he who has not crossed attachment” (atīrṇasaṅga) in the last pāda. The parallel Sn 7656 is different in the last two pādas, which state, “the monk leaves this shore and the far shore as a snake leaves its old worn-out skin” (trans. Norman 2001: 2).Tanjo (1967: 468–471) mentioned that T212657 (Chūyào jīng 出曜經) and T213658 (Fǎjíyàosòng

651 Derge: rtog.
652 Derge: rtog.
653 yasya vitarkā vidhūpitā, adhyātmam avikalpitā aśeṣaṃ | saṃgaṃ so 'tītya rūpasaṃjñāṃ, caturyogāpagato na jātim eti || (Uv-S 53.25–32).
654 yassa vitakkā vidhūpitā, ajjhattaṃ suvikappitā asesā | taṃ saṅgam aticca arūpasaññī, catuyogātigato na jātim etī 'ti || (Ud 71.10–12).
655 yasya vitarkā vidhūpitās, tv ādhyātmaṃ vinivartitā hy aśeṣam | sa hi saṅgam atītya sarvasaṃjñāṃ, yogāpetam atīrṇasaṅgam eti || (Uv-B 390.5–8).
656 yassa vitakkā vidhūpitā, ajjhattaṃ suvikappitā asesā | so bhikkhu jahāti orapāraṃ, urago jiṇṇamivattacaṃ purāṇaṃ || (Sn 2.1–3).
657 若欲滅其想 內外無諸因 亦無過色想 四應不受生 (T212, 4.752c9– 10).
658 若欲滅其想 內外無諸因 亦無過色想 四應不受生 (T213, 4.794a11–13).
 
jīng 法集要頌經) are also parallels to Uv-B. However, from the ambiguous translation,659 it is difficult to determine whether these two parallels are directly related to the eighteenth set.

4.1.19. The nineteenth set: giving (byin pa); 4 pādas / 1 verse

1. Three versions with their translations660
(1) Chinese version:
惠施令福增 防非滅怨害 修善捨諸惡 惑盡得涅槃 (T1579, 30.385a5) By giving kindly, one causes fortune to increase;
by avoiding wrongdoing, one eliminates animosity; by cultivating goodness, one discards all evil;
when all afflictions are exhausted, one achieves nirvāṇa.


(2) Sanskrit version:
dadataḥ puṇyaṃ pravardhate, vairaṃ saṃyamato na cīyate | kuśalī prajahāti pāpakaṃ, kleśānāṃ kṣayatas tu nirvṛtaḥ || (Enomoto 1989: 33 [YBh 133a2])
Merit increases for one giving;
hatred does not accumulate for one who is restrained; a good person abandons what is bad;
and from the destruction of the defilements one attains nirvāṇa.


(3) Tibetan version:
/ byin pa'i bsod nams rab tu 'phel / /yang dag sdom la khon mi skye /
/ dge ldan sdig pa rab tu spang / / nyon mongs rnams zad mya ngan 'das / (Peking 5536.295b3–4; Derge 4035.253b7)
The merit of giving fully flourishes;

659 These two texts share the same content. The English translation of the content is as follows: “if [one] wants to extinguish his thoughts of internal and external causes, which is also without the form-crossed thoughts, he is not [re]born with four concomitants”.
660 In the Chinese version, the translator(s) seems to use a different way to express the meaning, especially the second and third pādas. Overall, these three versions are basically the same.
 
no hatred grows for one genuinely restrained; evil deeds are relinquished by one with virtue; with defilements exhausted, one attains nirvāṇa.

2. Comparison of the verse with other texts
The corresponding Udānavarga verse is Uv-S 28.2.661 Parallel verses may be found at: Uv-B 28.2;662 Ud 8.5;663 DN 16.4.43;664 T212665 (Chūyào jīng 出曜經); T213666
(Fǎjíyàosòng jīng 法集要頌經). However, one pāda is obviously different. In Uv-S, the
last pāda mentions the destruction of the defilements, but the other parallels mention the destruction of desire, hatred and ignorance. These parallels specify the types of defilements. The description of defilements in Uv-S is closer to the texts of the Mūlasarvāstivādins, like T1455 and T1458.667

4.1.20. The twentieth set: all evils (sdig pa kun); 4 pādas / 1 verse

1. Three versions with their translations668
(1) Chinese version:
諸惡者莫作 諸善者奉行 自調伏其心 是諸佛聖教 (T1579, 30.385a28) Do not commit any evil;
cultivate all goodness; personally tame your mind—
such are the holy teachings of the Buddha.



661 dadataḥ puṇyaṃ pravardhate, vairaṃ saṃyamato na cīyate | kuśalī prajahāti pāpakaṃ, kleśānāṃ kṣayatas tu nirvṛtaḥ || (Uv-S 53.33–54.2).
662 dadataḥ puṇyaṃ pravardhate, vairaṃ na kriyate ca samyamāt | kuśalī prajahāti pāpakaṃ, rāgadoṣamohakṣayāt tu nirvṛtiḥ || (Uv-B 354.1–4).
663 dadato puññaṃ pavaḍḍhati, saṃyamato veraṃ na cīyati | kusalo ca jahāti pāpakaṃ, rāgadosamohakkhayā parinibbuto ti || (Ud 85.21–22).
664 dadato puññaṃ pavaḍḍhati, saṃyamato veraṃ na cīyati | kusalo ca jahāti pāpakaṃ, rāgadosamohakkhayā parinibbuto ti || (DN II 136.22–24).
665 惠施獲福報 不藏恚怒懷 以善滅其惡 欲怒癡無餘 (T212, 4.741c13–14).
666 惠施獲福報 不藏恚怒懷 以善滅其惡 欲怒癡無餘 (T213, 4.792a19–21).
667 T1455 (Gēnběnshuōyīqièyǒubù bìzōuníjiè jīng 根本說一切有部苾芻尼戒經): 若人能惠施 福增怨自息修善除眾惡 惑盡至涅槃 (T1455, 24.517a24–25); T1458 (Gēnběnsàpóduōbù lǜshè 根本薩婆多部律攝): 若人能惠施 福增怨自息 修善除眾惡 惑盡至涅槃 (T1458, 24.609b26–27).
668 These three versions are fundamentally the same.
 
(2) Sanskrit version:
sarvapāpasyākaraṇaṃ, kuśalasyopasaṃpadā | svacittaparyavadamanam, etaṃ buddhānuśāsanaṃ || (Enomoto 1989: 33 [YBh 133b2])
Not doing all that is bad, the undertaking of the good,
controlling one’s own mind—
this is the teaching of the Buddha(s).


(3) Tibetan version:
/ sdig pa thams cad mi bya ste / / dge ba phun sum tshogs par bya /
/ rang gi sems ni yongs su gdul / / 'di ni sangs rgyas bstan pa yin / (Peking 5536.296a7; Derge 4035.254b1)
Do not commit any evil; Accomplish virtue;
Thoroughly tame your own mind— This is the teaching of the Buddha.

2. Comparison of the verse with other texts
The corresponding Udānavarga verse is Uv-S 28.1.669 Parallel verses may be found at: Uv-B 28.1;670 Dhp 183;671 PDhp 357;672 DN II 14.3.28673; Nett III.A.7, III.A.15,
III.D;674 EĀ 1.1,675 48.2;676 T210677 (Fǎjù jīng 法句經); T212678 (Chūyào jīng 出曜經);



669 sarvapāpasyākaraṇaṃ, kuśalasyopasaṃpadā | svacittaparyavadamanam, etaṃ buddhānuśāsanaṃ || (Uv-S 54.3–6).
670 sarvapāpasyākaraṇaṃ, kuśalasyopasampadaḥ | svacittaparyavadanam, etad buddhasya śāsanam || (Uv-B 353.1–2).
671 sabbapāpassa akaraṇaṃ, kusalassa upasampadā | sacittapariyodapanaṃ, etaṃ buddhāna sāsanaṃ ||
672 sabbapāpassa akaraõaṃ, kuśalassa apasaüpadā | sacittapariyodamanaṃ, etaṃ buddhāna śāsanaṃ ||
673 sabbapāpassa akaraṇaṃ, kusalassa upasampadā | sacittapariyodapanaṃ, etaṃ buddhāna sāsanaṃ || (DN II 49.25–26).
674 sabbapāpassa akaraṇaṃ, kusalass' upasampadā | sacittapariyodapanaṃ, etaṃ buddhāna sāsanan ti ||
(Nett 43.12–13; 81.9–10; 171.22–23; 186.23–24).
675 諸惡莫作 諸善奉行 自淨其意 是諸佛教 (T125, 2.551a13–14).
676 一切惡莫作 當奉行其善 自淨其志意 是則諸佛教 (T125, 2.787b1–2).
677 諸惡莫作 諸善奉行 自淨其意 是諸佛教 (T210, 4.567.b1–2).
678 諸惡莫作 諸善奉行 自淨其意 是諸佛教 (T212, 4.741b24– 25).
 
T213679 (Fǎjíyàosòng jīng 法集要頌經); T374, T375680 (Dàbōnipán jīng 大般涅槃經). They all share the same content.

4.1.21. The twenty first set: falling down (ltung ba); 4 pādas / 1 verse

1. Three versions with their translations681
(1) Chinese version:
難調伏輕躁 淪墜於諸欲 善調伏其心 心調引安樂 (T1579, 30.385b14) Hardly taming fickleness [of mind],
falling down in desires.
[one] excellently tames the mind. A tamed mind elicits happiness.

(2) Sanskrit version:
durnigrahasya laghuno, yatrakāmanipātinaḥ |
cittasya dāmanaṃ sādhu, cittaṃ dāntaṃ sukhāvaham || (Enomoto 1989: 33 [YBh 133b5])
Good is the taming of the mind, which is hard to restrain, swift, falling on where it desires.
The tamed mind brings happiness.


(3) Tibetan version:
/ tshar gcad dka' zhing myur ba dang / / gang du 'dod par ltung ba yi682 /
/ sems dul ba ni legs pa ste / / sems dul bde ba bsgrub pa yin / (Peking 5536.296b6–7; Derge 4035.254b6–7)
Hard to restrain and swift, falling on wherever it desires,
disciplining such a mind is excellent.

679 諸惡業莫作 諸善業奉行 自淨其意行 是名諸佛教 (T213, 4.792a17–19).
680 諸惡莫作 諸善奉行 自淨其意 是諸佛教 (T374, 12.451c11–12; 12.693c12–13).
681 These three versions are basically the same.
682 Derge: yin.
 
The tamed mind leads to happiness.


2. Comparison of the verse with other texts
The corresponding Udānavarga verse is Uv-S 31.1.683 Parallel verses may be found at: Uv-B 31.1;684 Dhp 35;685 PDhp 345;686 Jā 1.7.10, 1.10.6;687 T210688 (Fǎjù jīng 法句經);
T212689 (Chūyào jīng 出曜經); T213690 (Fǎjíyàosòng jīng 法集要頌經); T721691
(Zhèngfǎniànchù jīng 正法念處經).

4.1.22. The twenty second set: skilled (mkhas); 4 pādas / 1 verse

1. Three versions with their translations692
(1) Chinese version:
於心相善知 能餐遠離味 靜慮常委念 受無染喜樂 (T1579, 30.385c4) [One who] well understands manifestations of the mind
can taste the flavour of seclusion.
One who meditates, is constantly comprehensive, and mindful receives undefiled joy and happiness.

(2) Sanskrit version:
cittanimittasya kovidaḥ, pravivekasya ca vindate rasaṃ |
dhyāyī nipakaḥ pratismṛto, bhuṃkte prītisukhaṃ nirāmiṣaṃ ||


683 durnigrahasya laghuno, yatrakāmanipātinaḥ | cittasya dāmanaṃ sādhu, cittaṃ dāntaṃ sukhāvaham || (Uv-S 54.8–9).
684 durnigrahasya laghuno, yatrakāmanipātinaḥ | cittasya damanaṃ sādhu, cittaṃ dāntaṃ sukhāvaham || (Uv-B 408.1–2).
685 dunniggahassa lahuno, yatthakāmanipātino | cittassa damatho sādhu, cittaṃ dantaṃ sukhāvahaṃ || 686 dunniggrahassa laghuno, yatthakāmanipātino | cittassa damatho sādhu, cittaṃ dāntaṃ sukhāvahaṃ || 687 dunniggahassa lahuno, yatthakāmanipātino | cittassa damatho sādhu, cittaṃ dantaṃ sukhāvahaṃ || Jā 312.15–16; 400.12–13.
688 輕躁難持 唯欲是從 制意為善 自調則寧 (T210, 4.563.a5–6).
689 輕難護持 為欲所居 降心為善 以降便安 (T212, 4.758c12– 13).
690 心輕難調伏 為欲所居懷 降心則為善 以降便輕安 (T213, 4.795b6–8).
691 輕轉難調伏 處處妄攀緣 若善調伏心 調伏則安樂 (T721, 17.156b14–16).
692 In the Chinese version, the term “undefiled” (wúrǎn 無染) is very likely to correspond to nirāmiṣaṃ in
Sanskrit. Overall, these three versions are basically the same in content. However, the term “receives/experiences” (shòu 受) in the last pāda of the Chinese version is closer to vetti than bhuṃkte. Therefore, the Chinese version seems close to the version of Uv-B.
 
(Enomoto 1989: 33 [YBh 134a4])
One [who] is skilled in the signs of the mind and knows the flavour of seclusion, meditating, wise [and] mindful,
enjoys spiritual joy and happiness.


(3) Tibetan version:
/ sems kyi mtshan ma mkhas pa dang / / rab tu dben pa'i nyams thob dang /
/ rtag 'grus dran ldan bsam gtan pa / / dga' bde zang zing med pa spyod / (Peking 5536.297a8-b1; Derge 4035.255a7)
One who is skilled in the signs of the mind and who acquires the taste of total seclusion, who meditates, is ever-diligent and mindful, enjoys spiritual joy and happiness.

2. Comparison of the verse with other texts
The corresponding Udānavarga verse is Uv-S 31.51.693 Parallel verses may be found at: Uv-B 31.51;694 Th 85;695 T212696 (Chūyào jīng 出曜經); T213697 (Fǎjíyàosòng jīng法集要 頌 經 ).698 But the term “enjoy” (bhuṃkte) in the last pāda of Uv-S is different from “obtain” (vetti, adhigaccheyya, huò 獲 ) in other texts.











693 cittanimittasya kovidaḥ, pravivekasya ca vindate rasaṃ | dhyāyī nipakaḥ pratismṛto, bhuṃkte prītisukhaṃ nirāmiṣaṃ || (Uv-S 54.10–14).
694 cittanimittasya kovidaḥ, pravivekasya rasaṃ prajānakaḥ | dhyāyī nipakaḥ pratismṛto, vetti prītisukhaṃ nirāmiṣam || (Uv-B 426.4–7).
695 cittanimittassa kovido, pavivekarasaṃ vijāniya | jhāyaṃ nipako patissato, adhigaccheyya sukhaṃ nirāmisan ti ||
696 行人觀心相 分別念待意 以得入禪定 便獲喜安樂 (T212, 4.763c22–23).
697 行人觀心相 分別念待意 以得入禪定 便獲喜安樂 (T213, 4.796b1–3).
698 In T212 and T213, the terms fènbié 分別, niàn 念 and dàiyì 待意 seem to be the counterparts of “wise/comprehensive”, “mindful”, and “the taste of seclusion”.
 
4.1.23. The twenty third set: craft (bzo); 4 pādas / 1 verse

1. Three versions with their translations699
(1) Chinese version:
無工巧活輕自己 樂勝諸根盡解脫 無家無所無希望 斷欲獨行真苾芻 (T1579, 30.385c21) Not living by the exercise of crafts, and taking himself lightly,
happily conquering all faculties of senses, completely liberated, without a home, without a place, without expectations, terminating desires and wandering alone; such is the true monk.

(2) Sanskrit version:
aśilpajīvī laghur ātmakāmo, jitendriyaḥ sarvato vipramuktaḥ | anokasārī hy amamo nirāśaḥ, kāmān prahāyaikacaro yas sa bhikṣuḥ || (Enomoto 1989: 34 [YBh 134b3f])
He who lives not by the exercise of crafts, light, wishes himself well, whose senses have been conquered, is completely liberated,
wanders without a house, is without selfishness, without expectation, having abandoned desires, wandering alone, he is a monk.

(3) Tibetan version:
/ bzo700 mi 'tsho dang yang dang bdag la legs 'dod dang /
/ dbang po thul dang kun las rnam par grol ba dang /
/ khyim na mi gnas bdag gir mi 'dzin re ba med /
/ 'dod spangs gcig pu rgyu ba de ni dge slong yin / (Peking 5536.298a1–2; Derge 4035.255b6–7)
One who does not practice a craft to live, and wishes goodness for himself, who has subdued the senses and is completely liberated from all,
who does not dwell in a household, has no sense of “mine” or expectation, abandons desire and roams alone, that is a monk.


699 The Tibetan version lacks the term “light” in the first pāda. Overall, these three versions are basically the same.
700 Derge: bzos.
 
2. Comparison of the verse with other texts
The corresponding Udānavarga verse is Uv-S 32.5.701 Parallel verses may be found at: Uv-B 32.5;702 Ud 3.9.703

4.1.24. The twenty fourth set: travelling far away (ring du 'gro); 4 pādas / 1 verse

1. Three versions with their translations704
(1) Chinese version:
心遠行獨行 無身寐於窟 能調伏難伏 我說婆羅門 (T1579, 30.386a20)
The mind which has traveled far and alone without a body lives in the cave.
[One] can tame that [[[mind]]] which is hard to tame— I call him a Brahman.

(2) Sanskrit version:
dūraṃgamam ekacaram, aśarīraṃ guhāśayaṃ |
damayati durdamaṃ cittaṃ, brāhmaṇaṃ taṃ bravīmy ahaṃ || (Enomoto 1989: 34 [YBh 135a5f])
One who goes far, wanders alone, without a body lies in the cave,
restrains his mind which is hard to restrain— I declare him to be a Brahman.

(3) Tibetan version:
/ ring du 'gro zhing gcig pu rgyu / / lus yod705 ma yin phug gnas pa /
/ gdul dka'i sems ni 'dul ba de / / bram ze yin zhes nga smra'o / (Peking 5536.298b8; Derge 4035.256b3–4)
701 aśilpajīvī laghur ātmakāmo, jitendriyaḥ sarvato vipramuktaḥ | anokasārī hy amamo nirāśaḥ, kāmān prahāyaikacaro yas sa bhikṣuḥ || (Uv-S 54.15–22).
702 yas tv alpajīvī laghur ātmakāmo, yatendriyaḥ sarvagatiḥ pramuktaḥ | anokasārī hy amamo nirāśaḥ, kāmaṃ jahaś caikacaraḥ sa bhikṣuḥ || (Uv-B 432.9–12).
703 asippajīvī lahu atthakāmo, yatindriyo sabbadhivippamutto | anokasārī amamo nirāso, hatvā māraṃ ekacaro sa bhikkhū ti || (Ud 32.15-17).
704 These three versions are fundamentally the same.
705 Derge: yongs.
 
One who goes far and travels alone, abides in a cave without a body, tames the mind that is hard to tame— I declare to be a Brahman.

2. Comparison of the verse with other texts
The corresponding Udānavarga verse is Uv-S 33.55.706 Parallel verses may be found at: Uv-B 33.55;707 T212708 (Chūyào jīng 出曜經); T213709 (Fǎjíyàosòng jīng 法集要頌經). Parallels may also be found at Dhp 37,710 GDhp 344,711 and T210712 (Fǎjù jīng 法句經); however, their fourth pāda instead states “will be freed from Māra's fetter.”

4.1.25. The twenty fifth set: obscuring (bsgribs); 40 pādas / 10 verses

1. Three versions with their translations713
(1) Chinese version:
誰能覆世間 誰能令不顯 誰復能塗染 誰為大怖畏無明覆世間 放逸令不顯 戲論能塗染 苦為大怖畏諸流處處漏 是漏誰能止 當說誰防護 眾流誰所偃世間諸流漏 是漏念能止 我說能防護 由慧故能偃念慧與名色 今問是一切 何當永滅盡 唯願為我說念慧與名色 我說是一切 若諸識永滅 於斯永滅盡云何念所行 諸識當永滅 今請垂方便 為釋令無疑
706 dūraṃgamam ekacaram, aśarīraṃ guhāśayaṃ | damayati durdamaṃ cittaṃ, brāhmaṇaṃ taṃ bravīmy ahaṃ || Uv-S 54.23–32.
707 There are two versions; the more similar one is the second.
1. yasyālayo nāsti sadā, yo jñātā niṣkathaṃkathaḥ | dūraṃgamaś caikacaro, bravīmi brāhmaṇaṃ hi tam || Uv-B 489.1–4.
2. dūraṃgamam ekacaram, aśarīraṃ guhāśayam | durdamaṃ ye damiṣyanti, tenai[?]kasya brāhmaṇam (v.l. brāhmaṇam taṃ bravīmy ahaṃ) || Uv-B 489.1–4.
708 遠逝獨遊 隱藏無形 難降能降 是謂梵志 (T212, 4.774a24– 25).
709 遠逝獨遊行 隱藏無形影 難降能自調 是名為梵志 (T213, 4.779a26–28).
710 dūraṅgamaṃ ekacaraṃ asarīraṃ guhāsayaṃ | ye cittaṃ saññamessanti mokkhanti mārabandhanā ||
711 dūraṃgamaṃ ekacaraṃ, aśarīraṃ guhāśayaṃ | ye cittaṃ saṃyyamehinti, mokkhaṃte mārabaṃdhanā ||
712 獨行遠逝 覆藏無形 損意近道 魔繫乃解 (T210, 4.563a8–9.)
713 These three versions are basically the same. Only one word is somewhat divergent. The Sanskrit term jalpa seems to be interpreted as “idle discourse” (xì lùn 戲 論 ) in Chinese and “thought” (rtog pa) in Tibetan.
 
於內外諸受 都不生欣樂 如是念所行 諸識當永滅若諸善說法 及有學異類 彼常委能趣 請大仙為說
不耽著諸欲 其心無濁染 於諸法巧念 是苾芻能趣 (T1579, 30.386b15–c5) [[[Ajita]] asked:]
“What could cover the world? What could make it un-manifest? What could also defile it?
What is the great fear?” [The Buddha replied:] “Ignorance covers the world.
Carelessness makes it un-manifest. Idle discourse could defile it.
Suffering is the great fear.” [[[Ajita]] asked:]
“The streams have outlets everywhere. Who could stop these outlets?
What should be said to prevent them? Who could make the streams be stopped?” [The Buddha replied:]
“The streams of the world have outlets;
these outlets could be stopped by mindfulness. I say it could be prevented,
that they could be stopped by wisdom.” [[[Ajita]] asked:]
“Mindfulness, wisdom and name-and-form, they are all I inquire about now.
How should they be annihilated permanently? I sincerely wish that you preach this for me.” [The Buddha replied:]
“Mindfulness, wisdom and name-and-form, I say, are everything.
If consciousness ceases permanently,
 
thus they are annihilated permanently.” [Ajita714 asked:]
“What is the operation of mindfulness,
so that consciousness is permanently ceased?
I now request that you would explain to me with any means and cause me to be without doubt.”
[The Buddha replied:]
“In internal and external sensations, one does not give rise to delight.
This is the operation of mindfulness,
[so that] consciousness would cease permanently.” [[[Ajita]] asked:]
“As those who have well expounded the dharma
and the learners of different types
constantly comprehend and proceed toward liberation,
I request the Buddha to preach to me [how they did that].” [The Buddha replied:]
“They do not indulge in sensual pleasures; their minds are without defilements;
and they are skillfully mindful of all dharmas.
Therefore, such monks could proceed toward [[[liberation]]].”


(2) Sanskrit version:
kenāyaṃ nivṛto lokaḥ, kenāyaṃ na prakāśate |
kiṃ cābhilepanaṃ brūṣe, kiṃ ca tasya mahad bhayam || avidyānivṛto lokaḥ, pramādān na prakāśate | jalpābhilepanaṃ brūmi, duḥkhaṃ tasya mahad bhayaṃ || sravanti sarvataḥ srotāḥ, srotasāṃ kiṃ nivāraṇaṃ | srotasāṃ saṃvaraṃ brūhi, kena srotaḥ pidhīyate ||
yāni srotāṃsi lokasya, smṛtiḥ teṣāṃ nivāraṇaṃ |


714 Sn 1110 states that this verse was asked by Udaya, but the commentary of this verse set in the
Śarīrārthagāthā states it was asked by Ajita.
 
srotasāṃ saṃvaraṃ brūmi, prajñayā hi pidhīyate || prajñāyāś ca smṛteś caiva, nāmarūpasya sarvaśaḥ | ācakṣva pṛṣṭa etan me, kutraitad uparudhyate || prajñā caiva smṛtiś caiva, nāmarūpaṃ ca sarvaśaḥ | vijñānasya nirodhād dhi, atraitad uparudhyate || kathaṃ smṛtasya carato, vijñānam uparudhyate | ācakṣva pṛṣṭa etan me, yathātatham asaṃśayaḥ || adhyātaṃ ca bahirdhā, ca vedanāṃ nābhinandataḥ | evaṃ smṛtasya carato, vijñānam uparudhyate ||
ye ca saṃkhyātadharmāṇo, ye ca śaikṣāḥ pṛthagvidhāḥ | teṣāṃ me nipakasyeryāṃ, pṛṣṭaḥ prabrūhi mārṣa || kāmeṣu nābhigṛdhyeta, manasānāvilo bhavet |
kuśalaḥ sarvadharmeṣu, smṛto bhikṣuḥ parivrajet || (Enomoto 1989: 34 [YBh 136a1–4])
[[[Ajita]] asked:]
“By what is this world enclosed? Why does it not shine?
And what do you say is its plaster? And what is its great fear?”
[The Buddha replied:]
“The world is enclosed by ignorance.
On account of negligence it does not shine. I say its plaster is desire.715
Suffering is its great fear.” [[[Ajita]] asked:]
“Streams flow everywhere. What hinders streams?
Tell [me] about the restraint of streams. By what is a stream dammed?”
[The Buddha replied:]


715 The term jalpa can mean “desire” or “discourse”/“disputation”. The latter meaning is used in the Chinese version.
 
“Mindfulness hinders the world’s streams.
I will tell [you] about the restrain of streams. By wisdom is [a stream] dammed.”
[[[Ajita]] asked:]
“Of wisdom and mindfulness and name-and-form altogether, tell me about this when asked. Where is this stopped?”
[The Buddha replied:] “Wisdom and mindfulness
and name-and-form altogether. From the cessation of consciousness, this is stopped here.”
[[[Ajita]] asked:]
“How is consciousness stopped for one who wanders mindful? Tell me about this when asked, precisely and without doubt.” [The Buddha replied:] “Internally and externally,
from not finding pleasure in sensation, consciousness is thus stopped
for one who wanders mindful.” [[[Ajita]] asked:]
“Honourable one, when asked,
tell me about the behaviour of the wise,
of those who have directly experienced the dharma, and who are those learners of different types.”
[The Buddha replied:]
“He would not desire sensual pleasures. He would be pure with regard to his mind. Skilled in all mental states,
 
mindful, a monk would wander”


(3) Tibetan version:
/ 'jig rten 'di ni gang gis bsgribs/ /'di ni gang gis mi gsal bgyis /
/ bsgos zhes bgyis ba ci716 la bgyi / / de ’i717 'jigs chen gang zhig lags /
/ ma rig pas ni 'jig rten bsgribs718 / / bag med phyir ni gsal ma yin /
/ rtog pas bsgos zhes nga smra ste / / de yi 'jigs chen719 sdug bsngal yin /
/ rgyun rnams kun nas 'dzags720 pa yi / / rgyun rnams bzlog pa gang zhig lags /
/ rgyun rnams sdoms721 pa bka' stsol cig / / rgyun ni gang gis dgag par bgyi /
/ 'jig rten rgyun ni gang yin pa / / de dag bzlog par dran pa ste /
/ rgyun rnams sdom pa722 bshad bya na / / shes rab kyis ni dgag par bya /
/ shes rab dang ni dran pa dang / / ming dang gzugs ni rnams723 kun du /
/ de ni gang du 'gag 'gyur ba / / zhu na de ni bdag la gsungs /
/ shes rab dang ni dran pa dang / / ming dang gzugs ni rnam kun du /
/ rnam par shes pa 'gag724 gyur na / / der ni de yang 'gag par 'gyur /
/ dran pa spyod pa'i rnam par shes / / ji ltar 'jug 'gyur zhu lags na /
/ ji bzhin the tshom ma mchis par / / de ni bdag la bshad du gsol /
/ nang dang phyi yi tshor ba la / / mngon par dga' bar mi byed na /
/ de ltar dran pa spyod pa yi / / rnam par shes pa 'jug par 'gyur /
/ gang dag legs gsungs chos can dang / / gang dag slob pa tha dad pa /
/ de dag nang nas 'grus pa'i spyod / / zhu na bzod ldan bdag la gsungs /
/ 'dod pa rnams la chags med cing / / yid kyis rnyogs725 pa med par byed /
/ chos rnams kun la mkhas pa dang / / dran pa726 dge727 slong kun du rgyu / (Peking 5536.299b4–300a3; Derge 4035.257a6–b4)

716 Derge: cis.
717 Derge: yi.
718 Peking: bsgrubs. 719 Derge: 'jig rten. 720 Derge: 'dzag.
721 Derge: sdom.
722 Derge: par.
723 Derge: rnam.
724 Derge: gags.
725 Derge: rnyog.
726 Derge: pas.
727 Peking: bdge.
 
[[[Ajita]] asked:]
“What is it that obscures this world? What is it that makes it unclear?
By what is it pervaded?
What indeed is its great fear?” [The Buddha replied:]
“By ignorance the world is obscured. Due to carelessness is it not clear.
By thought it is pervaded, I say; and suffering is its great fear. [[[Ajita]] asked:]
“Streams flow everywhere;
What is it that can reverse the streams? Please explain the damming of streams. What is it that stops a stream?”
[The Buddha replied:]
“Those which are the streams of the world, are reversed by mindfulness.
As to explaining the damming of these streams, they are stopped by wisdom.”
[[[Ajita]] asked:]
“Wisdom and mindfulness, name-and-form in all aspects, from where do they cease,
tell me about this when asked.” [The Buddha replied:] “Wisdom and mindfulness, name-and-form in all aspects, when consciousness ceases,
then they also come to cease therein.” [[[Ajita]] asked:]
“How is consciousness stopped
 
for one who proceeds endowed with mindfulness?
Please explain that to me exactly as it is, without doubt.” [The Buddha replied:]
“In internal and external sensations, if not finding pleasure,
in that way consciousness is stopped
for one who proceeds endowed with mindfulness.” [[[Ajita]] asked:]
Those who have experienced the dharma, and the various learners
among them, [what is] the practice of wisdom, tell me when asked, honourable one.
[The Buddha replied:]
Not desiring sensual pleasures, [whose] mind is undefiled,
is skilled in all dharmas
with mindfulness, the monk roams everywhere.


2. Comparison of the verse with other texts
This verse set includes ten verses. Corresponding verses have not yet been found. While there are a few parallel verses, the content of these is not precisely the same.728 Only the Suttanipāta contains parallels to all ten verses; however, their sequence differs.729 In the second pāda of the second verse, the Suttanipāta parallel refers to

728 The examples: SN 12.31.1: ye ca snṅkhātadhammāse, ye ca sekhā puthu idha | tesam me nipako iriyaṃ, puṭṭho me brūhi mārisāti || (SN II 47.12–13; 47.22–23); SĀ-G 1011: 誰掩於世間 誰遮絡世間 誰結縛眾生何處建立世 (T99, 2.264b23–24); 衰老掩世間 死遮絡世間 愛繫縛眾生 法建立世間 (T99, 2.264b26–27). The details of others can be seen in Enomoto 1989: 34.
729 The counterparts of the verses are Sn 1032–1037 (six verses), Sn 1110–1111 (two verses) and Sn 1038– 1039 (two verses). However, the asker in Sn 1110 is Udaya rather than Ajita.
Sn 1032–1037:
kenassu nivuto loko, (iccāyasmā ajito) kenassu nappakāsati | ki 'ssābhilepanaṃ brūsi, kiṃ su tassa mahabbhayaṃ ||
avijjāya nivuto loko, (ajitāti bhagavā) vevicchā pamādā nappakāsati | jappābhilepanaṃ brūmi, dukkham assa mahabbhayaṃ ||
savanti sabbadhi sotā, (iccāyasmā ajito) sotānaṃ kiṃ nivāraṇaṃ | sotānaṃ saṃvaraṃ brūhi, kena sotā
 
“avarice and negligence,” but in the twenty fifth verse set, the corresponding text only refers to “negligence.” Otherwise, they are mostly the same.

4.1.26. The twenty sixth set: sensual pleasures (’dod); 24 pādas / 6 verses

1. Three versions with their translations730
(1) Chinese version:
於諸欲希求 或所期果遂 得已心定喜 至死而保愛諸樂欲眾生 若退失諸欲 其色便變壞 如毒箭所中若遠離諸欲 猶如毒蛇首 彼於愛世間 正念能超度田事與金銀 牛馬珠環釧 女僕增諸欲 是人所耽樂攀緣沈下劣 變壞生諸漏 從此集眾苦 如船破水溢
若永絕諸欲 如斷多羅頂 棄捨諸愁憂 猶蓮華水滴 (T1579, 30.387b4–15) In pursuing sensual pleasures,
one fulfils them as expected.
Having done so, the mind is stable and joyful. One keeps one’s attachments until death.
If sentient beings who delight in desires lose their sensual pleasures,
their form will deteriorate


pidhiyyare ||
yāni sotāni lokasmiṃ, (ajitāti bhagavā) sati tesaṃ nivāraṇaṃ | sotānaṃ saṃvaraṃ brūmi, paññāy ete pithiyyare ||
paññā c' eva sati ca, (iccāyasmā ajito) nāmarūpañ ca mārisa | etaṃ me puṭṭho pabrūhi, katth' etaṃ uparujjhati ||
Sn 1110–1111:
kathaṃ satassa carato, viññāṇaṃ uparujjhati | bhagavantam puṭṭhuṃ āgamma, taṃ suṇoma vaco tava ||
ajjhattañ ca bahiddhā ca, vedanaṃ nābhinandato | evaṃ satassa carato, viññāṇaṃ uparujjhatī ti || Sn 1038–1039:
yam etaṃ pañhaṃ apucchi, ajita taṃ vadāmi te | yattha nāmañ ca rūpañ ca, asesaṃ uparujjhati |
viññāṇassa nirodhena, etth' etaṃ uparujjhati ||
ye ca saṃkhātadhammāse, ye ca sekhā puthū idha | tesaṃ me nipako iriyaṃ, puṭṭho pabrūhi mārisa ||
kāmesu nābhigijjheyya, manasānāvilo siyā | kusalo sabbadhammānaṃ, sato bhikkhu paribbaje ti ||
730 The first and fifth verses in the Chinese version seem to be different to the other two versions. In the first verse, the obvious difference is the last pāda, which states “one keeps one’s attachments until death.” It appears to represent an interpretation of translator(s) of martyo yad īpsitaṃ in Sanskrit, although it is grammatically different. As for the fifth verse, it is hard to know if the difference is to be attributed to the interpretation of translator(s) or the manuscript exemplar.
 
as if they were shot by a poisonous arrow.
If they detach themselves from desires
as if they were heads of poisonous snakes, attachments to this world
they are able to transcend with mindfulness. Farm estates, bullion,
oxen and horses, jewels and bracelets,
females and servants, and [many other things which] increases desires; such are what people wallow in.
What one clings to makes [him] degraded, becoming deteriorated gives rise to outflows.731 From which sufferings accumulate,
like a broken boat and inundated by water.
If one permanently cuts off all desires like severing the top of the palm tree, one discards all worries
like water drops from a lotus flower.


(2) Sanskrit version:
kāmān kāmāyamānasya, tasya cet tat samṛdhyati | addhā prītamanā bhavati, labdhvā martyo yad īpsitaṃ || tasya cet kāmāyamānasya, chandajātasya jaṃtunaḥ |
te kāmāḥ parihīyaṃte, śalyaviddha iva rūpyate || yaḥ kāmāṃ parivarjayati, sarpasyeva śirāt padaṃ | sa imāṃ viṣaktikāṃ loke, smṛtaḥ samativartate || kṣetravastuhiraṃnyaṃ ca, gavāśvamaṇikuṇḍalaṃ |
striyo dāsān pṛthakkāmān, yo naro hy abhigṛdhyati || abalaṃ vā balīyāṃso, mṛdnaṃty enaṃ parisravāḥ | tata enaṃ duḥkham anveti, bhinnāṃ nāvam ivodakaṃ || yasya tv etat samucchinnaṃ, tālamastakavad dhataṃ | śokās tasya nivartante, udabindur iva puṣkarāt ||

731 “Outflow” (lòu 漏 in Chinese) in this context is close to the meaning of “affliction”.
 
(Enomoto 1989: 35 [YBh 137a5–b1])
If it prospers for one desiring sensual pleasures,
a mortal truly becomes joyful in mind, having obtained his wish.
If those sensual pleasures decrease
for that person desiring [[[sensual pleasures]] and] producing desire,
he was represented as if pierced by an arrow. He who avoids sensual pleasures,
as if [keeping away] one’s foot from a snake’s head,
mindful, escapes this attachment to the world.
For a man who greedily strives after fields, property, gold,
cows, horses, jewels, bracelets,
women, slaves and different sensual pleasures, more powerful difficulties
pound that weak one. Then suffering enters him,
as water does a broken boat.
But [if] this [[[desire]]] of his is uprooted, cut off like the top of a palm tree,
his sorrows roll off
like a drop of water from a lotus [leaf].


(3) Tibetan version:
/ 'dod pa rnams ni 'dod pa dang / / gal te de yi de grub cing /
/ mis732 ni ci dgar thob gyur na / / yid ni mchog tu dga' bar 'gyur /
/ skye bo 'dun733 pa skyes gyur la / / tshor ba de yi 'dod de dag /


732 Derge: ming.
733 Peking: bdun.
 
/ gal te yongs su nyams gyur na / / zug rngu zug bzhin gnod par 'gyur /
/ sbrul mgo las ni rkang pa bzhin / / 'dod rnams spong ba gang yin de /
/ 'jig rten dag na sred 'di las / / dran bzhin du ni 'da' bar 'gyur /
/ zhing dngos dbyig dang ba lang dang / / rta dang nor bu rna cha dang /
/ bud med bran734 gyi735 'dod736 rgyas la / / mi ni mngon par zhen pa gang /
/ stobs chen nyam chung bab pa bzhin / / de ni nyes 'dzag rnams kyis bcom /
/ de phyir chur ni gru zhig ltar / / de ltar sdug bsngal 'byung bar 'gyur /
/ ta la gtan nas gcad pa ltar / / gang gis de ni bcad byas pa /
/ de ni mya ngan ldog 'gyur te737 / / pad+ma la ni chu thigs bzhin / (Peking 5536.302a1–5; Derge 4035.259a5–b1)
All that is desired is sensual pleasures, and if they are fulfilled,
if humans obtain what they wish,
their minds become completely happy. Sentient beings developing desires, hurt as if pierced by an arrow,
if those sensual pleasures [they] perceived, happen to decease.
like feet [keeping away] from the head of a snake, One avoids sensual pleasures.
Thereupon attachments to the worlds,
they are able to transcend with mindfulness. field, property, treasure and cattle,
horses, jewels and earrings,
women, servants and increasing desires for these are the manifest obsession of humans.
Just as the great strength brought down to weakness, the drip of their faults defeated him
as a ship is destroyed by water,


734 Peking: brang.
735 Peking: gi.
736 Peking: 'od.
737 Peking: ba.
 
so does suffering emerge.
Just like the top of a palm tree completely cut off, that [[[desire]]] is cut off by that.
[He] shall thereby cast away sorrow, like a drop of water on a lotus.

2. Comparision of the verse with other texts
This verse set includes six verses. A complete parallel of all six verses is yet to be found; however, many partial parallels exist.738 The following two texts have the highest number of parallel verses and correspond to the first five verses of the twenty sixth set: Sn 766–770739 and T198740 (Fóshuō yìzú jīng 佛說義足經). The content of these two parallels is basically the same.741 The obvious difference is that the fourth verse of these two parallels contains a different list of elements.742

4.1.27. The twenty seventh set: auspicious (bzang po); 6 pādas / 1 verse

1. Three versions with their translations743
(1) Chinese version:
於過去無戀 不希求未來 現在諸法中 處處徧觀察 智者所增長 無奪亦無動 (T1579, 30.387c28–388a1)
Without immersing in the past, not longing for the future,
738 See Enomoto 1989: 35.
739 kāmaṃ kāmayamānassa, tassa ce taṃ samijjhati | addhā pītimano hoti, laddhā macco yad icchati ||
tassa ce kāmayānassa, chandajātassa jantuno | te kāmā parihāyanti, sallaviddho va ruppati ||
yo kāme parivajjeti, sappasseva padā siro | so imaṃ visattikaṃ loke, sato samativattati ||
khettaṃ vatthuṃ hiraññaṃ vā, gavāssaṃ dāsaporisaṃ | thiyo bandhū puthu kāme, yo naro anugijjhati || abalā va naṃ balīyanti, maddante naṃ parissayā | tato naṃ dukkham anveti, nāvaṃ bhinnam ivodakaṃ || 740 增念隨欲 已有復願 日增為喜 從得自在
有貪世欲 坐貪癡人 既亡欲願 毒箭著身
是欲當遠 如附蛇頭 違世所樂 當定行禪田種珍寶 牛馬養者 坐女繫欲 癡行犯身
倒羸為強 坐服甚怨 次冥受痛 船破海中 (T198, 4.175c17–23)
741 However, the first pāda of the fifth verse is grammatically different.
742 The elements are fields, property, gold, cows, horses, servants, men, women, relatives in the Suttanipāta and T198, but in the Śarīrārthagāthā, they are fields, property, gold, cows, horses, jewels, bracelets, women, slaves/servants.
743 The three versions are basically the same.
 
for all present states
one observes here and there.
What is cultivated by the wise person cannot be appropriated or moved.

(2) Sanskrit version:
atītaṃ nānvāgamayen, na pratikāṃkṣed anāgataṃ | pratyutpannāś ca ye dharmās, tatra tatra vipaśyakaḥ | asaṃhāryam asaṃkṣobhyaṃ, tad vidvān anubṛṃhayet || (Enomoto 1989: 35 [YBh 138a4f])
One should not revive the past, should not long for the future. thoroughly observing present states here and there,
a wise man should practice that, Unfailingly, unshakably.

(3) Tibetan version:
/'das la rjes su 'gro mi byed / / ma 'ongs pa la re ba med /
/ da ltar byung chos gang yin pa / / de dang de la rnam lta la /
/ mi 'phrogs kun du mi 'khrugs pa / / de ni mkhas pas 'phel bar byed / (Peking 5536.303b3–4; Derge 4035.260b3–4)
Do not follow after what has passed, without expectations for the future. Whatever dharma arises in the present, one observes each and every one of them. What the wise cultivates is
Unassailable, completely imperturbable.
 
2. Comparision of the verse with other texts
According to Enomoto (1989: 35), the corresponding text is MĀ 165–167.744 However, the number of the verses (three verses containing twelve pādas) in MĀ is more than in the twenty seventh verse set (one verse containing six pādas). Also, only the first two pādas in the first and second verse corresponds to the first four pādas in the twenty seventh verse set. Moreover, the corresponding verses are also found at MN 131– 134.745 The content of the verses (except for the extra two pādas of the first verse) in MN is much closer to the verses of the twenty seventh verse set. Therefore, I assume that a corresponding parallel is yet to be found, although MĀ is viewed as a text of the Sarvāstivādins.746 According to the previous verse sets, the verses of the twenty seventh set might also relate to the version attributed to the Mūlasarvāstivādins (possibly a later version)747 which has not yet been found.


744 慎莫念過去 亦勿願未來 過去事已滅 未來復未至現在所有法 彼亦當為思 念無有堅強 慧者覺如是
若作聖人行 孰知愁於死 我要不會彼 大苦災患終 (T26, 1.697a18–23; a29–b5; b17–22; 698b8–13; c22– 27; 699b1–4; b29–c5; 700a15–20). The translation by Anālayo (2012: 426) is as follows:
“Be careful not to think about the past, And do not long for the future.
Matters of the past have already ceased, The future has not yet come.
As for phenomena in the present moment, One should contemplate
With mindfulness [their] lack of stability. The wise awaken in this way”.
Obviously, it can be seen that the verses do not totally match those of the Śarīrārthagāthā.
745 atītaṃ nānvāgameyya, nappaṭikaṅkhe anāgataṃ | yad atītaṃ pahīnan taṃ, appattañ ca anāgataṃ || paccuppannañ ca yo dhammaṃ, tattha tattha vipassati |
asaṃhīraṃ asaṃkuppaṃ, taṃ vidvā manubrūhaye || (MN III 187.22–26; 189.17–18; 190.18–22; 191.1–2;
191.11-2; 191.26–27; 193.10–14; 193.25–26; 194.17; 194.23; 195.26; 198.9; 198.23; 200.16–20; 201.27–28;
202.6–7). The translation by Bodhi (2005:1045) is as follows: Let not a person revive the past
Or on the future build his hopes; For the past has been left behind
And the future has not been reached. Instead with insight let him see Each presently arisen state;
Let him know that and be sure of it, Invincibly, unshakeably.
746 See Lü 1963: 242; Waldschmidt 1980: 136; Enmoto 1984; Mayeda 1985: 98; Enomoto 1986: 21; Minh
Chau 1991: 27; Oberlies 2003: 48; Anālayo 2002: 516–517.
747 Enomoto (1986: 21–22) demonstrated that the Chinese MĀ belongs to an older recension of the Sarvāstivādins.
 
4.2. 1. The analysis of Group B verses

From the preceding information, the Group B verses will be analysed mainly in two ways. Firstly, the distinct nature of the Chinese Group B verses will be demonstrated from internal comparisons based upon the three versions of the Śarīrārthagāthā.
Secondly, external comparisons will be made between the Śarīrārthagāthā to parallel verses in other texts. Additionally, in the last section, the relationship between Uv-B and T212, and the term prapañca will be examined.

4.2.1. The distinctive nature of the Chinese Group B verses

Among the three versions of the Śarīrārthagāthā, the distinctive nature of the Chinese version can be seen from the following three examples:
(1) the ninth verse set
The wording of the third pāda in the Chinese version is closer to the third pāda in Uv- B 17.12 than the corresponding Sanskrit or Tibetan Śarīrārthagāthā verses. The third pāda in Uv-B748 and Dhp 95749 mentions that the pond is rid of mud which is a simile for the wise. The Chinese instead contains a positive simile which compares the wise with a clean pond, instead of a pond without mud. A similar description can also be seen Dhp 95 (rahado va apetakaddamo), T210 (淨如水無垢) and T212 (澄如清泉). The Sanskrit and Tibetan versions instead contain a simile which compares a lake abounding with mud to saṃsāra. Therefore, the Chinese seems to be a different version from the other two but close to most of the parallels.
(2) the seventeenth verse set
In the second pāda the Chinese and Tibetan versions contain the word “moat” (qiàn塹,
'obs), while the corresponding word in the Sanskrit version is “obstacle” (parigham). This difference can also be seen in other texts. “Obstacle” is found in Uv-S, Ud and Nett; however, “moat” is found in Uv-B. The Sanskrit version and Uv-S are close to the Pāli parallels. The Chinese and Tibetan versions are related to Uv-B.
(3) the twenty second verse set

748 hrada iva hi vinītakardamo (Uv-B 237.5).
749 rahado va apetakaddamo.
 
In the last pāda the Chinese version contains the word “receives/experiences” (shòu 受 ), while the corresponding word in the Sanskrit and Tibetan versions is “enjoy” (bhuṃkte, spyod). The Chinese version is close to Uv-B (vetti)750 and other parallels (Th 85 and T212).
According to Schmithausen (1970), Uv-B belongs to the Sarvāstivādins and Uv-S belongs to the Mūlasarvāstivādins. Although most verse sets in the Chinese version are close to Mūlasarvāstivādin parallels, from the above examples we can see that some Sarvāstivādin heritage has been left in the Chinese version.751

4.2.2. The features of the Group B verses of the Śarīrārthagāthā

In this section, the features of the Group B verses will be divided into three aspects.
The first is the different terms in use. In the Group B verses, the terms are sometimes different from those in other texts (except the corresponding text, Uv-S). The second is the order of the verses. The arrangement of the verses in Group B is different from some parallels. The third is the number of the verses. The omission of material in two examples will be discussed later.

4.2.2.1. Different terms in use

Five cases will be discussed in which the Group B verses of the Śarīrārthagāthā contain a different term than its parallels. Firstly, the ninth verse set contains the word “sky” in the first pāda, but the corresponding word is “earth” in all other texts. Secondly, the twelfth verse set contains the word “sinews” in the second pāda, while the corresponding word is “blood” in all other texts. We find the term “blood” is even in the Mūlasarvāstivādin text called Gēnběnshuōyīqièyǒubù pínàiyē záshì 根本說一切有部毘奈
耶雜事 (Mūlasarvāstivādavinaya-kṣudrakavastu). Therefore, the usage of “sinews” is
quite unique in this case. Thirdly, in the fifteenth verse set, “desire” is described as “mud.” However, the corresponding terms are “flood” (ogha) in Uv-B and “fire” (aggi,


750 cittanimittasya kovidaḥ, pravivekasya rasaṃ prajānakaḥ | dhyāyī nipakaḥ pratismṛto, vetti prītisukhaṃ nirāmiṣam || (Uv-B 426.4–7).
751 In Tibetan, there is one case close to the Sarvāstivādin version as well.
 
huǒ 火 ) in Dhp, and T210. Fourthly, the nineteenth verse set mentions the destruction of “defilements” in the last pāda. However, other texts mention the destruction of “desire, hatred, and ignorance.” The usage of the word “defilements” in the nineteenth verse set is also found in parallels attributed to the Mūlasarvāstivādins, like T1455 and T1458.752 It seems to be a reading specific to the Mūlasarvāstivāda school. Lastly, the term “enjoy” is found in the last pāda of the twenty second verse set, yet the corresponding word in other texts is “obtain” (vetti, adhigaccheyya, huò 獲).
To summarise, one case in which the Group B verses of the Śarīrārthagāthā contain
a different term to its parallels can only be found in the Śarīrārthagāthā, one case is a specific reading of the Mūlasarvāstivādins, while the other three cases are different to their parallels. It is hard to know if the special usages of the three latter cases exist only in the Śarīrārthagāthā or in other texts attributed to the Mūlasarvāstivāda school since parallels to these particular verses have not yet been found in other Mūlasarvāstivādin texts.

4.2.2.2. The sequence of the verses

The following three examples show differences in the sequence of the verses of the Śarīrārthagāthā and its parallels. Firstly, in the sixth verse set, the fourth verse corresponds to the second verse in Uv-B 8.11, SN 8.5 and Sn 450. So, the second and third verses of the Śarīrārthagāthā connect with the third and fourth in these parallels. However, the order of the verses in SĀ-G 1218, SĀ-U 253 and another two texts of the (Mūla)sarvāstivādins753 is the same with that in the sixth verse set. Therefore, (Mūla)sarvāstivādin texts seem to share the same arrangement of the verses. The sequence of the verses in Uv-B, which is attributed to the Sarvāstivādins, is not a common situation. Secondly, in the tenth verse set, the sequence of verses is different to parallels in Uv-B, AN, Th, T212, and T213. The detail can be seen in § 4.1.10 above.
However, the sequence of verses in the tenth set is the same as that of T1451, a vinaya


752 T1455 (Gēnběnshuōyīqièyǒubù bìzōuníjiè jīng 根本說一切有部苾芻尼戒經): 若人能惠施 福增怨自息修善除眾惡 惑盡至涅槃 (T1455, 24.517a24–25); T1458 (Gēnběnsàpóduōbù lǜshè 根本薩婆多部律攝): 若人能惠施 福增怨自息 修善除眾惡 惑盡至涅槃 (T1458, 24.609b26–27).
753 T1545 (Āpídámó dàpípóshā lùn 阿毘達磨大毘婆沙論, the Chinese Abhidharmamahā-vibhāṣā) and T1546 (Āpítán pípóshā lùn 阿毘曇毘婆沙論, the Chinese Abhidharma-vibhāṣā),
 
text of the Mūlasarvāstivādins.754 It might be an internal arrangement found in Mūlasarvāstivādin texts alone. Thirdly, in the twenty fifth set—named Ajitapraśna—has a total of ten verses. Only the Suttanipāta has parallels to all the verses; however, their sequence is different, being Sn 1032–1037 (six verses), Sn 1110–1111 (two verses) and Sn 1038–1039 (two verses). The verses of Sn 1032–1039 are in a section called Ajitamāṇavapucchā, but the verses of Sn 1110–1111 are in another section called Udayamāṇavapucchā. Mūlasarvāstivādin parallels have not been found.755

4.2.2.3. Omitted material

Two cases will now be discussed in which there is omitted material. The first instance occurs in the twenty sixth set, which has six verses. In Sn 766–770756 and T198757 (Fóshuō yìzú jīng 佛說義足經), only five similar verses can be found. The sixth verse in Sn and T198 is the same,758 but it does not appear in the twenty sixth set; instead, another verse replaces it in the twenty sixth set. Why does only one verse appear divergent in this series of verses? Enomoto (1989: 23) conjectured that this verse was somehow replaced by a foreign verse which can be found in Uv-B 3.10cd and 10.13ab. It is difficult to resolve this issue satisfactorily because complete corresponding verses have not yet been found and there are not many parallels to the individual verses. The


754 It is Gēnběnshuōyīqièyǒubù pínàiyē záshì 根本說一切有部毘奈耶雜事, the Chinese
Mūlasarvāstivādavinayakṣudrakavastu.
755 According to Enomoto (1989: 23), in the Eastern Turkestan version, the Ajitaḥ Praśnaḥ has added two verses which have no counterparts found in the Suttanipāta nor in the Śarīrārthagāthā. He believes that the twenty fifth verse set in the Śarīrārthagāthā should be a later recension of the Mūlasarvāstivādins.
756 kāmaṃ kāmayamānassa, tassa ce taṃ samijjhati | addhā pītimano hoti, laddhā macco yad icchati ||
tassa ce kāmayānassa, chandajātassa jantuno | te kāmā parihāyanti, sallaviddho va ruppati ||
yo kāme parivajjeti, sappasseva padā siro | so imaṃ visattikaṃ loke, sato samativattati ||
khettaṃ vatthuṃ hiraññaṃ vā, gavāssaṃ dāsaporisaṃ | thiyo bandhū puthu kāme, yo naro anugijjhati || abalā va naṃ balīyanti, maddante naṃ parissayā | tato naṃ dukkham anveti, nāvaṃ bhinnam ivodakaṃ || (Sn 766–770)
757 增念隨欲 已有復願 日增為喜 從得自在有貪世欲 坐貪癡人 既亡欲願 毒箭著身是欲當遠 如附蛇頭 違世所樂 當定行禪田種珍寶 牛馬養者 坐女繫欲 癡行犯身
倒羸為強 坐服甚怨 次冥受痛 船破海中 (T198, 4.175c17–23).
758 The sixth verse is tasmā jantu sadā sato kāmāni parivajjaye, te pahāya tare oghaṃ nāvaṃ siñcitvā pāragū ti in Sn 771, and 故說攝意 遠欲勿犯 精進求度 載船至岸 in T198. Norman (2001: 103) translated this verse as follows, “A mindful person should always avoid sensual pleasure. Having abandoned them he would cross over the flood, like one who had gone to the far shore after baling out his boat.”
 
second case is the twenty seventh verse set, which has a total of six pādas. This set appears to be missing two pādas which are present in the parallels at MĀ 165–167 and MN 131–134. The content of these two missing pādas are not out of place in MĀ 165– 167 and MN 131–134.759 Also, the number of the verses in the twenty seventh set is six, which is rare for the Śarīrārthagāthā. There is a descrption in T198760 (Fóshuō yìzú jīng 佛說義足經, the Chinese Arthavargīgyasūtra), which states, “you (the Buddha) said eight verses.”761 Therefore, it is very likely that the two pādas of the first verse were somehow omitted by mistake. Besides, the content of the twenty seventh verse set is closer to the content of the verses in MN than that in MĀ. Based on the previous verse sets, the verses
of the twenty seventh set might also be based upon a text attributed to the Mūlasarvāstivādins which has not yet been found. This is different from the idea of Enomoto (1898: 35).762 It is thus difficult to know if the two pādas originally appeared in the text.

4.2.3. Other issues

In this section, two topics will be analyzed. The first is the relationship between Uv- B and T212. The second is the definition of prapañca. The term prapañca appears twice in the Group B verses, but the meanings of the term in the two verse sets are different.
Thus, this issue will be explained by investigating the definition of prapañca in the
Yogacārabhūmi.











759 The content is as follows: “For the past has been left behind, and the future has not been reached.”
760 增念隨欲 已有復願 日增為喜 從得自在有貪世欲 坐貪癡人 既亡欲願 毒箭著身是欲當遠 如附蛇頭 違世所樂 當定行禪田種珍寶 牛馬養者 坐女繫欲 癡行犯身
倒羸為強 坐服甚怨 次冥受痛 船破海中 (T198, 4.175c17–23).
761 汝說八偈 (T198, 4.175c4).
762 Enomoto indicated that the verses in MĀ165–167 correspond to the verses in twenty seventh verse set in the Śarīrārthagāthā.
 
4.2.3.1. The relationship between Uv-B and T212

Among the twenty four sets, the verses in T212 (Chūyào jīng 出 曜 經 ) mostly correspond to the Group B verses. Compared with Uv-S, Uv-B is closer to T212.763 The school to which T212 is affiliated is still in question. Hiraoka (2007: 186) believed that the verses in T212 are mainly from the Sarvāstivādins, but also collected from other schools as well. Su (2015) arranges the chapters with verses. Uv-B and T212 share the same sequence of chapters, and the titles of chapters. He strongly suggests that T212 is a version close to Uv-B, especially the part containing verses.764 In this section, the relationship between Uv-B and T212 (Chūyào jīng 出曜經) will be investigated based on the content of the twenty four verse sets.
Only eighteen verse sets in T212 can be found to compare with the verses in Uv-B. There are six differences as listed in the table below. The differences in the third and seventeen verse sets seem to have arisen from interpretation or mistranslation, which, if true, means the verses of the source are identical. The missing words and verses, like in the tenth and thirteenth sets, may have arisen when a scribe omitted material by accident while copying a manuscript. The most obvious divergences between the two texts are found in the second and twelfth sets. The two differences are content related. Overall, the versions of Uv-B and T212 are close. However, because six verses are missing and there are two obvious differences in content, it is still difficult to say that these two texts are from the same source; nonetheless, it seems they belong to the same school.


Uv-B and T212
1 Same






2 Difference:
1. Uv-B: carefulness is the path of the deathless; T212: “moral conduct” (jiè 戒) is the path of the death.
2. Uv-B: those who are careless are always dead; T212: “craving”
(tān 貪) is the key for death.


763 For example, the terms used in the ninth, ninteenth, and twenty second verse sets in T212 are closer to Uv-B than Uv-S.
764 In T212, there is also a part containing commentaries on verses, which might be mixed with the interpretation and explanation of translator(s). See Su 2015: 161.
 




3 Difference in the first two pādas:
Uv-B: for a living creature who is disturbed by discursive thoughts has strong lust regarding the beautiful.
T212: one who is careless, regards lust as the pure object.
4 Same
5 No verse
6 No verse
7 Same
8 No verse
9 Same
10 Difference: one less verse in T212.
11 Same




12 Difference in the last two pādas:
Uv-B: he is distressed by lust, hatred, conceit and hypocrisy.
T212: the sense facilities all open up, which is like making friends with thieves and being careless.

13 Difference in the second pāda: Uv-B, the terms “monk” (bhikṣu) and
“mind” (manas) are not mentioned.
14 Same
15 No verse
16 No verse



17 Difference in the second pāda:
Uv-B: has overcome the fetter [and] moat.
T212: [one] plants the seed of suffering in the body (shēnqiàn身塹).
18 Too ambiguous to analyse
19 Same
20 Same
21 Same
22 Same
23 No verse
24 Same
 
4.2.3.2. The definition of prapañca

The term prapañca has multiple meanings.765 In the Group B verses, the term prapañca can be seen twice with different meanings. Therefore, this section will investigate the definition of prapañca in the Yogacārabhūmi.
First case of the term prapañca is in the sixteenth verse set. Regarding the third pāda the commentary indicates that “delighting in xì lùn 戲 論 (prapañca)” is longing for anything, like sensual desire (yù qiú 欲 求 ), [continued] existence (yǒu qiú 有 求 ), and pure practice [of external path] (fànxíng qiú 梵行求).766 “xì lùn 戲論” (prapañca) in this case is “longing” or “desire.” Therefore, the next pāda of the verse mentions that the Tathāgata abandons all desires, thus he has no xì lùn 戲論 (prapañca).767 The second case is in the seventeenth verse set. The commentary explains that “arising evil thoughts, such as executing with a knife and stick” is “falling into xì lùn 戲論 (prapañca).” In this case, “xì lùn 戲 論 (prapañca)” is equated with “evil thought.” From the above two cases, the meaning of the term is quite divergent.
The difference between the two verse sets can be explained by another statement in the Yogacārabhūmi, which reads, “what is xì lùn 戲論 (prapañca)? It is said to be all the defilements and insignificant defilements.”768 Clearly, the two meanings of prapañca, desire and evil thought, are both attributed to defilements. This is the reason why there are two different meanings of the same term. In the sixteenth and eighteenth set, the term prapañca is specifically emphasized in context as “desire” and “evil thought” from the general definition “defilements” in the Yogacārabhūmi.

4.2.4. Summary

In this section, three aspects of the Group B verses have been discussed. Firstly, although most Group B verses in the Chinese version are related to the

765 A well-known publication which discusses this term is Ñāṇānanda 2012, which covers the term’s meanings from early Buddhist texts to Mahāyāna texts. However, most publications do not include the definition from the Yogacārabhūmi.
766又此一切三門所攝 或欲求門 或有求門 或梵行求門 如是皆名樂著戲論 (T1579, 30.384b8–10).
767 如來棄捨一切所有悕求 故無戲論 (T1579, 30.384b10–11).
768 云何戲論 謂一切煩惱及雜煩惱 (T1579, 30.345c20).
 
Mūlasarvāstivādins, it can be seen that they preserve some Sarvāstivādin heritage. Secondly, three special features of the Group B verses in the Śarīrārthagāthā have been found. Thirdly, two issues are involved. The first issue is that the versions of Uv-B and T212 are rather close with regards to these twenty four verse sets. However, due to six missing verse sets and two obvious differences in content, it is still difficult to say that these two texts represent the same version; nonetheless, it seems they belong to the same school. Another issue is the meaning of the term prapañca. In the sixteenth and eighteenth sets, the term prapañca means “desire” and “evil thought” which stem from the general definition “defilements” in the Yogacārabhūmi.

4.3. Conclusion

In this Chapter, the three versions of the Group B verses with their translations were presented, as well as a comparison of the verses from the Śarīrārthagāthā with other texts. Following this, based on previous information, it was argued that the Chinese version of the Group B verses is distinct from the other two versions and the differences of its version of the Śarīrārthagāthā verses were highlighted. Additionally, after comparison, it has been argued that the versions of Uv-B and T212 are rather close in content based on the twenty four verse sets. Finally, the meanings of the term prapañca in the Yogacārabhūmi have been shown.
Also from the above information, in these Group B verses of the Śarīrārthagāthā, the first twenty four sets actually correspond to the verses of the Udānavarga, and another three sets are related to the verses of the Pārāyanavarga (Ajitaḥ Praśnaḥ), Arthavargīyavarga (kāmān ārabhya) and Bhadraikarāgāthā769 (the parallel is found in MĀ and MN), although corresponding parallels for the last three verse sets have yet to be found. Parallels to these Group B verses also appear in early Indian non-Buddhist literature. For example, some verses of the twenty fifth set are also found in the Mahābhārata and an old Jain sūtra called Isibhāsiyāin (Aramaki 1976). These particular verses seem to have been popular at an early time in India and appear to have been incorporated into some Buddhist texts, such as the Suttanipāta and the Yogacārabhūmi,

769 This Sanskrit title (see Enomoto 1989: 35) is similar with the Chinese zàoxiánshàn 造賢善 in the
Śarīrārthagāthā. However, the Tibetan mtshan mo bzang po is closer to the title bhaddekaratta in MN.
 
with Buddhist philosophy for educational means. In other words, each verse (or verse set) might have been chanted broadly and seperately amongst the population, but was gradually gathered together into specific Buddhist, Jain and Brahmanical texts.
 
Chapter five: Texual issues of the Śarīrārthagāthā

In this chapter, three issues will be investigated. The first is the formation of the Śarīrārthagāthā including source materials, formation within the overall development of the Yogacārabhūmi, and attribution of Buddhist school. The second issue concerns the summary verse (uddāna). In the Śarīrārthagāthā, the summary verse appears to classify the verses of the Śarīrārthagāthā;770 however, the classification of the verses is treated differently between Tibetan and Sanskrit/Chinese versions. Possible reasons for this will be discussed. The third issue which will be investigated is the principles that may have guided the selection of verses in the Śarīrārthagāthā. As mentioned in chapter two, the verses are related to two aṅgas, namely, geya and vyākaraṇa; however, only particular geya and vyākaraṇa verses were selected by the compilers of the Śarīrārthagāthā. I will explore the possible reasons for this.

5.1. The formation of the Śarīrārthagāthā

In order to understand the formation of the Śarīrārthagāthā, it is important to first understand the formation in the Yogacārabhūmi. According to Schmithausen (1987a: 14), there are three main layers of the Yogacārabhūmi. The first layer makes no reference to ālayavijñāna (“storehouse consciousness”), the second sporadically mentions ālayavijñāna but has no reference to the Saṃdhinirmocanasūtra, and the third contains detailed discussion of ālayavijñāna and information about the Saṃdhinirmocanasūtra.
Based on this hypothesis, Deleanu (2006: 155) suggested six phases in the formation of the Yogacārabhūmi, including the text of the Saṃdhinirmocanasūtra: 1. the Śrāvakabhūmi; 2. the Bodhisattvabhūmi; 3. the rest of the Maulyo bhūmayaḥ, the Vastusaṃgrahaṇī, the Vyākhyāsaṃgrahaṇī and the Paryāyasaṃgrahaṇī;771 4. the Saṃdhinirmocanasūtra, closely related to the Yogacārabhūmi; 5. the early parts of the Viniścayasaṃgrahaṇī; 6. the citation of the Saṃdhinirmocanasūtra in the


770 The summary verses (uddāna) basically extract a few words from each paragraph or section so it is easier for people to memorize the content of the text.
771 Deleanu (2006) mentioned that the formation of the Vastusaṃgrahaṇī might have begun earlier than any other text belonging to phase three.
 
Viniścayasaṃgrahaṇī, and the final redaction of the entire Yogacārabhūmi.772 According to this account of the formation of six phases,773 the Śarīrārthagāthā belongs to phase three, which occasionally mentions ālayavijñāna. However, this needs further examination.
In the Śarīrārthagāthā, there are mainly two parts, namely, verses and the commentary on the verses. As discussed in previous chapters, the verses of the Śarīrārthagāthā are all cited from early Buddhist texts; thus, the commentary is the key which needs to be carefully examined. The commentary can be divided into two sections. The first is the explanation of the verse(s). This clarifies the meaning of technical terms, pādas and/or whole verse(s). The commentator(s) do not explain every single word or every single verse; rather, only significant parts of the verse(s) are discussed, which implies that the commentator(s) felt that the meaning of much of this material was already readily understandable. The second part is the brief explanation (luèbiàn shàng suǒshuō yì 略辨上所說義, bshad pa 'did'i mdor bstan pa). The commentator(s) briefly further clarify the previous explanation on the verse(s) or relate relevant theory and practices which are also associated with the verse(s).774 For example, in the first verse set of the Śarīrārthagāthā, two accomplishments (mǎn 滿 , 'byor pa)775 are discussed and connected with the verse in the brief explanation. These two accomplishments are not actually found within the verse. However, not every commentary includes the brief explanation.776
If one looks at the content of the commentary in the Śarīrārthagāthā, it can be found
that the theory and philosophy do not contain the idea of ālayavijñāna or relate to the
Saṃdhinirmocanasūtra. It simply presents Śrāvakayāna practices and ideas, such as the

772 This includes cross-references, interpolations, structural modifications, etc. See Deleanu 2006: 155. Aramaki (2013: 427) proposed a similar development (with different details). He mentioned that the oldest texts (the Śrāvakabhūmi and Bodhisattvabhūmi) were developed together with three bhūmis (Śrutamayī-, Cintāmayī-, and Bhāvanāmayī-) “playing the role of being practical manuals on the level of the mundane truth (saṃvr̥tisatya) to foster liberation from the deepest pursuing (ādāna) or cumulative subconsciousness (ālayavijñāna).”
773 Deleanu (2006: 155) specifically mentioned that this is not a perfect chronological or linear development. These phases might have partially overlapped.
774 略義者 謂宣說諸法同類相應 (T1579 30.752c21).
775 The two accomplishments are the accomplishment of high status (zēngshàngshēng mǎn 增上生滿,
mngon par mtho ba 'byor ba), i.e. rebirth in a higher level of existence in saṃsāra, and the accomplishment of the ultimate good (juédìngshèng mǎn 決定勝滿, nges par legs pa 'byor pa), i.e. the attainment of nirvāṇa. 776 The examples are the second, eleventh, and twelve verse sets in Group A, as well as the twenty third verse set in Group B.
 
three instructions (sān xué 三 學 , ṭrīṇi śikṣāṇi), four noble truths (sì dì 四 諦 , catvāri āryasatyāni), eightfold path (bā zhèngdào 八正道, āryāṣṭāṅgikamārga), etc. For instance, the twenty fourth set of the Group B verses states that the two consciousnesses cannot simultaneously arise,777 yet the idea of ālayavijñāna does not apply to this verse.
The author(s) have liberally employed Mahāyāna theory in their interpretation of the verse(s) in other sections, i.e. the Paramārthagāthā778 in the Cintāmayībhūmi. Why do the commentators(s) not do this in their interpretation of the Śarīrārthagāthā verses?
This question relates to the textual development of the Yogacārabhūmi. Scholars currently believe that the Yogacārabhūmi was not created by one author.779 Instead, Asaṅga (or Maitreya) may have been a representative of an editorial group which collected and made editorial efforts, even extending over generations.780 The author(s) of the Śarīrārthagāthā and Yogacārabhūmi might therefore have been Śrāvakayāna yogācāras.781 Yìnshùn (1968: 641; 1988: 243–246) suggested that Kaśmīra (Sarvāstivādin) yogis had a major influence on the Mahāyāna yoga development.782 Moreover, Schmithausen (1969b: 817; 1987a: 13–14) also convincingly argued that the Yogacārabhūmi contains heterogenic material in both content and structure, which was not arranged in a perfectly balanced whole in the final redaction. He believed that the Yogacārabhūmi was not based on a preparatory plan but was gradually developed from complex and separate textual units and materials.783
Based on the above information, I suggest that the Śarīrārthagāthā was created in the

777 於現在一一而轉 第二伴心所遠離故 一切種心不頓轉故 (T1579, 30.386a25–26); de yang da ltar gyi dus la brten te sems kyi zla gnyis pa med pa dang sems thams cad cig car du mi 'byung ba'i phyir re re nas 'jug go (Derge 4035.256b5–6).
778 Ui (1958: 316–317) and Wayman (1961: 164) have partly found the source of the verses in the early sūtras. The commentary mentions twice the notion of ālayavijñāna, which has been discussed by Schmithausen (1987a: 138–142; 160–166; 236–239).
779 See Nishi 1960: 672–673, Frauwallner 1969: 265, Schmithausen 1969b; 1987a: 13–14, 183–185; 2000,
May 1971: 281, Hirakawa 1974–1979: 2.96, Kudo 1975, Yokoyama 1976: 82–83, Takasaki1982a: 32,
Yamabe 1989: 46 note 8, Potter 1999: 398, Aramaki 2000, Ahn 2003: 1–4, Kritzer 2005, Deleanu 2006:
154. However, some scholars hold different views. The discussion of two side views can be found in Deleanu 2006: 154.
780 See Demiéville 1954: 381 note 4 and 434 note 9, Deleanu 2006: 154.
781 See Ui 1964: 370–372, Yìnshùn 1968: 41, Fukuhara 1975: 335–423. The term yogācāra is found in many texts of various schools. Different schools have their yogācāras specialising in spiritual development. See Silk 2000.
782 Although Yìnshùn (1988) believed that Maitreya is possibly the author of the Yogacārabhūmi, after comparing the theory from other texts attributed to Maitreya, he proposed that Maitreya was a real person who inherited the study from a Revata (Jiélìfáduō 頡隸伐多) temple in Kaśmīra.
783 Schmithausen 1987a: 267–269 note 124.
 
early stages of the formation of the Yogacārabhūmi, perhaps around the same time as the Śrāvakabhūmi. A primary reason for this assertion is that there are only Śrāvakayāna concepts found in the commentarial explanations and no distinctively Mahāyāna concepts.784 This is also the reason why the two aṅgas, geya and vyākaraṇa, are connected with the two verse groups in the Śarīrārthagāthā. The two aṅgas happen to be two major parts of the Saṃyuktāgama, which is an early major sūtra collection. The author(s)/compiler(s) knew the importance of these Śarīrārthagāthā verses connected with geya and vyākaraṇa, which in later times seems to have been gradually forgotten or developed with different definitions of the aṅga system.785 It is likely that the Cintāmayībhūmi also gradually developed from individual textual units. This means that textual units like the Śarīrārthagāthā, Paramārthagāthā and Ābhiprāyikārthagāthā786 were incorporated into the Cintāmayībhūmi.787 Although the Cintāmayībhūmi involves the idea of ālayavijñāna, the Śarīrārthagāthā textual unit is based upon Śrāvakayāna material complied or transmitted at an early stage. As for the monastic ordination, the redactor(s) of the Śarīrārthagāthā very possibly belonged to the Mūlasarvāstivādin lineage because the verses correspond to the verses found mostly in Mūlasarvāstivādin texts.788 Similarly, the Vastusaṃgrahaṇī is thought to be Mūlasarvāstivādin and is also connected with the sūtra aṅga part of the Saṃyuktāgama (Yìnshùn 1983 and Mukai 1985).

5.2. Differences in the summary verse (uddāna) between the Tibetan and Sanskrit/Chinese versions

The main function of the summary verse (uddāna) is to provide a descriptive framework for the verses in the Śarīrārthagāthā. This section will discuss the possible reasons why there are differences in the summary verse between the Tibetan and

784 Undeniably, it is also possible that the author(s) simply wanted to interpret the verses using only Śrāvakayāna concepts. However, it would be difficult to explain why the ālayavijñāna concept is contained in the previous section, Paramārthagāthā and not in the Śarīrārthagāthā. Another alternative explanation is that the Śarīrārthagāthā verses were compiled and explained by a person or group of persons who favoured Śrāvakayāna concepts.
785 As discussed in chapter two.
786 The Paramārthagāthā and Ābhiprāyikārthagāthā precede the Śarīrārthagāthā.
787 The structure of the Cintāmayībhūmi was introduced in chapter one.
788 See chapters three and four.
 
Sanskrit/Chinese versions.
In order to make sure this issue really exists between Sanskrit/Chinese and Tibetan versions, it is important to check the Śarīrārthagāthā summary verse in different Buddhist text collections. In Sanskrit, the summary verse has been edited from two manuscripts by Enomoto (1989: 21).789 In Chinese, the Yogacārabhūmi can be found in at least twenty Buddhist text collections.790 A thorough assessment of all these collections is beyond the scope of this thesis; however, Deleanu (2006: 131) mentioned two basic modern lineages, namely, Taishō and Zhonghua.791 In chapter one, the Chinese summary verse is from the Taishō edition. Therefore, in addition, I will assess the Fangshan stone- carved collection (abbreviated as FS),792 which belongs to the Zhonghua edition.793 In Tibetan, the Yogacārabhūmi is found in the Tanjur. The Buddhist text collections (four classical xylograph editions) of the Tanjur can be roughly divided into two lineages: Narthang, Peking against Derge and Cone.794 Therefore, I will compare the Peking and Derge editions. Additionally, the Golden Tanjur manuscript version will also be taken into account.795
The summary verse in the two Chinese collections and the two Sanskrit manuscripts are mostly the same, stating, “evil, what can be expressed, lust, flood, being frightened, caste, obtaining fame,796 current, (two) floods,797 lust and hatred, what ought to be done,

789 The Śarīrārthagāthā is found in Śrāvakabhūmi and Yogacārabhūmi manuscripts.
790 See Deleanu 2006: 130 for a diagram which illustrates the relationship between these Chinese collections. These collections can be divided into three lineages, namely, central, Northern, and Southern. See Chikusa 1993: 10–17 and Deleanu 2006: 116–117.
791 These two modern lineages are commonly used to make textual comparisons in the academic field. 792 The work of the stone-carved scriptures started in Fángshān 房 山 (in Beijing) in the Suí Dynasty (Suíchá 隋朝) by a Buddhist monk, Jìngwǎn 靜琬, and his team. This work lasted around 1040 years passing through the six dynasties of the Suí 隋, Táng 唐, Liáng 梁, Jìn 晉, Yuán 元 and Míng 明. The 14278 stone slabs contain 1122 Buddhist scriptures in 3572 volumes. The scriptures provide important
materials concerning politics, economy, culture, and especially the history of Buddhism in China. See Zhōngguó fójiào túshū wénwù guǎn 中國佛敎圖書文物館 2000.
793 Nine collections, which belong to the Zhonghua edition, are Fangshan Stone-Carved collection, Zifu
collection, Jin collection, Qisha collection, second edition of the Korean collection, Puning collection, Southern collection of the Yongle Era, Jingshan collection, and Qing collection.
794 See Vogel 1965: 25–30, Eimer 1978: 77, Pasadika 1987: 45, Schoening 1995: 123–124, 132–134, 141–
143, Skilling 1991, Mathes 1996: 37–40, Erb 1997: 261–268, Dietz 2000: 176, Mochizuki 2004, Deleanu 2006: 78. Further, Deleanu (2006: 89) provided a detailed diagram and discussion of the relationships between twenty collections.
795 The Golden Tanjur has been examined by Skilling (1991), who suggested that it is textually close to the Narthang and Peking editions.
796 In Chinese, the word yǔ 與 appears to be a scribal error. It perhaps ought to read yù 譽 from the content. Yù 譽 also fits with the Sanskrit and Tibetan versions.
 
struggle, attainment of the aim, and specification; [it is a total of] fourteen”.798 This classification is consistent with the Śarīrārthagāthā verses. Surprisingly, the last division “specification” includes within it twenty seven verse sets.799
In Tibetan, the summary verse (uddāna) states the following:800
sdig dang brjod bya 'dod801 chags dang / / chu bo dang ni 'jigs dang rigs / / grags ldan mtsho802 dang chu bo gnyis / / 'dod chags zhe sdang bya ba dang / / 'dzin pa dang ni don thob dang / / mi rtag803 pa804 dang bag yod dang / / 'dod chags shas che chos la gnas / / nges rtag805 legs par806 smras pa dang / / dad pa dang ni thos pa dang / / nam mkha'807 mtshungs dang gzugs dang rgyal / / rus pa'i ra ba rus808 sbal dang / / mtshungs dang 'dam dang gnas gnyis dang / / bsal809 gyur dang ni byin pa dang / / sdig pa kun dang ltung ba dang / / mkhas dang bzo810 dang ring du 'gro / / bsgribs dang 'dod dang bzang po yin // (Peking 5536.304b3–b6; Derge 4035.261a7–b2; Golden 138 756.6–757.3.) Evil, what is expressed, lust, flood, being frightened, caste, famous, current, two floods, lust and hatred, what is to be done, grasping, attainment of the goal, impermanence, carefulness, strong lust, dwelling in the dharma, contemplating the truth, good saying,
faith, learning, being like the sky, physical appearance, king, city of bone, turtle, equality, mud, two dwellings, having cleaned up, giving, all evils, falling down, skilled, craft, travelling far away, obscuring, sensual pleasures, and auspicious.




797 The Chinese version does not include the word “two”, but rather simply states liú 流 (flood). But “two floods” fits better because Sanskrit term is “oghau” (two floods) and there are two verse sets about floods in this division.
798 The content in Chinese is 惡說貪流怖 類與池流貪 作劬勞得義 論議十四種 (T1579, 30.388a29–b1; FS 18.162a20). The content in Sanskrit is pāpam ākhyeyarāgaś ca oghotrastābhivarṇṇatā / yaśasvī sara oghau rāgadveṣau ca kāryatā / āyūhanam arthaprāpti upadeśaś caturddaśaḥ // (Enomoto 1989: 21)
799 The detail is shown in chapter four. The reason why these twenty seven verses sets are gathered and named as “specification” is not explained in the Yogacārabhūmi. However, in chapter two I suggest that this should be related to the “vyākaraṇa” (explanation).
800 See Peking 5536.304b3–b6; Derge 4035.261a7–b2; Golden (TBRC) 138 756.6–757.3.
801 Golden: ’dong.
802 Golden: mtshon.
803 Golden and Peking: rtags.
804 Golden: ba.
805 Golden: rtags.
806 Golden: bar.
807 Peking: kha'.
808 Golden: ru.
809 Golden and Peking: brtsal.
810 Golden and Peking: gzo.
 
The three Tibetan text collections show the same result although there are some small differences in wording between the three collections.

The obvious difference is that the fourteenth division in the Chinese and Sanskrit versions is called “specification”, but the Tibetan version does not include such a division and instead provides a list of twenty six divisions in its place.811 In the Śarīrārthagāthā, the forty one verse sets are accompanied by the commentary.
Compared to the Chinese and Sanskrit versions, the Tibetan summary verse seems to be a more accurate description of the verse contents. There are two reasons for stating this. Firstly, it is rather difficult to understand why in the Chinese and Sanskrit versions, twenty seven verse sets (twenty six titles ) are collectively labelled “specification” (lùnyì 論議, upadeśa) without any explanation. Secondly and importantly, it is confusing that in the Tibetan version, the title of the last set (twenty seventh) in Group B is “auspicious” (bzang po). In contrast to the other verse titles, this term cannot be found in the corresponding verse. In the summary verse, each term is basically from the first or second verse pāda in the Śarīrārthagāthā. The term “auspicious” (bzang po) possibly implies that this verse set is cited from the sūtra called “auspicious night” (’phags pa mtshan mo bzang po zhes bya ba'i mdo).812 Apparently, the Tibetan uddāna is more deliberately designed.
Why, then, do the Chinese/Sanskrit and Tibetan versions contain different summary verses? There are two directions by which this matter may be investigated. Firstly, the variant structures of the Yogacārabhūmi may provide an explanation. The Chinese version of the Yogacārabhūmi is divided into five main parts,813 while the Tibetan version is divided it into six parts.814 In the Tibetan version, it is because the Vinayasaṃgrahaṇī has been taken out of the Vastusaṃgrahaṇī, but there is no explanation for this. However, a potential clue may be found in the structure of the

811 The sixteenth and seventeenth sets of Group B share the title “two dwellings” (gnas gnyis).
812 This Tibetan title might correspond to the Sanskrit “Bhadrakarātrīnāma-ārya‐sūtra”. The corresponding
sūtra is MĀ 165. See Skilling 1997: 585, Anālayo 2012: 422.
813 Maulyo bhūmayaḥ (Běn dì fèn 本 地 分 ), Viniścayasaṃgrahaṇī (Shè juézé fèn 攝 決 擇 分 ;), Vyākhyāsaṃgrahaṇī (Shè shì fèn 攝釋分), Paryāyasaṃgrahaṇī (Shè yìmén fèn 攝異門分), Vastusaṃgrahaṇī (Shè shì fèn 攝事分).
814 Maulyo bhūmayaḥ (sa'i dngos gzhi), Vastusaṃgrahaṇī (gzhi bsdu ba), Vinayasaṃgrahaṇī ('dul ba bsdu ba), Viniścayasaṃgrahaṇī (rnam par gtan la dbab pa bsdu ba), Paryāyasaṃgrahaṇī (rnam grangs bsdu ba), Vyākhyāsaṃgrahaṇī (rnam par bshad pa bsdu ba).
 
Tibetan version, which is consistent with a description of its structure in the Chinese version, namely, that seventeen bhūmis (shíqī dì 十 七 地 , sa bcu bdun) and four saṃgrahaṇīs (sìzhǒng shè 四 種 攝 , bsdu ba bzhi) belong to the mātr̥kā (categorising list).815 The seventeen bhūmis refers to the Maulyo bhūmayaḥ and the four saṃgrahaṇīs refers to the Vinayasaṃgrahaṇī, Viniścayasaṃgrahaṇī, Paryāyasaṃgrahaṇī, and Vyākhyāsaṃgrahaṇī. In other words, the Tibetan version of the Yogacārabhūmi is
structured according to the description also found in the Chinese version. Moreover, Suguro (1989: 280–283) revealed that the Tibetan Vinayasaṃgrahaṇī does not include the closing formulae which usually stays in the end of the chapter. Additionally, the two Tibetan text catalogues ldan-dkar-ma and Bu-ston also list five main parts, excluding the Vinayasaṃgrahaṇī ('dul ba bsdu ba).816 The reason why the Vinayasaṃgrahaṇī become classified as an individual textual unit appears to have been an attempt to make the structure correspond to the “Tripiṭaka”, this is, the Vastusaṃgrahaṇī (gzhi bsdu ba), Vinayasaṃgrahaṇī ('dul ba bsdu ba), the Vyākhyāsaṃgrahaṇī (rnam par bshad pa bsdu ba), matching the Sūtra, Vinaya, and Abhidharma piṭakas.817 Moreover, the order of the four saṃgrahaṇīs in the Chinese and Tibetan versions is different. The suggestion by Suguro (1976: 25–31, 1989: 273–279) and Jinapurtra’s Yogacārabhūmiyvākhyā818 show that the order of the four saṃgrahaṇīs are the same as in the Chinese Yogacārabhūmi. In sum, the above information indicates that the Tibetan Yogacārabhūmi has been reorganised.819 From this, it may be surmised that the Tibetan summary verse (uddāna) might have been changed to fit the structure of the Śarīrārthagāthā verses or for other reasons.820 It seems to be a modification of the earlier redaction in the Tibetan textual transmission.
Another direction by which this matter may be investigated is two aspects of the summary verses in the Yogacārabhūmi. Firstly, according to Kanakura (1977: 115–117), the summary verses in the Śrāvakabhūmi were added after the prose content had been

815 It is located in the Viniścayasaṃgrahaṇī and the content is de la ma mo bsdu ba ni sa bcu bdun dang bsdu ba bzhi'o in Tibetan version (Peking 5539.195b7); 摩呾理迦所攝者 謂十七地及四種攝” in Chinese version (T1579, 30.654b6).
816 See Lalou 1953: 334 and Nishioka 1981: 56.
817 Cf. Deleanu (2006: 46).
818 See T1580, 30.885a8–12.
819 Deleanu (2006: 47) suggested that the reorganisation took place at a late date, after Bu-ston.
820 In chapter two, I assume that the Tibetan compiler(s) might have intended to emphasize the common feature of all Śarīrārthagāthā verses, that is, geya.
 
finished.821 However, he also could not fully explain why the summary verses sometimes do not perfectly match the content of the text, like the cases in the Śrāvakabhūmi822 and the Manobhūmi.823 Secondly, Deleanu (2006: 151) also pointed out that the summary verses are not always located at the end of a section or chapter; they are sometimes located in the middle of the content to which it refers. An example is the Sanskrit and Chinese Yogasthāna I.824 However, the Tibetan version does not name the kind of verses as uddāna (sdom), but rather as antaroddāna (bar gyi sdom). In the Sanskrit and Chinese Śrāvakabhūmi and Yogacārabhūmi, antaroddānas (middle summary verse) exist.825 It is quite confusing that the verses are still called uddāna when they are located in the middle of sections in the Sanskrit and Chinese Yogasthāna I.
Deleanu (2006: 152–153) provided the following possible explanations which might solve the above two issues:
1. The verses already appeared differently in the Sanskrit manuscript transmission in India due to the presence of concurrent different recensions.
2. Tibetan translators/compilers used a manuscript different from the extant Sanskrit one and the one the Chinese translator(s) had.
3. Tibetan translators/compilers made some modifications to provide more coherence to the text.
4. The Śrāvakabhūmi might have developed from separate and individual textual units.826 When the textual units were combined together to form the Śrāvakabhūmi, the editors did not remove some uddānas even if they did not suit the structure or content. The editors distinguished them by labelling them antaroddāna or piṇḍoddāna827 instead of uddāna.828

821 Kanakura analysed the Sanskrit and Chinese Śrāvakabhūmi, and proposed that the wording of the uddāna does not matter between the different versions, and do not affect the text’s structure or content. Different wording might have been caused by the transmission of the text.
822 The detail can be seen in Kanakura 1977: 115–117 and Deleanu 2006: 150–152.
823 In the Manobhūmi, an uddāna is separated from the discussion to which it refers. See Bhattacharya 1957: 71.
824 See ŚrBh-Gr 98, T1579, 30.402a21–23.
825 Examples of antaroddāna (middle summary verse) can be seen at ŚrBh 106a7–106b1, T1579, 30.458b13–22.
826 An individual textual unit might be based on one or a few summary verses (uddāna) (Deleanu 2006: 152).
827 It is a “general summary verse” (zǒng wàtuōnán 總嗢拕南, bsdus pa’i sdom).
828 Although keeping irregular uddānas disturbs stylistic uniformity, it provided an advantage by serving as a mnemonic device tracking the content of long doctrines in the traditional environment. See Deleanu 2006:
 
5. The oral transmission might have played a part in the formation of the Yogacārabhūmi, especially the early materials. Plausibly, the yogic doctrines at first were orally spread amongst small groups of practitioners and were gradually taken into writing.829 The summary verses may have served as mnemonic methods to memorise the content of the text, especially during the oral transmission phase. More and more oral textual units continued to be collected and transmitted, but the need to write them down increased. Deleanu (2006: 153) stated, “Later oral expositions could then rely and elaborate upon these written notes and materials. And in their turn, these expanded explanations would also be written down.” The irregular uddānas were created between the oral and writing transmission process.
Deleanu attempted to solve the issues with reference to the process of textual transmission. The above five points seem to suitably explain why there is a different summary verse (uddāna) between the Chinese/Sanskrit and Tibetan Śarīrārthagāthā.
The first direction provides an explanation which is similar to Deleanu’s third point.
In brief, Deleanu’s five explanations can provide explanations for the issues.830 The different summary verse (uddāna) between the Chinese/Sanskrit and Tibetan Śarīrārthagāthā may be caused by different recensions in India (the first and second), the formation of the text from disparate textual units831 (the fourth), the oral and written transmission (the fifth), and/or modification during the textual transmission in Tibet (the third).832
Based on evidence provided below, I suggest the third explanation is the most likely to be correct, that is, the modification happened in Tibet. In Group B, the title of the sixteenth and seventeenth sets is “two dwellings” (gnas gnyis). This means that there is a dwelling (gnas) in each set (shown in the first pāda). However, the source terms in the two sets are different. In the seventeenth set, the term is zhù 住 in Chinese, sthiti in Sanskrit, and gnas in Tibetan. The three versions share the same meaning, “dwelling.”

153.
829 However, Deleanu (2006: 153) mentioned that the early materials of the Yogacārabhūmi might have been composed as written text(s). This suggestion cannot be completely excluded.
830 I suggest that these five explanations can also be applied to the question of why there is a different abbreviation formula in the thirteen set of Group A verses between Chinese and Sanskrit/Tibetan versions. 831 In the case of the Śarīrārthagāthā, the complier(s) of the summary verse (uddāna) were different for the Chinese/Sanskrit and Tibetan recensions, so the content was changed in India.
832 Deleanu (2006: 47) mentioned that it might also possibly be the editorial effort of the translator(s) Jinamitra and Yes shes sde, etc. But he also mentioned that this idea is highly speculative.
 
However, in the sixteenth set, the term gnas (dwelling) in Tibetan is different from the other two versions.833 In the Chinese and Sanskrit versions, the terms are instead “bird’s trace” (niǎojī 鳥 跡 ) and “footprint” (pada), respectively. The Chinese and Sanskrit versions are similar. The footprint in the sky could refer to a bird’s trace. The term gnas (dwelling) in Tibetan might indicate a similar meaning to “bird’s trace” (niǎojī 鳥 跡 ) or “footprint” (pada). The source term may have been “pada” in the manuscript(s) that the Tibetan translator(s) held. Specifically, the title of the sixteenth and seventeenth sets of Group B in the uddāna of the Sanskrit manuscript that Tibetan translator(s) held could not have been “two dwellings” (sthitī or padau). Also, the Tibetan translator(s) might have known the source terms, so they would not have labelled the two verse sets as “two dwellings”. It is very likely that the Tibetan uddāna is based on the Tibetan translation. It may be that the Tibetan compiler(s) read the term gnas in each set, but did not notice that the original terms were different, so they copied the pattern of the ninth and tenth set of Group A verses, which share the same title “two floods”, and named the sixteenth and seventeenth sets of Group B verses “two dwellings” (gnas gnyis). In other words, I suggest that the Tibetan uddāna represents a modification after the translation was completed, which, if correct, would mean that the extant Tibetan uddāna is not the effort of the translators, Jinamitra, Yes shes sde, etc.

5.3. Potential principles of selecting verses in the Śarīrārthagāthā

In cases where the Śarīrārthagāthā verses were attributed to the two aṅgas, geya and vyākaraṇa, as discussed in chapter two, why is it that not all geya and vyākaraṇa verses were included in the Śarīrārthagāthā? Were these verses ever incorporated into a particular text, so that the author(s) of the Śarīrārthagāthā commented upon the text verse by verse, just like the relationship between the Vastusaṃgrahaṇī and the Saṃyuktāgama?834 Or is the Śarīrārthagāthā an anthology of particular selected verses from other texts. These questions are still difficult to answer due to limited information. However, from an analysis of the content of the Śarīrārthagāthā verses with other texts,


833 The first pāda states, “there is no dwelling in the sky” (mkha' la ni gnas med ltar).
834 The Vastusaṃgrahaṇī comments on the sūtra aṅga section of the Saṃyuktāgama. See Yìnshùn 1983 and Mukai 1985.
 
we may postulate some potential principles which may have been applied in verse selection.
In the Saṃyuktāgama, we may roughly divide the verses into four different categories on the basis of their content. The first are those which directly express the Buddha’s teachings, such as the methods of practice and learning. The majority of verses fall into this category. The second category of verses praise the Buddha, and sometimes his disciples. Such verses are mostly found in the Brahmā saṃyukta (fàntiān xiàngyìng 梵天相應) and Vaṅgīsa saṃyukta (póqíshě xiàngyìng 婆耆沙相應). In the third category, the Buddha publicly compliments Śakra and these verses are found in the Śakra saṃyukta (dìshì xiàngyìng 帝 釋 相 應 ). In the fourth category, the Buddha’s disciples (mostly bhikṣus) are encouraged to practice better and these verses are found in the Vanasaṃyukta (lín xiàngyìng 林相應). The last three categories are verses expressed by perceptual experience. The verses of the four categories are basically attributed to the geya and vyākaraṇa aṅgas.835
If applying this categorisation scheme to the content of the Śarīrārthagāthā verses, they seem to belong to the first category only. In Group A, verses are quoted from the saṃyuktas of Devatā, Bhikṣu, Vaṅgīsa, Māra, and Yakṣa.836 While it is true that the third verse set is from the Vaṅgīsa saṃyukta where many verses are about praising the Buddha or his disciples, the verses of this set primarily concern the practices by which sensual



835 The range of geya and vyākaraṇa aṅgas is described by Yìnshùn (1983) and has been discussed in chapter two.
836 The Group A verses may be classified as follows:
Verse set SĀ-G number saṃyukta Verse set SĀ-G number saṃyukta
1 1270–1274 Devatā 8 601 Devatā
2

1078 Bhikṣu 9 603
1326
1329 Devatā Yakṣa Yakṣa
3


1214 Vaṅgīsa 10 1269
1326
1329
1316 Devatā Yakṣa Yakṣa
Devatā
4
1092 Māra 11 1314
1324 Devatā Yakṣa
5 596 Devatā 12 1311 Devatā
6 597 Devatā 13 1267 Devatā
7 1282 Devatā 14 1092 Māra
 
lust may be conquered.837 This is related to the Buddha’s teachings, which belong to the first category discussed above. All the other verses of Group A are relevant to the Buddha’s teaching as well. The Group B verses are quoted from four different texts.838 While these verses fall into various categories, they are also mostly about the Buddha’s teaching. From this we may deduce that the author(s) apparently selected verses on the basis of their direct connection with the Buddha’s teachings.
In the Yogacārabhūmi, there is little information which explicitly states how particular verses were selected for citation. However, since the Śarīrārthagāthā verses frequently come from (Mūla)sarvāstivādin texts, it seems likely that that the (Mūla)sarvāstivādin tradition might have influenced verse selection. Assessments of verses by (Mūla)sarvāstivādin author(s) tend to be rather conservative. There are a few statements which demonstrate this from the Āpídámó dàpípóshā lùn 阿毘達磨大毘婆沙
論 (the Chinese Abhidharmamahā-vibhāṣā), a (Mūla)sarvāstivādin text:


法善現頌當云何通。答此不必須通。以非素怛纜毘奈耶阿毘達磨所說。但是造制文頌。夫造文頌或增或減。不必如義何須通耶。 (T1545, 27.866b23–25).
How should one understand Fǎshànxiàn’s839 verses? Answer: There is no need to understand them because they are not stated in the sūtra, vinaya, or abhidharma. They are merely literary verses by creation. The literary verses by creation [may contain] more or less and do not necessarily correspond to the doctrine.

諸讚佛頌言多過實。如分別論者。讚說世尊心常在定。善安住念及正知故。又讚說佛恒不睡眠離諸蓋故。如彼讚佛實不及言。(T1545, 27.410b25–28).
The verses praising the Buddha go beyond reality. It is like what has been praised by the Vibhajyavādin: the Buddha’s mind is always in concentration, well-settled, mindful, and of right understanding. The Vibhajyavādin also praises the Buddha for not ever sleeping and for being free from concealments (skandha). The truth does not measure up to such praises of the Buddha.



837 See chapter three for more detail.
838 See chapter four.
839 According to Yìnshùn (1967: 324), Fǎshànxiàn 法善現 is the same person as Aśvaghoṣa (Mǎmíng 馬鳴).
 
尊者達羅達多所說當云何通。答彼不須通非三藏故。文頌所說或然不然。達羅達多是文頌者言多過實故不須通」 (T1545, 27.358b27–29).
How should one understand what the Venerable Dharadatta said? Answer: There is no need to understand it because it is not of the Tripiṭaka. What is mentioned in the literary verses might be true or not. Dharadatta is the creator of literary verses. [His work] goes beyond the reality. Thus, there is no need to understand it.

Apparently, (Mūla)sarvāstivādins were particularly conservative regarding which verses they accepted as Buddhavacana. They believed that verses could exaggerate the real state of the Buddha or that they do not necessarily correspond to the doctrine. In other words, they felt that this kind of literary material is derived from people’s affection which might idolise the Buddha or overstate the truth. If this attitude of the (Mūla)sarvāstivādins was held by the author(s)/compiler(s) of the Yogacārabhūmi, it then follows that the Śarīrārthagāthā verses do not involve verses praising the Buddha or his disciples and included only those which directly relate to the Buddha’s teaching. The Śarīrārthagāthā verses may have been selected particularly in order to eliminate perceived exaggerations.
Besides, in the group A verses, at least 10 sets come from the Devatā saṃyukta, as stated in the above discussion. This might reflect the situation that belief in deities was still prevalent in India at that time and in that way lead to Buddhism. A similar instance is found in the twenty fifth set of Group B. These verses, which were popular in India, were interpreted according to Buddhist doctrine.840 This may indicate that some verses were selected on the basis of popularity. In other words, because the verses were influential amongst the population, they were incoporated into the Śarīrārthagāthā.

5.4 Conclusion

There were three main points of discussion in this chapter. The first concerned the formation of the Śarīrārthagāthā. The Śarīrārthagāthā, which appears to represent a coherent textual unit, was created during an early stage of the development of the Yogacārabhūmi, such as the phase of the Śrāvakabhūmi. Although the Cintāmayībhūmi

840 The verses can be found in Indian non-Buddhist literature, such as the Mahābhārata and an old Jain
sūtra called Isibhāsiyāin, which was discussed in chapter four.
 
involves the idea of the ālayavijñāna, the Śarīrārthagāthā is based upon Śrāvakayāna materials complied or transmitted at an earlier time. The second main point of discussion was the analysis of the different summary verse (uddāna) between the Chinese/Sanskrit and Tibetan Śarīrārthagāthā. The five explanations of Deleanu (2006: 152–153) can be applied to solve this issue, but I suggest that the difference found in the Tibetan summary verse represents a modification which occurred in Tibet (Deleanu’s third explanation).
The Tibetan summary verse appears to be based on the Tibetan translation, so it is not the effort of the translators, Jinamitra, Yes shes sde, etc. The third main point of discussion was the potential principles of verse selection. The (Mūla)sarvāstivāda tradition might have a great influence, since it does not accept all verses as Buddhavacana. They believed that verses could exaggerate the real state of the Buddha or do not necessarily correspond to the doctrine. This might be the reason why all the Śarīrārthagāthā verses are relevant to the Buddha’s teaching. Some verses may also have been selected on the basis of their popularity in an attempt to bring more people in contact with Buddhism.
 
Chapter six: Conclusion

In this chapter, the content of previous chapters will be briefly reviewed and summarised. From this summary, and from a comparison with relevant studies of the Śarīrārthagāthā, the contribution this study has made to the topic will be clarified.
Avenues for future research on the Śarīrārthagāthā will also be suggested.

6.1. Summary of previous chapters

A close reading the Śarīrārthagāthā, which is composed of forty-one verse sets with its commentary, elicits a series of questions, such as, where do these verses come from? Why have these verses been collected in the Śarīrārthagāthā? Why is the text called the Śarīrārthagāthā? The answers to these questions have been provided in chapters one and two, and summarized in § 6.1.1 below. If these Śarīrārthagāthā verses are drawn from early Buddhist texts, have they been quoted verbatim or have they been changed? Do they have any special features? These questions have been investigated in chapters three and four, and are further developed in § 6.1.2 below. What is the place or status of the Śarīrārthagāthā in the development of the Yogacārabhūmi? Why is the uddāna treated differently in the Tibetan version compared to the Chinese and Sanskrit versions? Why were some verses in early Buddhist texts selected for inclusion in the Śarīrārthagāthā, while others were not selected? Possible answers to these questions have been explored in chapter five and have summarised in § 6.1.3 below.
Briefly, this study can be fundamentally summarised into the following three significant aspects.

6.1.1. Relationship between the two aṅgas of geya and vyākaraṇa and the
Śarīrārthagāthā verses

In this study, it was shown that the Śarīrārthagāthā verses can be divided into Groups A and B. Group A verses were cited from the Eight Assemblies section of the Saṃyuktāgama while Group B verses are from the Udānavarga, Pārāyaṇa, Arthavargīya and Madhyamāgama. As stated in chapter two, Group A verses are attributed to the geya
 
aṅga on the basis of one of the meanings of geya, namely, “collection/gathering” (jiéjí 結集 ), while Group B verses are attributed to the vyākaraṇa aṅga, particularly due to one of the meanings of vyākaraṇa “detailed explanation.” This vyākaraṇa meaning is “to elucidate the content of geya verses.” This also suits the literal meaning of the Group B title, upadeśa “specification.” In other words, Group A and B of the Śarīrārthagāthā verses are geya, but of different types. One type is like the verses in the Eight
Assemblies section of the Saṃyuktāgama. Another was sung popularly and then later collected in texts such as the Udānavarga, Pārāyaṇa, and Arthavargīya. These two aṅgas are also supposedly the reason why the author(s) of the Yogacārabhūmi chose these verses for inclusion in the Śarīrārthagāthā.
Having analysed the verses, the title Śarīrārthagāthā is more comprehensible. The Śarīrārthagāthā is regarded as “verses on the meaning of ‘relics’, or ‘corpus’ or ‘collection’.” The key word śarīra (literal meaning “body”) has been explained as relics (the Buddha’s teaching), corpus (collecting the verses), and collection (geya verses).
Importantly, the meaning of śarīra from √śri is also valued by Tibetan translator(s).841

6.1.2. Distinctive characteristics of Śarīrārthagāthā verses

Analyses of the Śarīrārthagāthā verses was presented in chapters three and four.
Their distinctive characteristics may be divided into three aspects, as follows:


1. Unique term and description


The Śarīrārthagāthā verse sets A-4, A-14, B-9, B-12, B-15, B-19 and B-22842 contain unique terms not found in early parallels. Unique phrases (one or two pādas) also exist in A-6 and A-12, which demonstrates that these Śarīrārthagāthā verses are cited from a different recension.




841 This is why the Tibetan translator(s) chose 'dus pa (aggregation, collection) as the translation.
842 For sake of intelligibility, A-4 is an abbreviation of “the fourth verse set of Group A”, for example. A and B mean Group A and B in the Śarīrārthagāthā. The number refers to the verse set in each group, which were provided in chapters three and four.
 
2. Omitted material and different sequences of verses


A-13, B-26 and B-27 omit material which is found in parallels belonging to other texts.843 Theoretically speaking, it is likely that these represent mistakes which occurred in the transcription or textual transmission.844 Moreover, the sequences of the verses in
B-6, B-10, and B-25 are somehow different from their parallels. However, in the cases of B-6 and B-10, Mūlasarvāstivādin text(s) can be found in which the order of verses are the same as that in B-6 and B-10. A corresponding Mūlasarvāstivādin text cannot be found for B-25. Overall, Mūlasarvāstivādin texts seem to share the same arrangement of the verses.

3. Distinctive characteristics of the Chinese version


A-14, B-9, B17, and B-22 are obvious cases which show the distinctive features of the Chinese Śarīrārthagāthā. The Chinese version of A-14 is similar to the Mūlasarvāstivādin version, while the Sanskrit and Tibetan versions, on the contrary, are close to the Pāli texts (Theravāda school). The other three cases (B-9, B-17, and B-22) demonstrate that the Sarvāstivādin heritage has been left in the Chinese version.845 The Sanskrit and Tibetan versions in these three cases are relevant to the Mūlasarvāstivāda school.846
If the Saṃyuktāgama (SĀ-G) is a Mūlasarvāstivādin text, the Śarīrārthagāthā verses as a whole are very close to the verses belonging to the Mūlasarvāstivādins. Based on this hypothesis, one possibility to explain the distinctive characteristics of the Chinese version can be found in Enomoto 1989: 22–23. Enomoto analysed four verses from different verse sets by comparing them with at least one Pāli parallel and at least one Mūlasarvāstivādin parallel. For example, in the sixth set of Group A, he compared the first and third pādas with counterparts at SN 2.2.4, SĀ-G 597, and SĀ-U 182. The

843 Enomoto (1989: 23) claimed that one verse in B-27 was somehow replaced by a foreign verse from Uv- B 3.10cd and 10.13ab.
844 Detail explanations may be found in chapters three and four.
845 B-9 and B-22 in the Chinese version are also close to Pāli and other texts. See § 4.1.9 and 4.1.22 for more details.
846 B-17 in the Sanskrit and Tibetan versions is also close to Pāli and other texts. See § 4.1.17 for more details.
 
descriptions in SN and SĀ-G are similar to each other, but different from that in the Śarīrārthagāthā. Thus, he suggested that the Śarīrārthagāthā was possibly a late Mūlasarvāstivādin recension which seems to have been changed after citation from Mūlasarvāstivādin sources. However, this explanation is not satisfactory in all instances. For instance, his suggestion cannot explain the case in which the Chinese Śarīrārthagāthā reproduces terms from a Sarvāstivādin text as well as some Pāli texts (instead of Mūlasarvāstivādin text(s)).847 These terms appear in early texts. The Chinese Śarīrārthagāthā is also earlier than the Sanskrit and Tibetan versions.848 Enomoto’s theory of a later Mūlasarvāstivādin recension is not consistent with this example.
Similarly, unique terms appear in the Śarīrārthagāthā. For example, the term “yoke” (yoktra) appears in A-4, but not in SN, SĀ or any parallels. The description in SN is different from that in SĀ. Is the occurrence of unique terms or phrases the reason that Enomoto argued that the Śarīrārthagāthā represents a later Mūlasarvāstivādin recension? Also, the school of SĀ-G still remains an open question.
Therefore, I would like to propose another possibility: the Śarīrārthagāthā represents a Mūlasarvāstivādin recension which was derived from “proto”-Sarvāstivāda and other saṃghas, including “Sthavira” (Skilling 1997: 100).849 This suggestion is based on the third model of the two schools’ development made by Skilling (1997: 100), as shown in the following chart:850


847 In B-9 and B-22, the terms can also be found in Dhp 95 and Th 855, two Pāli texts.
848 The Chinese version dates to around the first half of the seventh century, which is earlier than the Tibetan version (the end of the eighth century) and the earliest manuscript witness of the Sanskrit versions (around eleventh century). See § 1.6.
849 The relationship between the two schools (Sarvāstivāda and Mūlasarvāstivāda) is still controversial. The majority opinion is that they are two independent schools (Bareau 1955: 131–152, Heinz Bechert 1955: 153–154, Frauwallner 1956, Warder 1980: 393–394, Skilling 1997: 96–105) which I basically agree with.
The opposite idea is that of Yìnshùn (1971: 35), Lamotte (1976: 178) and Enomoto (2000: 247–248). 850 The first two models are (a) a straightforwad linear descent of the Mūlasarvāstivāda from the Sarvāstivāda; (b) the Mūlasarvāstivāda and the Sarvāstivāda share a common source, the proto- (Mūla)Sarvāstivāda. However, Skillings (1997: 101) mentioned that they cannot provide a satisfactory explanation to some variations occurring between texts.
 
This Mūlasarvāstivādin recension was developed from two sources: the proto- Sarvāstivāda and north Indian saṃghas, including the early Sthavira vinaya lineage.851 This model not only explains the historical situation, but also clarifies the various issues found in the Śarīrārthagāthā, especially the heritage from early Sarvāstivādin texts left in the Chinese Śarīrārthagāthā and the unique terms or phrases which may have been influenced by north Indian saṃghas. The differences between the Chinese and the Sanskrit/Tibetan versions might also reveal the process of textual transition in content. According to Deleanu (2006: 55–57, 76, 106–107), the Chinese version is the earliest, the Tibetan ranks second, and then the Sanskrit is thought to be the most recent. In the case of Group B, for example, the Sanskrit version is closest to the (Mūlasarvāstivādin) Uv-S; however, in the case of at least three occasions of the Chinese Śarīrārthagāthā,852 the wordings match that of the (Sarvāstivādin) Uv-B. In other words, before being converted into a Mūlasarvāstivādin text, vestiges of “proto-Sarvāstivāda”853 were left in the Chinese and Tibetan versions, which is the circumstance of the textual transition just as shown in the above chart.

6.1.3. Textual development of the Śarīrārthagāthā

Three aspects pertaining to the development of the Śarīrārthagāthā were discussed in chapter five. Firstly, as stated in chapter five, the Śarīrārthagāthā was supposedly an early textual unit formed during an early stage of the development of the Yogacārabhūmi, perhaps around the same time as the Śrāvakabhūmi. Secondly, the Tibetan uddāna represents a modification which occurred in Tibet. The Tibetan uddāna appears to be based on the Tibetan translation, so it is not the effort of the translators, Jinamitra, Yes shes sde, etc. Lastly, the (Mūla)sarvāstivādin tradition might have had a great influence on the principles of verse selection. Within this tradition, there existed the belief that verses could exaggerate the real state of the Buddha or do not necessarily correspond to the doctrine. This might be the reason why all the Śarīrārthagāthā verses are only



851 A detail explanation can be seen in Skilling 1997: 100–105.
852 Yet one occasion in the Tibetan version.
853 The terms/phrases of the (Sarvāstivādin) Uv-B reproduced in the Chinese and Tibetan versions may be the vestiges of “proto-Sarvāstivāda”.
 
relevant to the Śrāvakayāna practices and ideas (the Buddha’s teaching).854 Some verses may also have been selected on the basis of their popularity in an attempt to bring more people in contact with Buddhism.

6.2. Contribution

Each study has its own value and makes its own contribution. The following points describe the accomplishments of this study.

1. First study focusing on the analysis of the Śarīrārthagāthā


As shown in the introductory chapter, most studies concerning the Śarīrārthagāthā focus on the sources of the Śarīrārthagāthā verses. Other studies, which focus on other topics, passingly mention the Śarīrārthagāthā, such as the literal meaning of the title. Therefore, previous studies have not comprehensively investigated the Śarīrārthagāthā. The present thesis is the first study which has brought the relevant separate studies into a coherent whole and comprehensively researched the Śarīrārthagāthā.

2. Solving a long outstanding issue of why there are a variety of verse types in the
Śarīrārthagāthā


After Schmithausen (1970) and Enomoto (1989), who provided convincing information about the school association and sources of the Śarīrārthagāthā verses, it seems that no further research has been done on this particular topic. Most scholars mention the Śarīrārthagāthā as a Yogacārabhūmi section citing various canonical verses or which contains verses possibly belonging to the Mūlasarvāstivādins.855 In this study, knowledge on the historical development of the geya and vyākaraṇa aṅgas has been applied to the Śarīrārthagāthā verses. Through this, I suggest that the geya and vyākaraṇa aṅgas are key elements which explain why the author(s) collected them in the Śarīrārthagāthā.
854 See § 5.3 for details.
855 Such as Kragh (2013: 96), who introduced the Śarīrārthagāthā as a canonical passage by using the studies of Schmithausen (1970) and Enomoto (1989).
 
3. The importance of the three aṅgas


This study provides further evidence to support the three aṅgas (sūtra, geya, vyākaraṇa) theory.856 Yìnshùn (1971, 1983) discovered that the three aṅgas of sūtra, geya, and vyākaraṇa are the foundation of the Saṃyuktāgama. This discovery was based upon evidence mostly from the Yogācārabhūmi, so indicating that these three aṅgas were valued by the author(s) of the Yogācārabhūmi also. However, only the sūtra aṅga section in the Yogācārabhūmi had been identified in the studies of Yìnshùn (1971, 1983) and Mukai (1985). That is, the Vastusaṃgrahaṇī (the last chapter of the Yogācārabhūmi) is the mātr̥kā of the sūtra aṅga section of the Saṃyuktāgama. The geya and vyākaraṇa aṅgas had not been identified in the Yogācārabhūmi prior to this study. Importantly, by applying the three aṅgas theory of Yìnshùn (1971, 1983), the Śarīrārthagāthā verses have been identified in this study as belonging to the geya and vyākaraṇa aṅgas.
Therefore, the evidence not only solidifies the importance of the three aṅgas theory, but also supports Yìnshùn’s ideas that the three aṅgas can be considered as a fundamental organisational principal of early Buddhist texts, and that each aṅga can represent an actual collection, not just a literary style.

4. Providing effective explanations to the issue of the title Śarīrārthagāthā


Due to the fact that the title Śarīrārthagāthā is not explained in the Yogacārabhūmi, most studies provide the literal meaning of the Śarīrārthagāthā, which has still left confusion regarding the connection with the Śarīrārthagāthā verses.857 The Chinese and Japanese scholars in particular, who did not obtain the Sanskrit counterpart for comparison, have not provided particularly accurate or convincing explanations. By analysing the key word śarīra and connecting it with the Śarīrārthagāthā verses and geya aṅga, this study has provided effective explanations to the issue of the title Śarīrārthagāthā.





856 Some scholars, such as Anālayo (2011: 697 note 69), still express doubt on this theory.
857 See § 1.1.3 for further detail.
 
5. Convincing evidence regarding the issue of the Tibetan uddāna


There are some differences between the Chinese/Sanskrit and Tibetan versions of the Yogacārabhūmi, such as its structure, and the name(s) of some uddānas.858 However, previous explanations to these issues are basically hypothetical, such as the five explanations of Deleanu (2006: 152–153).859 In this study, significant evidence was provided which suggests that the Tibetan uddāna represents a modification after the translation was completed.

6.3. Future investigation

The following three aspects concerning the Śarīrārthagāthā commentary are recommended for future research.

1. Trilingual edition of the Śarīrārthagāthā commentary


As mentioned in chapter one, the Sanskrit commentary part of the Śarīrārthagāthā is not yet available. Although a tentative Sanskrit edition of the whole commentary has been made by Enomoto, he is reticent to publish it because many sections of the Yogācārabhūmi manuscript are difficult to read. However, a Japanese scholar, Takako Abe (personal communication, 9 September 2016), has been editing the Sanskrit commentary using not only Sāṅkṛityāyana’s photographs but also Tucci’s photographs of the two Sanskrit manuscripts.860 Because of its enourmous size, preparing an edition of the commentary is a huge task for one scholar. If the resources of Sāṅkṛityāyana’s and Tucci’s photographs can be shared with the public, an edition of the Sanskrit commentary could be completed more quickly by a team of editors. I have been editing the Chinese and Tibetan commentary, but this task is still ongoing due to the enormous




858 See § 5.2 for further detail.
859 See § 5.2 for further detail.
860 Enomoto did not obtain Tucci’s photographs for the Sanskrit edition. Delhey (2013: 505) mentioned that Tucci's collection is still not available to the public. Takako Abe acquired electronic copies of Tucci’s photographs privately from Delhey, who obtained them from Francesco Sferra in August 2009.
 
content. It is desirable that a trilingual edition (Chinese, Sanskrit and Tibetan) of the
Śarīrārthagāthā commentary be edited in the near future.


2. Analysis of the verse commentary


After the trilingual edition of the Śarīrārthagāthā commentary has been completed, some aspects of the verse commentary can be analysed, such as the special features of the commentary on these verses, whether there are particular doctrinal philosophies embedded in the commentary and so on.

3. Comparison of the verse commentary


Finally, there are some interpretations or commentaries on the verses from which the Śarīrārthagāthā was cited. Differences or developments of thoughts could be investigated by comparing these commentaries and interpretations. Through this, the distinguishing characteristics of the Śarīrārthagāthā commentary might also be clarified.
 
Table 2.1 Geya function

(R: respond, S: summarize, N: not summarize or respond, X: no verse)

SĀ-U Function of
verse SĀ-G Function of
verse Saṃyutta of SN
001 S 1062 S Bhikkhu
002 S 1063 S Bhikkhu
003 S 1064 S Bhikkhu
004 S 1065 S Bhikkhu
005 S 1067 S Bhikkhu
006 S 0275 S Bhikkhu
007 S 1068 S Bhikkhu
008 S 1069 S Bhikkhu
009 R 1070 R Bhikkhu
010 S 1071 S Bhikkhu
011 R 1072 R Bhikkhu
012 S 1073 S Bhikkhu
013 N 1074 N Bhikkhu
014 R 1075 R Bhikkhu
015 R 1076 R Bhikkhu
016 S 1077 S Bhikkhu
017 R 1078 R Bhikkhu
018 S 1079 S Bhikkhu
019 S 1080 S Bhikkhu
020 S 1081 S Bhikkhu
021 S 1082 S Bhikkhu
022 S 1083 S Bhikkhu
023 S 1084 S Māra
024 R 1085 R Māra
025 R 1086 R Māra
026 R 1087 R Māra
 
027 R 1088 R Māra
028 R 1089 R Māra
029 R 1090 R Māra
030 R 1091 R Māra
031 R 1092 R Māra
032 X 1093 R Māra
033 S 1104 S Sakka
034 S 1105 S Sakka
035 X 1106 S Sakka
036 S 1107 S Sakka
037 R 1108 R Sakka
038 R 1109 R Sakka
039 R 1110 R Sakka
040 R 1111 R Sakka
041 S 1112 S Sakka
042 R 1113 R Sakka
043 R 1114 R Sakka
044 R 1115 R Sakka
045 R 1116 R Sakka
046 S 1117 S Sakka
047 S 1118 S Sakka
048 R 1120 R Sakka
049 S 1222 N Sakka
050 R 1119 R Sakka
051 R 1223 R Sakka
052 R 1224
1225 R Sakka
053 S 1226 S Kosala
054 S 1227 S Kosala
055 S 1228 S Kosala
056 S 1229 S Kosala
 
057 S 1231 S Kosala
058 S 1230 S Kosala
059 S 1232 S Kosala
060 S 1233 S Kosala
061 R 1234 R Kosala
062 R 1235 R Kosala
063 R 1236 R Kosala
064 R 1237 R Kosala
065 S 1238 S Kosala
066 S 1239
0882 S Kosala
067 S 1240 S Kosala
068 S 1145 S Kosala
069 S 1146 S Kosala
070 S 1147 S Kosala
071 S 1148 S Kosala
072 R 1149 R Kosala
073 R 1150 R Kosala
074 R 1151 R Brāhmaṇa
075 R 1152 R Brāhmaṇa
076 R 1153 R Brāhmaṇa
077 R 1154 R Brāhmaṇa
078 R 1155 R Brāhmaṇa
079 S 1156 S Brāhmaṇa
080 R 1157 R Brāhmaṇa
081 R 1158 R Brāhmaṇa
082 R 1159 R Brāhmaṇa
083 R 1160 R Brāhmaṇa
084 N 1161 N Brāhmaṇa
085 S 1162 S Brāhmaṇa
086 S 1163 S Brāhmaṇa
 
087 S 1163 S Brāhmaṇa
088 S 0088 S Brāhmaṇa
089 S 0089 S Brāhmaṇa
090 S 0090 S Brāhmaṇa
091 S 0091 S Brāhmaṇa
092 S 1178 S Brāhmaṇa
093 R 1179 R Brāhmaṇa
094 R 1180 R Brāhmaṇa
095 R 1181 R Brāhmaṇa
096 R 1182 R Brāhmaṇa
097 R 1183 R Brāhmaṇa
098 R 1185 R Brāhmaṇa
099 R 1184 R Brāhmaṇa
100 N 1186 N Brāhmaṇa
101 S 1188 S Brahmā
102 S 1189 S Brahmā
103 N 1190 N Brahmā
104 R 1191 R Brahmā
105 R 1192 R Brahmā
106 R 1193 R Brahmā
107 R 1194 R Brahmā
108 R 1195 R Brahmā
109 S 1196 S Brahmā
110 S 1197
0979 S Brahmā
132 N 0995 N Devatā
133 N 0996 N Devatā
134 N 0997 N Devatā
135 N 0998 N Devatā
136 S 0999 S Devatā
137 N 1000 N Devatā
 
138 N 1001 N Devatā
139 N Devatā
140 N 1312
1002 N Devatā
141 N 1003 N Devatā
142 N 1004 N Devatā
161 N 0576 N Devatā
162 N 0577 N Devatā
163 N 0578 N Devatā
164 N 0579 N Devatā
165 N 0580 N Devatā
166 N 0581
0582 N Devatā
167 N 0583 N Devatā
168 N 0584 N Devatā
169 R 0585 R Devatā
170 N 0586 N Devatā
171 N 0587 N Devatā
172 N 0588 N Devatā
173 N 0599 N Devatā
174 N 0600 N Devatā
175 N 0598 N Devatā
176 N 0601 N Devatā
177 N 0602 N Devatā
178 N 1269
1316 N Devatā
179 N 1268 N Devatā
180 N 1267 N Devatā
181 N 0596 N Devatā
182 N 0597 N Devatā
183 N 0589 N Devatā
 
184 R 0590 R Devatā
185 R 0591 R Devatā
186 R 0592 R Devatā
187 R 0593 R Devatā
188 S 0594 S Devatā
189 N 0595 N Devatā
214 R 1198 R Bhikkhunī
215 R 1199 R Bhikkhunī
216 R 1200 R Bhikkhunī
217 R 1201 R Bhikkhunī
218 R 1202 R Bhikkhunī
219 R 1203 R Bhikkhunī
220 R 1204 R Bhikkhunī
221 R 1205 R Bhikkhunī
222 R 1206 R Bhikkhunī
223 R 1207 R Bhikkhunī
224 R 1208 R Vaṅgīsa
225 R 1209 R Vaṅgīsa
226 R 1210 R Vaṅgīsa
227 R 1211 R Vaṅgīsa
228 S 1212 S Vaṅgīsa
229 R 1213 R Vaṅgīsa
230 R 1214 R Vaṅgīsa
231 N 1005 N Devatā
232 N 1006 N Devatā
233 N 1007 N Devatā
234 N 1008 N Devatā
235 N 1008 N Devatā
236 N 1009 N Devatā
237 N 1010 N Devatā
238 N 1011 N Devatā
 
239 N 1012 N Devatā
240 N 1013 N Devatā
241 N 1014 N Devatā
242 N 1015 N Devatā
243 N 1016 N Devatā
244 N 1017 N Devatā
245 N 1018 N Devatā
246 N 1019 N Devatā
247 N 1020 N Devatā
248 N 1021 N Devatā
249 N 1022 N Devatā
250 R 1215 R Vaṅgīsa
251 R 1216 R Vaṅgīsa
252 R 1217 R Vaṅgīsa
253 R 1218 R Vaṅgīsa
254 R 1220 R Vaṅgīsa
255 R 1221 R Vaṅgīsa
256 R 0993 R Vaṅgīsa
257 R 0994 R Vaṅgīsa
258 R 0092 R Brāhmaṇa
259 S 0093 S Brāhmaṇa
260 S 0094 S Brāhmaṇa
261 R 0095 R Brāhmaṇa
262 R 0096 R Brāhmaṇa
263 R 0097 R Brāhmaṇa
264 R 0098 R Brāhmaṇa
265 R 0099 R Brāhmaṇa
266 R 0100 R Brāhmaṇa
267 R 0101 R Brāhmaṇa
268 R 0102 R Brāhmaṇa
269 N 1270 N Devatā
 
270 N 1271 N Devatā
271 N 1273 N Devatā
272 N 1274 N Devatā
273 N 1275 N Devatā
274 N 1276 N Devatā
275 N 1277 N Devatā
276 R 1278 R Devatā
277 N 1279 N Devatā
278 N 1280 N Devatā
279 N 1281 N Devatā
280 N 1282 N Devatā
281 N 1283 N Devatā
282 R 1284 R Devatā
283 N 1285 N Devatā
284 N 1286 N Devatā
285 N 1287 N Devatā
286 N 1288 N Devatā
287 R 1289 R Devatā
288 N 1290 N Devatā
289 N 1291 N Devatā
290 N 1292 N Devatā
291 N 1294 N Devatā
292 N 1293 N Devatā
293 N 1295 N Devatā
294 N 1296 N Devatā
295 N 1297 N Devatā
296 N 1298 N Devatā
297 N 1299 N Devatā
298 N 1300 N Devaputta
299 N 0577 N Devaputta
300 N 1301 N Devaputta
 
301 N 1302 N Devaputta
302 N 1303 N Devaputta
303 N 1304 N Devaputta
304 N 1305 N Devaputta
305 S 1306 S Devaputta
306 S 1307 S Devaputta
307 N 1308 N Devaputta
308 N 1309 N Devaputta
309 N 1310 N Devaputta
310 N 1311 N Devaputta
311 N 1312 N Devaputta
312 R 1313 R Devaputta
313 N 1314 N Devaputta
314 N 1315 N Devaputta
315 N 1316 N Devaputta
316 R 1317 R Devaputta
317 R 1317 R Devaputta
318 N 1319 N Yakkha
319 R 1320 R Yakkha
320 R 1321 R Yakkha
321 R 1322 R Yakkha
322 S 1323 S Yakkha
323 R 1324 R Yakkha
324 N 1325 N Yakkha
325 N 1326 N Yakkha
326 R 1328 R Yakkha
327 R 1327 R Yakkha
328 R 1329 R Yakkha
329 R 1130 X Yakkha
351 R 1331 R Vana
352 R 1332 R Vana
 
353 R 1333 R Vana
354 R 1334 R Vana
355 R 1335 R Vana
356 N 1336 N Vana
357 R 1337 R Vana
358 R 1338 R Vana
359 R 1339 R Vana
360 R 1340 R Vana
361 R 1341 R Vana
362 R 1342 R Vana
363 R 1343 R Vana
364 R 1344 R Vana
 
Table 2.2 Geya function by Saṃyutta


Saṃyutta of
SN SĀ-U Function SĀ-G Function
Bhikkhu 1-22 S(20), R(2) 1062-1083,
0275 S(20), R(2)
Māra 23-32 R(9), X(1) 1084-1093 R(10).
Sakka 33-52 R(12), S(7), X(1) 1104-1225 R(14), S(7).
Kosala 53-73 R(6), S(15) 0882, 1126-
1240, 1145-
1150 R(6), S(16).
Brāhmaṇa 74-100, 258-
268 R(25), S(11), N(2) 1151-1163,
088-091, 1178-
1186
0092-0102 R(25), S(11), N(2)
Brahmā 101-110 R(5), S(4), N(1) 1188-1197,
0979 R(5), S(5), N(1)
Bhikkhunī 214-223 R(10) 1198-1207 R(10)
Vaṅgīsa 224-230,
250-257 R(14), S(1) 1208-1214,
1215-1221,
993-994 R(14), S(1)
Devatā 132-142,
161-189,
231-249,
269-297 R(5), S(2), N(33) 0995-1004,
576-0595,
0596-0602,
1267-1269,
1312, 1316 R(5), S(2), N(35)
Devaputta 298-317 R(3),
S(2),N(15) 1300-1317, 577 R(3),
S(2),N(15)
Yakkha 318-329 R(8), S(1), N(3), 1319-1330 R(7), S(1),
N(3), X(1)
Vana 351-364 R(13), N(1) 1331-1344 R(13), N(1)
 

Table 2.3. The findings of four scholars (Group A)

Title Ui (1958) Enomoto (1989) Zōngzhèng (2005) Chén (2006)
1 evil SĀ-G 1270 SĀ-G 1270-1274 SĀ-G 1270-1274 SĀ-G 1270-1274
2 what can be expressed SĀ-G 1078 SĀ-G 1078 SĀ-G 1078 SĀ-G 1078
3 lust SĀ-G 1214 SĀ-G 1214 SĀ-G 1214 SĀ-G 1214
4 flood SĀ-G 1092 SĀ-G 1092 SĀ-G 1092 SĀ-G 1092
5 being frightened SĀ-G 596 SĀ-G 596 SĀ-G 596 SĀ-G 596
6 caste SĀ-G 597 SĀ-G 597 SĀ-G 597+SĀ-G 1315 SĀ-G 597

7 obtaining fame
SĀ-G 1282
SĀ-G 1282
SĀ-G 1282 SĀ-G 1282
SĀ-G 1326

8 current SĀ-G 595
SĀ-G 1329 SĀ-G 601
SĀ-G 1329 SĀ-G 601
SĀ-G 1329 SĀ-G 601
SĀ-G 1329

9 two floods
SĀ-G 603 SĀ-G 603
SĀ-G 1326
SĀ-G 1329 SĀ-G 603
SĀ-G 1326
SĀ-G 1329

SĀ-G 1316
SĀ-G 1326
SĀ-G 1329

10
SĀ-G 1269
SĀ-G 1269
SĀ-G 1326
SĀ-G 1329 SĀ-G 1269
SĀ-G 1316
SĀ-G 1326
SĀ-G 1329

11 lust and hatred SĀ-G 1314
SĀ-G 1324 SĀ-G 1314
SĀ-G 1324 SĀ-G 1314
SĀ-G 1324 SĀ-G 1314
SĀ-G 1324
12 what ought to be done SĀ-G 1311 SĀ-G 1311 SĀ-G 1311 SĀ-G 1311
13 struggle SĀ-G 1267 SĀ-G 1267 SĀ-G 1267 SĀ-G 1267
14 attainment of the aim SĀ-G 549 SĀ-G 1092 SĀ-G 1092 SĀ-G 1092
 

Table 2.4 The correspondence between the verses of Group A and SĀ-G and SĀ-U

Title SĀ-G SĀ-U Assemblies from SĀ Saṃyutta of SN
1 è 惡 (pāpa)
SĀ-G 1270-1274 SĀ-U 269-
272 Cātummahārājika
( Zhūtiān 諸天) Devatā
2 shuō 說 (ākhyeya) SĀ-G 1078 SĀ-U 17 Samaṇa (Bǐqiū 比丘) Bhikkhu
3 tān 貪 (rāga) SĀ-G 1214 SĀ-U 230 Gahapati (Póqíshě 婆耆舍) Vaṅgīsa
4 liú 流 (ogha) SĀ-G 1092 SĀ-U 31 Māra (Mó 魔) Māra
5 bù 怖 (utrasta)
SĀ-G 596 SĀ-U 181 Cātummahārājika
( Zhūtiān 諸天) Devatā
6 lèi 類 (abhivarṇṇatā)
SĀ-G 597 SĀ-U 182 Cātummahārājika
( Zhūtiān 諸天) Devatā
7 yù 譽 (yaśasvin)
SĀ-G 1282 SĀ-U 280 Cātummahārājika
( Zhūtiān 諸天) Devatā
8 chí 池 (sara) SĀ-G 601
SĀ-G 1329 SĀ-U 176
SĀ-U 328 Cātummahārājika
( Zhūtiān 諸天 and yèchā 夜叉) Devatā Yakkha
9 liú 流 (ogha) SĀ-G 603
SĀ-G 1326
SĀ-G 1329 X
SĀ-U 325
SĀ-U 328 Cātummahārājika
(yèchā 夜叉)
Yakkha Yakkha
10 liú 流 (ogha) SĀ-G 1269 SĀ-U 178 Cātummahārājika Devatā
 

SĀ-G 1316
SĀ-G 1326
SĀ-G 1329 SĀ-U 325
SĀ-U 328 ( Zhūtiān 諸天 and yèchā 夜叉) Yakkha Yakkha
11 tān 貪 (rāgadveṣa) SĀ-G 1314
SĀ-G 1324 SĀ-U 313
SĀ-U 323 Cātummahārājika
( Zhūtiān 諸天 and yèchā 夜叉) Devaputta Yakkha
12 zuò 作 (kāryatā)
SĀ-G 1311 SĀ-U 310 Cātummahārājika
( Zhūtiān 諸天) Devaputta
13 qú láo 劬勞 (āyūhana)
SĀ-G 1267 SĀ-U 180 Cātummahārājika
( Zhūtiān 諸天) Devaputta
14 dé yì 得義 (arthaprāpti) SĀ-G 1092 SĀ-U 31 Māra (Mó 魔) Māra
 

Table 2.5 The findings of four scholars (Group B)

heading Ui (1958) Enomoto (1989) Zōngzhèng (2005) Chén (2006)

1 impermanence FJ 1.2 Uv-S 1.3 SĀ-G 576
SĀ-G 956
SĀ-G 1197 FJ 1.2
2 carefulness Dhp 2.1 Uv-S 4.1 SĀ-G 996
SĀ-G 1289 FJ 10.1
Dhp 2.1

3 strong lust Dhp 24.6 Uv-S 3.1 FJ 32.18
FJP 32.2
CY 3.1
FJY 3.1 Dhp 24.6

4 dwelling in the
dharma Dhp 16.9 Uv-S 5.24 FJ 24.7
FJP 24.4
CY 6.21
FJY 5.21 Dhp 16.9
5 contemplating the
truth X Uv-S 9.6 X CY 29.11

6 good speech SĀ-G 45 Uv-S 8.11 FJ 38.14
CY 9.13
FJY 8.13 CY 9.13
FJY 8.13
 


7 faith FJ 4.1 Uv-S 10.1 FJ 4.1
CY 11.1
FJY 10.1 FJ 4.1
FJP 4.2
CY 11.1

8 learning FJ 3.4 Uv-S 12.6 FJ 3.4
FJP 3.4
FJY 22.15 FJ 3.4
FJP 2.1, 3.1
CY 23.1

9 being like the sky close to Dhp 7.6 Uv-S 17.12 FJ 15.6, 36.24
CY 18.12
FJY 17.13 Dhp 7.4, 7.6
10 physical
appearance X Uv-S 22.12-16 CY 23.11-15
FJY 22.14, 22.16-18 CY 23.13-15
11 king CY 17.22 Uv-S 16.22 CY 17.22
FJY 16.19 CY 17.22

12 city of bone Dhp 11.5 Uv-S 16.23 FJ 19.5
CY 17.23
FJY 16. 20 Dhp 11.5
CY 17.23
13 turtle Dhp 11.12 Uv-S 26.1 CY 27.1
FJY 26.1 CY 27.1
FJY 26.1
14 equality DĀ 2 Uv-S 26.30 DĀ 2
CY 27.28 DĀ 2
 

FJY 26.35
15 mud close to Dhp 18.17 Uv-S 29.37 X Dhp 18.17
16 two dwellings Dhp 18.20 Uv-S 29.38 FJ 26.18
FJY 29.34 Dhp 18.20
17 Udāna 7.7 Uv-S 29.51 CY 30.40
FJY 29.45 Udāna 7.7
18 having cleaned up Udāna 6.7 Uv-S 29.56 CY 30.43
FJY 29.48 Udāna 6.7
19 giving Udāna 8.5 Uv-S 28.2 CY 29.2
FJY 28.2 Udāna 8.5
CY 29.2

20 all evils Dhp 14.5 Uv-S 28.1 FJ 22.11
CY 29.1
FJY 28.1 Dhp 14.5
FJ 22.11
CY 29.1
FJY 28.1

21 falling down Dhp 3.3 Uv-S 31.1 FJ 11.2
CY 32.1
FJY 31.1 Dhp 3.3
22 skilled Theragāthā 84 Uv-S 31.51 CY 32.41
FJY 31.40 Theragāthā 85
 


23 craft Udāna 3.9 Uv-S 32.5 Udāna 3.9

24 travelling far away Dhp 3.5 Uv-S 33.55 FJ 11.4
CY 34.57
FJY 33.55 Dhp 3.5
CY 34.57
25 obscuring Sn 1032-1039 Pārāyaṇeṣv
Ajitapraśnaḥ Sn 1032-1039
Sn 1110-1111 Sn 1032-1039

26 sensual pleasures Sn 766-771
Chinese Arthavargīyasūtra (T198) Arthavargīyeṣu kāmān ārabhya Sn 766-771
Chinese Arthavargīyasūtra
(T198) Sn 766-771
Chinese Arthavargīyasūtra
(T198)
27 auspicious SĀ-G 955 MĀ 165-167 SĀ-G 955
close to SĀ-G 1072 SĀ-G 955
 
Bibliography

I. Primary Sources
A Thousand Books of Wisdom. CD-ROM Release IV. 1998. New York: The Asian Classics Input Project.
Āpídámó dàpípóshā lùn 阿毗達磨大毗婆沙論 (Abhidharma Mahāvibhāṣā-śastra). Taishō xīnxiū dàzōkyō 大正新修大藏經, Vol. 27, No 1545. Taipei: Xīn wénfēng 新文豐.
Abhidharmakośabhāśya. P. Pradhan, ed. 1975. Patna: K.P. Jayaswal.
Abhidharmakośavyākhyā. U. Wogihara, ed. 1932–1936. Tōkyō: Publishing Association of Abhidharmakosavyākhyā.
Aṅguttaranikāya. R. Morris & E. Hardy, eds. 1885–1900. London: Pali Text Society.
Apadāna. M. E. Lilley, ed. 1925–1927. London: Pali Text Society.
Chéngshí lùn 成實論 (Satyasiddhi-śāstra). Taishō xīnxiū dàzōkyō 大正新修大藏經, Vol.
32, No 1646. Taipei: Xīn wénfēng 新文豐.
Cintamayī bhūmiḥ, extant Sanskrit fragments are edited and translated in: Shōmon ji Kenkyūkai 声 聞 地 研 究 会 ed. and tr., 1993. “Bonbun Shōmon ji (12): Hon ji bun chū Monshojō ji (1), Shishojō ji, wayaku, kamon” 梵文声聞地 (十二): 本地分中聞所成地 (2)・思所成地, 和訳・科文. Taishō daigaku sōgō bukkyō kenkyūjo nenpō 大正大学綜合佛教研究所研究叢書 15: 1–49.
Derge Tibetan Tripiṭaka (bsTan-hgyur division) デルゲ版チベット大蔵経 論疎部.
Takasaki Jikidō 高崎直道, Zuihō Yamaguchi 山口瑞鳳 and Noriaki Hakamaya 袴谷憲昭, eds. 1980. Tokyo: the Faculty of Letters, University of Tokyo.
Dhammapada. Oskar Von Hinüber and Kenneth Roy Norman, eds. 1994. London: Pali Text Society.
The Facsimile Edition of the “Śrāvakabhūmi” Sanskrit Palm-leaf Manuscript 瑜伽師地論聲聞地梵文原文影印本. Joint Publication of Taishō University (Tokyo) and China Library of Nationalities (Beijing), ed. 1994. Beijing: Minzu chubanshe.
 
Dga'-ldan bstan 'gyur (Golden Manuscript Tanjur). Facsimile reproductions (CD-ROM).
New York: Tibetan Buddhist Resource Center.
Fangshan Stone-carved Collection 房山石經. Facsimile reproduction: 趙樸初, editor-in- chief. 1988. 22 vols. Beijing 北京: Zhōngguó Fójiào Túshū Wénwù Guǎn 中國佛教圖書文物館.
Fóshuō yìzú jīng 佛說義足經 (Arthapada Sūtra). Taishō xīnxiū dàzōkyō 大正新修大藏經, Vol. 4, No 198. Taipei: Xīn wénfēng 新文豐.
Kokuyaku daizōkyō 國譯大藏經 [[[Japanese]] Translation of the Canon]. 1986 [1927–1928].
31 vols. Tokyo: Kokumin Bunko Kankōkai.
Kokuyaku issaikyō: Indo senjutsubu 國譯一切經：インド撰述部 [[[Japanese]] Translations of the Scriptures: Works Composed in India]. 1926–1936. 156 vols. Tokyo: Daitō Shuppan.
Kokuyaku issaikyō: Wa-Kan senjutsubu 國 譯 一 切 經 ： 和 漢 撰 述 部 [[[Japanese]] Translations of the Scriptures: Works Composed in China & Japan]. 1936–1945.
66 vols. Tokyo: Daitō Shuppan.
Gāndhārī Dhammapada: edited with an introduction and commentary. J. Brough, ed.
1962. London: Oxford University Press.
Gēnběn shuōyīqiēyǒubù pínàiyē 根本說一切有部毗奈耶 (Mūlasarvāstivāda Vinaya).
Taishō xīnxiū dàzōkyō 大正新修大藏經, Vol. 24, No 1451. Taipei: Xīn wénfēng
新文豐.
Isibhāsiyāiṃ. W. Schubring, ed. 1969. Hamburg: Routledge Curzon Press.
Jātaka. V. Fausbøll, ed. 1877–1896. London: Pali Text Society.
Mahābhārata. V. S. Sukthankar et al., eds. 1933–1966. Poona: Bhandarkar Oriental Research Institute.
Mahāvastu. E. Senart, ed. 1882–1897. Paris: Imprimerie Nationale.
Majjhimanikāya. R. Chalmers & V. Trenckner, eds. 1977–1979. London: Pali Text Society.
Nettipakaraṇa. E. Hardy, ed. 1902. London: Pali Text Society.
 
Papañcasūdanī: Majjhimanikāyaṭṭhakathā of Buddhaghosācariya, James Haughton Woods (1864–1935, ed.), D. D. Kosambi (1907–1966, joint ed.), I. B. Horner (1896–1981, ed.) 1922–1938. 5 vols. London: Pali Text Society.
Peking Tibetan Tripiṭaka 北京版チベット大藏經. D. T. Suzuki 鈴木大拙, ed. 1955–1962.
168 vols. Kyoto: Tibetan Tripiṭaka Research Institute.
Peṭakopadeśa. A. Barua, ed. 1949. London: Pali Text Society.
Rnal 'yor spyod pa'i sa (Yogācārabhūmiśāstra), translated by Prajñāvarman, Jinamitra, Ye-Shes-Sde, et al. Derge Tibetan Tripiṭaka, Vol. Thsi, No 4035–4042. Tokyo: the Faculty of Letters, University of Tokyo.
Rnal 'yor spyod pa'i sa (Yogācārabhūmi). translated by Prajñāvarman, Jinamitra, Ye- Shes-Sde, et al. Peking Tibetan Tripiṭaka, Vol. Dsi, No 5536–5543. Kyoto: Tibetan Tripiṭaka Research Institute.
Saṃyuttanikāya. M. Léon Feer, ed. 1884–1898. 5 vols. London: Pali Text Society.
Saṃyuttanikāya. G. A. Somaratne, ed. 1998. 1 vol. London: Pali Text Society.
Samantapāsādikā: Buddhaghosa's commentary on the Vinaya Piṭaka. J. Takakusu and M. Nagai, eds. 1975. London: Pali Text Society.
Sāratthappakāsinī. F. L.Woodward, ed. 1929– 1937. London: Pali Text Society.
Shùnzhènglǐ lùn 順 正 理 論 (Abhidharma Nyāyānusāra-śāstra). Taishō xīnxiū dàzōkyō 大正新修大藏經, Vol. 29, No 1652. Taipei: Xīn wénfēng 新文豐.
Śrāvakabhūmi of Ācārya Asaṅga. Karunesha Shukla, ed. 1973. Patna: Kashi Prasad Jayaswal Research Institute.
Śrāvakabhūmi of Ācārya Asaṅga. Part Ⅱ. Karunesha Shukla, ed. 1991. Patna: Kashi Prasad Jayaswal Research Institute.
Sutta-nipāta. Dines Andersen and Helmer Smith, eds. 1913. London: Pali Text Society.
Taishō shinshū dai zōkyō 大正新脩大藏經 [Revised version of the canon, compiled during the Taishō era]. Takakusu Junjirō 高楠順次郎 and Watanabe Kaikyoku 渡辺海旭, et al, eds. 1991 [1922–1933]. 85 vols. Tokyo: Taishō Issaikyō Kankōkai.
Udāna. P. Steinthal, ed. 1885. London: Pali Text Society.
Udānavarga Ⅰ. Franz Bernhard, ed. 1965. Göttingen: Vandenhoek & Ruprecht.
 
Udānavarga Ⅱ. Franz Bernhard, ed. 1968. Göttingen: Vandenhoek & Ruprecht.
Yújiā shī dì lùn 瑜伽師地論 (Yogācārabhūmi). Taishō xīnxiū dàzōkyō 大正新修大藏經, Vol. 30, No 1579. Taipei: Xīn wénfēng 新文豐.
Yújiā lùn jì 瑜伽論記 (Notes on the Yogacārabhūmi). Taishō xīnxiū dàzōkyō 大正新修大藏經, Vol. 42, No 1828. Taipei: Xīn wénfēng 新文豐.
Yújiāshī dì lùn luèzuǎn 瑜伽師地論略纂 (A Summary Concise Compendium of the Yogacārabhūmi). Taishō xīnxiū dàzōkyō 大正新修大藏經, Vol. 43, No 1829. Taipei: Xīn wénfēng 新文豐.
The Yogācārabhūmi of Ācārya Asaṅga: The Sanskrit Text Compared with the Tibetan Version Vidhushekhara Bhattacharya, ed. 1957. Calcutta: University of Calcutta.
Zá āhán jīng 雜阿含經 (Saṃyuktāgama). Taishō xīnxiū dàzōkyō 大正新修大藏經, Vol. 2, No 99. Taipei: Xīn wénfēng 新文豐.
Biéyì zá āhánjīng 別譯雜阿含經 (Saṃyuktāgama). Taishō xīnxiū dàzōkyō 大正新修大藏經, Vol. 2, No 100. Taipei: Xīn wénfēng 新文豐.
zhǎng āhán jīng 長阿含經 (Ekottarikāgama). Taishō xīnxiū dàzōkyō 大正新修大藏經, Vol. 1, No 1. Taipei: Xīn wénfēng 新文豐.
Zhōng āhán jīng 中阿含經 (Madhyamāgama). Taishō xīnxiū dàzōkyō 大正新修大藏經, Vol. 1, No 26. Taipei: Xīn wénfēng 新文豐.

II. Dictionaries, Indexes, and Catalogues
Akanuma Chizen 赤沼智善.
1929/1990. Kanpa shibu shiagon goshōroku 漢 巴 四 部 四 阿 含 互 照 錄 [The Comparative Catalogue of Chinese Āgamas and Pāli Nikāyas]. Nagoya: Hajinkaku Shobō 破塵閣書房.
Anesaki Masaharu 姉崎正治.
1908. The Four Buddhist Āgamas in Chinese: A Concordance of their Parts and of the Corresponding Counterparts in the Pāli Nikāyas. Transactions of the Asiatic Society of Japan 35.3. Yokohama: Kelly & Walsh.
 
Hirakawa Akira 平川彰.
1997. Bukkyō kanbon daijiten 佛教漢梵大辭典 [[[Buddhist]] Chinese-Sanskrit Dictionary]. Tokyo: The Reiyukai.
Jeffrey Hopkins.
[Online] 2003. Tibetan-Sanskrit-English Dictionary. Available at the Tibetan and Himalayan Library, University of Virginia
Kunsang, Erik Pema, James Valby, Ives Waldo, and Jeffrey Hopkins.
2003. The Rangjung Yeshe Tibetan-English Dictionary of Buddhist Culture, Version 3 [CD-ROM]. Kathmandu: Rangjung Yeshe.
Lokesh Chandra.
1992–1994. Tibetan-Sanskrit Dictionary. 7 vols. New Delhi: International Academy of Indian Culture and Adiya Prakashan.
Monier Monier-Williams.
[1899] 1986. A Sanskrit-English Dictionary. Etymologcically and Philologically Arranged with Special Reference to Cognate Indo-European Languages. Tokyo: Meicho Fukyūkai.
Nakamura Hajime 中村元.
1981. Bukkyōgo daijiten 佛教語大辞典. Tokyo: Tōkyō shoseki 東京書籍.
Edgerton, Franklin.
1953. Buddhist Hybrid Sanskrit Dictionary and Grammar. 2 vols. New Haven: Yale University Press.
Otani Unversity supervised.
[1961] 1985. The Tibetan Tripitaka (Peking Edition): Catalogue and Index. 影印北京版 西藏大藏経総目録・索引. Kyoto: Rinsen shoten 臨川書店.
Sarat Chandra Das.
[1902] 1988. A Tibetan-English Dictionary with Sanskrit Synonyms. Kyoto: Rinsen shoten 臨川書店.
T. W. Rhys Davids & William Stede (eds.)
1921–5. The Pali Text Society’s Pali–English Dictionary. Chipstead: Pali Text Society.
 
Taishō Shinshū Daizōkyō Kankōkai 大正新脩大藏經刊行会 ed.
1975. Dàzhèng xīnyǒu dàcángjīng mùlù 大正新脩大藏經目録 [Comprehensive Catalogue of the Taishō Revised Canon]. Taipei: Xinwenfeng Chubanshe.
Rozenberg, Otton Ottenovich.
1916. Introduction to the Study of Buddhism According to Material Preserved in Japan and China, Part 1: Vocabulary. A survey of Buddhist Terms and Names Arranged According to Radicals with Japanese Reading and Sanskrit Equivalents. Supplemented by Addition of Terms and Names relating to Shinto and Japanese History. Tokyo: Faculty of Oriental Languages of the Imperial University of Petrograd.
Ui Hakuju 宇井伯壽 et al.
1934. Chibetto daizōkyō sōmokuroku 西藏大藏經總目錄 [A Complete Catalogue of The Tibetan Canons] (bKaḥ-ḥgyur and bsTan-ḥgyur). Sendai: Tōhoku Emperial University (Tōhoku teikoku daigaku hōbun gakubu 東北帝國大學法
文學部)

Ⅲ. Secondary works
Ahn, Sung-doo.
2003. Die lehre von den Kleśas in der Yogācārabhūmi. Stuttgart: Franz Steiner Verlag.
Akanuma Chizen 赤沼智善.
1939. Bukkyō kyōten shiron 佛教經典史論 [The History of Buddhist Literature].
Nagoya: Hajinkaku Shobō 破塵閣書房.
Anālayo, Bhikkhu.
2009. “Veyyākaraṇa.” In W.G. Weeraratne, ed., Encyclopaedia of Buddhism, 8.3: 549–554. Sri Lanka: Department of Buddhist Affairs, Ministry of Religious Affairs.
2011. A Comparative Study of the Majjhima-nikāya. (Dharma Drum Buddhist College Research Series 3.) Taipei: Dharma Drum Publishing Corporation.
 
2012. Majjhima-āgama studies. (Dharma Drum Buddhist College Research Series 5.) Taipei: Dharma Drum Publishing Corporation.
2014. Saṃyukta-āgama Studies. (Dharma Drum Institute of Liberal Arts Research Series 2.) Taipei: Dharma Drum Publishing Corporation.
Ānandajoti, Bhikkhu.
2004. A Comparative Edition of the Dhammapada.
Aramaki Noritoshi 荒牧典俊.
2000. “Toward an Understanding of the Vijñaptimātratā.” In Jonathan A. Silk, ed., Wisdom, Compassion, and the Search for Understanding: The Buddhist Studies Legacy of Gadjin M. Nagao, 39–60. Honolulu: University of Hawaii Press.
2013. “Two Notes on the Formation of the Text-Complex.” in Ulrich Timme Kragh, ed., The Foundation for Yoga practitioners: the Buddhist Yogācārabhūmi treatise and its adaptation in India, East Asia, and Tibet, 398–439. Cambridge, Mass: Department of South Asian studies, Harvard University.
Azami Noriaki 薊法明.
2007. “Yuishiki bunken niokeru kyōten no juyō: Yuga ron kangaeru sho naru chi o chuushin toshite” 唯識文献における経典の受容: 瑜伽論思所成地を中心として [On Āgama in the Yogācārabhūmi]. Indogaku bukkyōgaku kenkyū 印度學佛教學研究 [Journal of Indian and Buddhist Studies] 55.2: 950–946.
Bacot, Jacques.
1954. “Titres et colophons d'ouvrages non-canoniques tibétains: Textes et traduction.”
Bulletin de l'École Française d'Extrême- Orient 44.2: 275–337.
Bandurski, Frank.
1994. “Übersicht über die Göttinger Sammlungen der von Rāhula Sāṅkr̥tyāyana in Tibet aufgefundenen buddhististichen Sanskrite-Texte (Fundebuddhistischer Sanskrit-Handschriften, Ⅲ).” In Frank Bandurski et al., eds., Untersuchen zur buddhistischen Literatur, 9–126. (Sanskrit-Wörterbuch der buddhistischen
 
Texte aus den Turfan-Funden, Beiheft 5.) Göttingen: Vandenhoek & Ruprecht.
Bareau, André.
1955a. Les premiers conciles bouddhiques. Paris: Presses Universitaires de France. 1955b. Les sectes bouddhiques du Petit Véhicule. Saigon: École française d'Extrême-
Orient.
1966. “Le bouddhisme indien.” In André Bareau, Walter Schubring & Christoph von Furer-Haimendorf, Les Religions de l'Inde : III. Bouddhisme, Jaīnisme, religions archaïques, 13–246. Paris: Payot,
Bechert, Heinz.
1970. “Theravada Buddhist Sangha: Some General Observations on Historical and Political Factors in Its Development.” Journal of Asian Studies 29.4: 761–778.
1978. “Beginnings of Buddhist historiography: Mahavamsa and political thinking.” In Bardwell L. Smith, ed., Religion and Legitimating of Power in Sri Lanka, 1–
12. Chambersburg, Pennsylvania: Anima Books. Bingenheimer, Marcus.
2011. Studies in Āgama Literature– With special reference to the Shorter Chinese Saṃyuktāgama. Taipei: Xinwenfeng 新文豐.
Bodhi, Bhikkhu.
2000. The Connected Discourses of the Buddha: A Translation of the Samyutta Nikaya.
Boston: Wisdom Publications.
Bronkhorst, Johannes.
2009. Buddhist Teaching in India. Boston: Wisdom Publications.
Brough, John
2001. The Gandhari Dharmapada. Delhi: Motilal Banarsidass Publishers Private Limited.
Bucknell, Roderick S.
2007. “The Structure of the Sagātha-Vagga of the Saṃyutta-Nikāya.” Buddhist Studies Review 24.1: 7–34.
 
2008. “The Two Versions of the Other Translation of Saṃyuktāgama.” Chung-hwa Journal of Buddhist Studies 21: 23–54.
2011. “The Historical Relationship Between the Two Chinese Saṃyuktāgama Translations.” Chung-hwa Journal of Buddhist Studies 24: 35–70.
Chén Qióngcuǐ 陳瓊璀.
2006. “Yúgāshīdì lùn běndìfèn sīsuǒchéng dì tǐyìgātā jīng lùn duìzhào shàngpiān” 瑜伽師地論本地分思所成地體義伽他經論對照 上篇 [The Contrast between Śarīrārthagāthā and Śrāvakayāna Canonical Texts, Part 1]. Zhìlián wénhuà jíkān 志蓮文化集刊 [[[Wikipedia:Culture|Culture]] Journal of Zhilian] 2: 59–131.
2007. “Yúgāshīdì lùn běndìfèn sīsuǒchéng dì tǐyìgātā jīng lùn duìzhào xiàpiān” 瑜伽師地論本地分思所成地體義伽他經論對照 下篇 [The Contrast between Śarīrārthagāthā and Śrāvakayāna Canonical Texts, Part 2]. Zhìlián wénhuà jíkān 志蓮文化集刊 [[[Wikipedia:Culture|Culture]] Journal of Zhilian] 3: 37–132.
Chikusa Masa’aki 竺沙雅章.
1993. Kanyaku daizōkyō no rekishi: Shakyō kara kangyō he 漢訳大蔵経—写経から刊経へ. Kyoto: Ōtani Daigaku 大谷大学.
Choong Munkeat.
2000. The Fundamental Teachings of Early Buddhism: A Comparative Study Based on the Sūtrāṅga Portion of the Pāli Saṃyutta-nikāya and the Chinese Saṃyuktāgama. Beiträge zur Indologie, Band 32. Wiesbaden: Harrassowitz.
2007. “A Comparison of the Pali and Chinese Versions of the Vaṅgīsa-thera Saṃyutta, a Collection of Early Buddhist Discourses on the Venerable Vangīsa.” Buddhist Studies Review 24.1: 35–45.
2010. “Problems and Prospects of the Chinese Saṃyuktāgama: Its structure and content.” In Konrad Meisig, ed. Translating Buddhist Chinese: Problems and Prospects. East Asia Intercultural Studies, Interkulturelle Ostasienstudien 3: 53–64. Wiesbaden: Harrassowitz.
 
Chowdhury, Hemendu Bikash, ed.
1994. Mahapandita Rahula Sankrityayana Birth Centenary Volume: A Jagajiyot Publication, Special Number on Rahula Sankrityayana. Calcutta: Bauddha Dharmankur Sabha (Bengal Buddhist Association).
Clarke, Shayne.
2001. “The Mūlasarvāstivāda Vinaya Muktaka 根本説一切有部目得迦.” Bukkyō kenkyū 佛教研究 [[[Buddhist Studies]]] 30: 81–107.
2002 “The Mūlasarvāstivādin Vinaya: A Brief Reconnaissance Report.” In Early Buddhism and Abhidharma Thought: In Honor of Doctor Hajime Sakurabe on His Seventy-seventh Birthday, 45–63. Kyoto: Heirakuji shoten.
Collins, Steven.
1990. “On the Very Idea of the Pali Canon.” Journal of the Pali Text Society 15: 89– 126.
1982. Selfless Persons: imagery and thought in Theravada Buddhism. London: Cambridge University Press.
Cousins, Lance S.
1983. “Pali Oral Literature.” In P. Denwood & A. Piatigorsky eds., Buddhist Studies: Ancient and Modern, 1–11. London: Curzon.
de Jong, J. W.
1976. A Brief History: Buddhist Studies in Europe and America. Varanasi: Bharat- Bharati.
Deleanu, Florin.
The Chapter on the Mundane Path (Laukikarmārga) in the Śrāvakabhūmi. Tokyo: International Institute for Buddhist Studies of the International College for Postgraduate Buddhist Studies.
Delhey, Martin.
2013. “The Yogācārabhūmi Corpus: Source, Editions, Translations, and Reference Works.” in Ulrich Timme Kragh, ed., The Foundation for Yoga practitioners: the Buddhist Yogācārabhūmi treatise and its adaptation in India, East Asia,
 
and Tibet, 498–563. Cambridge, Mass: Department of South Asian studies, Harvard University.
Demiéville, Paul.
1951. “A propos du concile de Vaiśālī.” T'oung Pao 40: 239–296.
Dhammajoti, Kuala Lumpur.

ṣṭāntika, Sautrāntika and Yogācāra— Some Reflections on Their Interrelation.” Journal of the Centre of Buddhist Studies Sri Lanka 4: 184–212.
Abhidharma Doctrines and Controversies on Perception. Hong Kong: Centre of Buddhist Studies, University of Hong Kong.
Sarvāstivāda Abhidharma. Hong Kong: Centre of Buddhist Studies, University of Hong Kong.
Dietz, Siglinde.
2000. “Mātr̥ceṭas Kaliyugaparikathā.” In C. Chojnacki, J. U. Hartmann, and V. M. Tschannerl, eds., Vividharatnakaraṇḍaka: Festgabe für Adelheid Mette, 230–238. Swisttal- Odendorf. Germany: Indica et Tibetica Verlag.
Dutt, Nalinaksha.
1978. Mahāyāna Buddhism. Delhi: Motilal Banarsidass.
Dutt, Sukumar.
1957. The Buddha and Five After-centuries. London: Luzac.
Eimer, Helmut.
1978. Bodhipathapradīpa: Ein Lehrgedicht des Atiśa (Dīpaṃkaraśrījñāna) in der Tibetischen Überlieferung. Wiesbaden: Otto Harrassowitz.
Enomoto Fumio 榎本文雄.
1980. “Udānavarga shohon to Zōagongyō, Betsuyaku-zōagongyō, Chūagongyō no buhakizoku” Udānavarga 諸本と雑阿含経、別訳雑阿含経、中阿含経の部派帰 属 [On the Recessions of the Udānavarga and the School to which
Saṃyuktāgama, its Alternative Translation, and the Madhyamakāgama are to
 
be ascribed]. Indogaku bukkyōgaku kenkyū 印度學佛教學研究 [Journal of Indian and Buddhist Studies] 28.2: 55–57.
1984. “Agon kyōten no seiritsu” 阿含経典の成立 [The formation of the Original Text of the Chinese Āgamas]. Tōyō gakujutsu kenkyū 東洋学術研究 [The Journal of Oriental Studies] 23.1: 93–108.
1986. “On the Formation of the Original Texts of the Chinese Āgamas.” Buddhist Studies Review 3.1: 19–30.
1989. “Śarīrārthagāthā: A Collection Verses in the Yogācārabhūmi, Part 1: Text.” In Fumio Enomoto et al., eds., Sanskrit-Texte aus dem buddhistischen Kanon: Neuentdeckungen und Neueditionen, 17–35. (Sanskrit-Wörterbuch der buddhistischen Texte aus den Turfan-Funden, Beiheft 2.) Göttingen: Vandenhoek & Ruprecht.
1994. A comprehensive study of the Chinese Saṃyuktāgama: Indic texts corresponding to the Chinese Saṃyuktāgama as found in the Sarvāstivāda-Mūlasarvāstivāda literature. Kyoto: Kacho Junior College 1994 .
1998. “Konpon setsuissai’u-bu to Setsuissai’u-bu 根本說一切有部と説一切有部.”
Indogaku bukkyōgaku kenkyū 印度學佛教學研究 [Journal of Indian and Buddhist Studies] 47.1: 111–119.
2000. “Mūlasarvāstivādin and 'Sarvāstivādin.” In C. Chojnacki, J. U. Hartmann, and V.
M. Tschannerl, eds., Vividharatnakaraṇḍaka: Festgabe für Adelheid Mette, 239–250. Swisttal- Odendorf. Germany: Indica et Tibetica Verlag.
Fǎzhào 法照.
2010. Tànjiū yúgāshīdì lùn tǐyìgātā yǐnyòng chūqījīngdiǎn jìsòng zhī lǐyóu 探究瑜伽師地論體義伽他引用初期經典偈頌之理由 [[[Reasons]] for Quoting Canonical Verses in the Śarīrārthagāthā of Yogācārabhūmi]. Taipei: Master Thesis of the Chung-hwa Institute of Buddhist Studies.
Fukuhara Ryōgon 福原亮厳.
1975. Bukkyō gairon 仏教概論. Kyoto: Nagata Bunshō-dō 永田文昌堂.
 
Franke, R. O.
1908. “The Buddhist councils at Rājagaha and Vesālī, as Alleged in Cullavagga XI., XII.” Journal of the Pali Text Society 1–80.
Frauwallner, Erich.
1956, 1969. Die Philosophie des Bukkhismus. Berlin: Akademie-Verlag.
1994. “On the Historical Value of the Ancient Ceylonese Chronicles.” In J. Soni (tr.),
Erich Frauwallner's Posthumous Essays, 7–33. New Delhi: Aditya Prakashan.
Gethin, Rupert.
2005. “Buddhist Books and Texts : Canon and Canonization : Abhidharma.” In L. Jones, M. Eliade, C. J. Adams, eds., Encyclopaedia of Religion, 15: 10020–23. New York: Macmillan Reference.
Gombrich, R.F.
1988/1990. “How the Mahāyāna began.” In Tadeusz Skorupski, ed., Buddhist Forum Volume I: Seminar Papers 1987–1988, 21–30. London: University of London. (First Published Paper: The Journal of Pali and Buddhist Studies, 1988, 1: 29–46)
Goyal, S. R.
2003. Indian Buddhism After the Buddha. Jodhpur: Kusumanjali Book World.
Hallisey, Charles.
1992/2005. “Councils as Ideas and Events in the Theravada.” In Paul Williams, ed., Buddhism: Critical Concepts in Religious Studies, 2: 171–185. London: Routledge. (First Published Paper: The Buddhist Forum, 1992, 2: 133–148)
Hakamaya Noriaki 袴谷憲昭.
1982. “Yugagyō-ha no bunken” 瑜伽行派の文献 [The Literature of the Yogācāra School]. In Hirakawa Akira 平 川 彰 , Kajiyama Yūichi 梶 山 雄 一 , and Takasaki Jikidō 高崎直道, eds., Yuishiki shisō 唯識思想 [The Philosophy of Mind Only], 43–76. (Kōza Daijō bukkyō 講座大乗仏教, 8.) Tokyo: Shunjūsha 春秋社.
1985. “Tonkō shutsudo Chibetto go yuishiki bunken” 敦煌出土チベット語唯識文献
[The Tibetan Literature of Mind Only from Dunhuang]. In Yamaguchi Zyuhō
 
山口瑞鳳 ed., Kōza Tonkō VI: Tonkō kogo bunken 講座敦煌 6: 敦煌胡語文献, 207–264. Tokyo: Daitō shuppansha 大東出版社.
Hán Jìngqīng 韓鏡清.
1947. “Xīcáng chuánběn yújiāshīdìlùn sīsuǒchéngdì tǐyìjiātā wàtuōnán zhī fènxī” 西藏文本瑜伽師地論．思所成地體義伽他嗢拕南之分析 [The Analysis of Uddāna in the Śarīrārthagāthā of the Cintāmayībhūmi in the Tibetan Yogācārabhūmi]. Biānjiāng rénwén 边疆人文 [[[Humanity]] of borderland] 4: 31–36.
Hán Qīngjìng 韓清淨.
[1983] 1996. Yújiāshī dì lùn kējù pīxún jì huìbiān 瑜伽師地論科句披尋記彙編 [The Collection of the Detailed Catalogue and Comprehensive Exegesis on the Yogācārabhūmi]. 4 vols. Taipei: Xīnwénfēng 新文豐.
Harrison, Paul.
2002. “Another Addition to the An Shigao Corpus? Preliminary Notes on an Early
Chinese Saṃyuktāgama Translation” in The Sekurabe Ronshu Committee (櫻部建博士喜寿記念論集刊行会) ed., Shoki Bukkyou kara abidaruma e: Sakurabe Hajime Hakase koki kinen ronshou 初期仏教 からアビダルマへ: 櫻部建博士喜寿記念論集 [[[Early Buddhism]] and Abhidharma thought: in honor of doctor Hajime Sakurabe on his seventy-seventh birthday], 1–32.
Tokyo: Heirakuji Shoten 平楽寺書店.
Hayashiya Tomojirō 林屋友次郎.
1928. 關於十二部經的研究 [The Studies of Twelve Classification ]. Shūkyō kenkyū
宗教研究 [Journal of Religion Studies] 5–6.
1936. Bukkyō kenkyū 佛教研究 [[[Buddhist Studies]]]. Tokyo : Sanseidō 三省堂.
Hinüber, Oskar von.
1994. “Die neun Aṅgas: ein früher Versuch zur Einteilung buddhistischer Texte.” WZKS 38: 121–35.
1996. A Handbook of Pāli Literature. Indian Philology and South Asian Studies, Volume 2. Berlin: Walter de Gruyter.
 
Hirakawa Akira 平川彰.
1964. “Ku bun jūni bukyō no gen kei to daiji you kyō ten to no kan kei” 九分十二部経の原型と大乗経典との関係 [The Relationship between Mahāyāna Sūtras and the Original Navāṅgabuddhavacana and Dvādaśāṅgabuddhavacana].
Bukkyō shisō shi ronshū 仏教思想史論集 (結城教授頌寿記念) [The Collections of Buddhist Thought History] 93–110. Tokyo: Ōkura 大 蔵 .
1974–1979. Indo bukkyōshi インド仏教史. 2.vols. Tokyo: Shunjūsha 春秋社.
1990. A History of Indian Buddhism from Śākyamuni to early Mahāyāna. Translated and edited by Paul Groner. Asian Studies at Hawaii, 36. Honolulu: University of Hawaii Press.
Hofinger, M.
1946. Étude sur la concile de Vaiśālī. Louvain.
Huimin, Bhikkhu 釋惠敏.
1991. “Two Yogācāra Verses in the Śrāvakabhūmi Related to the Venerable
Revāta.” Indogaku bukkyōgaku kenkyū 印度學佛教學研究 [Journal of Indian and Buddhist Studies] 40.1: 18–22.
1994. Shōmonji ni okeru shoen no kenkyū 聲聞地における所縁の研究 [The Studies on the Objects of Perception in the Śrāvakabhūmi]. Tokyo: Sankibō 山喜房.
Jayawickrama, N. A.
1959. “Biddhaghosa and the traditional classification of the Pali canon.” University of Ceylon Review 17.1–2: 1–17.
Kalupahana, David J.
1965. “Aṅga (2).” In G.P. Malalasekera, ed., Encyclopaedia of Buddhism, 1.4: 616– 619. Colombo: Government of Ceylon.
Kamata Shigeo 鎌田茂雄.
1999. Zuitō no bukkyō (ge) 隋唐の仏教 (下) [The Buddhism in the Dynasty of Sui and Tang, Part 2]. (Chūgoku bukkyōshi, 6:中国仏教史 第六卷 [The History of the Chinese Buddhism Vol. 6].) Tokyo: Tōkyō Daigaku Shuppankai 東京大学出版会.
 
Kanakura Enshō 金倉圓照.
1939. Indo kodai seishinshi 印度古代精神史 [History of the Spirit in Ancient India].
Tokyo : Iwanami Shoten 岩波書店.
1977. “Yuga ron shōmon chi no genten” 瑜伽論声聞地の原典 [The Sanskrit Text of the Śrāvakabhūmi of the Yogācārbhūmi]. Suzuki gakujutsu zaidan kenkyū nenpō 鈴木学術財団研究年報 [Annual of Oriental and Religious Studies] 14: 115–118R.
Katō Seishin 加藤精神.
1930. “Yugashijiron Kaidai” 瑜伽師地論解題 [The decipherment on the title of Yogācārabhūmi]. Indo senjutsubu 國譯一切經：インド撰述部 [[[Japanese]] translations of the Scriptures: Works composed in India] 58: 1–19.
Katsumata Shunkyō 勝又俊教.
1975. “Yugashijiron ni okeru shikan” 瑜 伽 師 地 論 に お け る 止 観 [[[Śamatha]] and Vipaśyanā in the Yogācārabhūmi]. In Sekiguchi Shindai 関 口 真 大 ed., Sh ikan no kenkyū 止 観 の 研 究 [The Studies on the Śamatha and Vipaśyanā], 85–
112. Tokyo: Iwanami shoten 岩波書店. Kragh, Ulrich Timme
2013. “The Yogācārabhūmi and Its Adaptation: Introductory Essay with a Summary of the Basic Section.” in Ulrich Timme Kragh, ed., The Foundation for Yoga practitioners: the Buddhist Yogācārabhūmi treatise and its adaptation in India, East Asia, and Tibet, 22–289. Cambridge, Mass: Department of South Asian studies, Harvard University.
Kritzer, Robert.
2005. Vasubandhu and the Yogācārabhūmi: Yogācāra Elements in the Abhidharmakośabhāṣya. Tokyo: International Institute for Buddhist Studies.
Kudō Jōshō 工藤成性.
1975. “Yuga shi ji ron no seiritsu ni kansuru shiken” 瑜伽師地論の成立に関する私見 (Opinions on the formation of the Yogācārabhūmi). Bukkyōgaku kenkyū 佛教学研究 31:1–20.
 
La Vallée Poussin, L.
1910. “Councils (Buddhist).” In James Hastings, ed., Encyclopedia of Religion and Ethics, 15: 179–185.
1976. The Buddhist Councils. Calcutta: K. P. Bagchi & Company.
Lamotte, Étienne.
[1976] 1958. Histoire du Bouddhisme indien: des origines à l'ère Śaka. Louvain: Université de Louvain.
1980. Le traité de la grande vertu de sagesse. Tome 5. Publications de l'Institut Orientaliste de Louvain. Louvain: Université de Louvain.
1988. History of Indian Buddhism: From the Origins to the Śaka Era. (Translated by Sara Webb-boin) Publications de l'Institut Orientaliste de Louvain. Louvain: Université de Louvain.
Lín Chóngān 林崇安.
2002. “Yújiāshī dì lùn nà de yìyì jí qí biānchéng” 瑜伽師地論內的異義及其編成
[The Heteronymous Phenomenon in the Yogācārabhūmi and Its Compilation]. Fǎ guāng 法光 [[[Dharma]] Light Monthly] 158: 1.
Lǚ Chéng 呂澂.
1924. “Záāhán jīng kāndìngjì” 雜阿含經刋定記 [The Revision of Saṃyuktāgama].In Ōuyáng jìng wú 歐 陽 竟 無 et al., Nèixué 內 學 [The Learning of Inside], 1: 106–127. Taipei: Dǐngwén shūjú 鼎文書局.
19??. “Fǎjù jīng jiǎngyào” 法句經講要 [A Talk on the Essence of the Dharmapāda].
In Lǐ ān 李安 ed., Lǚ Chéng foxue lùnzhe xuǎnjí 呂澂佛學論著選集 [The Collection of Lǚ Chéng Buddhist Studies], 2: 643–674. Chinan, China: Qílǔ shūshè 齊魯書社.
Masunaga Reihō 增永靈鳳.
1948. Konpon bukkyō no kenkyū 根本佛敎の硏究 [The Studies of Original Buddhism].
Ichikawa : Kazama Shobō 風間書房.
 
Mathes, Klaus-Dieter.
1996. Unterscheidung der Gegebenheiten von ihrem wahren Wesen (Dharmadharmatāvibhāga): Eine Lehrschrift von der Yogācāra-Schule in tibetischer Überlieferung. Swisttal-Odendorf: Indo et Tibetica Verlag.
Matsunami Yasuo.
1992. “On the Manuscript of the “Śrāvakabhūmi.” Indogaku bukkyōgaku kenkyū 印度學佛教學研究 [Journal of Indian and Buddhist Studies] 40.2: 29–34.
May, Jacques.
1971. “La philosophie bouddhique idéaliste.” Asiatische Studien/ Études Asiatiques 25: 265–323.
Mayeda Egaku 前田恵学.
1964. Genshi bukkyō seiten no seiritsushi kenkyū 原始佛教聖典の成立史研究 [A History of the Formation of Original Buddhist Texts]. Tokyo: Sankibō Busshorin.
Mayeda Takashi 前田崇.
1991. “Yuga ron bonbun kenkyū” 瑜伽論梵文研究 Abhiprayikarthagathanirdeca (釈意 趣 義 伽 他 ) [The Studies of Sanskrit Yogācārabhūmi: Abhiprayikarthagathanirdeca]. Bunka 文化 [Jounal of Culture] 55: 71–101.
Mayrhofer, Manfred.
1976. Kurzgefasstes etymologisches Wörterbuch des Altindischen. Heidelberg: Winter.
Minayeff, I. P.
1984. Recherches sur le bouddhisme. Traduit du russe par R. H. Assier de Pompignan.
Paris: Ernest Leroux.
Mino Kojun 美濃晃順.
1926. “Kubun jūnibu kyō no ken kyū-ge” 九分十二部教の研究 (上), (中), (下) [The Studies of Nine and Twelve Classification (Part1), (Part2), (Part3)]. Bukkyō kenkyū 仏教研究 [Journal of Buddhist Studies] 22: 60–111; 23: 76–130; 24:
13–60.
 
Mizuno Kōgen 水野弘元.
1952. “Arthapada sūtra ni tsuite” Arthapada sūtra( 義 足 経 ) に つ い て [Arthapada sūtra]. Indogaku bukkyōgaku kenkyū 印度學佛教學研究 [Journal of Indian and Buddhist Studies] 1: 87–95.
1988. “Zōagonkyō no kenkyū to shuppan” 雜阿含經の研究と出版 [Research and Publication on the Saṃyuktāgama]. Bukkyō kenkyū 仏教研究 [Journal of Buddhist Studies] 17: 1–45.
Mukai Akira 向井亮.
1979. “Kenyō shōgyō ron to YugashiJiron” 『顕揚聖教論』と『瑜伽師地論』 [On the Śāsanodbhāvana and the Yogācārabhūmi]. Bukkyōgaku 佛教學[Journal of Buddhist Studies] 8: 39–68.
1980. “Zōagon no kōsei ni tsuite” 雑阿含の構成について [On the Structure of Samyukta-Āgama]. Shūkyō kenkyū 宗教研究 [Journal of Religion Studies] 53.3–180–181.
1981. “Yugaron no seiritsu to Asanga no nendai” 瑜伽論の成立とアサンがの年代[The Time of Yogācārabhūmi and Asaṅga ]. Indogaku bukkyōgaku kenkyū 印度學佛教學研究 [Journal of Indian and Buddhist Studies] 29.2: 198–204.
1985. “The Vastusaṃgrahaṇī of the Yogācārabhūmi and the Saṃyuktāgama.”
Hokkaidō daigaku bungakubu kiyō 北海道大學文學部研究紀要 [The Annual Reports on Cultural Science] 32.2: 1–41.
Nakamura Hajime 中村元.
1987. Indian Buddhism: A Survey with Bibliographical Notes. Delhi: Motilal Banarsidass.
Nattier, Jan.
2004. “The Twelve Divisions of Scriptures ( 十 二 部 経 ) in the Earliest Chinese Buddhist Translations.” Annual Report of the International Research Institute for Advanced Buddhology at Soka University for the Academic Year 2003
167–196.
 
Nishi Giyū 西義雄.
1959. “Yuga shi ji ron ni okeru san jō: toku ni Shōmon ji ni tsuite” 瑜伽師地論における三乘—特に声聞地について [Trīṇi yānāni of the Yogacārabhūmi, especially Śrāvakayāna]. Indogaku bukkyōgaku kenkyū 印度學佛教學研究[Journal of Indian and Buddhist Studies] 7.2: 666–673.
Norman, A.K.
1997a. “Buddhism and Sanskritisation.” In the Institute of Buddhist Studies of Tring (UK), and the Institute of Buddhist Studies, Berkeley, California (USA), eds., Buddhist Forum Volume V: Philological Approach to Buddhism, 95–112. (The Bukkyō Dendō Kyōkai Lectures 1994.) London: University of London.
1997b. “Buddhism and Canonicity.” In the Institute of Buddhist Studies of Tring (UK), and the Institute of Buddhist Studies, Berkeley, California (USA), eds., Buddhist Forum Volume V: Philological Approach to Buddhism, 131–148. (The Bukkyō Dendō Kyōkai Lectures 1994.) London: University of London.
1983. Pāli literature: Including the Canonical Literature in Prakrit and Sanskrit of All the Hīnayāna Schools of Buddhism. Wiesbaden: Otto Harrassowitz.
Nishi Giyū 西義雄.
1953. Genshi bukkyō ni okeru hannya no kenkyū 原始佛敎に於ける般若の研究 [The Studies of Prajñā in Original Buddhism ]. Yokohama: Ōkurayama Bunka Kagaku Kenkyūjo.
Oldenburg, Hermann.
1927. Buddha: His Life, his Doctrine, his Order. Calcutta.
Ogiwara Unrai 荻原雲來.
1869–1937. Ogiwara Unrai bunshū 荻原雲來文集 [The collections of Ogiwara Unrai].
Tokyo : Ogiwara Hakushi Kinenkai.
Panglung, Jampa Losang.
1979. “Preliminary Remarks on the Uddānas in the Vinaya of the Mūlasavāstivādin.”
In M. Aris et al., ed., Tibetan Studies in
honour of Hugh Richardson, 226–232. Warminster: Aris & Phillips.
 
Potter, Karl H. ed.
1999. Encyclopedia of Indian Philosophies. Volume VIII. Buddhist Philosophy from 100 to 350 A.D. Delhi: Motilal Banarsidass Publishers.
Przyluski, J.
1926–1928. Le concile de Rājagr̥ha. Paris: Paul Geuthner.
Prebish, Charles S.
1974. “A Review of Scholarship on the Buddhist Councils.” The Journal of Asian Studies 33.2: 239–254.
1987. “Councils: Buddhist.” In Mircea Eliade, ed., Encyclopaedia of Religion, 4: 119–
124. New York: Macmillan Publishing Company. Renou, Louis. and Filliozat, Jean.
1953. L'Inde classique, tome II, avec la collaboration de P. Demiéville et M. Lalou.
Paris : Impr. Nat., École française d'Extrême-Orient, Hanoi.
Rockhill, William Woodville.
1975. Udānavarga : a collection of verses from the Buddhist canon compiled by Dharmatrāta. Amsterdam: Oriental Press.
Ruegg, Seyfort Ruegg.
1999. “Remarks on the Place of Narrative in the Buddhist Literatures of India and Tibet.” In Alfredo Cadonna, ed., India, Tibet, China: Genesis and Aspects of Traditional Narrative. Orientalia Venetiana 7: 193–27 (esp. pp. 200–6).
Florence: Leo S. Olschki.
Sāṅkr̥tyāyana, Rāhula.
1935. “Sanskrit Palm-Leaf MSS. in Tibet.” Journal of the Bihar and Orissa Research Society 21.1: 21–43.
1937. “Second Search of Sanskrit Palm-Leaf MSS. in Tibet.” Journal of the Bihar and Orissa Research Society 23.1: 1–57.
1938. “Search for Sanskrit MSS. in Tibet.” Journal of the Bihar and Orissa Research Society 24.4: 137–163.
 
1984. Selected Essays of Rahula Sankrityayana. New Delhi: People's Publishing House.
Schmithausen, Lambert.
1969a. Der Nirvāṇa-Abschnitt in der Viniścayasaṃgrahaṇī der Yogācārabhūmmi.
Vienna: Ӧsterreichische Akademie der Wissenschaften. Philosophische- Historische Klasse. Sitzungsberichte, 264. Band, 2. Abhundlung.
Verӧffentlichungen der Komission für Sprachen Süd-und Ostasiens. Heft 8.
1969b. “Zur Literaturgeschichte der Alten Yogācāra-Schule.” Zeitschrift der Deutschen Morgenländischen Gesellschaft, Supplementa I: 812–819.
1970. “Zu den Rezensionen des Udānavargaḥ.” Wiener Zeitschrift für die Kunde Südasiens 14: 47–124.
1987a. Ālayavijñāna: On the Origin and the Early Development of a Central Concept of Yogācāra Philosoph. Tokyo: International Institute for Buddhist Studies.
1987b. “Beiträge zur Schulzugehörigkeit und Textgeschichte kanonischer und
postkanonischer Materialien.” In Heinz Bechert, ed., Zur Schulzugehörigkeit von Werken der Hīnayāna-Literatur, Symposien zur Buddhismusforschung III, 2: 304–403. Göttingen: Vandenhoek & Ruprecht.
2000. “On Three Yogācārabhūmi Passages Mentioning the Three Svabhāvas or Lakṣaṇas.” In Jonathan A. Silk, ed., Wisdom, Compassion, and the Search for Understanding: The Buddhist Studies Legacy of Gadjin M. Nagao, 245–264. Honolulu: University of Hawaii Press.
2007. “Aspects of Spiritual Practice in Early Yogācāra.” Journal of the International College for Advanced Buddhist Studies 國際仏學教大學院大學
研究紀要 11:67–98.
Schoening, Jeffrey D.
1995. The Śālistamba Sūtra and Its Indian Commentaries. 2 vols. Vienna: Arbeitskreis für Tibetische und Buddhistische Studien, Universität Wien.
 
Seyfort Ruegg, David.
1981. “Autour du lTa ba'i khyad par de Ye šes sde (Version de Touen-houang, Pelliot Tibétain (814).” Journal Asiatique 269: 208–229.
Shiio Benkyō 椎尾辨匡.
1935. “Kokuyaku issaigyō kaitei” 國譯一切經解題 [Introdution to the Edition of the Kokuyaku issaigyō 國譯一切經]. In Benkyō Shiio 椎尾弁匡, Kōgen Mizuno 水野弘元, and Mitsuyoshi Saigusa 三枝充 eds., Kokuyaku issaigyō 國譯一切經: Agonbu ichi 阿含部一 [The Section of Āgama part 1], 1: 21. Tokyo: Daito Shuppansha 大東出版社.
Simonsson, Nils.
1957. Indo-tibetische Studien: Die Methoden der tibetischen Übersetzer, untersucht im Hinblick auf de Bedeutung ihrer Übersetzung für die Sanskritphilologie.
Vol. I. Upsala: Almqvist & Wiksell Boktryckeri Ab.
Skilling, Peter.
1991. “A brief Guide to the Golden Tanjur.” Journal of the Siam society 79.2: 138– 146.
1994–1997. Mahāsūtras: Great Discourses of the Buddha. Vol. I: Texts. Critical editions of the Tibetan Mahāsūtras with Pāli and Sanskrit counterparts as available. 1994. Vol. II: Parts I & II. 1997. Oxford: The Pali Text Society (Sacred Books of the Buddhists XLIV, XLVI).
2009. “Redaction, Recitation, and Writing: Transmission of the Buddha’s Teachings in India in the Early Period.” In Stephen C. Berkwitz, Juliane
Schober, and Claudia Brown, eds., Buddhist Manuscript Cultures: Knowledge, Ritual, and Art, 53–75. London: Routledge.
2013. “Nets of Intertextuality: Embedded Scriptural Citations in the Yogācārabhūmi.” in Ulrich Timme Kragh, ed., The Foundation for Yoga practitioners: the Buddhist Yogācārabhūmi treatise and its adaptation in India, East Asia, and Tibet, 772–791. Cambridge, Mass: Department of South Asian studies, Harvard University.
 
Śrāvakabhūmi Study Group
1994. Introduction to the facsimile edition of the “Śrāvakabhūmi” Sanskrit palm- leaf manuscript. Tokyo: The Institute for Comprehensive Studies of Buddhism, Taisho University.
Steinkellner, Ernst.
2004. A Tale of Leaves. On Sanskrit manuscripts in Tibet, their Past and their Future. Amsterdam: Royal Netherlands Academy of Arts and Science.
2007. Dharmakīrti's Pramāṇaviniścaya: Chapters 1 and 2. Beijing: China Tibetology Publishing House; Vienna: Austrian Academy of Sciences Press.
Sugawara Yasunori 管原泰典.
2013. “The Bhāvanāmayī Bhūmiḥ: Contents and Formation.” in Ulrich Timme Kragh, ed., The Foundation for Yoga practitioners: the Buddhist Yogācārabhūmi treatise and its adaptation in India, East Asia, and Tibet, 792–851. Cambridge, Mass: Department of South Asian studies, Harvard University.
Suguro Shinjō 勝呂信静.
1976. “Yugaron no seiritsu ni kansu ru shiken” 瑜伽論の成立に関する私見 [On a Establishment of the Yogācāryabhumi. Osaki gakuho 大崎学報 [The Journal of Nichiren and Buddhist Studies] 129: 1–50.
1989. Shoki yuishiki shisō no kenkyū 初期唯識思想の研究 [The Studies on the Early Theory of Yogācāra-Vijñānavāda]. Tokyo: Shunjūsha.
Takasaki Jikidō 高崎直道.
1982. “Yugagyō-ha no keisei” 瑜伽行派の形成. In Hirakawa Akira 平川彰, Kajiyama Yūichi 梶山雄一, and Takasaki Jikidō 高崎直道, eds. Kōza Daijō bukkyō, Vol. VIII, Yuishiki shisō 講座大乘仏教 8 唯識思想. Tokyo: Shunjūsha 春秋社.
Tilakaratne, Asaṅga.
2000. “Saṅgīti and Sāmaggī: Communal Recitation and the Unity of the Saṅgha.”
Buddhist Studies Review 17.2: 175–193.
 
Tsukamoto Keisho 塚本啟祥.
1963. “Mahakasyapa's Precedence to Ananda in the Rajagrha Council.” Indogaku bukkyōgaku kenkyū 印度學佛教學研究 [Journal of Indian and Buddhist Studies] 11.2: 53–60.
Ui Hakuju 宇井伯壽.
1924. “Fumito teki jinbutsu toshite no miroku oyobi mu cho no chojutsu” 史 人 的 人物としての彌勒及び無著の著述 [The Books of Maitreya and Asaṅga].
Journal of Indian Philosophy Studies 1: 359–372.
1925. “Genshi bukkyō shiryōron” 原始佛教資料論 [The Information of Early Buddhism]. Indo tetsugaku kenkyū 印度哲学研究 [A Study of Indian Philosophy] 2: 115–260.
1958. Yugaron kenkyū 瑜伽論研究 [A Study of the Yogācārabhūmiśāstra]. Tokyo: Iwanami Shoten 岩波書店.
Varma, Vishwanath Prasad.
1973. Early Buddhism and Its Origins. Delhi: Munshiram Manoharlal.
Vogel, Claus.
1965. Vāgbhāṭa’s Aṣṭāṅgahr̥dayasaṃhitā: The First Five Chapters of Its Tibetan Version. Wiesbaden: Franz Steiner GmbH.
Walleser, Max.
1914. Prajñāpāramitā: Die vollkommenheit der erkenntnis, nach indischen, tibetischen und chinesischen quellen. Göttingen: Vandenhoeck & Ruprecht.
Warder, A. K.
1965. “On the Relationship between Early Buddhism and Other Contemporary Systems.” Bulletin of the School of Oriental and African Studies 18: 43–63.
1970. Indian Buddhism. Delhi: Motilal Banarsidass.
Watanabe Baiyū 渡邊楳雄.
1948. Jōdai Indo bukkyō shisōshi 上代印度佛敎思想史 [The History of Early Indian Buddhist Philosophy]. Tokyo : Shūkyō Jihōsha 宗教時報社.
 
Wayman, Alex.
1961. Analysis of the Śrāvakabhūmi Manuscript. Classical Philology 17. Berkeley and Los Angeles: University of California.
1984. Buddhist Insight. Delhi: Motilal Banarsidass.
Williams, Raymond B.
1970. “Historical Criticism of a Buddhist Scripture: The Mahāparinibbāna Sutta.”
Journal of the American Academy of Religion 38.2: 156–167.
Witanachchi, C.
2006. “Saṅgīti.” In W.G. Weeraratne, ed., Encyclopaedia of Buddhism, 7.4: 720–728.
Sri Lanka: Department of Buddhist Affairs, Ministry of Religious Affairs.
Willemen, Charles, Bart Dessein and Collett Cox.
1998. Sarvāstivāda Buddhist Scholasticism. Leiden; New York; Kӧln: Brill.
Woodville Rockhill, William.
1975. Udānavarga: a collection of verses from the Buddhist canon. Amsterdam: Oriental Press (reprint).
Yamabe Nobuyoshi 山部能宜.
1989. “Shūji no hon’ to shingun no mondai ni tsuite” 種子の本有と新熏の問題について (Regarding the issue on the original exist or later production of seeds). Nihon bukkyōgakkai nenpō 日本佛教学会年報 63: 43–58.
1990. “Bīja Theory in the Viniścayasaṃgrahaṇī.” Indogaku bukkyōgaku kenkyū 印度學佛教學研究 [Journal of Indian and Buddhist Studies] 38.2: 13–15.
2003. “On the School Affiliation of Aśvaghoṣa: ‘Sautrānitika’ or ‘Yogācāra’?.”
Journal of the International Association of Buddhist Studies 26.2: 225–254.
2013. “Parallel Passages between the Manobhūmi and the Yogācārabhūmi of Saṃgharakṣa.” in Ulrich Timme Kragh, ed., The Foundation for Yoga practitioners: the Buddhist Yogācārabhūmi treatise and its adaptation in India, East Asia, and Tibet, 596–737. Cambridge, Mass: Department of South Asian studies, Harvard University.
 
Yáng Báiyī 楊白衣.
1984. “Dàolún yújiāshī dì lùn jì yánjiū” 道倫《瑜伽師地論記》之研究 [Daolun's Notes on the Yogācārabhūmi]. Huágǎng fóxué xuébào 華崗佛學學報 [Hua- Kang Buddhist Journal] 7: 113–134.
Yìnshùn 印順.
1971. Yuánshǐ fójiāo shèngdiǎn zhī jíchéng 原始佛教聖典之集成 [The Formation of Early Buddhist Texts]. Taipei: Zhèngwén chūbǎnshè 正聞出版社.
1983. Záāhán jīng lùn huìbiān 雜阿含經論會編 [Combined Edition of Sūtra and Śastra of Saṃyuktāgama]. Taipei: Zhèngwén chūbǎnshè 正聞出版社.
1988. Yìndù fójiāo sīxiǎng shǐ 印度佛教思想史 [The History of Indian Buddhist Theory]. Taipei: Zhèngwén chūbǎnshè 正聞出版社.
Yokoyama Kōitsu 横山紘一.
1976. Yuishiki shisō nyūmon 唯識思想入門. Tokyo: Daisan bunmei-sha 第三文明社.
Yūki Reimon 結城令聞.
[1931] 1999. “Yugarongi no chosha mei ni taisuru gigi” 『瑜伽論記』の著者名に対す る 疑 義 [The Question on the Author of Yújiā lùn jì 瑜 伽 論 記 ]. In Yūki Reimonchosaku senshu 結城令聞著作選集. vol. I: Yuishiki shisō 唯識思想 [The Theory of Yogācāra-Vijñānavāda]. Tokyo: Shunjūsha.
[1962] 1985. Yuishikigaku tensekishi 唯 識 學 典 籍 志 [The Record of Yogācāra- Vijñānavāda Books]. Tokyo: Tōkyō daigaku tōyō bunka kenkyūjo 東 京 大 學 東洋文化研究所.
Zimmermann, Michael.
2002. A Buddha Within: The Tathāgatagarbhasutra the Earlist Expositions of the Buddha-Nature Teachings in India. Tokyo: International Research Institute for Advanced Buddhology, Soka University.
Zōngzhèng 宗證.
2006. “Záāhán děng jīng yǔ yúgāshīdì lùn tǐyìgātā zhī xiànghù duìzhào” 雜阿含等經與瑜伽師地論體義伽他之相互對照 [The Contrast between Śarīrārthagāthā and
 
Saṃyuktāgama and Others]. Fúyán fóxué yánjiū 福嚴佛學研究 [The Buddhist Studies of Fuyan] 2: 46–100.