Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

A Study of Haribhadra's Abhisamayālaṁkārālokā Prajñāpāramitā-vyākhyā Sparham, Gareth 1989

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


831-UBC_1989_A1 S68.pdf [ 25.16MB ]
JSON: 831-1.0098337.json
JSON-LD: 831-1.0098337-ld.json
RDF/XML (Pretty): 831-1.0098337-rdf.xml
RDF/JSON: 831-1.0098337-rdf.json
Turtle: 831-1.0098337-turtle.txt
N-Triples: 831-1.0098337-rdf-ntriples.txt
Original Record: 831-1.0098337-source.json
Full Text

Full Text

A S T U D Y O F H A R I B H A D R A ' S ABHISAMA YALAMKARALOKA PRAJMPARAMJTTA- VYAKHYA: B y G A R E T H S P A R H A M B.A., M c G i l l University, 1970 A T H E S I S S U B M I T T E D I N P A R T I A L F U L F I L L M E N T O F T H E R E Q U I R E M E N T S F O R T H E D E G R E E O F D O C T O R O F P H I L O S O P H Y in T H E F A C U L T Y O F G R A D U A T E S T U D I E S (Department of Asi a n Studies) W e accept this thesis as conforming to the required standard T H E U N I V E R S I T Y O F B R I T I S H C O L U M B I A A p r i l 1989 © Garelh Sparham, 1989 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of AS'Q-A. S - r ^ c U e S The University of British Columbia Vancouver, Canada Date V 3 DE-6 (2/88) A B S T R A C T Haribhadra was an Indian Buddhist writing i n about the year 800 during the reign of Dharma-pala, the greatest of the Pala kings who held sway over north-east India f r o m the 8th to the 12th centuries. H i s works have been largely ignored, up until this point, by Western scholarship, even though they had a great influence on the course of later Indian and Tibetan scholastic Buddhism. Haribhadra's major work is the Abhisamayalamkaraloka Prajha-paramita-vyakhya, Commentary on the "Perfection of Wisdom" With the Light [Provided by Maitreya's] "Ornament for Clear Realization" (abbreviated title A A A ) . It is not so much a single book as a composite of three: (a) the Asta-sahasrika-prajna-paramita, Perfection of Wisdom in 8.000 Lines (A), considered to be a revelation, (b) the Abhisamayalamkara, Ornament for Clear Realization (AA), an aphoristic codification, at least according to Haribhadra's point of view, of the topics i n the revelation, and (c) the A A A itself, Haribhadra's explanatory sentences or commentary which links the A and A A together. T h e A can be characterized as a mystic-religious work. It sets forth the doctrine of emptiness (gunyata) or no-self (anatman), also called lack of own-being, absence of sva-bhava, a perfect wisdom which consists in complete detachment from all phenomena, such that all phenomena are seen to be illusory (maya). Directed towards people who have an intuitive understanding of the value of altruistic endeavor the A says, paradoxically, that there ultimately are no l i v i n g beings to be helped, no motivation to help them, and no state of perfection by means of which their needs are met. The A A is a highly scholastic systematization concerned, first and foremost, with the delineation of the path (marga = inner spiritual development) particularly at its higher levels. There is a striking contrast between the complete lack of systematization i n the A and the detailed, step by step, delineation of the stages of the path, levels (bhumi) and Buddha bodies (kaya) which one meets with in the A A . T h e difference i n the general tenor of the two texts is carried over into the A A A , which seeks to explain the former with the help of the latter. Those sections i n the A A A directly explaining the sutra which have no parallel i n the A A are more religious in tone, more faith-oriented. Those sections i n the A A A i i detailing a topic of the AA, and explaining how that topic is found in the sutra, are more scholastic and difficult to penetrate. These latter sections presuppose considerable familiarity with terminology and theories about the path and its structure. W hen the topic of the A A naturally corresponds with a division i n the A the structural tension that comes from trying to combine two different books together in a single unit is not apparent. Sometimes, however, there is no apparent correspondence. A t such points in the text there is a structural tension and the thread, which Haribhadra, at least, felt kept the different parts of his work together as a coherent whole, may slip from the reader's hand. T o place this thread firmly in the reader's hand the dissertation clarifies (a) the different layers of text which are embodied i n the A A A and (b) issues concerning the path which Haribhadra's comments are indirectly or directly addressing. In regard to the latter, the language of discourse has two different vocabularies: (a) the terminology of Vasubandhu's Abhidharma-kote and (b) the terminology found in the AA. There are considerable difficulties involved in attempting to systematically present the A A A , for the first time, in a Western language. These difficulties are mostly the result of the many layers of historical material merged together into the single text of the A A A , as well as of loss of some information and Haribhadra's Sanskrit style. T o overcome the difficulty that results from the need to retain i n a translation (a) the integrity of the book as a religious text, and (b) the historical reality that words, and particularly technical terms, often mean different things at different periods, the following strategies are employed: (i) The Sanskrit text of the translated portion (two-thirds of the first chapter or a little over 80 of the A A A ' s nearly 1000 pages) has been presented in Devanagari based on a comparison of the earlier editions and the Tibetan translations, (ii) The text itself is presented, as far as possible, in an unbroken literal translation that lets Haribhadra speak for himself, (iii) B y using modern technology the different layers of the text are consistently identified and separated by use of bold, italic, underline and quotation marks, (iv) The important historical issues the text raises, and the central concepts discussed in the three texts, are identified and discussed in a sufficiently detailed introduction i i i C O N T E N T S ABSTRACT TABLE OF CONTENTS ACKNOWLEDGMENTS TECHNICAL NOTE ABBREVIATIONS PREFACE PART ONE: INTRODUCTION I THE AUTHOR 2 Haribhadra 2 II HARIBHADRA'S TEXT 8 The title 8 Structure of the A A A 9 A more detailed description of the first chapter 14 III THE A AND AA 26 T h e Asta-sahasrika 26 The Abhisamayalariikara 31 IV INDIAN LITERARY GENRES AND THE AAA 36 T h e commentarial tradition after Haribhadra 50 V HARIBHADRA'S RELIGIOUS PHILOSOPHY 56 D o m a i n of Haribhadra's philosophical enquiry 58 The three nature doctrine 67 A T h e consciousness-transformation doctrine 71 The influence of Dignaga and Dharmaklrti 77 VI DOCTRINES OF YOGA AND BHUMITN THE AA AND AAA 83 Presuppositions 84 Paths 89 Seven attentions and three realms 106 Leve l s 113 V THE PATH AS DESCRIBED IN PP LITERATURE 116 F o u r practices 123 T h e dharma-kaya 127 i i i v v i i i ix x x i i i v T h e ten topics 131 Spiritual Community 134 VIII EXPLANATION OF SOME IMPORTANT TERMS USED IN THE TEXT 143 T h e meaning of the term abhisamaya 143 Th e meaning of the term akara 147 Th e meaning of the term pratipatti 150 Haribhdra's use of the term prajna-paramita 152 T h e terms irutamayi, cintamayi, and bhavanamayi prajha 153 IX CHARTS 155 T h e eight basic categories (padartha) 155 T e n topics figuring i n a description of the sarvakarajnata 155 Three rearms (dhatu) 156 Th e path of meditation (bhavana-marga) according to the A K and its correspondence with the three realms 157 Seven attentions (manaskara) and ninefold division of the path of meditation (bhavana-marga) 158 Correspondence between the levels (bhumi) and paths of meditation 159 Correspondence between the levels (bhumi) and ten perfections (paramita) 160 Lis t of twenty samgha members i n AA23-24 according to A r y a Vimuktisena 161 Lis t of twenty samgha members i n AA23-24 according to Haribhadra 162 PART TWO: SANSKRIT TEXT I PRELIMINARY REMARKS Homage Anubandha-catustaya Setting the scene Retinue T h e scene and retinue according to the longer Sutras T h e contents in brief 164 165 169 172 175 178 v II PERFECT WISDOM 184 Bodhicitta 185 D i v i s i o n s of bodhicitta 187 Preceptual advice 189 Te n topics o f preceptual advice 192 Discursus on the twenty members Of the A r y a Samgha 195 HI APPROACHING ULTIMATE REALITY 197 Warmed proximate penetration 197 Peaked proximate penetration 205 Forbearance proximate penetration 211 Supreme dharma proximate penetration 215 Four conceptualizations 218 Conceptualization of an apprehender 222 Mentor 226 IV LINEAGE, OJECTlVE SUPPORT AND AIMS 229 Lineage 229 Objective support 230 Ai m s 232 PART THREE: TRANSLATION I PRELIMINARY REMARKS 237 Homage 237 Anubandha-catustaya 238 Setting the scene 245 Retinue - 251 The scene and retinue according to the longer Sutras 259 The contents i n brief 265 II PERFECT WISDOM 275 Bodhicitta 279 D i v i s i o n s of bodhicitta 282 Preceptual advice 286 Te n topics of preceptual advice 293 Discursus on the twenty A r y a Samgha 299 v i III APPROACHING ULTIMATE REALITY 303 Warmed proximate penetration 305 Peaked proximate penetration 319 Forbearance proximate penetration 331 Supreme dharma proximate penetration 339 Four conceptualizations 345 Conceptualization of an apprehender 352 Mentor 357 IV LINEAGE, OJECTIVE SUPPORT AND ATMS 362 Lineage 362 Objective support 366 A i m s 369 BIBLIOGRAPHY 376 v i i A C K N O W L E D G M E N T S Here it is my pleasure to acknowledge my debt, and express my gratitude to the following people. T o my learned and patient teacher of Sanskrit at the University of British Columbia, Dr. A. Aklujkar, who very kindly consented to be my thesis director. T o the two other members of my thesis committee Dr. L. H u r v i t z and Dr. S. Iida. Whatever there is of scholarly merit i n this thesis is the result of the guidance of these scholars. I also wish to express my profound thanks to Dr. M. Futrell, professor i n the University of British Columbia Department of Slavonic Studies, who selflessly encouraged me and read the early drafts of my work. M y indebtedness is also to Dr. J. Hopkins of the University of V i r g i n i a Department of Religious Studies. Dr. Hopkins, and his wife Dr. Elizabeth Napper, scholar in residence at the University of V i r g i n i a , have been tirelessly helpful to me. I would never have begun this dissertation, and once begun, would never have finished it, without their continual encouragement. Th e Devanagari and Tibetan fonts used in this thesis were designed by Dr. K. Bryant and Mr. Pierre Robillard, respectively. I would like to thank them for allowing me to make use of their work. Finally, I would like to thank Janice K i r k e r who typed part of an earlier draft of the translation. The research on which this thesis is based was carried out under doctoral fellowships from the Norman Mackenzie M e m o r i a l Fund, the K i l l a m Trust and the S o c i a l Sciences and Humanities Research C o u n c i l of Canada. v i i i T E C H N I C A L N O T E Th e sections of the thesis are numbered consecutively i n order to facilitate cross-referencing. A reference to a section of the thesis is indicated by numbers in parenthesis preceded by §, e.g., [§3.2.1]. In the translation all words not found in the text itself are enclosed in square brackets []. T h e only exception to this rule is the section heading i n bold type preceded by a number. These headings, which usually reflect a natural break in the text, have been added to make the text easier to read. T h e edition of the Skt text is basically a presentation of Wogihara's (1973) version, with some changes based on T u c c i (1932), Obermiller (1933-36), V a i d y a (1960) and the Tibetan translations. Its primary function is to allow readers ready access to the original text and to obviate the need for the insertion of Skt into the body of the translation itself. It is presented in Devanagari, with hyphenation and some punctuation, designed to help, but not dictate, the probable correct reading. In both the translation and the Skt text the corresponding pages of Wogihara's version are indicated by numbers preceded by W in smaller type, e.g., [W28.1]. Indented connected passages in italics indicate the Asta-saliasrika-prajna-pframita, as do words and phrases in bold embedded in the text. Underlined words embodied i n the text indicate the Abhisamayalamkara. In order to avoid excessive clutter, paraphrases of, or direct quotation from, A r y a Vimuktisena's Abhisamayalamkara-vrtd are indicated only occasionally i n footnotes. The symbol * indicates that the following word or title is the probable equivalent o f a Tibetan translation. T h e symbol -» indicates an acceptable emendation. ix A B B R E V I A T I O N S A [AryaJ-Asta-sa^asrika-prajna-paramita A A Abhisamayalanrtaya-[prajm-paramitopade§a-§astra] A A A Abhisamayalamkaraloka Prajna-paramita. Vyakhya A A G s e r Shes rab kyi pha rol tu phyin pa'i man ngag gi bstan bcos mngon par rtogs pa'i rgyan 'grel ba dang bcas pa'i rgya cher bshad pa'i legs bshad gser phreng zhes by a ba A A P r a s Abhisamayalamkara-bhagavad-prajiia-paramitopadeSa-§astra-pka-prasphuta-pada A A R g y a n Shes rab kyi pha rol tu phyin pa'i man ngag gi bstan bcos mngon par rtogs pa'i rgyan zhes by a ba'i 'grel ba'i rnam bshad snying po'i rgyan A A P D o n Pan chen spyi don =Rnam bshad snying po'i rgyan gyi don rigs lam bzhin du 'chad pa'i yum don yang gsal sgron me A A S p h u A bhisamayalamkkra-karika-[nama prajna-paramitopadesa-sastraj-vrtti A A V Abmsamayalamkara-vrtd A K Abhidharma-kos'a A K B Abhidharma-kosa-bhasya A K S Abhidharma-kos'a-vyakhya Sphutartha A K T h a r Dam pa'i chos mngon pa'i mdzod kyi rnam par bshad pa thar lam gsal byed. A S Abhidharma-samuccaya A S B Abhidharma-samuccaya-bhasya B C A Bodhicaryavatara B K Bhavana-krama B S O A S Bulletin of the School of African and Oriental Studies Edgerton Buddhist Hybrid Sanskrit Grammar and Dictionary H Haribhadra I s M E O Istituto Italiano per il Medio ed Estremo Oriente J A O S Journal of the American Oriental Society J I A B S Journal of the International Association of Buddhist Studies JIBS Indogaku Bukkyogaku Kenkyu [Journal of Indian and Buddhist Studies] x lc L o k e s h Candra, Tib Skt Dictionary M A V Madhyanta-vibhaga M C B Melanges Chinoise et Bouddhique M H K Madhyamaka-hrdaya-karika mvp Mahavyutpatti M M K Mula-madhyamaka-karikas de Nagarjuna avec la Prasannapada Commentaire de CandrakLrti (la V a l l e e Poussin 1903-13) M p p s he Traite de la Grande Vertu de Sagesse de Nagarjuna (Lamotte 1967-80) M S Mahay ana-samgraha M S A Mahayana-sutralamkara mw M o n i e r Williams, A Skt. English Dictionary P Tibetan Tripitaka, Peking Edition. (Suzuki 1955-61) Panca Panca-vimsati-sahasrika-prajm-paramita P E W . Philosophy East and West P P P i Prajna-p&amita-pindartha-[sariigraha] P P prajna-paramita P V Pramana-varttika P V T h a r Rnam 'grel thar lam gsal byed R G S Ratna-guna-sariicaya-gatha Sata Sata-sa^asrika-prajna-paramita S B h Sravaka-bhiimi S D V Satya-dvaya-vibhanga Subodhini Bhagavad-ratna-guna-samcaya-gatha-panjika Subodhini Skt Sanskrit T i b Tibetan T o h A Complete Catalogue of the Tohuku University Collection of Tibetan Works on Buddhism. (Kanakura 1934-5) T S P Tattva-samgraha-panjika T u T u c c i (1932) V V a i d y a (1960) W Wogihara W Z K S Wiener Zeitschrift fur die Kunde Sud-(und Ost)asiens x i P R E F A C E Part one of the present work is a general introduction to the first chapter of Haribhadra's AbMsamayalamkaraloka ( A A A ) and a specific discussion of the first chapter. Part two is an edition of the Sanskrit text and part three a translation. Because of the sheer size o f the text it has been necessary, for the present, to limit the scope of the work to a translation of two-thirds of the first chapter or a little over 80 of the A A A ' s nearly 1000 pages. T h e translation covers the first six o f the ten topics discussed i n the first chapter of the A A . It breaks at the end of the exchange between the Buddha, Subhuti and Sariputra, marked by the entry o f P u ma Maitrayani-putra. Since the last four topics of chapter one fo r m a unit separate f r o m the first six it is a convenient place to break the text. F o r the sake of consistency, the introduction to the translation covers all the topics dealt with in the first chapter. Haribhadra's A A A is not so much a single book as a composite of three: the Asta-sa^asrika-prajtt-p&amita (A), a revelation, the Abhisamayalamkara ( A A ) , an aphoristic codification o f the topics i n the revelation, and the A A A itself, a commentary which links the A and A A together. There are considerable difficulties involved i n translating Haribhadra. These difficulties are mostly the result of the many layers of historical material merged together into the single text of the A A A , as w e l l as of loss o f some information and Haribhadra's Sanskrit style. T h e A A was probably written as much as three or four hundred years before the A A A , and the earlier portions of the A as much as five hundred years before that. Haribhadra, however, sets forth his explanation o f the words and ideas o f both earlier texts as an authentic (but claimed to be unoriginal) exposition of their original intent. T o overcome the difficulty that results from the need to retain i n a translation (a) the integrity of the book as a religious text, and (b) the historical reality that words, and particularly technical terms, often mean different things at different periods I have used the following strategies: (i) T h e text itself is presented, as far as possible, i n an unbroken literal translation that lets Haribhadra speak for himself, (ii) The different layers o f the text are consistently identified and separated by use of bold, italic, underline and quotation marks, (iii) T h e x i i important historical issues the text raises, and the central concepts that are discussed in the three texts, are identified in the introduction that precedes the translation. T hough the translation into a modern Western language of a text the size and complexity of the A A A does present considerable difficulties the time has now come for an attempt to be made. The ground has been smoothed by earlier scholarship. Preparation of an edition of the text close to what Haribhadra originally wrote presented few difficulties. 1 M u c h of the A and A A has already been made available in English by Obermiller and Conze. 2 Though it has been necessary to restore the parts of the A that Conze, in the interests of readability left out in his translation of the text, his work on the sutra made the task of translating the A A A much easier. In regard to Conze's translation of the AA, it was less helpful. Conze's aim was simply to systematize the many lists and insert just enough glosses to make it intelligible. It has been necessary, therefore, to completely retranslate the text in a manner that allows the reader to recognize its original structure and language. Obermiller's unfinished Analysis of the A A, which translates parts of the A A A , the A A S p h u and considerable T i b material has been less helpful for the present undertaking. 3 Editions of the A, AA, AAA and AAV by Tucci (1932), Obermiller (1933-36, 1970), Wogihara (1973), Pensa (1967) and Vaidya (1960) have been used. 2 Obermiller (1933-36), Conze (1954a, 1973, 1979). 3 Obermiller's excellent work introduces much material from the later Tib scholastic tradition, particularly as found in the [Dga' ldan-»] Dge lugs sect. Although I have read the Tib material Obermiller consulted, besides Tib commentaries Obermiller did not consult, I have felt it better to try, as much as possible, to avoid reference to it in the interests of making less complicated an already complicated subject. x i i i PART ONE: INTRODUCTION 1 I THE AUTHOR 1.1 Haribhadra Haribhadra (Seng ge bzang po) 1 was writing i n about the year 800 during the reign of Dharmapala (rg. c. 770-810; Ruegg 1981:101, n. 320), the greatest of the Pala kings. W e know this from the colophon of the A A A (W994.15-22): s i n w w I" cTf^rW-fwfMT fa<facii ^-sPh^m *t*tt ii I have written this fine panjika (an exposition) which throws light on the truly real, having lived, with the support of the glorious Dharmapala, in this excellent, glorious Trikatuka monastery, mine of all good qualities, which is rightly famed for virtue and ornamented by learned persons ... a distinguished place which has all prerequisites for happiness. In the same colophon H a r i appears to relate the Bhadra (= good) part of his name with the upadhyaya Bhadra Vairocana (or Vairocana-bhadra) who, according to Tafanatha (Chattopadhyaya and Chimpa 1980:277) and B u ston (Obermiller 1932:156-9), taught Haribhadra the PP. Whether he does this as a simple play on words (cf. W4.22-3), or because this teacher-disciple line had Bhadra as the second member of their names, is unclear. Spfe II Through the power of those well-intentioned and generous souls who follow the view of A r y a Asanga, and also of [my] much learned and knowledgable good (bhadra) guru Vairocana, one called H a r i the good (bhadra) got a little spark of intelligence and has, with devotion, composed this noble and sacred [commentary] in which the topics are totally clear. W e know little else about Vairocana, or for that matter Hari. According to Taranatha and B u ston, Haribhadra was from a royal family. H i s knowledge of Sirhha-bhadra was an earlier incorrect sanskritization of Seng ge bzang po. 2 the P P was so great, according to Taranatha, that he saw A j i t a (=Maitreya, the author of the A A according to tradition) i n a dream. H e learned Madhyamaka from Santaraksita and, perhaps as a younger disciple, studied alongside KamalaSlia, a circumstance that would explain the quotations from, or paraphrases of, KamalaSlla's Tattva-samgraha-pahjika (TSP) at W2.3 (Amano 1969), Bhavana-krama ( B K ) at W93.22, Sarva-dharma-nihsvabhava-siddhiat W969ff (Moriyama 1984) and, possibly, Sapta-Satfea-prajna-pframita-tikaiaX W5.16 in his work. Haribhadra is not to be confused with the earlier Jaina writer Haribhadra Suri (700-770), the author of, amongst other works, the Sad-darSana-samuccaya, a compendium of Indian philosophies. 1.1.2 There is no record of any text on tantra by Haribhadra either i n Skt or in the T i b Bstan 'gyur. This is surprising because he li v e d at a time when tantric Buddhism enjoyed royal patronage. During the reign of king Dharmapala the Guhya-samaja Tantra, in particular, flourished and Haribhadra's main disciple Buddha-jnana was the originator of a major stream of explanation of tantra (Wayman 1983:25-6). Haribhadra was clearly familiar with tantric mythology. 1 But i f he wrote any works on tantric Buddhism nothing survives. Anything he did write on tantra has been lost or was, possibly, published under a different name. He wrote four texts on the PP: the A A A , AASphu, 2 a recast of the Paiica i n eight chapters {Le'u brgyad ma) and Subodhini (Bhagavad-ratna-guna-samcaya-gatha-pahjhkaVSamcaya-gatha-vrtti-sub O f these four books, which comprise the total known works of Haribhadra, only the A A A , E.g., his description of Vajrapani as Maha-vajra-dhara (W5.6-10) or his exhortation to meditators in the Subodhini "to visualize their deity in the form of Vajradhara" ( ^ ' ^ ' ^ ' i ' r a a ^ ' ^ l ^ ' ^ ' l K ^ ' 3 * r ^ l P5190:273.2.1). 2 The AbhisamayMamkara-nania-prajna-paranutopadeSa-S^^ Sphutartha (P5191) or, Abhisamayalamkara-karika-Sastra-vnti (Amano 1975), also known in Tib as 'grel cung "Small Commentary" as distinct from 'grel chen "Big Commentary (=AAA). The AASphu has the merit of expanding on the AA just enough, making it penetrable without again losing the reader in lists and sub-topics. It is, unfortunately, at present only available in Tib translation and two Skt reconstructions (Amano 1975 and Tripathl 1977). Unpublished fragments of a Skt manuscript of the AASphu are mentioned by Tucci (1932:v) and Amano (1975:1, n. 5). 3 Haribhadra's most important work and the topic of this dissertation, is still definitely extant in Skt. 1.1.3 The A A S p h u is best understood as an abridgement o f the A A A (Obermiller 1932-3, Amano 1975), and is smaller in scope. It omits detailed reference to the P P surras and, apart from some occasional expansion, supplies only the necessary framework for understanding the A A . 1.1.4 The translation of the recast Panca is Haribhadra's Le'u brgyad ma ('Eight Chapters'). 1 It gets its T i b name because the long and diverse discourses of the Panca are organized along the lines of the eight abhisamayas, literally 'clear realizations,' but in this context 'chapter' (§8.1), set forth i n the AA. It is not an original work, in the usual sense of the term, but a re-organization of the Sutra. The topics listed i n the A A are spelled out clearly after each section. F o r example, AA1.19-20 lists 22 analogies for "production of an [altruistic] mind [set on] enlightenment" {bodhi-cittotpada). The first of these analogies is given as earth (prthivi). In Haribhadra's recast version, the text of the Panca (Dutt 1934:19.4-6) is followed by a few words indicating the corresponding part of the A A : HKpfcimi ^ I d ^ H ^?-?^-»Icr: " j l^MM: II Furthermore, Sariputra, a great Bodhisattva who wishes to completely come face to face with all dharmas 2 should stand in perfect wisdom. [This is the bodhi-cittotpada] exemplified by earth and accompanied by longing. 1.1.5 The Subodhini is an aid to understanding the oldest of the P P sutras, the Ratna-guna-samcaya-gatha (RGS, Conze I960). 3 Haribhadra employs the Listed as P5188. This Pafica-PP would be very similar to manuscript Pl in Dutt's edition of the Panca (1934), listed by Conze (1960:36) as Paris Bibl. Nat. Skt 68-70. 2 Unless 'dharmas,' in this context, refers only to living beings I do not know what this means. The authorship of the Subodhini appears to be questioned by Conze (1960:55), apparently on the basis of an obscure passage in Tsongkha pa's AAGser: 14.4-6: I ^ S F ^ I ^ " ^ ! ^'•g'^'ash'*^'1^' 4 scheme of the A A here, as i n his other books, but only quotes the first seventeen verses which summarize the chapters and topics. Based on internal evidence it was written after the AAA. 1 F i n a l l y B u ston (Obermiller 1932:156-9) mentions, besides these four works, "also the PP-bhavana, etc." as being written by Haribhadra. There is no other record of these works. 1.2 The theme which runs through Haribhadra's writing is that the A A is equally a commentary on all the main P P sutras (in particular, on the A, Pafica and Sata). T h i s is a particular insight Haribhadra believes to have been vouchsafed to h i m by Maitreya, and is one of the primary reasons he wrote the books he did, as is apparent in the opening lines of his A A S p h u where he pays homage to the PP, "[wanting] to delineate the stanzas of the Ornament [=AA] as an ornament for all [the sutras]," 2 Dignaga, whose PrajM-paramita-pindartha (PPPi:7, Wl2.7-8) Haribhadra quotes, had already made the point that the P P sutras, though differing i n length, embody the same teaching. "In regard to the Samcaya-gatha-vrtti-subodhim, Sunaya-sn says that it is similar in style to the AAA and is therefore H's (slob dpon) work. Rngog bio ldan shes rab and Sthira-pala explain that it is not, saying, 'It is not since it asserts the abhisamayas alone (mngon rtogs rkyang pa ?).' Nevertheless, since in the Marma-kaumudi we find 'It says in H's Subhodini...' (which is probably to refute this [wrong idea of it not being H's work]), I think it is." A few remarks about the names Tsongkha pa mentions. According to the Blue Annals (Roerich 1949:226 and 228) Sunaya-Srl was a contemporary of Abhayakara-gupta. Blue Annals (Roerich 1949:332) refers to Rngog bio ldan shes rab as Mahatman = Bdag nyid chen po; also Obermiller (1932-33:10, n. 5). Roerich (325) says that 'bum phrag gsum pa is the same as Brtan skyong or Sthira-pala, whose unusual name came from his having memorized 300,000 Slbkas (789). Taranatha, (1980:311 n. 14) mentions "a Sthirapala, alias Trilaksa" who is said to have explained the PP at Vikramas'ila, and in the colophon of the AAA, of which he is listed as a translator, he is mentioned as Maha-pandita Dhlra-pala, who retranslated the AAA at a later time with Rngog. 2 ^frg^£^n)a*,g*r<r'^ ^ ' S ' S 3 ^ ' ^ ' 1 ^ ! Reconstructed by Tripathi (1976:3) as ...d«W*K-+|R+Iull*ii I ^f<>H>kull«f f ... The admittedly crucial gloss "all sutras" is based on AAGser:24.6. 5 Thus this A is not short of any topics spoken about [in the longer P P sutras]. It is asserted to be a condensed text. Its topics are just those that have been spoken of there. A n d A r y a Vimuktisena, the most important influence on Haribhadra after Maitreya, had already used the A A to explain the Panca. Haribhadra, however, relates the AA's delineation of the path (marga = inner spiritual development) directly to other P P sutras like the A. H e appears to have reasoned that i f the shorter A includes the entire teaching on emptiness (sunyata) found in the longer sutras, it must also include the teaching on the path which the A A sets forth i n such careful detail. It is difficult to be certain about al l the implications Haribhadra drew fr o m this. Nevertheless, by stressing the intimate relationship between emptiness (the explicit topic of the sutras) on the one hand, and the path (the hidden topic of the sutras) on the other, Haribhadra does seem to be subtly adjusting the meaning of emptiness. H i s concern appears to be to ensure that emptiness, and the perfect wisdom that knows it, are defined in such a way that they do not lose the capacity to give rise to the entire range of mental states, starting from those of the religious neophyte, up to and including the state of a Buddha. Stil l , Haribhadra's contribution is found not so much in original thought. H e was primarily an organizer of others' work, and a master at setting the ideas of other writers in a helpful juxtaposition. One should not, however, on that account make light of his contribution. For, by l i n k i n g together what were previously quite diverse strands of Buddhist religious thought, he brought P P study based on the A A into the mainstream, and initiated a tradition of commentarial literature that continues amongst Tibetans even until the present day. 1.2.2 T h e P a l a kings held sway over north-east India f r o m the 8th to the 12th centuries, after the demise of the Gupta dynasty and before M u s l i m rule finally ended the extended classical period of Indian history. A t the apogee of its power under king Dharmapala, the Pala dynasty controlled an area, centered around the triangle formed by the modern cities of Patna, VaranasI and Gaya, that stretched as far west as present day L u c k n o w and south into the northern reaches i of the D e c c a n (Basham 1954:70). T o the north the H i malayan k i ngdom of Tibet 6 had reached the peak of its secular power under K h r i srong lde btsan (742-794 or 804; Stein 1972:63-64). Some 30 years before Haribhadra composed his A A A , Tibetan cavalry had penetrated as far as the northern areas of the Gangetic plain and received tribute from the Pala rulers. It was an important and fortuitous juxtaposition of events. T h e twilight of Indian B u d d h i s m heralded the dawn of 1000 years of Tibetan Buddhist culture. Tibetan ecclesiastical civilization, with its strong emphasis on tantra and the PP, bears the unmistakable imprint of the Pala dynasty, and the importance of Haribhadra's contribution is fully appreciated only when his influence on later Tibetan scholasticism is fully realized. T h i s Buddhism, which gives retrospective importance to Haribhadra, reached its fullest development i n Tibet, but it began to develop even before Haribhadra's death. T h e scholastic study of the P P i n large monastic universities like Nalanda and Vikramasila accorded Haribhadra considerable stature. There were a number of commentaries and summaries of his work (§4.12), many unfortunately now lost. 7 II HARIBHADRA'S TEXT 2.1 Here, the reader w i l l find (a) a brief discussion of the A A A ' s title, (b) a brief summary of the structure of the commentary and (c) a more detailed consideration of the structure and contents of the first chapter. 2.2. The title vyakhya (Commentary on the "Perfection of Wisdom" With the Light [Provided by Maitreya's] "Ornament for Clear Realization"). The different possible meanings of abhisamaya are discussed elsewhere (§8.1-8.10). Wogihara, Obermiller, T u c c i and Conze usually cite the work as Aloka, and though the title is sometimes emended to Abhisamayalamkaraloka (short final, masculine gender) [Prajna-paramita-vyakhya\ or Aloka, under the influence of titles such as Dhvanyaloka or Madhyamakaloka, there is no manuscript evidence for this emendation. There is some evidence i n the colophon, (W994.20-21) however, for a bahuvrihi compound (abhisamayalamkarasya alokah yasyam), with a feminine ending. In explaining the title, Haribhadra says commentary which has a good future and which makes things clear with the light of the Alamkara C Ornament') for clear realization [i.e., A A ] , is finished." Here the compound tad-aloka-prakaiika is best construed as a genitive tat-purusa, but that does not mean that A A A should also be interpreted as a tat-purusa. In fact, tadaloka-prakas'ika suggests that the AA's light is used in the A A A and hence the title A A A is intended as a bahuvrihi. In the line (W994.19) dr«Ml+-faOTfoft ft<Rkil ^ - T f N ^ t W , "this fine panjika which throws light on ultimate reality has been written by me," which precedes the line just cited, aloka is used as in Dhvanyaloka or Madhyamakaloka, meaning 'light on' or 'illumination o f a particular topic or text. But Haribhadra does not appear to be specifically glossing his title at this point. Hence the title should be taken as meaning, "Commentary on the 'Perfection of Wisdom' With the Light [Provided by Maitreya's] 'Ornament for Clear Realization.'" A n d this is exactly what we find the A A A to be. The f u l l title of the A A A is Abhisamayalamkaraloka Prajha-paramita-(KMH-y+ifJH+i I ^ - m f w - s i n w ww^A sr*ik*rr, "This P P 8 2.3 The structure of the commentary Haribhadra's A A A is an explanation of the A based on the A A , an aphoristic codification of the seventy topics said to be hidden i n the P P sutras. T h e A A A is a long book, almost a 1000 pages i n Wogihara's 1932 edition. It incorporates both the A and A A . In the opening verse of his smaller commentary, the A A S p h u , Haribhadra says that he wishes to demonstrate that the A A is not only an explanation of the longer P P sutras (the Panca and Sata), but equally an explanation of the A and other shorter P P sutras as well. T h i s aim seems to have detenriined the overall structure of his work. H o w is this so? In the A A A , each word of the A, starting f r o m the opening words evam may a srutam ('Thus have I heard') right up to the story of the Bodhisattva Sada-prarudita and the fin a l entrusting (parlndana) of the sutra to Ananda are explained. A considerable portion of the A A A is devoted solely to this task. Based on this, one can divide the A A A along the lines of the thirty-two divisions (parivarta) of the A, a strategy that Wogihara partially follows. T h e divis i o n of the A A A into thirty-two chapters is, however, somewhat superficial. Its deeper structure is given by the divisions o f the A A , first into eight chapters (adhikara) and then into the seventy sub-topics of these eight chapters (listed AA1.5-17). A summary of each of the eight chapters and of the first ten topics (which comprise are the sub-topics of the first chapter) is given below (§2.15, §7. Iff). 2.3.2 That his A A A was an authentic and unoriginal exposition of both the A and A A cou l d have been questioned, i n Haribhadra's own time, only on sectarian grounds, i.e., on the grounds that his systematic d i d not reveal fully the breadth and profundity of the message contained i n the earlier texts. Strictly p h i l o l o g i c a l observations about the proper reading of the earlier texts would have been admissible, but only insofar as they added to the purity of the earlier texts and thereby enhanced their authenticity. Remarks about the earlier texts f r o m a historical perspective would have been quite inappropriate. Such historical remarks would have presupposed a lower status than Haribhadra, as a Mahay ana Buddhist c o u l d have accepted. H e would have had to accept that statements i n the earlier texts could, possibly, be properly accounted for by 9 ordinary human limitations, limitations which the earlier authors, caught in their own time and intellectual climate could not transcend, but which later thinkers, aided by the passage of time and historical comparative methodology could discern. 1 Such a historical perspective would not, in Haribhadra's time at least, have been able to co-exist with traditional faith i n the omniscience of a Buddha and would have obviated, f r o m the start, the perceived need for a book like the A A A . 2.4 In general, the A can be characterized as a mystic-religious work and the A A as a highly scholastic commentary concerned, first and foremost, with the path particularly at its higher levels. T h i s difference i n the general tenor of the two texts is carried over into the A A A , which seeks to explain the former with the help of the latter. Those sections i n the A A A directly explaining the sutra which have no parallel i n the A A are more religious i n tone, more faith-oriented as it were (cf. W8-16). Those sections i n the A A A detailing a topic of the AA, and explaining how that topic is found i n the sutra, are more scholastic and impenetrable (cf. W36.18ff). These latter sections presuppose considerable familiarity with terminology and theories about the path and its structure, some of which may be unfamiliar to the modern reader. W h e n the chapter of the A A naturally corresponds with a d i v i s i o n i n the A (as it does, e.g., i n the case of the first parivarta which ends at W128) the structural tension that comes fr o m trying to combine two different books together i n a single unit is not apparent. Sometimes, however, there is no apparent correspondence, (as i n the case of the fourth chapter o f the A A , W445, which comes i n the middle of the ninth parivarta). A t such points in the text there is a structural tension and the modern reader, i n particular, may come to feel overwhelmed by a mass of what seems to be unconnected information. The thread, which Haribhadra, at least, felt kept the different parts of the work together as a coherent whole may slip from the reader's hand. One of the aims of the first part of this dissertation is to place this thread f i r m l y i n the reader's hand by clarifying (a) the different layers of text which are It scarcely needs mentioning that the absence of such a perspective, which is an ornament of modern scholarship, does not detract from the worth of the AAA. 10 embodied i n the A A A and (b) issues concerning the path current at the time of Haribhadra, and which Haribhadra's comments are indirectly or directly addressing. In regard to the latter, as Obermiller, T u c c i and C o n z e have all pointed out, the A A is primarily an exposition o f the path, and the language of discourse employed by Haribhadra i n his A A A is a language developed to explain and chart that path. There are two different vocabularies i n that language: (a) the terrninology already standardized by Vasubandhu i n his A K and (b) a terminology unique to the A A . Both of these the reader w i l l f i n d explained and used i n this introduction. 2.5 Besides the different layers of textual material i n the A A A there is also the stylistic layering often seen i n Skt commentaries. In each part of the text, whether based on the A or A A , Haribhadra usually first discusses the issue at hand fr o m a general perspective. T h e n he turns to the specific words of the texts and explains them individually. F i n a l l y he makes use of the exegetical devices bequeathed to h i m by bis cultural heritage, particularly c o d i f i e d rules of Skt grammar and the different strategies for giving an etymology to important words (for an example of this use see below §2.14.2-3). 2.6 Haribhadra calls the first chapter of the A A A , (i.e., W21-128), about "the thorough total knowledge of a l l things, i n absolutely every aspect" a "Turn (parivarta) of the Activities of the State of Being a Knower of A l l Aspects." "Here," i n this chapter, he says "just the state of being knower of a l l aspects associated with the Tathagatas turns up (parivartate) again and again by way of [ten] cause and effect 'activities,' starting with production of an [altruistic] mind, which one has to put into practice" (W128.12-15: ^F+k^h8)awhc3 '1 "FTi-cR": "TfWc^c^rr J P T - hR - " ^ : *Nf<*>ktfcfI-^ff-M P<4cf: l). D e l v i n g into Haribhadra's first chapter one is struck by the amount borrowed f r o m other writers. The body of Haribhadra's 'original' work gets smaller and smaller the more one searches for it. T h i s was perhaps, at least partially, the result of the only printing technology available for the transmission of ideas, the handwritten manuscript. Access to the compositions of earlier writers was extremely limited, and a writer knew that either he or she 11 included in the manuscript the entire passage from an earlier writer, or else most other readers would have no access to it. T h e composite text then served a double purpose: it became a sort of Readers Digest and also, perhaps, on the piggy-back principle, served as a vehicle to get one's own work out to a larger audience. Nearly a quarter of the first chapter (approximately 29 of 127 pages) comprises the text of the A; 1 56 verses are quoted from the A A ; large excerpts from A r y a Vimuktisena's AbhisamayMamkara-vrtti ( A A V ) are either directly incorporated (W79.9ff; 84.16ff; 95.1-97.5) or else function like a template, giving structure to Haribhadra's discussion (W24.13ff, 76.21ff). There is a long recast of KamalasTla's T S P (W2.3-3.24). A similar paraphrase of the B K occurs at W93.22-94.1 and M o r i y a m a (1984) has pointed out the parallels i n Kamalasila's Sarva-dharma-nihsvabhava-siddhi. Although quoted explicitly only twice in the first chapter, there are many echos of Vasubandhu's Abhidharma-kos'a ( A K ) , e.g., in the explanation of the twenty members of the Spiritual Community (W35-36). Dharmaklrti's Pramana-varttika (PV) is quoted (W44.25, 67.15) and the Mahayma-sutrMamkara ( M S A ) is quoted frequently, often marking the point of departure for an explanation (W27.10, W31.2, etc.). There are striking similarities between Haribhadra's treatment of the Buddha bodies and the presentation in the Mahayana-samgraha (MS, Lamotte 1973:11, notes p. 49). Other Yogacara texts like Dignaga's P P P i (W23.10), the Madhyanta-vibhaga ( M A V , W110.2) and Vasubandhu's Vyakhya-yukti (W15.25) are also quoted, as are Madhyamika texts: Nagarjuna's Ratnavali (W66.16), Santideva's Bodhicaryavatara ( B C A , W24.8), Jnana-garbha's Satya-dvaya-vibhanga (W45.6) 2 and Bhavya's Madhyamaka-hrdaya-karika (MHK, W40.13). One can, therefore, as I have already suggested, best describe Haribhadra's commentary as an insightful arrangement of earlier writing whose value lies not so much i n a particular 'original' doctrine, as in providing a focus on the PP. Exactly what H included in his original manuscript is, of course, unknown. The text of the AAA, with the complete A embedded within it, is W's creation. 2 See also Hisao Inagaki (1977). 12 2.7 A pattern found frequently i n the first chapter and throughout the commentary, which weaves Buddhist texts representative of a l l four Buddhist schools (Vaibhasika, Sautrantika, Yogacara and Madhyarnika) into an explanation of the PP, is a discussion of the A A based on A r y a Vimuktisena's A A V (Madhyamaka), introduced by a quotation f r o m a Yogac a r a text, like the M S A , which locates the specific part of the path within a general framework. Particular words, which are part of the general Buddhist vocabulary, are then explained with the help of the A K , and problems i n epistemology are dealt with based on the stream o f Buddhist reasoning and epistemology originating from D i gnaga and Dharmakirti. 1 T h e perspective which lends coherence to the variety o f different quoted material is the understanding that the A A is not only a commentary on the longer sutras, but equally on the A. T h i s is Haribhadra's insight, mentioned earlier (§1.2). E v e n though the relationship between the A and A A is by no means self evident and sections of commentary apparently unrelated to the A, and sections unrelated to the A A are frequent, Haribhadra always writes f r o m the perspective of their being related, and, on the theoretical level at least, deals with both books simultaneously. T h i s gives order to what would otherwise be a mass of detailed explanation about seemingly unrelated topics, and indeed, represents his real and lasting contribution. 2.7.2 Haribhadra's discussion of the chapter's first topic, production of an [altruistic] m i n d (W21.22-27.25), illustrates both this perspective and structure. T h e order of the explanation is given by A A 5 a (faTfic<ii<4h«HK^ ^ ) , which is further expanded upon i n AA18-20. T h e corresponding section of the A (W21.22-22.10), which bears little apparent relation to the verses of the A A, says: Eckel (1987) has identified this stream of reasoning and epistemology originating from Dignaga and Dharmakirti as an influence on Jnana-garbha, the teacher of H's teacher Santaraksita. 13 M a y it be apparent for you, Subhuti, how concerning the P P of great Bodhisattvas, the great Bodhisattvas progress in the PP. Haribhadra explains each word in the A in an introductory passage, and then begins an apparently unconnected explanation of the A A which he declares to be an 'expansion' (vistaratah tu, W24.1) on the same, as yet unmentioned, first topic, production of an [altruistic] mind. Haribhadra's explanation of the A A is based on A r y a Vimuktisena's A A V supplemented by a Yogacara text, Maitreya's M S A , and the work of a later Madhyamika, Santideva. A t the end of the explanation, Haribhadra is left with the necessity of relating the section of the A A he has explained to the earlier section of the A. T o do so he quotes the sections of the Panca corresponding to the beginning of the A and cites Dignaga's P P P i verse 7 (W27.24, quoted above §1.2) to justify the procedure: The signifance and importance o f production of an altruistic mind is discussed below (§7.9.2). 2.8 A more detailed description of the first chapter The introductory section of the A A A (Wl-21.22) is clearly separate from the main body of Haribhadra's text. It comprises four distinct segments: a) a short verse homage to earlier gurus b) a complicated highly scholastic discourse on the principles of book writing (anubandha-catustaya) c) a detailed word by word explanation of the opening lines of the A supplemented by a description of the corresponding section in the Panca d) a brief expansion on AA3-17 listing the eight abhisamayas and seventy topics. 2.9 Wogihara mistakenly includes the homage of the A A (W 1.7-10) amongst Haribhadra's own five verses of homage. This verse of homage (discussed in more detail §7.3, §7.5) says: *rr y4*kim HMCWIM S I M M : * I H + I I , *TT Tpfoprr srq^RT-fcrr ^ - M i f e r i 14 Homage to the Mother [=PP] of the Buddha with an assembly of Sravakas and Bodhisattvas: which, as the state of being knower of all, leads Sravakas seeking peace to tranquility; which, as the state of being knower of the paths of those who work for the benefit of the world, is accomplisher of the aims of the masses; [and], perfectly endowed with which, the Sages give expression to this universe in all its aspects. 2.10 Turning to Haribhadra's own verses of homage at the beginning of the A A A , we find that they chart the evolution of a stream in Indian Buddhism that Ruegg (1981) helpfully calls the Madhyaniika-Prajnaparamita synthesis. Haribhadra ( W l . 11-22) mentions five persons: Maitreya, Asanga, Vasubandhu, A r y a Vimuktisena and Bhadanta Vimuktisena as having preceded him: The merciful L o r d Maitreya composed an especially clear brief verse commentary (karika tika) [the A A] pertaining to guidance i n the form of perfect wisdom, so that he himself could fully lead mortals, overwhelmed by the multitude of faults i n the bowels of the world, to the supreme freedom of the yogins [and yoginls] who know that all things are like forms of illusion. T h e leader among the wise, A r y a Asanga, wrote a commentary (bhasya), the fame of which is especially glorious on account of the capability of its author, to ascertain the truly real. 1 Towering on account of his pride because of a minute knowlege of views regarding the division of being and non-being, A c a r y a Vasubandhu attained a [firm] position in the area of describing topics i n [his] well-trodden path. 2 Great was the effort made i n the form of the Vrtti by one with good intelligence, A r y a Vimuktisena, who acquired extraordinary wisdom from the churning of yogic practice with the truly real i n things. Concerned with his Vrtti, another intelligent man is known in the world through the Or, following Bu ston, "a commentary called tattva-viniScaya." Bu ston understood paddhati as the name of Vasubandhu's commentary. 15 designation o f Vimuktisena, who made, i n his Varttika, a treatise which quenched ri v a l opinions, standing i n extremes. O f the writing o f these five persons whom Haribhadra mentions as having preceded him, much has been lost. O n l y the A A of Maitreya and the A A V of A r y a Vimuktisena remain i n Skt. T h e origins of the Madhyarnika-Prajnapararnita synthesis itself are traced back no further than A r y a Vimuktisena who is the first writer, so far as we know, to use the A A as a vehicle for explaining the P P sutras. It is this utilization of the A A when setting out one's views that defines the synthesis, not a particular Madhayarhika or Yogacara doctrine. 2.11 T h e first extant commentary on the A A is the A A V by A r y a Vimuktisena, relatively well known through the critical edition o f the first abhisamaya prepared by Pensa (1967). T h e A A V relates the A A closely to the Panca and ignores the A as a separate work. It is also called Pancaloka (Obermiller 1932-3) reflecting its brief T i b title Nyi khri snang ba. A few details about the life and time of A r y a Vimuktisena are known from the Buddhist historiographers B u ston and Taranatha (Obermiller 1932:155, Ruegg 1968-69:305-7). T h e former reproduces the information contained in the colophon to the A A V where it says A r y a Vimuktisena was a student and relative of a certain Buddha-dasa, and "belonged to the Kaurukulla branch of the A^a-Sarhmitiyas" (Conze 1960:112). Taranatha records traditions which hold that A r y a Vimuktisena was a disciple o f Vasubandhu or Dignaga and was a contemporary o f Bhavya. T h e historical accuracy o f these traditions is not as important as is the presence o f early Tibetan doxographers recognizing a) that an extensive commentary on the A A was extant from an early period and b) that certain o f Bhavya's and A r y a Vimuktisena's doctrines were similar. Haribhadra, i n his AASphu, 1 describes A r y a Vimuktisena as a person "who delineated [the A A ] with an awareness o f abiding i n the middle way." T he Tibetan D k o n mchog 'jigs med dbang po (Hopkins and Sopa 1976:123) calls 1 Amano(1975:3) a ^ ' ^ r v p t ; ^ - ^ ^ ' 5 ' ! ^«'^-=s ,gnj''T^'3JN^C ,| ^g«rg*r<rS»aj'*ufc-aj«i| ^ S ' ^ ' ^ ' ^ ' r l ^ ' ^ ' B l Reconstructed by Tr ipathi (1977:3) 3TPTf^-*TftTsT: Wmt R^Rh< f % fM-^mj ; I ^ cjf JtZWT ^ V - M J WTJTT farrr || 16 him a Yogacara-Svatantrika Madhyamika, who, unlike Bhavya, accepted the Yogacara denial of external existence. In the T i b tradition the views of A r y a Vimuktisena and Haribhadra are considered to be largely similar, and a sub-set of Svatantrika referred to as 'phags seng gi lugs ('the position of the A r y a and Lion') is sometimes mentioned. They are differentiated, however, in terms of whether they do, or do not, incline to the Yogacara assertion of the satyakara and allkakara (§8.2). These positions can be summarized, in brief, as accepting or denying "that an appearance of a gross object to a sense consciousness exists as it appears" (Hopkins and Sopa 1976:108, 123-124). 2.12 Little is known about the author of the Varttika mentioned in Wl.20-22. Haribhadra evidently has some reservations about the understanding of this person. These reservations are made even clearer i n the AASphu: 1 After him, Vimuktisena, who abides on the faith level, d i d not discover all the §astras and made an explanation concordant with his [incomplete discovery]. A c c o r d i n g to Tsongkha pa (AAGser:35.1ff), the faith level is the very lowest part of the Mahayana path, before correct realization is attained. This author is usually called Bhadanta Vimuktisena based on two references (W51.4, W71.15-20). T h e Varttika traditionally ascribed to Bhadanta Vimuktisena is now only extant in T i b (P5186). It is listed as 'phags pa shes rab kyi pha rol tu phyin pa stong phrag nyi shu lnga pa'i man ngag gi bstan bcos mngon par rtogs pa'i rgyan gyi tshig le'ur by as pa'i rnam par 'grel ba and comes immediately after A r y a Vimuktisena's A A V , and before Haribhadra's Le'u brgyad ma (i.e., his recast version of the Paiica). A c c o r d i n g to Bu-ston (Obermiller 1932:155) there is a tradition that Bhadanta was a disciple of A r y a Vimuktisena, and Ruegg (1968-69) says they were roughly contemporary. *5^'a;«rq^g«l||. Reconstructed by TripathI (1977:3) cleft R ^ R b d ^ «IH|u*IH|ui wfrlf: I *T5T-17 T h e first reference to Bhadanta Vimuktisena i n the A A A (WW51.4), " A c c o r d i n g to Bhadanta Vimuktisena, [Srenika] was a renunciate of worldly greed with a comprehension based on either the first or second concentration (dhyana)" gives us no clue about the writer. T h e second reference (W71.7-9) is only slightly more helpful. There Bhadanta Vimuktisena relates each word of a passage i n the A to a different stage of the path. T h e quotation comes immediately after a different interpretation o f the same passage by A r y a Vimuktisena, and to that extent follows the order o f the homage. Neither of these two passages are i n the T i b text, however, a circumstance which l e d T s o n g k h a pa to speculate (AAGser:36.6) that the T i b translation of the Varttika was either by another person with the same name, or else "by a Bhadanta known to be a student of A r y a Vimuktisena." Fin a l l y , the text of the Haribhadra's homage (Wl.21-22) itself raises some questions. It is somewhat unusual for vimuktisena-vacasa to mean simply vimuktisena-namna "called Vimuktisena," though Haribhadra's use of vacasa in the sense of namna 'expression, designation' is in fact confirmed by a similar usage four pages later i n the compound srotra-vijiiananubhava-vacasa (W5.16). Similarly, varttikah (emended to varttike) is problematic. Perhaps it means simply "of whom there is a work associated with the Vrtti" and refers to a student of A r y a Vimuktisena, a conjecture which goes w e l l with the idea that the Bhadanta Vimuktisena quoted later i n the A A A is a different person. Wl.21-22 would then read "Another intelligent man known i n the world by the designation of Vimuktisena, of w h o m there is a work associated with the Vrtti, made a treatise which quenched rival opinions, standing in extremes." 2.13 1 T h e second part of Haribhadra's introductory section (W2.1-5.4) is about the so-called 'four principles o f book writing' (anubandha-catustaya). It is a sort of advertisement for the rest of the book. T h e four 'principles' or limbs are the abhidheya ('subject matter'), prayojana ('purpose' or 'aim'), prayojana-prayojana ('purpose's purpose'-*final goal) and the connection (sambandha) which obtains between them. Haribhadra's remarks, w h i c h he perhaps intended as a direct comment on the A are occasioned by AA1-2, which says: Aklujkar was particularly helpful in clarifying the problems presented by this section of the text 18 M y purpose i n beginning is so that those with awareness would wish to behold the path, (not touched by others), of the state of being knower of all aspects, which has been taught here [in the PP] by the Teacher; and by settling i n recollection the surra's meaning would wish to easily take to the practice of dharma which consists of the ten activities. In doing so he employs the anubandha-catustaya, which, by his time was a standard way for writers of Sastra, especially commentaries, to begin their work. Haribhadra's, however, is a particularly complicated piece o f writing. 2.13.2 T h e origins of the anubandha-catustaya convention are not entirely clear. 1 T h e convention is associated in particular, however, with commentarial works of writers i n the lineage of the Buddhist epistemologist Dignaga. In this respect it is noteworthy that Haribhadra's explanation of the anubandha-catustaya, though based primarily on Kamala-sTla's Tattva-samgraha-panjika, incorporates passages f r o m both Vinlta-deva's and Dharmottara's Tikas on Nyaya-binduilA. In brief, Haribhadra, drawing on Vinita-deva, says the purpose (prayojana) is just the immediate result o f writing a text, i.e., the reader's comprehension of what the author writes. A s Haribhadra says (W4.16-20) Hence easy understanding of a l l briefly stated topics, pertaining to the continuum of what is to be taught, is the uncommon [result] and is, therefore, the purpose resulting from the action [of expressing the PP]. T h e purpose's purpose, or final goal, on the other hand, is what the writer intends the reader to finally attain. A s Haribhadra says (W4.21.-30), That one with accumulated merit due to giving, etc., who has devoted the mind to accomplishing one's own and others' many excellent purposes should, based on the surface level and ultimate truths, actualize the above-mentioned subject matter, the path of the knower of a l l aspects and so forth, i n a l l its variety, by means of the gradual attainment of such [parts of the The formulation appears to be anticipated, though not clearly stated, by Patanjali in his Mahabhasya. 19 path] as the proximate penetration, etc., is the [purpose's] purpose, the result of the result of the action [of expressing]. It has a matching means. Woven into Haribhadra's discussion of the first of the four parts of the anubandha-catustaya, the subject matter, are the opening lines of Arya Vimukti-sena's AAV (W3.25-4.15). There three wrong views about the subject matter of the PP sutras are set forth and refuted. Here is not the place to go into later scholastic speculation about the three wrong views and their refutation. Suffice it to say that, initially, Arya Vimuktisena was probably simply making a general statement about the PP sutras and saying a) that they did not simply systematise everything that exists, b) that they did not simply deal with what removes klesa; and c) that they did not simply list the various aspects of complete knowledge. Arya Vimuktisena's point was that the PP discussed, fully and conclusively, everything pertinent to the liberation seeker. The link between the subject matter and what the reader comes to understand, and the further link between that understanding and gaining the final goal, is what is meant by the sambandha or connection. 2.14 The next part of Haribhadra's introductory section (W5.6-15.23) is a detailed, word by word, explanation of the opening lines of the A. These lines are written in religious language often found in Mahayana sutras, where profundities are buried beneath detailed descriptions, somewhat tedious for the non-believer, of place, retinue and the events occasioned by the Buddha's presence and teaching. Here the considerable body of Mahayana sutras and earlier commentary on the sutras is particularly evident, and, in the later part of the introduction summarizing the corresponding section in the Panca and Sata, where the Buddha emanates light to summon the hosts of Bodhisattvas in other parts of the universe, one enters a miraculous world reminiscent of the Skt Epic and Puranic texts.1 Since I am concerned here w i t h H's thought and his delineation o f the path a more detailed study of this section o f the A A A must be left to a future occasion. 20 2.14.2 Here Haribhadra makes much use of the exegetical tools (mentioned earlier §2.5) supplied by Skt grammar and traditional etymology in this part of the introduction. He explains grammatical forms in the A according to Paninian grammar and his familiarity with the Astadhyayi is evident (e.g., aphorism 3.2.111 at W8.12). Unlike his commentary on the RGS, where 'mistakes' in grammar are explained away as a method to cut attachment to literary style and to focus the reader's mind on meaning,1 the AAA is written on the supposition that the grammatical forms employed in the A and AA are Paninian. 2.14.3 His use of traditional etymology is even more extensive. Unlike modern etymology with its concern for the historical, in this section, as in the rest of the text, etymology (derivation to draw out the essence of the word as the author sees it, would perhaps be a better description than etymology)2 is employed as one further didactic tool in exegesis. When Haribhadra gives the etymology of arhant, first as a participle derived from VarA 'to be worthy' (W9.8-9) and then again as a compound of ah 'enemy' and hata, (the past participle of ^ Ihan 'to kill, to destroy,' W10.18), he does so, not because he is in doubt about which is historically correct, but because the two etymologies are a convenient vehicle to convey to the reader relevant information. The sources of his etymologies are still to be identified. His explanation of the word parvata (mountain) at W8.5 echoes Yaska's Nirukta 1.20, but only faintly. 2.14.4 The importance of this introductory section of the AAA lies in the fact that it begins to point to the main theme mentioned earlier (§1.2): namely that the AA is equally a commentary on all PP sutras. By attaching the description of the corresponding section in the Panca, and by stressing that the difference between the sutras is only the result of Buddha's skill in means, not one of content, Haribhadra is saying that an explanation of one PP surra is a commentary on the others as well. The statement stands out in strong relief P5190:237.1.6-2.2, e.g., gives as one of the purposes of the RGS "the removal of attachment to refined and unrefined language (legs parsbyarba = saihskrta)" G^WJ^'V^'^WW^'Orsraist 2 This has been suggested by Aklujkar. 21 because there is no obvious connection between the first 17 verses of the A A and the few lines at the beginning of the A. 2.14.5 In the final section of his introduction, Haribhadra expands briefly on the AA'stwo introductory summaries: one under eight of the most general rubrics (AA3-4, W16.1-16) and a second more detailed summary by way of the so-called seventy topics (don bdun bcu; AA5-17, W16.17-21.14). This is Maitreya's grand scheme of transcendental psychology. Different aspects of the same stage of yogic development are described from different vantage points, using a variety of technical terminology, drawn from the speculative metaphysics of Yogacara theoreticians (§7.If, §7.9ff). 2.15 The AAA is an explanation of prajna-paramita ('perfect wisdom'). This perfect wisdom is described in terms of eight 'clear realizations'1 (see chart §9.1). The three 'states of being knower' (jiiata) form one group of clear realizations, the four 'practices' (pratipatti) another, and the resultant 'body of truths' (dharma-kaya) a third. The three states of being knower are wisdom in its perfect form. They are taught in the first three chapters as a goal to be attained. The four practices, taught in chapters four to seven, are the same three states of being knower as they are being learned. They are wisdom while it is still being acquired and becoming perfected. The body of truths is the result of the four practices. The eight clear realizations do not, then, each describe a different perfect wisdom, but rather describe the same perfect wisdom from different perspectives. 2.15.2 The scope of each clear realization is so vast that the meaning of the term is better conveyed, intially, by a word like culture. The first chapter is about the Buddha or the Buddhas's mind which is conceived as an over-arching principle defined by transcendent awareness and within which the knowledge of all cultures fit. There is no topic in the other seven chapters not discusssed, at least implicitly, in this first chapter. On the meaning of the word abhisamaya see §8.1. 22 The second chapter is about the Bodhisattva or Bodhisattva's mind, again an over-arching principle since it embraces knowledge of every religious path. The idea is that one must have first hand, experiential knowledge of the states of mind and practices which others (of various capacities and in various stages of development) will need to pursue a satisfactory and beneficial life. The 'clear realization' discussed in the third chapter is associated with the knowledge that leads Sravakas to nirvana. But it is not exactly the same as the Sravaka attainment because Bodhisattvas, who are said to deny themselves personal nirvana in Order to save the world, attain it. Bodhisattva must practice the third clear realization in order to look after those disciples who are inclined to the Sravaka path. Just as Sakya-muni guided his early disciples, apparently by way of the Sravaka path to nirvana, so too, future Bodhisattvas and Buddhas should be knowledgable in those same methods and practices.1 The fourth, fifth, sixth and seventh chapters describe the same three states of being knower as they unfold in practice. They explain the four practices (pratipatti): of 'all aspects' sarvakarabhisambodha, 'peaked' murdhabhi= samaya, 'serial' anupurvabhisamaya and 'in a single instant' eica-ksanabhisambodha. The eighth chapter describes the Buddha bodies (kaya) from the point of view of the result of practice. 2.16. The first chapter of the AAA describes the sarvakarajnata (literally, 'the state-of-being-a-knower-of-all-aspects'). This is the Buddha's omniscience. The ten topics which figure in the description of the state of being a knower of all aspects are listed at AA5-6 (W17.7-11) FHTJ\HI«IVHK«I ^  =rcTf%R;1 *Hf<t>K?idi ii (see chart §9.2). They are dealt with in greater detail below (§7.9-7.9.8). The ten topics may be thought of in three groups: - (a) the production of an altruistic mind, (b) preceptual advice and (c) proximate penetration. - (d) lineage, (e) objective support and (f) aims The differences between the sarvajnata of Sravakas and Bodhisattva is discussed in §7.4, §7.5.2. 23 - the four courses (pratipatti) in (g) armor, (h) going forth, (i) accumulations and (j) emergences. The internal logic to the order of these topics is as follows. Bodhisattvas (a) seek enlightenment to benefit others. They therefore request, and are given, (b) advice (avavada) about the path to attain it. The essence of that advice is that the Bodhisattvas should cultivate perfect wisdom (PP), in essence, understand emptiness. When that understanding is first properly cultivated (c) knowledge becomes proximate (nirvedha-bhagiya) to perfect wisdom. The deep wisdom of the Bodhisattva itself is the focus of this wisdom. The lack of own-being that underpins this wisdom, both when it is proximate to, and when it directly witnesses, ultimate reality is (d) the substratum, basis, lineage or dharma-element. It is nothing but the absence of a substratum with an own-being. It is the Bodhisattva viewed from the angle of ultimate reality, and hence unfindable. It is the reality irrfonning the surface world; a world which, in all its diversity, is (e) the objective support of the Bodhisattva's understanding because there is nothing, good or bad, which is not qualified by ultimate emptiness. Since emptiness is not something other than surface reality (as Subhuti will stress again to Sariputra "in that state of no mind there neither exists, nor does one apprehend either a state of 'there is' or 'there is not'") the objective support and dharma-element are like different sides of one and the same coin. What then is to be attained? It is realization of all these dharmas when one has attained the state of enlightenment. This is (f) the aim {samudde§a). It draws the Bodhisattvas on; yet it is no other than the reality at hand, turned, by a Buddha's skill, into the means to work for the welfare of countless others. The aim is achieved by the four courses {pratipatti). The metaphor is that of a warrior going forth to battle. The Bodhisatvas (g) arm themselves {sarhnaha) with generosity, morality, patience, etc., (h) mount and go forth (prasthana) upon the horse of stabilized mind, and, (i) accompanied by a vast gathering (sambhara) of followers to help in the task, (j) emerge {niryana) victorious in battle onto the ground of enlightenment. These ten topics, each of which includes within its scope the entire Mahayana path and result, illustrate the sarvakarajnata because the explanation of them 24 could have originated only from a person with total knowledge of all aspects of reality. 25 m THE ASTA-SXHASRTKA AND ABHISAMAYALAMKARA 3.1 Haribhadra's most important work is the A A A . It includes within itself the A and A A . Each text was written by a different author (or authors) and each presents the reader with considerable problems. Since it is impossible to penetrate the A A A without first bringing to mind the A and A A , it will be necessary to digress, as it were, into a discussion of these two earlier texts. The intention, in so doing, is to give the reader a better position from which to appreciate the A A A as a work that successfully telescopes different components into a single coherent unit. If, while recollecting the earlier texts, one begins to lose sight of the central concern, that is, understanding the A A A , this is a difficulty the modern reader cannot avoid. The original readers of the AAA knew at least some of the PP sutras by heart, and were learned in the whole range of yogic terminology presented in the A A . Haribhadra wrote his A A A on the presupposition of such familiarity and we, as modern readers, have therefore first to acquire that background knowledge. 3.2 The Asta-sahasrika m m The A is a PP surra. Our now considerable knowledge of the PP sutras is due in large part to Conze, who, over a period of thirty years edited and translated much of the most important primary material.1 The Sata-sahasrika (Sata), Panca and A (PP sutras in 100,000, 25,000 and 8,000 Lines) are the most important of the longer sutras, and the Vajracchedika (Diamond Cutter) and Hrdaya-sutra (Heart Sutra) the most important of the shorter summaries.2 The shortest PP surra is the *EkaksarI-mata (according to Vaidya 1960:xii) or *Sarva-tathagata-mata ekaksara nama [Single Letter Mother of All Tathagatas], consisting of a short introduction describing the place and retinue, followed by the single letter 3f 'a.' The longest is the Sata consisting of sixteen For a complete list of Conze's publications on the PP literature see Conze (1960:127ff, 1977:indices) and Lancaster (1977:indices). 2 For an exhaustive annoted bibliography of all known PP sutras see Conze (1960;31-92). 3 Yig ge gcig ma is no. 32 in Conze's list of tantric PP texts. 26 huge volumes. Between them the various PP sutras are ranged in terms of length. All are anonymous and attributed to the Buddha. 3.2.2 From its earliest stages of development the PP became associated with a worship of the primordial Mother, through its being described metaphorically as the Mother of the Victorious Ones (jinana mata or jina-jananl). Perhaps this ensured a more universal appeal for what might otherwise have become merely a complex philosophical doctrine. For whatever the reason, the PP sutras became the sacred literature of many cultures. 3.2.3 According to Conze (1960:1), the development of the PP sutras as a genre spans a period of nearly seven hundred years, from the initial formulation, ca. 100 B.C.E., until "the period of tantric influence and of absorption into magic" starting ca. 500 CE. As many as nine stages have been discerned in this evolution (Conze 1977:124ff), the first five of which describe an original creative period. There was then a period of Yogacara influence and a recasting of the larger sutras into smaller, more accessible versions, and finally a period of tantric and early Ch'an (Zen) influence. 3.3 The A is one of this group of Mahayana Buddhist texts called PP sutras. It first came to the attention of modern scholarship in 1888 when R. Mitra published his Skt edition of the A in the Bibliotheca Indica series. Mitra's text, re-edited by Wogihara, was republished alongside the AAA (Tokyo, 1932-35) and translated by Conze (1973). In its earliest form the A was probably of a piece with the RGS which summarized its contents in verse. Early Buddhist Mahayana sutras often took this form: a prose section in near perfect classical Skt, and a summary written in hybrid, less grammatically correct, Skt verse. From there the PP sutra developed into the largest branch, and most important part, of the Mahayana revelation. 3.3.2 In its earliest form the A was probably far shorter than the present version. It probably did little more than introduce what were to become key terms: a) great Bodhisattva, b) production of an [altruistic] mind [set on] enlightenment (bodhicittotpada), c) the state of omniscience (sarvajnata), a new goal surpassing the traditional goal of arhantship, and d) perfect wisdom. These terms are discussed in more detail elsewhere (§7.5, §7.9.2, §7.6.2, §7.16.2). 27 Soon after the appearance of the original sutra, further elaboration on the original terms seems to have been added, as were passages designed to prove the authenticity of the revelation. As a counterbalance, as it were, to the original, world-denying doctrine of the lack of self in all dharmas (dharma-nairatmya), passages were added explaining at greater length how the doctrine did not obviate the need for good deeds, for nuturing the doctrine, for skillful means and for high religious accomplishment. Later still, a large body of speculative metaphysics (abhidharma) was incorporated, accounting for the phenomenal growth in the size of the texts. Finally stories about popular mythological religious figures like Aksobhya and Avalokitesvara were added, popularizing the sutras and making them more accessible to the faith of ordinary people. By 150 CE. the A Sutra was probably in much the same form as we now find it. 3.3.3 Lewis Lancaster (1968, 1975) identifies three stages in the later development of the A based on his analysis of the Chinese translations. The first is reflected in the three earliest Chinese translations beginning with Chih lu chia ch'an's (=Lokaksema's) translation finished in about 179 C.E.1 He contrasts it with a second period of development, reflected in the translations of Kumara-jiva and Hsiian Tsang, and a final period when the text is identical with the Tib and Skt versions reflected in Danapala's translation of 985. Two points emerge from his analysis: First, the name of the text is not fixed. Chinese versions of the A have as many as seventeen different names given by what translators felt to be its distinctive doctrine or feature. Second, and more important, key terms found in the later versions of the sutra are missing from earlier versions. The terms dharma-kaya in the sense of an abstract Buddha body, bhuta-kod as a Mahayana term meaning something more than just nirvana, advaya ('non-dual') and dharma-element are not found in the earlier versions. 3.3.4 Haribhadra's word by word commentary on the first abhisamaya makes it clear that Mitra's edition of the A, based on Pala dynasty manuscripts dating to Other early translations were by Chih Ch'ien or An Hsiian 222-9 and by T'an-ma-pai (Dharma-priya) ca. 382 or Dharma-raksa 265-74. 28 ca. 1000 C.E., is nearly identical to the one he was working from. There are a few differences however. At W120.16, e.g., it reads anuprapta eva ayatnena ('obtained without effort') instead of simply anuprapta eva. Haribhadra mentions here, as at other places also, the different readings found in other mss. 3.4 The traditional explanation of the origin of the PP sutras holds that the A, along with all the other PP sutras, was taught in its entirety by the historical Buddha himself on Mount Grdhra-kuta in Raja-grha in what is now the Indian state of Bihar. The variety of sutras differing mainly, though not solely, in terms of length, is explained by the propensity of listeners, some of whom enjoyed extensive, others middling, and others brief exposition of the doctrine. The late appearance of the PP sutras in our world is explained by the fact that the sutras were transported, soon after the Buddha taught them, to the palace of the King of Dragons (naga=mythical serpent-like creatures who recognize, and ferociously cling to, valuable property). There the sutras adorned the crown of the Naga king deep in the ocean, a mythical place midway between the human realm and the world of Buddhas, until Nagarjuna plunged in and, understanding their meaning, out of his compassion restored them to the human realm. Who, say traditional thinkers, if not an Enlightened One, could have written such excellent and profound books? Surely we would at least have heard of the author's name were it not in fact the Buddha. 3.5 The opening chapter of the A (the major part of which is translated in Part III of this dissertation) contains at its core most of the original doctrines. It is presented somewhat like a play with a cast, as it were, consisting of four speaking people: the Lord (bhagavat), Subhuti, Sariputra and Purna the son of Maitrayani. Their names are consistently prefixed by formulaic honorifics which give a grandeur to the procedings. The opening few lines set the scene and describe the retinue. They are a standard literary convention in Mahayana sutras, used to mark the sutra as a Mahayanist work (Rawlinson 1977:7, 13). Then the Lord, who acts as moderator of the discussion, instructs Subhuti to teach the PP. Subhuti clearly has a good grasp of the (new?) doctrine and his discourse is an exercise in non-clinging. There is, he says, no Bodhisattva, no practice of perfect wisdom and no goal of all knowing. A Bodhisattva is not a person, in the usual sense of the word, because no person beyond 'the five 29 appropriating aggregates' (upadana-skandha) is to be found. This much the older Buddhism asserted. But now a Bodhisattva is not even the substratum of the five grasping aggregates, not even the non-existence of a substratum. The word 'Bodhisattva,' we are told, is just a word, naming nothing real at all. The same is true of perfect wisdom, the state of being a knower of all and indeed, all phenomena. All are mere names. Even the Bodhisattva's understanding of selflessness (anatmari), a perfect wisdom which consists in complete detachment from all phenomena, such that all phenomena are seen to be illusory (maya), is no understanding. Raising up his or her heart (cittotpada) to embrace all living beings as his or her own person, the Bodhisattva is aware that even when the goal, the state of all-knowing (sarvajnata) is attained, he or she will simply be like "a skillful magician [who] conjures up a great host of people at the intersection of two highways and then, having done so, may cause that very same great host of people to disappear" (W89.1-21). There are no living beings to be helped, no motivation to help them, and no state of perfection by means of which their needs are met. If Subhuti has a clear grasp of the new doctrine Sariputra has only an intuition. He tries to pin Subhuti down as to exactly what is meant by 'no Bodhisattva,' 'no perfection of wisdom' and 'no attainment,' but Subhuti parries his every question until Sariputra, too, begins to use language, like Subhuti, merely as a tool to pry the mind from cringing. Puma is a minor figure who speaks twice (W91.23, 108.19) and then only briefly. He voices his concern, first as the doctrine seems to be skirting dangerously close to the edges of nihilism, and, second, when he catches the hint of a Mahayana opposed to older Buddhism and conceived in sectarian terms. 3.6 There is a striking contrast between the complete lack of systematization in the A and the detailed, step by step, delineation of the stages of the path, levels (bhumi) and Buddha bodies which one meets with in the AA. This contrast stands out in stark relief in, e.g., the section of the A which Haribhadra relates (W37-74) to the four proximate penetrations (nirvedha-bhagiya; §6.10-13). In the AA, as we shall see, there is a careful analysis of the development of inner spiritual growth, carefully delineated divisions of the path and a name for each paticular stage. In the A, on the other hand, technical terms are almost entirely 30 absent. Even such terms as "the meditative stabilization called 'not grabbed hold of as any dharma'" (sarva-dharmaparigrha nama samadhir, W49.21) are rare. In the A the Bodhisattva's practice is described only in ordinary language with words like coursing, meditating and training. The technical vocabulary of later highly systematized Mahayana Buddhism is almost completely absent. 3.7 The Abhisamayalamkara The AA is an aphoristic, schematic summary of the PP sutras. It also is called Abhisamayalamkara-nama prajna-paramitopade^a-Sastra, e.g., in the title of Haribhadra's smaller commentary the AASphu. It is possible that it is called an upades'a-s'astra ('instructional treatise') because it is viewed as bringing out the doctrine implicit in the PP sutras (la Vallee Poussin 1931-32). Haribhadra considers it to be a codification of all the major PP sutras, but it appears to follow the Panca particularly closely, and the earliest extant commentary we have on the A A, Arya Vimuktisena's AAV, relates it only to that sutra. It was first introduced to Western scholarship by Stcherbatsky and his student Obermiller who together published the Skt text and Tib translation in the Bibliotheca Buddhica Series (1929). Tucci (1932:1-48) produced a helpful schema and indices, and Obermiller (1932-3) then produced a helpful summary of the text and an incomplete detailed analysis (1933-36). Wogihara included the complete text within his edition of the AAA, as did Vaidya (1960), and Wogihara translated the entire text into Japanese. Conze (1954) translated the text into English and added copious parenthetical material in an attempt to make it more intelligible.1 3.7.2 As Stcherbatsky (1929:iv) and la Vallee Poussin (1931-2:404-406 n. 21) pointed out many years ago, this rather enigmatic text raises puzzling questions. The Indian and Tib traditions attribute it to Maitreya and list it amongst his five works (Tib byams gzhung sde lnga): MSA, MAV, Dharma-dharmata-vibhaga, Uttara-tantra (also called Ratna-gotra-vibhaga) and AA. Amongst these five texts, four have commentaries by Asanga. A corresponding commentary by Asanga on the A A is missing. A possible reference (Wl.15-18) to a An undated manuscript by J. Hopkins and Denma Locho Rinpochay, "The Seventy Topics" contains a translation of the Tib version. 31 commentary (bhasya) by Asanga, as well as a commentary (paddhati) by Vasubandhu, in the homage to the lineage Gurus has been mentioned (§2.10), but not only are there no known Skt manuscripts of either text, there are also no Tib or Chinese translations. Moreover neither text seems to be quoted in either the commentary of Arya Vimuktisena, or in any of the works of Haribhadra. Also puzzling is the absence of any Chinese translation of the AA,1 particularly in view of the importance accorded to the PP sutras and the reverence extended to everything associated with Maitreya and Asanga in the Chinese traditions. Obermiller (1933-6:4) has remarked in somewhat choppy English: Indeed, how is it to be explained that the AA should have remained quite unknown to the Chinese? Suppose it would have been inexplicably strange if later on, at the time of the Yuan Dynasty, when the Tib Lama Lo-do-gyal-tsen played such an important part at the court of the Emperor Khu-bilai setsen khan as an ardent propagator of Buddhist science, the AA, which was then doubtlessly regarded by the Tib tradition as one of the most sacred texts, should have not found any entrance into China. 3.7.3 Tucci (1956:28-9) has pointed out that there are in fact two streams of commentary on the PP, one following the seven topics set forth by Asanga in his verse summary of the Vajracchedika and the other the eight topics of the A A. The former continued in India until the time of Haribhadra, but was then largely supplanted by the tradition based on the eight topics of the AA. In China direct explanation of the PP sutras appears to have found favour. Possibly the existence of the Ta chih tu lun (=Mpp§), a direct commentary on the longer PP sutra precluded any further interest in the scholastic speculations of the AA and its commentaries. Nevertheless, the absence of a translation of the AA is surprising, all the more when one considers that some editions of the Panca after Haribhadra (e.g., the Le'u brgyad ma mentioned above §1.1.4) actually incorporate the A A into the Sutra itself. It would seem likely that the text did exist somewhere in an early Jeh-chang Shih, of Los Angeles California has brought to my attention a recent translation of the AA into Chinese (from the Tib?). Nagao (1980:131) believes the AA to have been written later than usually accepted. 32 translation, incorporated within the Chinese translation of some other longer sutra, or under a different title. However it has yet to appear. 3.8 Modern scholarship divides the five texts traditionally ascribed to Maitreya into three groups, finding close similarities between the MSA, MAV, Dharma-dharmata-vibhaga (which are thought to be by the same Yogacara writer) and attributing the Uttara-tantra and AA to different authors. Ruegg (1969:39-70) has summarized earlier scholarship on the authorship issue and discussed at length the possible authors of these texts and the relation between them. This authorship issue is not a particular interest of mine and I have, therefore, not attempted to definitively resolve it. 3.9 The view that Asanga had a teacher Maitreya (either a real human being or mysterious divine creature) who taught him at least some of the five texts is accepted by both traditional and most modern scholarship. The position that I take here, that Maitreya or Maitreya-natha is the author of the AA, reflects that fact. It also accords with the opinion of Haribhadra, as evidenced in his mention of "a verse commentary (thca) composed by the merciful Lord Maitreya" (§2.10), and his comments (W75.17-22)at the end of the discussion of proximate penetration: m*l <PJ a f^ rq" wmm «m-mfwr-^r ^ i ^ i d ^ 3rf^y^^H ' ^K- * iR4> i -? r r ^ f ^ Even though the master Asanga had attained to the realization and had knowledge of the revelation's entire meaning, he did not recognize (in non-repetitious teaching as well) the precise signification of individual words, because of the profundity, and because there is so much repetition. So he could not figure out the PP's meaning and became depressed. Hence, with him in view, Maitreya the Lord explained the PP Sutras and composed the A bhisamayalamk&a-kMika-sastra. Having listened to that it was explained again by the master Asanga and by the master Vasubandhu and others. 33 3.9.2 In passing, mention should be made of a particularly interesting Indian (tantric?) tradition, predating Haribhadra, which connects the AA with Arya Maitreya, one of eight Bodhisattvas in the direct circle of the Buddha. This tradition is recorded in Abhayakaragupta's Marma-kaumudi (P5202:111.2.1ff). According to Abhaya, immediately after the Samgltikara Vajrapani compiled the sutras, the Bodhisattva Maitreya wrote the AA, not later at the request of Asanga. Both Bu ston and Tsongkha pa reject Abhaya's presentation, however, on the ground that the weight of tradition goes too strongly against such a point of view (Ruegg 1969:44). 3.9.3 The traditional view is explained in greater detail by Bu ston (Obermiller 1932:137-139) who says that Asanga, after meditating for eleven years in solitude, met Maitreya in the refined space between the ordinary world of common appearance and the celestial pure realm of the Buddhas. Miraculously he was transported to a divine realm where, in just a few days, he received the transmission of all five texts, and, out of his compassion, returned to this vale of tears to write them down for the benefit of mankind. 3.10 The contents of the AA are presented in eight chapters (adhikara) which —-are again broken down into seventy topics (Tib don bdun bcu). Each chapter describes an abhisamaya ('clear realization'). These have been already been mentioned in the context of the AAA (§2.15), and are discussed in detail below (§7.4f). 3.11 Finally, a comment about the emphasis the author of the AA places on a delineation of the particlar stages of the Mahayana path. 3.11.2 Considering, for the moment, only the first chapter, when one compares the lines devoted to particular topics in the AA with the corresponding sections in the A and Paiica one is struck by Maitreya's particular concern to delineate clearly these stages, and particularly the ten levels (bhumis). Of the 56 stanzas elaborating the ten topics of the first abhisamaya (AA18-73), 23 (AA48-70) are given over to a detailed presentation of these levels. This is surprising given the conception of Maitreya's work as a whole. For in the schema of the first chapter the levels are just one of the seventeen subdivisions of the ninth topic, the course in accumulation (sambhara-pratipatti). Even accepting the historically most 34 defensible position, that the AA was originally a commentary only on the Panca, one finds that the corresponding section in that work is only eleven pages long (Panca:214-224), out of a total of 252 pages. In the A the contrast is even starker: the corresponding section is only four lines (W98.7-11). 3.11.3 The other topic in the first chapter which Maitreya deals with at some length, the proximate penetration (AA25-36), is similar to the levels in that it is a further detailed breakdown of a particular section of the Bodhisattva's path. Maitreya, in his explanation of the proximate penetration, as in his detailed description of the levels, stresses the difference between the Sravaka-yana and Bodhisattva-yana, and the superiority of the latter path. Why, amongst the inner divisions of the Mahayana path, particular aspects are given emphasis in the AA remains to be detenrrined. 35 IV INDIAN LITERARY GENRES AND THE AAA 4.1 We have seen that the AAA, a late eighth century example of commentarial literature of the PP school, incorporates within itself three distinct texts: the A, AA and an exegetical portion explaining those two texts. In this, the AAA conforms to a general pattern found in Indian religious literature, a pattern which incorporates revelation, aphoristic summary and commentary within one text, and, macrocosmically as it were, within each system of thought. Specifically, in regard to the PP tradition itself, the revelation is contained in the PP (§3.2-3.4), a section of Mahayana sutras. The AA is the aphoristic summary (sutra) for the PP school concerned with Mahayana metaphysics. The commentarial tradition associated with this school begins with the AAV of Arya Vimuktisena, continues through the works of Haribhadra and ends, in India,1 with the works of the late Pala dynasty writers Ratnakara-santi (Santipa) and Abhayakara-gupta (Abhaya). 4.1.2 This pattern of revelation, sutra and commentary reflects a deep trend in Indian literary development. It is a pattern that influenced not only the way writers like_Haribhadra structured their texts, but also the way they approached the ideas of earlier writers. An investigation of this pattern in detail will provide the wider context for the AAA (incorporating as it does the A and AA) within the general frame of Buddhist religious literature, a literature which spans a period of about 1600 years on the Indian subcontinent. It will also illustrate how the revelation, sutra, commentary structure, which up until now has been mainly noticed in Brahmancial systems, is equally applicable to Buddhist systems as well. Following this investigation there is also a brief summary of the PP literature in India and Tibet after the time of Haribhadra. 4.2 The structure of Buddhist religious literature parallels what is found in the Brahmanical tradition, and it has even been argued that there was a conscious attempt, on the part of Buddhists, to model an alternative revelation and revelatory language in a sort of viable opposition to the Vedic Brahmanical one For Tibet see §4.12 36 (Despande 1986). Our knowledge of the development of this literature is considerable but still at a less than perfect level. 4.2.2 In both Brahmanical and Buddhist traditions there is evident a similar pattern of literary development. Fundamental is the unquestioned revelation, Brahmanical §mti('act or result of hearing' by extraordinary individuals) or buddha-vacana (literally that spoken by an Enlightened One), which combines popular elements together with theological and philosophical speculation. Although the centrality of this literature is never doubted, it is an open question how much actual importance it had for the systematic philosophers beyond endowing their works with the requisite orthodoxy. Then, crucial in the development of each system, is the aphoristic summary, the sutra.^ In no case can a later author go beyond the aphoristic statements except insofar as he or she can somehow interpret the words to mean what he or she wants. And finally there are the vrttis, varttikas, bhasyas and pkas explaining the meanings, words and concepts of the sutra (and where necessary the actual §ruti or buddha-vacana itself), frequently increasing in size as they incorporate more and more of the material from earlier writers. 4.3 At the heart of each orthodox system there is a hierarchy of revelations expressed through persons free from any blemish or shortcoming, who lost importance as persons. The rsis (Seers) who transmitted the Veda, or the Buddha who taught the truth, were subsumed into the general tenor of the revelation, a circumstance which led to delightful conclusions: ages making Methuselah look as if he died young, and capacities which make God's creation of the world in seven days a mere minor miracle. In the Indian context the human dimension is down-played (Ingalls 1965) and the individuality of the wellspring is always lost either in the concern with cyclic origination or in the multiplication of forms of the divinity. Both Brahmanical and Buddhist traditions accepted a certain body of literature as authentic utterance coterminous with a perfect ultimate source. One must express it thus, and not say 'from an omniscient god' because the Mlmamsakas ~ Vedic hermeneuticists concerned with the rules for achieving correct interpretations of apparently contradictory or layered Vedic utterances -- never understood the Vedic revelation as the utterance of a person, but rather as 37 an unoriginated reality heard by rsis. These rsis transmitted the revelation, through many lineages, to a spiritually capable section of humanity. Unlike the Mlmariisakas, the Buddhists (specifically the later Buddhists) understood their revelation in a more usual way, as having originated from an omniscient being. Yet Buddhists, who always considered special those works said to have been uttered by the Buddha Sakya-muni, continually incorporated so many texts into the body of their revelatory literature one must question exactly what they had in mind by the word 'Buddha.' 4.3.2 It might seem preposterous to question the existence of an historical Buddha or Tathagata Sakya-muni, but based on the evidence we now have it is perhaps not unreasonable to do so in a certain sense. Just as with an historical Jesus, we have no documents which can conclusively be shown to have been transmitted or written by such a being, and even those ideas most commonly associated with the name of a Buddha, the four noble truths and the eightfold noble path, have parallels and perhaps even precursors in popular, Samkhya and Upanisadic texts. Furthermore, the earliest Buddhist texts we now possess are layered both linguistically and stylistically. They contain traces of earlier linguistic forms and parallel, in their descriptions of the life of the founder, standard life stories of other religious leaders such as the Mahavlra. 4.3.3 Snellgrove (1987:8) points out, correctly, that "the notion of 'the Buddha,' as though there were only one, just as there is traditionally only one Christ is entirely a modern nineteenth century Western idea;" and, (43) that it is "impossible to conceive of Sakya-muni in any traditional religious sense as the founder of his religion in the way in which Jesus Christ is certainly the founder of Christianity and Mohammed of Islam." He also says (149) that "the Mahayana sutras ... increased to infinity the number of Buddhas and Bodhisattvas." 4.3.4 That there are spontaneous appearances of the godhead, be it Brahman or Buddha, in forms appropriate to the believer's understanding, was probably closer to the prevailing Indian concept, certainly by the time of the rise of Mahayana. The notion of a body of emanations (nirmana-kaya), which becomes prevalent in later Mahayana as a partial description of Buddha, is an example of 38 this psychological perspective. Throughout Indian religious literature one finds the forms of the omniscient described as a variable of individual human capacities, and the multiplicity of Buddhas, each capable of expressing authentic revelation, fits into this pattern. 4.3.5 The structure of Brahmanical revelation is in general more complex than that of the Buddhist, most probably because of the longer time frame within which it evolved and, perhaps, also because of its application to a wider spectrum of social groupings. At the pinacle of the revelation, in an extremely rarified sphere almost beyond the mythical, as it were, are the ancient collections of hymns (mantra or samhita). Even at a very early period of Indian history the four Vedas (with the Arthava being the last to gain complete acceptance) were firmly entrenched in a position of mystical glory proximate to or even one with the original source or godhead. These Samhitas in their various recensions form the solid undisputed Sruti. Gradually the Brahmanas, primarily direct and indirect explanations of rituals, associated with the Saihhitas achieved a more respectable status as the glory of the original rubbed off on them. In the Aranyakas (a body of literature reflecting the changing concerns of a less socially-oriented social group) the largely exoteric, though occasionally ritualistically esoteric, language of the Brahmanas begins to take on more and more of an esoteric significance. This 'forest literature,' and in particular the intimate instruction of guru to disciple found in the early Upanisads, also attained the mystical status of sruti, partially by the same process as the Brahmanas and partially because they contained unmistakably, in seed form, the sublime insights which were to be worked out in the later systems of Indian philosophy of the Classical period. At the apex of the Brahmancial revelation, then, and covered loosely by the term Veda are the Saihhitas, Brahmanas and the Aranyakas and Upanisads. Apart from some hesitation on the part of the Purva-mlmarhsa to invest full authority in the Aranyakas and Upanisads, this literature is undisputed sruti or revelation.1 Among non-revealed Brahmanical religious texts, the smrti enjoys a status almost equivalent to the highest Vedic revelation, and, on the practical level, is more influential in shaping the religious 39 4.4 In regard to the corresponding Buddhist revelatory literature, the buddha-vacana, there are, it seems to me, four parts. These parts overlap like concentric circles spreading out from a common core or center:' a) In the middle, the point around which the outer parts revolve, are early doctrines and rules: a core literature. This is a recent concept of buddha-vacana suggested by history. b) In the first concentric circle (including within itself all the literature of the core) is early Buddhist canonical literature, the Hinayana sutras. c) In the second concentric circle are syncretic Mahayana sutras. d) In the final outer, possibly unbounded, circle are Buddhist tantras. 4.4.2 It is probably meaningless to search for an original Buddhism beyond the ideas current in Indian society about 500 B.C.E. (Snellgrove 1987:20, Nakamura 1980:60). The notion of a core section of buddha-vacana comes, sensibility of the common man than even the Sniti which, apart from the Aranyakas and Upanisads, is largely relevant only to those Brahmins engaged in sacrificial rites and rituals. These 'works of the tradition' (smrti literally means 'recollection or the recollected') are roughly equivalent in status to the writings of the Church Fathers in Christian tradition. They have no exact Buddhist canonical equivalent though the Buddhist sutra (on the different meaning of the term in Buddhism, see below §4.9) incorporates much smrti-Uke material, and also the quasi-history of the puranas, into its new revelatory form. ' The syncretistic views of the Tib writer Tsongkha pa (1357-1410), particularly as found in his explanation of the sources for the Lam rim literature (Lam rim chen mo 8b-14b2, Napper 1985:27-31), are the basis for this quasi-historical explanation of buddha-vacana. He asserts: (1) that no one teaching of Buddha contradicts any other (2) that all Buddhist doctrine is personal instruction, and (3) that denying revelatory status to certain Buddhist doctrines is tantamount to abandoning the doctrine in its entirety. His views, influenced as they were by the exigencies of late fourteenth century Tib society, nevertheless have the merit of dealing exhaustively with works accepted by Buddhists as Buddhist revelation and, in a manner to which, in all probability, the author of the AAA would have assented. The roots of Tsongkha pa's views are to be found in the Saddharma-pundanka, Kamalaslla's BK and, in particular, Diparhkara Srijnana Atls"a's Prajna-pradlpa (Thurman 1985). Napper (1985:27) sums up Tsongkha pa's contribution in the following words. "Applying standards of reasoning and consistency, Dzong-ka-ba analyzed the vast corpus of Buddhist literature and derived an ordered system that attempts to uphold the validity of all the Buddha's diverse teachings within reconciling seeming contradictions between them." 40 prirnarily, from scholars influenced by the methods which New Testament scholars like R. Bultmann employed in the search for an historical Jesus.1 The largest collection of Buddhist sacred scripture extant in an Indian language is the Pali Tripitaka: the Sutra (Pali Sutta), Vinaya and Abhidharma (Pali Abhidhamma) baskets (pitaka). Discourses attributed to Sakya-muni are included in Sutra. Kerygma-like passages, from which are derived the rules of the Buddhist Order, are included in the Vinaya, Metaphysical speculations, primarily element-lists and their definitions, are included in the Abhidharma.. One traditional view is that Pali, the language of the canon of the Sthavira-vada (= Theravada) school, was the language of an original Buddhism, the word of the Buddha himself, and what he said is enshrined in the Pali canon. This traditional view, while defensible as a method to generate a positive attitude to the canonical works, does not appear to be historically defensible. Sakya-muni's own language, given that a question framed in such terms is meariingful, remains unidentified and therefore cannot be a criterion for determining original core documents. Nevertheless, some pre-canonical Buddhist language is known, and can help us determine which parts of the vast canonical literature are earlier and which later (Lamotte 1967:617-656). Of the collections (nikaya) in the Surra basket, the Maha-vagga, Atthaka-vagga and Parayana sections of the Sutta-nipata , itself a part of the Khuddaka-nikaya, are the oldest. At least some of this section of the Sutta-nipata would seem to contain the earliest Buddhist doctrines. In particular, the doctrine of misery, its cause and cessation found in the last part of the Maha-vagga (the Dvayatanupassana-sutta), the stress on internal righteousness rather than external factors being the arbiter of religious status (a doctrine which appears again and again throughout the Sutta-nipata and Dhammapada), and an unswerving rejection of the social world (renunciation) appear to be original core doctrines. One can determine this on metrical and linguistic grounds, by tracing quotation of earlier parts of the canon in later parts, and by comparing common passages in the different versions of the canon. 1 For an introduction to the voluminous historical scholarship concerning itself with early Christianity see, Perrin and Duling (1982). 41 Scholars like Frauwallner (1956) have also tried to identify a core Vinaya text, the ur-text of the various Vinayas of the later Buddhist sects, by comparing the different redactions in the various canons. But as P.S. Jaini (1977) has pointed out, the Vinaya is of less importance to Buddhists as revelation since its differences, touching on social conventions as distinct from metaphysical issues, do not reflect on questions which touch the heart of orthodox Buddhist beliefs. Most of the central Buddhist beliefs are located in texts placed in the Abhidharma basket. They are all historically much later works. 4.4.3 The notion that early Buddhist canonical literature is the Hrnayana sutras is the outcome of a complex maneouver in hermeneutics. Later Indian writers (Mahayana Buddhists) re-interpreted words like 'Buddha' and buddha-vacana, and described what earlier Buddhist believers had considered to be the exhaustive, authentic, revelation as 'deficient' or 'lacking' (hina). They applied the designation 'deficient' - to the persons toward whom the older doctrines were directed, - to the breadth of philosophical inquiry, - to the fruit and - to a religious practice. The persons were deficient because they lacked the confidence to strive for a goal the Mahayanists considered higher than nirvana. The breadth of inquiry was deficient because the earlier sutras dealt only with the absence of a soul as a distinct entity within the constituent elements (skandha) and not with general emptiness (SunyatS). The fruit was deficient since only personal liberation resulted from following the practices taught in those sutras, not universally beneficial enlightenment. And the religious practice was deficient because it included only the Sravaka-yana and Pratyeka-buddha-yana as distinct from the B odhisattva-y ana. A yana is a vehicle or journey. The internalized revelation is a vehicle (yana) taking the adept to the goal. The earlier internalized revelation was a hinayana ('deficient vehicle') since it was a practice which carried only one person to nirvana, unlike the mahayana ('grand vehicle') which transported the totality of beings to salvation. 42 4.4.4 From the death of the founder until about one hundred years C.E., a series of splits on account of ethical, doctrinal and geographical differences led to a situation in which there were many separate Buddhist sects. The earlier Mahayana sutras probably evolved out of the speculations of the different and by now practically autonomous Buddhist communities. Among Mahayana sutras most central and influential are the PP sutras. Other important branches of the Mahayana revelation are the Saddharma-pundarika, Ratna-kuta and A vatamsaka collections, Lankavatara, Vimala-klrti-nirdeSa and the Samdm-nLrmocana-sutra, to name only some. The Mahayana revelation became as extensive as the earlier "Hrhayana" revelation and eventually grew to astronomical proportions with the incorporation of apparently contradictory teachings and doctrines. A helpful list of the main "Mahayanist canonical sources of the Madhyamaka school" compiled from the works of Nagarjuna and his later disciple Candrakirti is given by Ruegg (1981:7). Each generated a commentarial tradition. During the Classical period, even though one invariably finds a sort of pro-forma genuflection to the accepted revelation, and its source, in both the Brahmanical and Buddhist texts there is a tendency to debunk accepted truths. This scepticism was the outcome, partly of the liberal environment which allowed free exchange of even 'culturally subversive' ideas, and partly of a drift in Indian speculative thinking over many centuries towards questioning the reality of appearances. This willingness to detract from even the most central and important parts of the ordinary person's religious sensibilities, which we see in Nagarjuna's removing special status from Buddhist doctrines as central as the four noble truths, and in Samkara's gentle devaluation of the karma kanda, was there in pre-classical times also. Krsna, in the Bhagavad-gha, e.g., first devalues the 'flowery recitations' of the orthodox ritualist and then even the religious philosophizing (samkhya) of the monk. And in the earliest Buddhist literature there are uncompromising attacks on religious institutions and accepted beliefs. These religious, or perhaps one should say anti-religious, speculations are most often associated with an intellectual elite or with cults cut off, to a greater or lesser extent, from the mainstream society and living in monastic type communities. They are an important element in the Mahayana sutra literature. 43 Another characteristic of the Mahayana sutras is their magical or supernatural component. They are positively caked in the paraphernalia of omniscience with every speaker and listener a part of a pervasive theophany and every utterance loaded with significance and power. The Mahayana sutras are the works attributed to Sakya-muni which earlier Buddhists rejected as spurious, but which later Mahayanists believed to be authentic. Mahayanists did not consider them better, or more profound, than the revelation in the Hrnayana sutras, merely no less authentic. Hence the Mahayana writers were not exclusive in their designs but syncretistic, and from their own point of view, all the sutras were equally revelation, equally directed to persons at various stages of development and capacity, and equally the spontaneous utterances of an enlightened being. From their vantage point in the ocean, as it were, Mahayana writers could afford to be tolerant. Every doctrine and teaching had become a method to direct Buddhists of different propensities and capacites along different tributaries. 4.4.5 In the tantras,1 the notion of revelation is expanded to include the sayings and worship of vast numbers of celestial and quasi-demonic 'Buddhas.' Esoteric practices and beliefs are integrated with fundamental Buddhist doctrines. These are then set forth as original and authentic sayings of one or other of a variety of Buddhas described as manifestations of a primordial Buddha Vajrapani or Vajradhara (Snellgrovel987:134). The time and place of the greatest spread of Buddhist tantra was during the Pala dynasty in north east India between the ninth and twelfth centuries. Although there is no record of H having written a tantric work, the time he lived was a time of tantra. Reference to tantric figures in his writing (e.g., Vajra-dhara at W5.6-8) have already been mentioned. I mention the tantric 'revelation,' at some length, even though the context is a study of the AAA, to give the modern reader enough material to construct something of the interior world of Haribhadra. It is surely impossible to understand a commentary like the AAA without an informed understanding of how Haribhadra approached a revealed text like the PP sutra. One must try to construct, as far as possible, from a knowledge of the historical period and Haribhadra's own statements, a notion of buddha-vacana like the author's own. 44 Small sutra-like texts, ritual formulae and mantras called mula ('root') tantras, together with explanatory tantras, were ascribed revelatory status (i.e., were written or said by the Buddha). The vast commentaries on the various mulas, attributed to adepts with names like Nagarjuna, Aryadeva and Candrakirti were said to be the work of people possessing incredible powers and capacity. If it seems out of place to include such texts in a description of Buddhist revelation, this is due, probably, more to the fact that the study of tantra has been hindered by preconception and prejudice about what religion in general, and Buddhism in particular, should say, than because of any dispassionate scholarly judgement. Winternitz (1934:3-4), e.g., whose considerable contribution to Western knowledge of Indian classical literature should not be belittled, nevertheless writes: I cannot find much wisdom in the gibberish of most of the Tantras... in regard to the strange and filthy language... is it not, to say the least, highly dangerous to use? ... On the whole we find in it [the Guhya-samaja-tantra] the same unsavory rnixture of mysticism, occult pseudo-science, magic and erotics as in most of the other Buddhist Tantras. It is not that Winternitz's description is wrong, just that it makes it difficult for the reader to understand that this 'gibberish' is as much buddha-vacana as is the Dhammapada) According to Taranalha the earliest Buddhist tantras are contained in the two lowest tantra sets.2 Benyotosh Bhattacharyya (1931) and Wayman (1973:15-19) have held that the Guhya-samaja-tantra was written by Asanga and dates back to the fourth century. In his discussion of the historical evolution of tantric Buddhism, from earlier Buddhist and Brahmanical traditions, Snellgrove (1987:117-128) places the emergence of most of the tantras in about Snellgrove (1987) has discussed the earlier opinions of Buddhist Tantra and gives the most researched and dispassionate presentation of this branch of Buddhist revelation. 2 The standard division of the tantras according to Tibetan sources is into kriya, carya, yoga and anuttara-yoga. 45 the eighth century, when the earliest Tib and Chinese translations begin to appear. 4.4.6 Of the earlier, traditional, discussions of the Buddhist equivalent of Brahmancial §ruti the most important, within the context of the study of a Yogacara/Madhyamika writer like Haribhadra, is the so-called three turnings of the wheel of doctrine (dharma-cakra). The theory is found in the Samdhi-nirmocana-sutra (Lamotte 1935) and presents a new hermeneutics based on the presupposition that in seeking the true middle way between eternalism and nihilism the Buddha finally revealed the three natures (tri-svabhava, §5.4-4.5). The different sorts of emptinesses, of three different sorts of characteristic marks (laksana), become a definitive statement around which are ordered the other statements (a) directed towards the eternalistically inclined Hmayanist and (b) directed towards the riihilistically inclined Mahayanist. In the first turning of the wheel of doctrine (based, apparently, on the historical first turning accepted by 'Hrnayana' schools) all dharmas were said by the Buddha to exist inherently (sva-bhavena). In the second taming (this, according to the Samdhi-nirmocana-sutra, is the teaching that seems to be found in the PP sutras) all dharmas were said to lack inherent existence. In the third, and definitive, turning of the wheel of the dharma the intention (dgongs pa =abhipraya) behind Buddha's two apparently contradictory statements is investigated and set forth as the three natures (parikalpita-, paratantra- and parinispanna-sva-bhava). The theory of the three turnings seems to be a-bistorical and is primarily important as a contribution to the rich field of Indian hermeneutics. Nevertheless, as Snellgrove (1978:103) says, "the three turnings of the Wheel of the Doctrine represent a perfectly valid historical development of Buddhist teachings," in the sense that there were in India, to some extent, three stages in the understanding of the Buddhist teachings and the theory of the three turnings mirros that reality to some extent. 4.8 All Buddhist religious literature can be called §astra.y Nevertheless, there is a clear separation of those texts attributed to Buddha. Only they can be called 1 Panchen bsod nams grags pa (AAPdon:81-84), however, considers bstan bcos (=Sastra) and sangs rgyas kyi bka' (=buddha-vacana) to be mutually exclusive categories. 46 sutras, i.e., actual revelation. Texts written by a later saint or scholar are subsumed under the catch-all rubric of sasrra. Amongst these iastras, however, as with the Brahmanical systems, there is a tendency, albeit less systematized, to propel one aphoristic text to a position of prominence. 4.9 In the strict Brahmanical usage of the term, sutra includes only that body of distinctively Indian aphoristic literature which seeks to assemble in a coherent chain the seminal ideas of a particular system.1 In Buddhist Skt the word sutra is used to refer to what Buddha said. It describes texts which are very often the exact opposite of what sutra means in the Brahmanical tradition; e.g., the Sata-sanasrika-prajM-pa^arnita, possibly the wordiest book ever written, can take one hundred pages to say what is condensed into one hundred lines in the A. One informed speculation is that the Pali word sutta, used to describe the discourses attributed to Buddha, is derived from sukta (well-spoken) and quite unrelated to the Skt word sutra. As Skt became a more important Buddhist language, and classical writers coined Skt words to express the ideas originally found in Pali and other now lost vernaculars, sutra came to stand for sutta. The Prakrit equivalent of both sukta and surra would be sutta. Alternatively, the early Indian Buddhists may have used the word to express their belief that no word of the Buddha was superfluous. 4.9.2 In general a sutra is even more compressed than the corresponding metric form which we find in the AA, the karika. It is reflection on the sutra and its extreme parsimony with words that a verse karika such as the AA, with its occasional verse fillers to meet metric requirements, is considered relatively wordy. 4.9.3 Each of the well-known six Brahmanical schools of philosophy (darsanas) of the Classical period had at its source a sutra or a set of aphorisms. The etymology from root so — cognate with the English worrfsew' and the tra suffix of means, i.e., 'a thread' — clearly shows that the Brahmanical usage reflects the original meaning. 47 These sutras are the Nyaya, Vaisesika, Mlmamsa, Brahma/Vedanta, and Yoga sutras.1 In much the same fashion Nagarjuna's MMK is a 'sutra,' as it were, for the entire Madhyamika school; the Madhyam aka va tar a -karika s of Candraklrti (according to Tib descriptions of Indian Buddhism) for the Prasangika sub-school of Madliyamika; Dharmaldrti's PV (cornmenting on Dignaga's Pramana-samuccaya) is the 'sutra' for the Buddhist equivalent to the Nyaya-VaiSesika and the AA is the 'sutra' for the PP school concerned with Mahayana metaphysics. The MSA, MAV and other smaller works of Asanga and Vasubandhu, as a group, are a sort of composite 'sutra' for the Yogacara school. Vasubandhu's AK also clearly attained a special, sutra-like, significance in later times, particularly in China and Tibet. 4.9.5 Both the Brahmanical and Buddhist sutra (i.e., aphoristic summary) present a certain difficulty, a difficulty met with when considering the syle of the AA which is clearly a work of the sutra genre. On the one hand they seem to be the original text providing the impetus and basis for the later commmentarial system. On the other hand, the level of codification and speculation presupposed by the sutra would surely require a long previous history. 4.10 From each of these 'sutras,' as in the Brahmanical tradition and in some cases directly parallel to it, (e.g., the history of the debate between the Nyaya-Vaisesika and the school of Dignaga between the fifth and twelfth centuries, approximately) a body of commentarial literature arose in which each subsequent commentary incorporated the ideas of the earlier writers, either as direct quotation, or, by adapting what they wrote, in such a way that criticisms leveled by other schools would be deflected or absorbed without damage. 4.11 What is the source of the Buddhist revelation's authority? Why, specifically, in both the AA and AAA, is the authority of the A unquestioned? There is no record of Isvarakrsna's Samkhya-karika, or any other Sarhkhya text, having attained to sutra status in its own time. Nevertheless, perhaps because of the lack of other authentic texts from the period, there is a tendency to treat the Samkhya-karika as a sutra for the Sarhkhya school as a whole. 48 4.11.2 Lamotte (1949:80 n. 2, 1983-84:4-15) identifies four stages in the devlopment of the idea of authority (pranianya/pramanatva) in Buddhist revelation. (1) In the first stage, when a text was set forth as being (a) Buddha's, (b) the Samgha's, (c) a group of elders' or (d) very ancient, it had to be checked that it was in the Vinaya and Sutra baskets. Already a slightly less restricted version of this criterion for authenticity is found in the Chinese versions of the Dirghagama where it also says buddha-vacana will not contradict reality. (2) In the second stage, the idea of the person, Buddha, having authority is removed. "Rely not on the person but on what he taught." "Avec le developpement de la litterature bouddhique, le critere d'autorite perd sa force. De plus en plus c'est la valeur mtrinseque d'un texte qui decidera s'il doit etre accepte ou rejete" (Lamotte 1949:80 n. 1, 536-540). (3) In this stage subjectivity becomes a criterion for revelation. The mention of a work as being by Buddha is no longer relevant. The question becomes only 4s it beneficial and useful?' The idea of subhasita, that "all Buddha said is said well" (Suttanipata 3.3) becomes "all that is said well is said by Buddha." (4) Lastly, when confronted with there being no authority for their Mahayana revelation, the Mahayanists say: "It is Buddha's word because it leads one to enlightenment." 4.11.3 The stance of the Brahmanical writer is that a fundamentally authorless (apaureseya) primordial statement of truths which is at the heart of the Brahmanical conception of Sruti does not require, and is indeed quite beyond, validation by even the most profound human religious experience. The sruti is a revelation similar to the Judeo-Christian or Islamic conception of the word of God that does not require further validation beyond that imparted to it by being revealed word (iabda) itself. Just as it would be blasphemous even to entertain the thought that God's word might require further authorization or validation, sirnilarly, Brahmanical systems, at least in theory, considered Sruti to be itself a separate authority (Sabda-pramana), answerable, as it were, only to itself. 4.11.4 Later Buddhist thinkers (e.g., Dharmakirti, PV Svarthanumana 225-231) rejected the Veda because they considered its conclusions, whether ethico- ' V -soteriological or philosophical, to be invalidated by direct personal experience 49 (pratyaksa) or by inference (anumana). Nevertheless, they believed the Tathagata's statements to be authoritative, and justified the apparent double standard by saying that his word (§abda), though not itself self-evidently authoritative (pramana), was nevertheless always in accord with the truth. It derived its power and authority, not from being the Tathagatha's word, but from not contradicting the nature of things. This is also the position set forth in MS A: 1.11 where the characterization is ^ -sr^  ...^fcrt ^  ^  f^rr*rof<T (§4.11.2), and, perhaps, what Haribhadra has in mind when he refers (W34.16) to scriptural authority (agama-pramana). 4.11.4 According to the Buddhists, then, the real domain of the Tathagata's authority was his explanation of the way to nirvana (e.g., PV Pramana-siddhi 34-38). An early canonical passage (Dhammapada 276) says "You yourself must make an effort. The Tathagatas are only preachers." Hence disciples were expected to validate the doctrine by deepening their own religious insights through study and introspection. As Carpenter (1985:186) has said, Buddhist revelation is "based upon a personal religious experience." This is probably the underlying reason why Buddhist sacred scripture is so difficult to delimit. 4.12 The commentarial tradition after Haribhadra The Madhyamika/Prajna-paramita synthesis up until the time of Haribhadra has already been discussed (§1.1-1.1.5, §2.10-2.14). After Bhadanta Vimuktisena there is a gap of perhaps two hundred years until the time of Haribhadra, whose four commentaries (§1.1.2) gave great impetus to the later PP tradition. 4.12.2 Soon after Haribhadra, the PP tradition was divided into two theoretically complementary streams. The first mention of these two streams is found in the AbMsamaya^amkara-bhagavad-prajM-pa^amitopade§a-Sastra-tlka Prasphuta-pada (AAPras) of Dharmamitra, a commentary on the AASphu of Haribhadra. There it refers to "Arya Nagarjuna elucidating the essential meaning of the PP sutras in his MMK, etc." This essential meaning is characterized as profound or clearly taught (gsal bar bstan). In the same place it says Maitreya elucidates the topic of the abhisamayas in a vast (rab tu rgyas pa) discourse (^!c,3rv^-^<w^Irgq-*irw^  *' 50 gawri ...AAPras:65.2.8-66.3.2). These two lineages, the Profound (zab mo =gambhlra) and Vast {rgya che ba =vistara), are both said to explain the meaning of the PP sutras. The former, traced back through the Madhyamika commentarial tradition to the MMK, does so by explicating their direct teaching of §unyata\ and the latter, traced back through the lineage of syncretic PP commentaries mentioned in Haribhadra's homage to Maitreya's AA (§2.10), by setting out in systematic and clear language the concealed teaching of the paths, levels and other metaphysical components subsumed under the general rubric of the eight abhisamayas. 4.12.3 This description of two lineages parallels the Brahmanical, post-Advaita accomodative view of the Mimamsa tradition, which divides Mimamsa into Earlier {purva) and Higher (uttara), though there is no evidence of any direct borrowing. According to this view, in the karma-kanda, or the texts of the purva-mimamsa, the activities of the religious person, prerequisite to penetration to the ultimate truth of brahman are set forth, while in the jnana-kanda, the texts of the uttara-mimamsa, one finds a description of the ultimate truth of the Vedic revelation. The uttara-mimamsa, with its concern for the ultimate, parallels the profound tradition which deals with the ultimate truth of iunyata, the dominating concern of Nagarjuna and his followers Aryadeva, Buddhapalita, Bhavya and Candrakirti. The purva-mimamsa, which delineates, in scholastic detail, the specifics of correct action and the principles of the science of interpretation, parallels the vast tradition, with its concern for systematization and detail as it deals with the conventional realities of the practices and mental development of the Bodhisattva on the religious path. Just as both the uttara and purva-mimamsa are understood by latter syncretists to set out necessary and non-contradictory parts of the Vedic revelation, similarly, the traditions of the Profound and the Vast are said (particularly by Tibetan scholastics) to set out the two (ultimate and conventional) realities in the PP sutras, both of which are necessary and non-contradictory. 4.12.4 None of the Indian commentaries based on the A A written after Haribhadra, most now lost in the original and known only through Tib translations, is a direct commentary on the AAA. They are not, however, on that 51 account irrelevant to a deeper understanding of Haribhadra's text. Some are explanations of the shorter AASphu which is itself an abridgement of the AAA. All are important for an understanding of the later Tib commentaries, where they are quoted and referred to by a set of abbreviations. Access to this later Tib PP tradition, which anyone seriously interested in the Indian PP tradition must gain, requires at least a farniliarity with the names. 4.12.5 The most definitive bibliography of PP so far is Conze (1978). He lists fourty sub-commentaries on the AA and gives, as well, a careful list of the various sutras and the commentaries connected to them. 4.12.6 A helpful scheme for considering the most important of the later commentaries is proposed by Tsongkha pa (AAGser:12.1ff) and followed by Obermiller (1932-3). Tsongkha pa proposes two general divisions of commentary: (a) those which attempt to connect the AA directly to a PP sutra and (b) those which do not. He places in the former category twelve texts, of which Arya Vimuktisena's AAV, Bhadanta Vimuktisena's Varttika and Haribhadra's Le'u brgyad ma connect the AA with the Panca. In this category is also the commentary, by the Cittamatrin Ratnakara-santi (Santipa), called Suddhamatior Suddhimati (Jaini 1979:3). Besides the AAA, connecting the AA to the A is another commentary, by the same Santipa, called Saratama (corrected in some catalogues to Sarottama), a major part of which has recently been edited by P.S. Jaini (1979). Ruegg (1969:65, 139) first mentions Santipa as interpreting the AA in accord with Cittamatra doctrine, and later mentions the opinion of Nya dbon kun dga' dpal (a teacher of Tsongkha pa), that one finds in the works of Santipa a Cittamatra-Madhyamika synthesis (dbu ma chen po). According to Bu ston (Obermiller 1932:31) and Taranatha (299-300) Santipa was a disciple of the tantric adept Naropa and a teacher of the celebrated Diparhkara-srijnana AtlSa who died in Tibet in 1053. The Saratama is written from the Yogacara point of view. Also connecting the AA with the A is the Marma-kaumudi by Abhayakaragupta, now extant only in Tib translation. Taranatha (313-14) accords Abhaya, along with Santipa, generous praise as most important of all the Buddhist writers after the 'Six Jewels' (Nagarjuna, Aryadeva, Asanga, 52 Vasubandhu, Dignaga and Dhamaklrti) because they passed on the doctrine to later times. Besides these seven commentaries connecting the AA to the Panca and A, there are four texts connecting the AA to other PP sutras. DharmaSrl's Sata-sahasrika-vivarana connects the topics of the A A to the Sata. No mention of this Dharmas"ri is found in either Taranatha or Bu ston but in the Blue Annals (Roerich 1949:859) reference is made to "the Kashmiri DharmaSr! called the 'one-eyed,'" the author of a commentary on the Sata-sahasrika and of the Kote-tala. Tsongkha pa (AAGser: 13.4-14.4) doubts this Vivarana is an authentic Indian commentary. Srnrti-jnaha-klrti's Samanarthastabhisamaya-^asana connects the eight abhisamayas to the Sata, Panca, and the AstMaSa-prajM-paramita-sutra? carrying Haribhadra's insight that the A A is an explanation of all the PP sutras to its logical conclusion. Wayman (1983:31-32), following Roerich (1949:60), places his activities in eastern Tibet where he was known for propagating the Manjugrl-nama-samgiti. Tsongkha pa (AAGser:14.4) again doubts the authorship of the Tib version of this commentary on the grounds that Smrti-jnaha-krrti was a learned scholar and the Tib version of the commentary is weak and full of simple errors. Connecting the A A with the RSG are Haribhadra's Subodhini, the small text by Dharmasri mentioned above called Koto-tala and lastly Buddha-jnana's (also called Buddha-jnana-pada and Buddha-frl-jnana; Taranatha:260) Prajna-paramita-pahjika. The nine 'unconnected' commentaries, which do not attempt to apply the A A categories directly to any of the sutras, are (1-3) Haribhadra's AASphu, and its two major commentaries, Dharmamitra's AAPras, and Dharma-krrti[sri]'s Durbodhaloka. Of Dharmamitra, Taranatha says only that he should not be confused with another Dharmamitra who was a disciple of the Mula-sarvastivadin writer Guna-prabha, writer of the Vinaya-sutra and contemporary with Arya Vimuktisena. Dharma-krrti[^ri] seems to be the name of the guru of 1 This should not be confused with the Brhattikas on the Sata and the Sata-pahca-astadaSa in the Bstan 'gyur attributed to Darhstrsena (Conze 1960:33-34) which are important texts for the Cittamatra-Madhyamika synthesis (dbu ma chen po) in Tibet 53 Dlparhkara-Srl-jnana AtlSa who came from Malaysia (suvarna-dvipa, gser ling; Roerich 1949:869, Obermiller 1932-3:11, Conze 1960:122). Bu ston (Lung gi Nye ma) calls him 'Kulanta' which Obermiller (1932-3:11, note 3) amends to a possible Kuladatta. There are also three summaries of Haribhadra's AASphu: Prajnakara-mati's Vrtti-pindartha, Ratna-kirti's Vrtti-klrti-kala and Buddha-jnana's Vrtti-pradlpavaii. The PrajM-paramita-pindartha attributed to Dipamkara-snjnana AtiSa and Kumara-£i[-bhadra]'s work by the same name are both summaries of the A A itself. The last work mentioned by Tsongkha pa, Abhayakara-gupta's Muni-matalamkara, is a long treatise, the last three chapters of which deal specifically with the subjects of the A A. 4.12.7 Brief mention of the Tibetan tradition of interpretation of the PP would not be out of place here. According to the Blue Annals (Roerich 1949:330) there are two main sources: (a) the exposition by Shes rab 'bar of 'Bre (usually called, simply, the 'Bre tradition), the best of the disciples of the famous translator Ngog bio ldan shes rab, whose teaching followed the tradition of the PP as taught during the period of the early spread of the doctrine, i.e., up until the time of Rin chen bzang po (958-1055), (b) the exposition by Byang chub ye shes of Ar (the Ar tradition). In Tibet, Haribhadra had a particularly strong influence on writers of the Sa skya and the [Dga' ldan-»]Dge lugs sects. The Sa skya sect was founded by Dkon mchog rgyal po in 1073 and the [Dga' ldan-»]Dge lugs sect founded by Tsongkha pa 1357-1419. Bu ston (a Sa skya), writing in his Lung gi nye ma, an important 13th century commentary on the AASphu, mentions that there were already too many Tib commentaries on the AA and AASphu to merit mention. This state of affairs in no way inhibited later [Dga' ldan-»]Dge lugs monks, however, who wrote even longer and more minutely detailed commentaries on Haribhadra's work. Most important of these are Tsongkha pa's AAGser, Rgyal tshab's AARgyan and the scholastic manuals (yig ca) written by Rje btsun pa, Pan chan bsod nams grags pa and 'Jam byangs bzhad pa. The largest of the Dge 54 lugs pa monasteries, containi