THE NAMES OF WISDOM. A CRITICAL EDITION AND ANNOTATED TRANSLATION OF CHAPTERS 1-5 OF VILASAVAJRA'S COMMENTARY ON THE

THE NAMES OF WISDOM. A CRITICAL EDITION AND ANNOTATED TRANSLATION OF CHAPTERS 1-5 OF VILASAVAJRA'S COMMENTARY ON THE NAMASAMGITI, WITH INTRODUCTION AN...
59 downloads 0 Views 8MB Size
THE NAMES OF WISDOM. A CRITICAL EDITION AND ANNOTATED TRANSLATION OF CHAPTERS 1-5 OF VILASAVAJRA'S COMMENTARY ON THE NAMASAMGITI, WITH INTRODUCTION AND TEXTUAL NOTES.

Thesis submitted for the degree of Doctor of Philosophy in the Faculty of Oriental Studies by A.H.F. Tribe, Wolfson College

University of Oxford Michaelmas 1994

Figure of Namasamglti (Svayambhunath, Kathmandu)

CONTENTS

Abstract Acknowledgements Abbreviations

v vii viii

I. Introduction 1. 2. 3. 4. 5.

Opening Remarks The NMAA & Modern Scholarship Significance of the NMAA for the Study of Buddhism The NMAA: Author and Date Structure and Contents of the NMAA

1 2 3 9 24

II. Translation of Chapters 1-5 of Vilasavajra's Namamantrdrthdvalokini Conventions Used in the Translation Chapter 1: Chapter 2: Chapter 3: Chapter 4: Chapter 5:

63

On 'The Request for Instruction' On 'The Reply' On 'The Survey of the Six Families' On 'The Method of Awakening According to the Mayajala'' On 'The Vajradhatu-Mahamandala of Bodhicittavajra'

66 90 95 98 117

III. Introduction to the Text 1. 2. 3. 4. 5.

Materials Description of MSS Method of Editing The stemma codicum The Tibetan Translation

146 149 163 169 189

IV. Critical Edition of Chapters 1-5 of Vilasavajra's Namamantrdrthdvalokini Abbreviations and Symbols in the Text, Apparatus and Textual Notes Adhikara 1 Adhikara 2 Adhikara 3 Adhikara 4 Adhikara 5

210 231 236 238 258

Insignificant Variants Adhikara 6. Text Collated to Establish the Stemma Codicum

281 293

V. Textual Notes

303

in

207

VI. Appendices I II III IV

Works and Authors Cited in the NMAA Works Attributed to Vilasavajra (sGeg pa'i rdo rje) NMAA Colophons Namasamgiti Commentaries Surviving in Sanskrit and in Tibetan Translation V NMAA Chapters: Folio Numbers for the Sanskrit Manuscripts and Tibetan Xylograph Editions VI Sakurai's Edition of Chapters 3 and 4 of the NMAA VII The Namasamgiti Commentary, Gudhapadd. VIII Namasamgiti verses 1-41. Text and Parallel Translation.

414 415 417 419

Bibliography

423

IV

403 407 410 412

Abstract A.H.F. Tribe. Wolfson College, Oxford.

The Names of Wisdom. A Critical Edition and Annotated Translation of Chapters 1-5 of Vilasavajra's Commentary on the Namasamgiti, with Introduction and Textual Notes.

D. Phil. Degree Michaelmas 1994

The Namamantrdrthavalokini ('An Explanation of the Meaning of the Namemantras') is an early, and major, commentary on the Namasamgiti ('The Chanting of Names'). Written by the eighth century Indian dcarya Vilasavajra, it survives in the original Sanskrit and in Tibetan translation. The Namasamgiti enumerates the 'Names' of Manjusii, the Mahayana figure embodying wisdom, and it exerted a strong influence on liturgy, ritual and meditation in the later phase of Buddhism in India (750-1200 CE). Vilasavajra's commentary is written from a Yogacara perspective and interprets the 'Names' within an elaborate ritual framework which consists in a mandala that has Manju6ri as its central deity. The central part of the thesis comprises a critical edition and annotated translation of the Sanskrit text of the first five chapters of Vilasavajra's commentary, approximately a quarter of the whole. The critical edition is based on eight Nepalese manuscripts for which a stemma codicum is established. Two blockprint editions of the Tibetan translation are consulted at cruces in the Sanskrit. Their readings, treated as those of any other witness, are incorporated into the apparatus as appropriate. The edition is followed by textual notes. Introductory material is divided into two parts. Matters relating to the Sanskrit and Tibetan materials are discussed in a section placed before the edition. These include a description of the manuscripts, discussion of the method of editing, establishment of the stemma codicum and an assessment of the Tibetan translation.

An introduction to the contents precedes the translation and is primarily concerned with an outline of the ritual structure of the commentary, giving particular attention to chapters 1-5. Evidence concerning the life and date of Vilasavajra is considered, suggesting he should be placed in the latter part of the eighth century. Assessing the work's significance for the study of Buddhism, 1 suggest that it is of historical importance in that it throws light on the process by which Tantric methods were being related to soteriology in this period; and that it contains material, especially in the sadhana of chapter 4, that contributes to an understanding of the development of Tantric forms of Buddhist meditation. The work is also the only known instance of a commentary of a Yogatantra type that survives in Sanskrit.

VI

Acknowledgements A number of people and institutions have aided and encouraged the present work: In Oxford, Drs Paul Harrison and Jim Benson shed light respectively on problems of Tibetan and of Sanskrit grammar; Or David Gellner searched the NepalGerman Manuscript Preservation Project catalogue in Berlin for manuscript titles on my behalf and also greatly assisted my researches in Nepal; Mr Haru Isaacson and Mr K. Tanaka helped me with Japanese material; Dr Michael Aris improved my Tibetan; Dr Gillian Evison, Librarian of the Indian Institute, and Adrian Hale, Librarian of Wolfson College, gave more than generously of their time when faced with bibliographical queries. To all of these, my thanks. Mr B. D. Dangol, Mr B. R. Citrarkar and Dharmaratna Bajracarya smoothed my path through the National Archives in Kathmandu. Others in Nepal to whom I am grateful include Bijay Shakya of Nag Bahal, Dibyabajra Bajracarya and Min Bahadur Shakya. Chinia Tamrakar, of the Department of Archaeology at the Ministry of Education and Culture, generously and unexpectedly sent me the photograph of the Malla period figure of Namasamgiti from the museum at Swayambhunath which I have used for the frontispiece. Drs Paul Williams and Steven Collins taught me as an undergraduate at the University of Bristol and first encouraged me to pursue Buddhological research. I remain very grateful both to them and to Richard Gombrich, Boden Professor of Sanskrit at Oxford, who showed a continual interest in and encouragement of my work. I am indebted to my partner, Linda Moody, for her kindness and forbearance as well as for her pertinent observations on my prose style. Above all I must thank my supervisor, Professor Alexis Sanderson, for the constant and detailed interest he took in my research and for the generosity with which he gave his time throughout its course. The extent to which I have benefited from his scholarship and understanding is inestimable. Financial support from the British Academy and the Boden Fund, Oxford, made this research possible. A grant from the Boden Fund also enabled me to undertake a research visit to Nepal from November 1991 to January 1992.

Vll

Abbreviations

ABBREVIATIONS

Texts AA

Astddhydyi of Panini. Ed. and tr. S. C. Vasu, 1891. Reprinted MBS, 1962.

AAA

Abhisamayalamkaraloka of Haribhadra. Ed. P. L. Vaidya in Astasdhasrika Prajndparamitd, with Haribhadra's commentary called Aloka. BST No. 4, 1960.

AK

Abhidharmakosa of Vasubandhu. Ed. Swami Dwarikadas Sastri. See AKBh.

AKBh

Abhidharmakosabhdsya of Vasubandhu. Ed. Swami Dwarikadas £astri, Bauddha Bharati Series 5-7, 9, Varanasi, 1987. (Includes the Abhidharmakosavydkhyd Sphutdrtha of YaSomitra.)

AKBh.Pou.

L'Abhidharmakosa de Vasubandhu, traduit et annote. Louis de La Vallee Poussin's French translation of the AKBh. Six volumes. Paris-Louvain, 1923-31. Reprinted as MCB, Vol. 16(1971).

AKBh.Pn/.

Abhidharmakosabhdsyam by Louis de La Vallee Poussin. Leo M. Pruden's English

translation of AKBhPou. Four volumes. Berkeley: Asian Humanities Press, 1988. AKBhVya

Abhidharmakosabhdsyavydkhyd of YaSomitra. Ed. Swami Dwarikadas Sastri. Included in AKBh.

AbhayaP

Abhayapaddhati of Abhayakaragupta. A commentary on the Buddhakapdlatantra (Toh. 1654). Unedited MS, National Archives, Kathmandu MS No. 5-10 (NGMPP A 48/2).

AdVaSam

Advayavajrasamgraha of Advayavajra. Ed. Haraprasad Shastri, Baroda, 1927. GOS 40.

ASPP

Astasdhasrika Prajndparamitd. Ed. P. L. Vaidya, BST No. 4, 1960.

KriySam

Kriydsamuccaya of Jagaddarpanacarya. Ed. Lokesh Chandra. A photographically reproduced manuscript. Satapitaka Series No. 237. Delhi, 1977.

GuSama

Guhyasamdjatantra. Ed. Yukei Matsunaga. Osaka, 1978.

GudhPa

Aryandmasamgititikd Gudhapadd. MS in the possession of the Royal Asiatic Society, London. Hodgson Collection, No. 34 (Cowell & Eggling, 25-26).

CSViv

Cakrasamvaravivrti of Bhavabhatta. A commentary on the Cakrasamvaratantra (Laghusamvara). Unedited MS. IASWR microfilm No. MBB-1971-33.

rNyingGB

rNying ma rgyud'bum. Ed. Jamyang Khyentse Rimpoche, Thimphu, 1973.

Trims

Trimsikd of Vasubandhu.

DBS

Dasabhumikasutra. Ed. P. L. Vaidya, BST No.7, 1967.

DhP

Dhdtupdthah. See Otto Bohtlingk, Panini's Grammatik. Part n, pp. 61-64. Leipzig, 1887.

vill

Abbreviations

DhSam

Dharmasamgraha. Ed. Max Mttller, Oxford, 1885. Reprinted in MSuSam 329339.

NispYA

Nispannayogdvali of Abhayakaragupta. Ed. Benoytosh Bhattacharyya, Baroda, 1949. GOS 109.

NMAA

Aryandmasamgltitikd Ndmamantrdrthdvahkini of Vilasavajra.

NS

Manjusrijndnasattvasya Paramdrthd Ndmasamglti. Ed. R. M. Davidson, MCB No. 20, Vol. l,pp. 1-69, 1981.

PancKr

Pancakrama of Nagarjuna. Ed. Louis de La Vallee Poussin, Gand, 1896.

PancKrTi

PancakramatippatiiofPsLTahiia.ra.kstiSi. Ed. with PancKr.

PPA

Prajndpdramitdpinddrtha of Dignaga. Ed. G. Tucci, JRAS 1947, p. 534ff. Reprinted with ASPP in BST No. 4, 1960 ( ed. P. L. Vaidya).

BuBhuS

Buddhabhumisdstra.

BoBh

Bodhisattvabhumi. Ed. Unrai Wogihara, Tokyo, 1930-36.

BCA

Bodhicarydvatdra of £antideva. Ed. Louis de La Vallee Poussin. See BCAP

BCAP

Bodhicarydvatdrapanjikd of Prajnakaramati. Ed. Louis de La Vallee Poussin, Calcutta, 1904-14.

Blue Annals

See Roerich, Bibliography.

MadhyS

Madhyamakasdstra. Ed. P. L. Vaidya, Darbhanga, 1960 (BST No. 10). This contains Candraklrti's commentary on Nagarjuna's Mulamadhyamakakdrika (MMK), the Prasannapadd, and also fragments of his Ratndvali.

MAVibh

Madhydntavibhdga of Maitreya.

MAVibhBh

Madhydntavibhdgabhdsya of Vasubandhu. Ed. G. M. Nagao, Tokyo, 1964.

MMK

Mulamadhyamakakdrika of Nagarjuna. Ed. P. L. Vaidya (in Madhy£), BST No. 10, 1960.

MV

Mahdvastu. Ed. E. Senart, Paris, 1882-97.

MVy

Mahdvyutpatti. Ed. Ryozaburo Sakaki, Kyoto, 1916.

MSam

Mahdydnasamgraha of Asanga. Tr. Etienne Lamotte, Louvain, 1938 (contains Tibetan and Chinese texts). (Reprinted 1973.)

MSuSam

Mahdydnasutrasamgraha. Part 1. Ed. P. L. Vaidya, Darbhanga, BST No. 17, 1961.

YRM

Yogaratnamdld of Kanha. Ed. D. Snellgrove in HVT U, 103ff.

VajAv

Vajrdvallof Abhayakaragupa. Ed. Lokesh Chandra. A photographically reproduced manuscript. Satapitaka Series No. 239. Delhi, 1977.

VijnMaSi

Vijnaptimdtratdsiddhi of Xuanzang. Tr. Louis de La Vallee Poussin, Paris, 1928-9.

SaMa

Sddhanamdld. Ed. Benoytosh Bhattacharyya, Baroda, 1925 and 1928. GOS 26, 41.

SekUdTi

Sekoddesatlkd of Naropa. Ed. Mario E. Carelli, Baroda, 1941. GOS 90.

STTS

Sarvatathdgatatattvasamgraha. Ed. Isshi Yamada, Satapitaka Series, 262, New Delhi, 1981.

IX

Abbreviations

SDPS

Sarvadurgatiparisodhana Tantra. Ed. and tr. Tadeusz Skorupski, MBS, 1983.

SpPr

Spandapradlpikd of Utpalabhagavata. Ed. Gopinath Kavraj, Tantrasamgraha pt.l

(Yogatantragranthamdla 3). Varanasi, 1970. SBSYDJS

Sarvabuddhasamdyogaddkinijdlasamvaratantra. (Toh 366-367.)

HBI

History of Buddhism in India [rGya gar chos 'byung] of Taranatha (1608). Tr. Lama Chimpa and Alaka Chattopadhyaya, 1970. (For edition of Tibetan see A. Schiefner, 1868.)

HVT

Hevajratantra. Ed. and tr. D. L. Snellgrove. Two volumes, London, 1959.

Catalogues Baroda

Raghavan Nambiyar (1950) An Alphabetical List of the Manuscripts in the Oriental Institute, Baroda. Vol. 2 (= GOS 114), Baroda, pp. 1458-1467, "Bauddha-Sastra".

Bendall

Bendall, Cecil (1883) Catalogue of the Buddhist Sanskrit Manuscripts in the University Library, Cambridge. Cambridge.

Bir

Sanskrit Seminar of Taisho University (1954): Buddhist Manuscripts of Bir Library, pp. 1-30.

BSP

SrlNepdlardjakiyavirapustakdlayasthapustakdndm Brhatsiicipatram, Bauddhavisayakah saptamo bhdgah,

khanda 1-3 (Purdtatvapra-

kdsanamdld 29, 38, 39), Kathmandu (1964-66). Cordier

Cordier, P. (1909 and 1915) Catalogue du Fonds Tibetain de la Bibliotheque Nationale, Parts II & HI. Paris.

Cowell & Eggling

Cowell, E. B. and Eggling, J. (1876) 'Catalogue of Buddhist Sanskrit Manuscripts in the possession of the Royal Asiatic Society (Hodgson collection)' JRAS pp. 5-50.

Dhlh

Dhlh, Durlabhabauddhagranthasodhanapatrikd (Review of rare Buddhist Texts). (1986-1993) 16 Vols. Central Institute of Higher Tibetan Studies, Sarnath.

Goshima & Noguchi

Goshima, K. and Noguchi (1983) A Succinct Catalogue of the Sanskrit Manuscripts in the possession of the Faculty of Letters, Kyoto University. Kyoto.

IASWR list

Buddhist Sanskrit Manuscripts: A Title List of the Microfilm Collection of the Institute for Advanced Studies of World Religions. New York. (1975)

IASWR

Bajracharya, M. B. & George, C. S. A Descriptive Catalogue of the Buddhist Sanskrit Manuscripts in the Microfilm Collection of the Institute for the Advanced Study of World Religions. Part 1. (six microfiches).

Abbreviations

Kaneko

Kaneko, Eiichi. (1982) (ed.) Kotantora zenshu kai mokuroku. [A complete cat. of the rNyingGB] Tokyo: Kokusho Kangyokai. Mironov, N. D. (1914) Catalogus codicum manu scriptorum Indicorum

Mironov

qui in Academiae Imperialis Scientiarum Petropolitanae Museo Asiatico asservantur. Fasc. I, Petropoli. Takaoka, Hidenobu (1981) A Microfilm Catalogue of the Buddhist Manu-

Takaoka

scripts in Nepal, Vol. 1. Nagoya: Buddhist Library. Hakuju Ui et al. (1934) A Complete Catalogue of the Tibetan Buddhist

Toh

Canons. Sendai: Tohoku Imperial University. Tsukamoto, K., Matsunaga, Y. and Isoda, H. (1989) (eds.) Bongo Butten

Tsukamoto

no Kenkyu IV, Mikkyo Kyoten Hen IA Descriptive Bibliography of The Sanskrit Buddhist Literature. Vol. IV: The Buddhist Tantra. Kyoto. Yoshizaki

Yoshizaki, Kazumi. (1991) A Catalogue of the Sanskrit and Newari Manuscripts in Asha Archives (Asha Saphu Kuthi), Cwasa Pasa, Kathmandu, Nepal. Kurokami Library, Japan.

Wintemitz & Keith

Wintemitz, M. and Keith, Arthur B. (1905) Catalogue of Sanskrit Manuscripts in the Bodleian Library, Vol. n. Oxford: Clarendon Press.

Other Abbreviations

annot.

annotated

BST

Buddhist Sanskrit Texts. Mithila Institute, Darbhanga, Bihar.

BHSD

Buddhist Hybrid Sanskrit Dictionary. Franklin Edgerton. New Haven: Yale University Press, 1953.

BHSG

Buddhist Hybrid Sanskrit Grammar. Franklin Edgerton. New Haven: Yale University Press, 1953.

ed.

edited

ERE

Encyclopaedia of Religion and Ethics. Ed. James Hastings. Edinburgh & New York, 1908.

Fol. / fol.

Folio(s)

ff.

following

GOS

Gaekwad's Oriental Series. Oriental Institute, Baroda.

IASWR

The Institute for Advanced Studies of World Religions, New York.

IsMEO

Istituto Italiano per il Medio ed Estremo Oriente, Rome.

JA

Journal Asiatique.

JRAS

Journal of the Royal Asiatic Society, London.

MBS

Motilal Banarsidass, New Delhi.

MCB

Melanges Chinois et Bouddhiques, Brussels.

XI

Abbreviations

MS / MSS

manuscript / manuscripts

MW

Sanskrit-English Dictionary. Sir Monier Monier-Williams. Oxford: Clarendon Press, 1899.

NGMPP

Nepal-German Manuscript Preservation Project.

OUP

Oxford University Press.

PTSD

The Pali Text Society's Pali-English Dictionary. London: The Pali Text Society, 1921-25.

SUNY

State University of New York Press, Albany.

Text

Used to refer to the present edition of the NMAA. Followed by chapter and line number, eg. Text 4.52.

rr.

translated

Phoneticizations of Tibetan words Kanjur bKa' 'gyur Tanjur

bsTan 'gyur

Derge

sDe dge

Narthang

sNar thang

XI1

Introduction

Opening Remarks

INTRODUCTION

1. Opening Remarks Vilasavajra's Namamantrarthavalokini (NMAA) - 'An Explanation of the Meaning of the Name-mantras' - in fourteen chapters is one of the first major commentaries on the Ndmasamgiti (NS), 'The Chanting of Names', and the only early NS commentary known to survive in Sanskrit. It is a fairly long work, equivalent to approximately 3000 anustubh verses, written in chaste Sanskrit. 1 That its author is both learned and widely read is witnessed by his reference to classical grammarians and by the range of sources cited in the course of the exegesis. The commentary interprets the NS within an elaborate framework of ritualised meditation, interweaving a Vijnanavada doctrinal perspective. Another NS commentator, Smrtijnanakirti, translated the NMAA into Tibetan in the early eleventh century. After revision it was incorporated into the xylograph editions of the Tanjur (bsTan 'gyur) where it takes up somewhat over 100 folios (T6h 2533) in the Peking edition. The NS itself is a short verse text (167 anustubh verses) accompanied by a prose section describing the qualities of the NS and the benefits accruing to the person who recites and meditates on it. The core verses consist of the 'Names' of Manju£ri, considered as the embodiment of wisdom. The extensive secondary literature stimulated by the NS shows that its admixture of doctrinal, devotional and tantric elements exerted a strong influence on the Buddhist Mahayana tradition (largely in its Tantric form) for more than four centuries in India (750-1200 CE). 2 Vilasavajra's NMAA initiated a distinctive tradition of NS exegesis. The central part of the present work comprises a critical edition and annotated translation of the Sanskrit text of chapters 1-5 of the NMAA, approximately a

lrThe colophon states that the text is equivalent to 2500 verses, an underestimate according to my calculations (see Appendix in for the Skt. and Tibetan text). 2Other early commentators on the NS included Manjus~rimitra and Vimalamitra. See Appendix IV for a list of NS commentaries surviving in Skt. or Tibetan translation.

Introduction

The NMAA & Modern Scholarship

quarter of the whole. The edition is based on eight Nepalese manuscripts and two blockprint editions of the Tibetan translation and is followed by textual notes. Matters relating to the Skt. and Tibetan materials - description of the MSS, method of editing, stemma codicum etc. - are discussed in the Introduction to the Text placed immediately before the edition.

2. The NMAA & Modern Scholarship The NMAA has been neither edited nor translated in its entirety. In 1988 Munenobu Sakurai published an edition of chapters 3 and 4 of the Sanskrit text, accompanied by the Tibetan translation.3 To my knowledge no other chapters have been edited and there is no published translation of any part of the work. Nonetheless, Vilasavajra's commentary has been familiar to modern scholarship since the late nineteenth century. In 1883 Bendall described a palm-leaf manuscript of the NMAA held by the University of Cambridge4 and I. P. Minaev consulted another manuscript in the preparation of his edition of the NS published in Russia in 1887. 5 Minaev also cited the NMAA in an essay published in the same year. 6 In the present century Louis de La Vallee Poussin quoted from the NMAA in the annotations to his translations of Vasubandhu's Abhidharmakosabhasya and Xuanzang's

3 Sakurai, Munenobu. (1988) 'A Study on the Namamantrarthavalokini (1) - A Critical Text of Its Chapters III & IV.' Bunka (Culture) 51, No. 3/4, pp. 323-354. I am grateful to H. Isaacson for bringing Sakurai's work on the NMAA to my attention and providing me with a copy of his edition of chapters 3-4. Sakurai bases his edition on five Skt. MSS, including the Cambridge palm-leaf MS, and two Tibetan xylograph editions. However, of his five MSS one is an apograph and four descend from a single hyparchetype (my 6) that is often corrupt; he also has a tendency to misread the script of the Cambridge MS, often on occasions when it provides the best reading. (For further remarks on Sakurai's edition see Appendix VI.) 4Bendall, C. (1883) Catalogue of the Buddhist Sanskrit Manuscripts in the University Library, Cambridge. Cambridge, pp. 203-4. This remains the oldest known MS of the NMAA (= MS A in the present edition). 5Minaev, I. P. (1887) Buddhizm. Izledovaniya i Materialui. St. Petersburg. Vol. II, pp. 137-159 (for the description of the NMAA manuscript see p. xii). Davidson (1981, 49) states that Minaev had intended to translate the Namasamgiti using Vilasavajra's commentary. ^Buddhizm. Izledovaniya i Materialui. Vol. I, p. 226. Minaev cites a verse from Nagarjuna's Mulamadhyamakakdrikd that is quoted in chapter 6 of the NMAA (comm. to NS 50). Buddhizm. Izledovaniya i Materialui. Vol. I was translated into French in 1894 by R. H. Assier de Pompignan and published under the title Recherches sur le Bouddhisme.

Introduction

Significance for the Study of Buddhism

Vijnaptimatratasiddhi, as well as referring to it in his article on the Adibuddha in Basting's Encyclopaedia of Religion and Ethics.1 More recently, Ronald Davidson used the Tibetan translation of the NMAA as one of the four NS commentaries forming the basis for the annotations to his translation of the NS published in 1981. 8 In 1987, the year before his edition of chapters 3 and 4 of the NMAA, Sakurai published two papers that, describing themselves as studies of the 'Jam dpal gsang Idan School, examine some aspects of these two chapters.9

3. Significance of the NMAA for the Study of Buddhism As stated above, the NMAA is one of the first major commentaries on the NS. It is the only Yoga Tantra commentary known to survive in the original Sanskrit and it may also be the earliest Buddhist Tantric exegetical work of any type surviving in the original. 10

7Although de La Vallee Poussin cites the NMAA on a number of occasions he never gives a title or author, but refers to it as the 'Namasamglti commentaire1 (VijnMaSi 616, note 2) or even 'Le commentaire, non tantrique, de Namasamglti' (VijnMaSi 802, vi.). 8Davidson, Ronald M. (1981) The Litany of Names of Manjusri.' in Tantric & Taoist Studies in honour of Professor R. A. Stein, Vol. 1. MCB No. 20, pp. 1-69. As well as the translation of the NS this contains a valuable introduction as well as an edition of the Sanskrit text of the NS that is a compilation of the editions of Minaev (1887), Vira (1960) and Mukherji (1963). 9Sakurai, Munenobu. (1987) 'A Study of the hJam-dpal-gsan-ldan School mainly based upon the Namamantrarthavalokini (1).' Mikkyogaku Kenkyu (The Journal of Esoteric Buddhist Study) 19, pp. 87-109. Sakurai, Munenobu. (1987) 'A Study of the hJam-dpal-gsan-ldan School mainly based upon the Namamantrarthavalokini (2).' Journal of Indian and Buddhist Studies (Indogaku Buddhogaku Kenkyu) 36.1, pp. 342-5. These two papers (in Japanese) are concerned with NMAA chapters 4 and 3 respectively. I am grateful to Mr K. Tanaka for writing on my behalf to Mr Sakurai who was generous enough to send copies of the above two articles. 10Unfortunately all the major commentaries on the fundamental Yoga Tantra, the Sarvatathagatatattvasamgraha (STTS) - Anandagarbha's Tattvdloka (Toh 2510), £akyamitra's Kosaldlamkdra (Toh 2503), Buddhaguhya's Tantrdrthdvatdra (Toh 250) - survive only in Tibetan translation. The description of the NMAA as a 'Yoga Tantra commentary' is based on its ritual content being centred on Mahavairocana and the Vajradhatumandala, though it also contains elements characteristic of works that came to be designated as Mahayoga or Yogottara (see below). In the Tanjur it is classified as a Yoga Tantra commentary (Toh 2533). The NS itself is classified as a Mahayoga Tantra in the rNyingGB (see Kaneko, 1982, No. 196) and the Skt. colophon of the NS states that is part of the Mahayoga Tantra, the Mdydjdla (Toh 466). Bu ston classified the NS as a Non-dual Tantra, due to its association with the Kalacakratantra, but the majority of the NS commentaries in the Tanjur are in the Yoga Tantra section, although others are found in the Mother Tantra or Father Tantra divisions of the Anuttara Tantras. Other possible contenders for the position of earliest surviving work of Buddhist Tantric exegesis are Indrabhuti's Jndnasiddhi and Anangavajra's Prajnopdyaviniscayasiddhi (Two Vajraydna Works,

Introduction

Significance for the Study of Buddhism

I believe the NMAA can be dated to the latter part of the eighth century with a degree of certainty (see the section below). If this is so, it was written at a time of important transition in the development of Buddhist Tantrism. Tantric methods - the use of mantras, mandalas, rites of consecration - were being related to soteriological rather than purely instrumental goals. The notion that enlightenment itself could be achieved through such means is first clearly articulated in the Sarvatathdgatatattvasamgrahasutra (STTS), a work that despite being termed a sutra came to be seen as the fundamental Yoga Tantra. The first commentaries on the STTS and related Yoga Tantra works were produced in the eighth century and like the NMAA they were concerned to elaborate the liberative potential of the Tantric approach. Vilasavajra's work thus falls into a period where for the first time there was a considerable literary output that sought to encode and interpret Tantric ritual in Mahayanist doctrinal terms, and his NS commentary strikingly exemplifies this process. The NMAA demonstrates not only how ritual structures can encode doctrinal positions but also illustrates how doctrinal categories may be modified to accommodate them to ritual structures. The NMAA falls at another point of transition in the development of Tantra in Buddhism, that of the appearance of the Yogottara Tantras, as they came to be called by later Indian Buddhist commentators. In these Tantras the Tathagata Aksobhya and

ed. B. Bhattacharyya, 1929) though the dates of neither author are at all certain. Bhattacharyya's dating of Anangavajra and Indrabhuti to the first quarter of the eighth century seems much too early. It is based on a teacher-disciple lineage (guruparampara) given in the dKar chag (index) volume of the Peking edition of the Tanjur (see Cordier XLVI 2-8) in which Anangavajra is said to be the teacher of an Indrabhuti. Bhattacharyya takes this Indrabhuti to be Indrabhuti the Great rather than one of the later ones. It is more likely that the Indrabhuti (III) being referred to is the brother of Laksimmkara (given next in the parampara), who was living in the late ninth century. Assuming one can rely on the Tibetan dKar chag guruparampara (which is by no means certain: see the final note in Appendix II) this would give a date of the ninth century for Anangavajra. This certainly accords with the Yoganiruttaratantra content of the Prajnopdyaviniscayasiddhi Indrabhuti III may well also be the author of the Jnanasiddhi since it is not clear that either Indrabhuti the Great (teacher of Padmasambhava) or Indrabhuti (II) of Zahor were authors, although they were of importance in Tantric lineages (see Dowman 1985, 232-234).

Introduction

Significance for the Study of Buddhism

deities associated with him come to stand as the central embodiment of enlightenment. The deities of Aksobhya's vajra family are often wrathful in appearance and thus in many of the Yogottara Tantras the status of these figures is greatly enhanced. Though they are not yet depicted in terms of the cremation-ground (kapalika) symbolism that came to dominate the later (yoganiruttara-) Tantras and the final period of Indian Buddhist Tantrism, they can embody enlightenment rather than be a servant and protector of it, or the medium for the acquisition of mundane (laukika) attainments. 11 In the NMAA the central deity in Vilasavajra's ritual structure, Manjus"rijnanasattva, is peaceful in form; yet the wrathful forms are portrayed as embodiments of his wisdom and in this respect the NMAA can be seen, like the Yogottara Tantras, as standing midway between the perspectives of the Yoga and Yoganiruttara Tantras. Though Vilasavajra distinguishes only three categories of Tantra - Yoga, Carya and Kriya - he cites a number of Yogottara works such as the [Guhyajsamdjatantra, the Vajrabhairavatantra and the Mayajalatantra, particularly in chapter eight, which is concerned with the Mirror-like Awareness (adars"ajnana) of Aksobhya. Study of the NMAA should shed more light on this phase in the development of Buddhist Tantra, which is as yet poorly understood. One of the most important genres of Tantric literature is the sddhana, descriptions of the methods by which the Tantric practitioner (sadhaka) achieves the goal of either Liberation or some lesser attainment (siddhi). The central feature of a

11 See A. Sanderson, Vajraydna: Origin and Function, note 1, for a description of the main divisions of Buddhist Tantras and their distinguishing features. I do not employ the commonly used classification of 'Anuttarayoga Tantra'. Sanderson has pointed out elsewhere - see The Dependence of the Herukatantras on the Saiva Tantra of the Vidydpltha - that the expression anuttarayogatantra is not found in any of the surviving Skt. enumerations of the different classes of Tantras. The term yoganiruttaratantra, however, is found (see note 47 below for examples); it is likely that anuttarayogatantra is thus an incorrect back-formation from the Tibetan rnal 'byor bla med kyi rgyud. 'Yoganiruttara' denotes Tantras of the Mother (ma rgyud) sub-division of Bu ston's Anuttara class (bla med kyi rgyud), whereas Yogottara denotes those of the Father (pha rgyud) sub-division. 'Anuttarayoga Tantra' is often (mis)used to describe both of these.

Introduction

Significance for the Study of Buddhism

sadhana is typically an evocation of a deity through visualisation and mantra recitation. Around the evocation are described ritual preliminaries that function to orient the sadhaka's state of mind and post-evocatory and concluding acts. Over time sadhanas came increasingly to follow a set pattern, with variety consisting mainly in the appearance and mantras peculiar to the evoked deity. 12 This standardisation was partly encouraged by the commentarial application of a number of interpretative categories to the various phases of the sadhana, leading to writing that conformed to exegetical expectations. The fourth chapter of the NMAA consists of a lengthy sadhana that forms the basis of the ritual aspect of Vilasavajra's commentary. The material in this chapter illustrates the interaction and mutual influence of ritual and doctrine mentioned previously, but the sadhana is also valuable as an early example of the genre in the original Sanskrit. 13 Study of its structure and terminology should further understanding of sadhanas of the Yoga Tantra type as well as of the development of sadhana literature as a whole. I touch briefly on this later in the introduction, summarising the organisation of the NMAA sadhana and examining the use of one of Vilasavajra's structuring categories, the Five Realisations (pancabhisambodhayah). Central to the NMAA is the figure of Manjus"ri. In Mahayana Sutra literature ManjuSri is familiar as the Bodhisattva pre-eminently associated with wisdom where he often functions as the major interlocutor14 as well as spiritual friend (kalyana-

12One consequence of this was that parts of sadhanas could be written in abbreviated form with no more than an allusion to some phases. 13The STTS and SDPS, the two Yoga Tantras known to survive in the original Skt., contain descriptions of mandalas and their production as well as accounts of consecrations (abhiseka) but do not contain sadhanas. It may therefore be the case that the NMAA sadhana is the earliest surviving in the original. 14Manjusri's role as £akyamuni's interlocutor (eg. the Saddharmapundarika Sutra chapters 12 & 14) shades into that of becoming his chief spokesman (eg. the Vimalakirtinirdesa, the Saptasatikaprajnaparamita Sutra, and the Manjusribuddhaksetragunavyuha Sutra).

Introduction

Significance for the Study of Buddhism

mitra), 15 a converter of beings to the Buddhist Dharma16 and an object of devotion and meditation. 17 His status is usually that of a Bodhisattva of the tenth bhumi, though in some texts he is described as the teacher of Buddhas and even as a fully enlightened Buddha. 18 In the NMAA ManjuSri, as Manjus'rijnanasattva, is equated by Vilasavajra with the non-dual wisdom (advayajnana) that is productive of the enlightenment of the Buddhas. He is therefore at the heart of all the Tathagatas both metaphorically and in terms of the visualisation of the sadhana of chapter 4 . In NS 100 ManjuSri is named as the Adibuddha and consistent with his interpretation of Manju^ri's nature Vilasavajra takes it that the Adibuddha is one of ManjuSri's aspects or forms (ie. the Adibuddha has Manjus'rijnanasattva as his nature). The depiction of ManjuSri as underlying wisdom, which is present in the NS, although unelaborated doctrinally and ritually, comes into its own only in the Tantric phase of the Mahayana. Although his role and status here have been little explored, ManjuSri continued to be an important figure throughout the period of the Tantras. The NS and the NMAA provide material for a fuller account of Manjusri's portrayal in this literature. 19

Gandavyuha Sutra is probably the most important source for showing ManjuSri in this role. Other sources are the Ajdtasatrurdja Sutra which describes him as the spiritual friend of the Bodhisattvas, and the Drumakimnarardjapariprcchd Sutra in which AjataSatru is told he has the great advantage of having Manju&i as a spiritual friend (Lamotte 1960, 95). 16In chapter 12 of the Saddharmapundanka Sutra Manjusn is said to have converted innumerable beings to the Dharma after visiting the underwater palace of Sagara, king of the ndgas. The Ratnakdrandavyuha Sutra describes how Manju&i converts some of the followers of the Jain teacher Satyaka Nirgranthaputra (Lamotte 1960, 39). 17Two important sources for this role are the Manjusribuddhaksetragunavyuha Sutra and the Manjusnparinirvdna Sutra. Manjusri's origins are obscure: unlike Vajrapani, for example, he does not appear in non-Mahayana literature, though Marcelle Lalou (1930, 66-70) has suggested that the Gandharva Panca&kha may be his antecedent. 18According to Lamotte (1960, 29), in the Angulimaliya Sutra MafijusrT is said to be a Buddha of the present, though not of our universe. He is the Buddha of a universe named Nityapramudita ('Always Happy'). 19While ManjusYrs portrayal in the Sutranaya has been examined by Lamotte in his monograph Manjusn (T'oung Pao 48, 1960, 1-96) his role in Tantric literature has received little attention. The observations of Lalou (1930) and MacDonald (1962) (part of their work on the Manjusnmulakalpa) and Raoul Birnbaum's excellent monograph on Manjusn in China (Studies on the Mysteries of

Introduction

Significance for the Study of Buddhism

The NMAA also contains a number of citations of other Buddhist texts, some of which survive otherwise only in Tibetan or Chinese translation. Vilasavajra cites a wide range of Tantric sources - some twenty-eight different works - and there are quotations from Prajnaparamita and Cittamatra scriptures as well as from authors such as Nagarjuna, Candrakirti and Vasubandhu (for the list of works and authors cited in the NMAA see Appendix I).20

Manjusri, 1983) are exceptions. ManjusYI retained an importance in later Tantras as well as in Kriya and Yoga Tantras. As Manjuvajra he is one of the central deities of the Guhyasamajatantra and in wrathful form is identified as Vajrabhairava and the Black and Red Yamaris. (Could the sddhana of the NMAA - Manju&ijnanasattva being at the heart of all the Buddha-families - have paved the way to interpreting Manjus'ri as being, in the form of Manjuvajra, the essence of the vajra family? The connection is not implausible, especially since Vilasavajra is said to have been the teacher of Jnanapada, the founder of one of the major schools of Guhyasamdja exegesis.) Forms of Manjuvajra are also found in the Yoganiruttara Tantras. In the 7th patala of the Abhidhdnottaratantra he appears out of Vajrasattva as the consort of Vajravarahl, saffron-coloured, six-faced and twelve-armed (I am grateful to Prof. A. Sanderson for this reference). 20Among the citations of Tantras a number are attributed to a Samvaraftantra] and elsewhere a CakrasamvaraftantraJ and Satprajndnayasamvaratantra are referred to, though not quoted from. Davidson (1981, 7-8) takes these to be citations of the Laghusamvaratantra (Toh 368) and as therefore demonstrating that the Yoginitantras (ie. Yoganiruttaratantras) existed in the latter half of the eighth century, well before the earliest dateable commentary. Prof. A. Sanderson informs me, however, that he has been unable to find any of these citations in the surviving Sanskrit of the Laghusamvaratantra. Mr K. Tanaka has examined four of the Samvara citations and has identified them as coming from the Sarvabuddhasamdyogaddkimjdlasamvaratantra (SBSYDJS; Toh 366-7) (see Appendix I for details). This Tantra, with two six-family mandates, one centring on the figure of Heruka, functioned as the exemplar for the mandala of the Hevajratantra and can be regarded as the original Yoginitantra (Yoritomi 1990, 693-716: English summary). It must have been composed no later than the early eighth century since at some point between 746 and 774 CE it was summarised by Amoghavajra in his Jin-gang-ding-jing-yu-jia-shi-ba-hui-zhi-gui (Taisho No. 869), a work on an enumeration of eighteen Tantras regarded as forming an enlarged version of the Kon-gd-cho-kyo (the Sarvatathdgatatattvasamgraha) in which the SBSYDJS is placed ninth (see K. Tanaka 1992, 276). Also, according to the rNying ma pa tradition it is one of the eighteen Tantras of the Mahayoga cycle translated into Tibetan during the first propagation (sNga dar) of Buddhism in Tibet. Vilasavajra was aware, therefore, of at least one Tantra of the Yoginitantra class, though not the Laghusamvara; nonetheless these 'Samvaratantra' citations are consistent with the view that all the citations in the NMAA are of works known or thought to be available in the latter part of the eighth century. A citation from the Mahdvairocandbhisambodhitantra in chapter 6 has been edited by M. Sakurai and published as an appendix to K. Tsukamoto et al. (ed.) Bongo Butten no Kenkyu W, Mikkyo Kyoten Hen / A Descriptive Bibliography of The Sanskrit Buddhist Literature. Vol. IV: The Buddhist Tantra. Kyoto, 1989.

8

Introduction

Author & Date

4. The NMAA: Author and Date A number of rather complicated issues surrounding the question of Vilasavajra's name should be considered before examining the evidence concerning his life and date. Three of the ten manuscripts of the NMAA (MSS ABH) used for the present edition give the author's name as Acarya Vilasavajra in their colophon. According to the stemma codicum (see the Introduction to the Text) the name has descended from the archetype (£}) since it is present in MSS A and B.21 The Tibetan translation of the NMAA gives the author's name - in translation only - as sGeg pa'i rdo rje, who is described in the Blue Annals as the teacher of Buddhajnana (8th century). (Buddhajnana, also known as Jnanapada, was a pupil of Haribhadra and founder of the eponymous Jnanapada tradition of Guhyasamdjatantra exegesis.) In his catalogue of the Peking xylograph edition of the Tanjur, Cordier gave 'Lllavajra' as the Skt. of sGeg pa'i rdo rje under the entry for the Tibetan translation of the NMAA, with 'Vilasavajra' as an alternative name.22 Since then sGeg pa'i rdo rje has been referred to as Lllavajra by a number of writers.23 Roerich, in his translation of the Blue Annals has 'Lllavajra' for two of the three references to sGeg

21 H is contaminated at this point. Its reading, krtir iyam dearyavildsavajrasya (H.77v7), is an addition in a second hand placed at the end of the original manuscript. 22Cordier 1909, 265. It is not clear from where Cordier derived the reconstruction 'Lllavajra 7 since none of the works attributed to sGeg pa'i rdo rje gives a transliterated Skt. form. 'Lilavajra' (sic), given in transcription, is found as the name of a translator on two occasions (XXII 35, XLII 5) and 'Lllavajra', again in transcription, is given in the index volume (dkar chag) as the name of the author in two works ascribed to Lalitavajra (also in transcription) in their colophons (XLIII 4, 5) but none of these have sGeg pa'i rdo rje as the Tibetan translation. Lalou, in her index to Cordier's catalogue inherits Cordier's identification of sGeg pa'i rdo rje and Lllavajra (Repertoire du Tanjur, 162). Cordier also gives 'Visvarupa' and 'Varabodhi' as different names of sGeg pa'i rdo rje, reconstructing them from the Tibetan translation's colophon to the NMAA. A possible source for 'Lllavajra' may be Schiefher's 1869 translation of Taranatha's rGya gar chos 'byung where (p. 214) he gives 'Lllavajra' as a reconstruction for sGeg pa'i rdo rje. The names Visvarupa and Varabodhi (= Agrabodhi) are discussed later in this section. 23Eg. Roerich, Blue Annals (see following note); HBI 271-272; Wayman 1977, pp. 56-57, 94, 107; Dowman 1985, pp. 42, 185, 203, 370, 392. However, Ronald Davidson who consults the NMAA for his translation of the NS uses the name Vilasavajra, following the colophon of the Cambridge manuscript described by Bendall. In his valuable introduction Davidson discusses the name and identity of Vilasavajra in some detail (Davidson 1981, 6-8).

Introduction

Author & Date

pa'i rdo rje but uses 'Lalitavajra' for the third.24 I take 'Lalitavajra' here to be a slip on Roerich's part since elsewhere he gives 'Lalitavajra' for Rol pa'i rdo rje, who is not the same figure as sGeg pa'i rdo rje in the Blue Annals, being elsewhere referred to as a disciple of both Maitripa, the llth century contemporary of Naropa25 , and Naropa's teacher Tillipa (= Tilopa).26 This might help account for 'Lalitavajra' being also found as a Skt. reconstruction for sGeg pa'i rdo rje in more recent writing.27 However, it may be that the disciple of Tilopa is not the only Lalitavajra. The colophon to the Tibetan translation of the Vajrabhairavatantra states that it was revealed "by the Acarya (slob dpon) £ri Lalitavajra (the Tibetan transcribes the name) from ... Urgyan".28 Since the NMAA cites the Vajrabhairavatantra composition of the former must post-date the latter. Assuming that the NMAA precedes Tilopa and Naropa, the revealer of the Vajrabhairavatantra cannot be the disciple of Tilopa. Could the Lalitavajra of the Vajrabhairavatantra be the author of the NMAA? According to Taranatha the two are different. He describes the legends associated with sGeg pa'i rdo rje separately from those of Lalitavajra, stating that the latter had a disciple called Lilavajra (sic, name given in transcription) who was distinct from "Lilavajra the great" (also in transcription).29 Leaving aside the complicating factor of these Lilavajras, Taranatha's account is given general support by the titles of the works ascribed to Lalitavajra (Rol pa'i rdo rje) in the Tanjur, which mostly refer to

24 'Lilavajra' for sGeg pa'i rdo rje: Blue Annals I 204, II 869; 'Lalitavajra' for sGeg pa'i rdo rje: Blue Annals I 367. 25'Blue Annals II 843. 26ibid. II 1030. 27Eg. Tarthang Tulku 1977, 202-204; Martin Willson 1986, pp. 189 & 271. However, Willson recognises that there is confusion surrounding the names Lalitavajra and Lilavajra and the individuals they refer to (p. 399, note 56). 28 dpal u rgyan gyi gnas chen po nas dpal 'jam dpal gyi rgyud las phyung (em. to byung) ba rgyud kyi rgyal po chen po // dpal rdo rje 'jigs byed chen po zhes bya ba bla ma dam pa'i dkyil 'khor chen po'i slob dpon dpal la li ta badzras bton nas mdzad pa rdzogs s.ho (em. to so) // "The end of [this tantra] revealed from the great and glorious land of Urgyan by the glorious Lalitavajra, the master of the mandala of the holy gurus called the glorious Vajramahabhairava, the great king of tantras arising from the Manjusrl-tantra." Siklos 1990, 249-250; 123 respectively. Siklos fails to translate slob dpon. 29Text: Schiefner 1868, 146, 9-10. Translation: Schiemer 1869, 191 (German); HBI 242-244 (English).

10

Introduction

Author & Date

Vajrabhairava and have different concerns to those attributed to sGeg pa'i rdo rje. 30 Lacking any further evidence to the contrary I shall assume that the names Lalitavajra and Vilasavajra denote (at least two) different individuals. sGeg pa'i rdo rje is a satisfactory translation of Vilasavajra. The Mahdvyutpatti has no equivalent for vilasa but gives sgeg pa for lasya (MVy 7132), a synonym of lasa and deriving from the same root (/las). The upasarga (vi-) is not translated one might expect 'rNam par sgeg pa'i rdo rje' - but Davidson points out that there are other instances of dropped upasargas in the translation of names into Tibetan. 31 Although Ilia and lalita are synonyms of vilasa and could equally well be translated by sgeg pa (and there would be no omission of an upasarga to explain in either case), I have not found it used as an equivalent for either. The Mahdvyutpatti distinguishes lalita (/lal) from Idsya giving rol pa as its equivalent,32 which is thus in accord with the translation of the name Lalitavajra by Rol pa'i rdo rje; and though there is no entry for lila MVy 7342 gives 'jo bag can or 'chos pa for the adjective salila. Given the above and given that 'Lilavajra' is not found in transcription as the Sanskrit name of any of the works attributed to sGeg pa'i rdo rje in the Tanjur, I shall proceed on the assumption that sGeg pa'i rdo rje is the translation of the name Vilasavajra, witnessed by the Sanskrit NMAA transmission. In both the Sanskrit and Tibetan translation the colophon to the NMAA states that Vilasavajra ('sGeg pa'i rdo rje' in the Tibetan) was known by a number of different names. I discuss these below, together with the fact that Taranatha's account of sGeg pa'i rdo rje largely centres on these names. Neither 'Lilavajra' nor 'Lalita-

30See Appendix II for a list of works attributed to sGeg pa'i rdo rje in the Tanjur. Of the nineteen works attributed to Lalitavajra (listed in HBI 411) eleven have the names Vajrabhairava, Bhairava, or Krsnayamari in their titles. These names do not appear in any of the works attributed to sGeg pa'i rdo rje and although he has a work on Yamantaka, it is a commentary (most of Lalitavajra's works are sddhana and mandalavidhi texts) whose title - Sriyamantakamulamantrdrthavajraprabheda (Toh 2014) - bears a resemblance to that of the NMAA. 3 Davidson 1981, p. 6 note 18. 32lalitavistara; rgya cher rol pa (MVy 1331, 7954).

11

Introduction

Author & Date

vajra' (or 'Rol pa'i rdo rje') are included in the list of different names and Taranatha states that the name 'sGeg pa'i rdo rje' is a Tantric name.33 'Vilasavajra' is highly probable as a name bestowed during Tantric consecration (abhiseka). It contains the key Tantric term vajra and is found in the Yoga Tantra, the Sarvatathagatatattvasamgraha (STTS) from which Tantric names would have been drawn; 34 it is also in the list of names that may be given during the ndmabhiseka found in Abhayakaragupta's Vajrdvall and also in the Kriydsamuccaya of Jagaddarpanacarya. Vilasavajra is one of the names in the Aksobhyakula as are both Lilavajra and Lalitavajra. 35 The presence of the three names in these lists suggests that they all were likely to be fairly common. Add to this their near identity of meaning and it is little wonder that there is confusion both in the Tibetan histories and in subsequent western scholarship. Sources of information concerning the life and date of Vilasavajra include entries in the work of the Tibetan historians (on sGeg pa'i rdo rje) and statements contained in a number of colophons, including that of the NMAA. Examination of other works attributed to him as well as the internal evidence of the NMAA itself cast further light on Vilasavajra's date. The Tibetan author Taranatha in his History of Buddhism in India (rGya gar chos 'byung) written in 1608, states that Vilasavajra (ie. sGeg pa'i rdo rje) lived during the reigns of the Pala kings Devapala and his son Rasapala. 36 This would put Vilasavajra in the mid-ninth century, King Devapala's reign usually being placed

33HBI 272 34I take this information from Davidson (1981, 6, note 18) who adds that he was unable to find a Lllavajra in the STTS. See K. Horiuchi, 'The Romanized Text of the Sarva-Tathdgata-samgraha .' Mikkyo Bunka 103 (July 1973): 72. 35For the lists see VajAv 203.2-205.1, KriySam 345.2-346.7. Vilasavajra, Lllavajra and Lalitavajra are mentioned as names at VajAv 203.6 & KriySam 345.6 (at KriySam 345.6 read lalitasvayambhu for lalisvayambhu). I am grateful to Prof. A. Sanderson for these references. 36"King Devapala ruled for forty-eight years. After him, his son Rasapala ruled for twelve years. ... During this period sGeg pa'i rdo rje ... spent ten years in &ri Nalendra (HBI 271)." It is not clear from this whether Vilasavajra is said to have been at Nalanda during the reign of Rasapala or in a period straddling the reigns of both kings.

12

Introduction

Author & Date

c. 810-850, following that of his father King Dharmapala (c. 770-810). However, Taranatha's attempt at providing a matrix of dates using the reigns of Indian kings is not always reliable and his dating should be treated with considerable caution. The Blue Annals of 'Gos lo tsa ba, composed between 1476 and 1478, discussing the life of Jnanapada (alias Buddhajnana), state that after studying with Haribhadra in TaksaSila he visited Uddiyana (the Swat Valley) where "in the presence of the Acarya sGeg pa rdo rje who was born in Nor-bu glin (Ratnadvipa) he heard many Kriya and Yoga Tantras, and studied them thoroughly".37 Jnanapada is also said to have been a pupil of Manju&imitra (Blue Annals I 369-370) who according to a tradition reported by Bu ston also taught Vilasavajra.38 Manjus'rlmitra and Jnanapada are usually dated to the second part of the eighth century.39 In support of such a date for Vilasavajra there is evidence, reported by Davidson, that links him with the translator rMa Rin chen mchog, known to be one of the first six or seven Tibetans ordained at bSam yas by £antaraksita (779 CE).40 The colophon of the Devljalamahamaydtantra ndma (Toh 836) in the rNying ma rgyud 'bum (rNyingGB) states that it was translated by Acarya sGeg pa'i rdo rje and rMa Rin chen mchog, and the same information is repeated in the dkar chag volume.41 The colophon to the translation of the commentary on the Guhyagarbha-

37This is a slightly emended version of Roerich's translation (Blue Annals I 367). He gives 'Lalitavajra' for 'sGeg pa'i rdo rje' and has 'Manidvipa' as a retranslation of 'Nor-bu glin'. The Sanskrit colophon of the NMAA - krtir acaryavilasavajrasya ratnadvipanivasinah (see Appendix HI) - indicates that 'Nor-bu glin' is likely to have translated 'Ratnadvipa'. The Blue Annals elsewhere (I 204) implies that sGeg pa'i rdo rje was responsible for initiating a tradition of NS exegesis: "At a later date, he [Smrtijnanakirti] prepared in Khams numerous translations of the ManjuSri-nama-sangrti according to the method of the Saint Lilavajra (sGeg-pa rdo-rje)". 38Davidson (1981, 6) refers to this tradition reported by Bu ston but gives no source. I have not been able to trace it in his Chos 'byung (History of Buddhism). 39Jnanapada can be dated by his disciple Buddhaguhya, the great Yoga Tantra commentator whose letter to the Tibetan king Khri Srong Ide btsan is preserved in the Tanjur (PTT vol. 129, 284.1.4-286.2.2). For a partial translation and discussion, see Snellgrove, 1987, 446-449. 40For the date of the first ordinations in Tibet and the names of the ordinees see Tucci, Minor Buddhist Texts, II: p.l2ff. For discussion of the list of ordinees in Bu-ston's Chos 'byung see Obermiller 1932, 190. 41 Colophon: rNyingGB vol. 15, 96.7; dkar chag: rNyingGB vol. 36, 512.5. The Derge Kanjur edition of the Devijdlamahamdyatantra does not identify its translator(s). See also Kaneko's Kotant-

13

Introduction

Author & Date

tantra (gSang ba'i snyingpo rgyud)42 attributed to Vilasavajra (sGeg pa'i rdo rje) also gives rMa Rin chen mchog as the translator.43 A third connection between the two figures, unnoted by Davidson, is recorded in the colophon to the Cittabindupadesa, stating that it was written by Sangs rgyas gsang ba (= Buddhaguhya), Vimalamitra and sGeg pa'i rdo rje, and translated by rMa Rin chen mchog.44 Davidson suggests that Vilasavajra and rMa Rin chen mchog may have worked together when the latter was staying in India. Though it is not certain that these colophons are free from corruption and that the sGeg pa'i rdo rje of the NMAA is the same as the sGeg pa'i rdo rje of the other works, at the same time there are no substantial reasons for doubt. The external evidence on the whole points towards the latter part of the eighth century as a date for Vilasavajra. Such a supposition is in harmony with the internal evidence from the NMAA. As has been stated previously Vilasavajra cites a wide range of both Tantric and nonTantric sources in the NMAA (see Appendix I). The citation of Tantras includes those known to have been composed by or before the early eighth century, such as

ora zenshu kai mokuroku. [A complete catalogue of the rNyingGB] p.241, No. 195, for the title and translators of the rNyingGB translation of this work, which opens the Mahayoga division of the collection. (The following work - Kaneko No. 196 - is a translation of the NS.) 42 Guhyagarbhatantra. The title of this work is problematic. It is a key text of the Tibetan rNying ma tradition (classified as the root Tantra in a set of eighteen Mahayoga Tantras) and is preserved in Tibetan translation in the Kanjur as well as in the rNyingGB but to my knowledge it is not referred to by name or cited in the extant Skt. literature including Vilasavajra's NMAA. This is surprising given its supposed importance. However, the title Sriguhyagarbhatattvaviniscaya is found in transcription in the colophon of a text in the volume of rNying ma works added at the end of the rgyud section of the Narthang Kanjur (Koros, 1836, 548, text 2) and for the commentary attributed to sGeg pa'i rdo rje Cordier has 'Mahdrdjatantrasriguyagarbhandmatika' as the 'mutilated' Skt. title given in the colophon to the version in the (Peking) edition of the Tanjur (LXXV 3). He notes that a dkar chag volume held in St. Petersburg reads, 'Mahd ... snguhya ... tika\ 43 Sanje Dorje (ed.) Commentaries on the Guhyagarbha Tantra and Other Rare Nyingmapa Texts from the Library ofDudjom Rinpoche, New Delhi, 1974, 222.5. This same colophon states that sGeg pa'i rdo rje was an Acarya from Nalanda and contains a benediction for the longevity of Khri Srong Ide btsan. ^Cordier LXXV 23. The Cittabindupadesa does not appear to have been included in the Derge edition of the Tanjur. Vimalamitra, the disciple of Buddhaguhya, visited Tibet during the reign of Khri Srong Ide tsan, though precisely when is a matter of some controversy. Davidson suggests a date of about 795 or a little later (Davidson, op. cit., note 23, p.9).

14

Introduction

Author & Date

the Susiddhikara, the Subdhu, the Mahavairocana and the Sarvatathdgatatattvasamgraha, but also includes some that were becoming more widely known towards the middle of the eighth century, such as the Guhyasamaja, the Vajmbhairava, and the Maydjala.45 Nothing thought to be later than the eighth century is cited. It is possible, though unlikely, that the NMAA was written later than the eighth century with no contemporary works cited; but authors like to be up-to-date. It is significant that Vilasavajra lists just three categories of Tantras, Yoga, Carya and Kriya.46 This suggests that classifications where the number of categories is expanded to four, then five, by the addition in turn of classes referred to as Yogottaratantra and Yoganiruttaratantra were not current when he was writing,47 and the suggestion gains strength since Vilasavajra nonetheless cites Tantras that did come to be classified as Yogortara and Yoganirurtara. Again, a date of the latter part of the eighth century is indicated, a period when Tantras such as the Guhyasamaja and Vajrabhairava were gaining influence but were not yet seen as belonging to a special category. The doctrinal and ritual structure of the NMAA also accords with what would be expected of this period. The sadhana of chapter 4 is based on the Vajradhatumandala of the Sarvatathdgatatattvasamgraha; it contains no division into generation (utpattikrama) and completion (nispannakrama) stages, and no distinction between Pledge (samaya-) and Knowledge (jnana-) deities (-satrva). The Vijnanavada doctrine, underpinned by references to Prajnaparamita literature is again characteristic of the

45 See note 20 above for discussion of citations from the SBSYDJS, referred to as the Samvara[tantra]. 46yogacaryakriyatantram tatha paramitanayam (NMAA 1.3); namani yogakriyacaryatantrapravacanasutranlabhidharmavinayalaukikalokottarani (NMAA 1.58-59). 47An example of the fourfold classification of Tantras is found in Parahitaraksita's Tippam on Nagarjuna's Pancakrama, a commentary on the completion stage of the Guhyasamajatantra: caturvidham tantram kriyacaryayogayogottaratantram (PancKrTi 39). For the fivefold see Kanha, Yogaratnamala on the HVT n viii 10: sarvamantranayam iti pancavidham kriyacaryayogayogottarayoganiruttarabhedena (Snellgrove 1959, p. 156). The NS commentary Gudhapada gives the same list: vajram pancajnanatmakam / iha paficajnanaSabdena laiyacaryayogayogottarayoganiruttaraS ca tantrany ucyante (GudhPa fol. 5r6-7).

15

Introduction

Author & Dale

time, as witnessed most notably in the more philosophically systematic work of £antaraksita, Kamalas"ila and Haribhadra.48 Most of Vilasavajra's Sastric sources are authors who unquestionably predate him, such as Nagarjuna, Candrakirti, Dignaga and Vasubandhu. Two citations, however, may be contemporary. There is a single citation attributed to Jnanapada who as stated above is described as a pupil of Vilasavajra in the Blue Annals.49 How likely would it be for a teacher to cite the work of a pupil? It is perhaps not out of the question here if we assume that Jnanapada was already distinguished as a nonTantric scholar before studying with Vilasavajra. The other citation is one attributed to Kamalacarya. 50 Though I have not found it elsewhere this seems a likely alternative name for KamalaSila. Another consideration in assessing Vilasavajra's date is whether the other works attributed to him (see Appendix II) could also have been written in this period, with the associated issue as to whether their content suggests they were written by the author of the NMAA. Since full examination of these questions is beyond the scope of the present work the following comments are merely preliminary.51 Apart from his NS commentary, Vilasavajra (as sGeg pa'i rdo rje) is best known in Tibetan Buddhism as the author of a short work, the Guhyasamajalantranidanagurupadesabhdsya (Toh 1910), on the niddna of the Guhyasamdjatantra and, particularly in the rNying ma school, for a commentary on the Guhyagarbhatantra.

48It is probable that the doctrinal position of the NMAA should be described as Yogacara rather than Yogacara-Svatantrika-Madhyamaka (a term applied retrospectively by the Tibetans to the doctrinal position of eighth-century writers such as &antaraksita). A full account would require study of the remaining chapters of the NMAA, especially chapters 6 and 8. 49NMAA 8 (B.58r9, D.133r2) - on the 'Name' sarvadharmasvabhavadhrk (NS 116): tatha coktam jnanapadaih / sambodhicittamutpadyamahamaitrimprayogatah / sarvadharma niratmana iti jnatva vimucyata iti // 50NMAA 8 (B.50r4, D.113v3) - on the 'Name' moksah (NS 95): kamalacaryenapy uktam / ragadimalinam cittam samsaras tadvimuktata / samksepat kathito moksah prahinavaranair jinair iti // 51 Problems of attribution complicate using a writer's works to determine his chronology. Only after an author's date has been established can it be decided on chronological grounds that a work has been falsely attributed. If a work does appear 'late' it may be because the author lived later than had been initially supposed rather than it being falsely attributed.

16

Introduction

Author & Date

A number of other works in the Tanjur are attributed to him, all just a few folios in length with the exception of two other commentaries, the Yamdntakamulamantrdrthavajraprabheda (Toh 2014), 83 folios long in the Peking blockprint edition, and the Pindikramatippam (24 folios), a work found in the Guhyasamaja section of the Derge Tanjur. That Vilasavajra should have written a commentary on the Guhyasamajatantra presents no obstacle to assigning him a date in the latter part of the eighth century. It accords with the tradition that he taught Jnanapada, whose school of Guhyasamajatantra exegesis is characterised by Wayman as contrasting with the Arya school in being more literary, in the sense of citing scripture as its authority, with particular fondness for citations from the Yoga Tantras. 52 The Jnanapada school also employs Vijnanavada terminology, something which Candrakirti of the Arya school, like his Madhyamika predecessor, is careful to avoid in his Pradipoddyotana commentary on the Guhyasamajatantra. 5^ Such an approach to exegesis certainly chimes with that of the NMAA, and it is not therefore surprising that Smrtijnanakirti, who translated the NMAA into Tibetan, also commented on the Guhyasamajatantra from the perspective of the Jnanapada school. 54 While study of the contents of other works attributed to Vilasavajra remains a desideratum,55 the evidence of their titles in general suggests it is not mistaken to

52 Wayman, 1973, 16. 53Wayman (1977, 94-5) writes, "This school [the Jnanapada], at least as far as its literary products are concerned, does not bother with the topics of three lights and the Clear Light so prevalent in the works of the 'Arya' school. If the Jnanapada school comes across a term in the Guhyasamajatantra like ' prakrtiprabhasvara', it would be prone to explain it just as in non-tantric Buddhism, to wit 'intrinsically clear' (said of the pure consciousness); while a writer of the Arya school would be likely to say it means (in what is called the 'pregnant sense') 'the Clear Light along with the (80) prakrtis (of the three lights')." The Jnanapada approach as described here is that of the NMAA. Wayman also notes (1977, 94) that sGeg pa'i rdo rje's Guhyasamajatantra commentary does not appear to distinguish generation and completion stages. As stated earlier, the NMAA contains no such distinction. 54Smrtijnanakirti was the author of a Sriguhyasamajatantrarajavrtti (PTT vol. 66). See Wayman, 1977,91-101. 55Wayman refers to a passage from the Guhyagarbhatantra commentary where the word guhya of the Tantra's title is discussed (1977, 56-57). In the course of his discussion sGeg pa'i rdo rje cites a

17

Introduction

Author & Date

ascribe them to the author of the NMAA or to a period later than the late eighth century. Two possible exceptions are, firstly, the Mahdtilakakrama (Toh 1290), which is placed in the Hevajratantra section of the Tanjur and described by Dudjom Rinpoche (1991, 464) as being concerned with completion stage practices, and secondly, a Vajravarahl sadhana attributed to a 'Mahapanditacarya Vilasavajra' that survives in the original Sanskrit. 56 Both of these, if composed by the author of the NMAA would indicate a date for him later than the latter part of the eighth century. Two remaining sources of evidence concerning the life of Vilasavajra are the colophon of the NMAA itself and the entry on sGeg pa'i rdo rje in Taranatha's History of Buddhism in India (rGya gar chos 'byung). The two appear to be related. The NMAA concludes with a short colophon stating the name of the author together with a small amount of biographical detail. Unfortunately the complete colophon is preserved only in a single manuscript (the Cambridge palm-leaf, MS A)57 which is in poor condition, making a number of syllables illegible. However, it is present in the Tibetan translation, in a somewhat expanded form, and the Tibetan is useful in suggesting a restored form of the Sanskrit (see Appendix III for discussion of the colophon text).

verse that mentions an 'utmost secret': Wayman writes, "In summary of his commentary on this verse, it turns out that the utmost secret is the non-dual, self-originated Wisdom (jnana), an effortless fount of good qualities while its own aspect is incognizable". This parallels the position in the NMAA where the purpose of the NS is said to be attainment of non-dual Wisdom (prayojanam ca ... advayajnanapratilambhah: NMAA 1.63-65). 56This is the Samksiptavajravarahisadhana , found in the Guhyasamayasamgraha, a collection of forty-seven works, mainly sadhanas, concerning VajrayoginI and Vajravarahi. It is one folio long (fol. 101rl-101v6) in the palm leaf MS in the possession of the Bodleian library, Oxford (see Winternitz & Keith No. 1455, p.264-5, where the work is titled Sddhanamala Tantra). The colophon of the sadhana attributed to Vilasavajra reads: sarnksiptavajravdrdhisddhanam samdptam krtir iyam mahdpanditdcdryavildsavajra...(the rest of the colophon is missing as fol. 102 is lost). 57Of the remaining manuscripts, BH2 have the name of the author and the rest have nothing.

18

^ uthor & Dale

Introduction

The Sanskrit (as emended) and Tibetan of the colophon may be respectively translated as follows: [Here ends] the work of Acarya Vilasavajra, inhabitant of Ratnadvipa, son of the sister (-bhagineya) of Sri Agrabodhi58 [and] whose name is [also] known as Sri ViSvarupa. [The work] called "An Explanation of the Meaning of the NameMantras", a commentary on the Aryandmasamgiti - composed by the Indian pandita Vilasavajra, dweller on the jewel island, 'Endowed with sun', *Srimadagrabodhibhagin, whose name is [also] celebrated as Sri ViSvarupa - is complete. 59 A number of observations arise from comparison of the two versions of the colophon. Firstly, though the name Visvarupa is partly restored from the Tibetan, it is clear that both Sanskrit and Tibetan agree in giving this as an alternative name for Vilasavajra. Secondly, the Skt. has nothing corresponding to nyi ma can ('Endowed with sun'), which could be taken as a description ('sunny Ratnadvipa'!) as well as a name. 60 Thirdly, the Sanskrit gives a useful piece of biographical information, namely that Vilasavajra's maternal uncle is called Agrabodhi, presumably the Agrabodhi to whom four works on the NS are attributed in the Tanjur. 61 The Tibetan, however, misconstrues the Sanskrit, reading -bhdgineyasya as two words, -bhdgine yasya (skal ba dang Idan pa gang gi), and taking dpal Idan byang chub mchog gi skal ba dang Idan pa ('Srimadagrabodhibhagin') to be in apposition to sgeg pa 'i rdo rje (Vilasavajra), leaving little alternative but to understand the expression as another name of Vilasavajra. This mistranslation may well be the source of the identification of Vilasavajra with Agrabodhi accepted by Bu ston and

^srimadagrabodhibhagineyasya. I am indebted to Prof. A. Sanderson for my understanding of this compound. He also pointed out that the relation between a child and his mother's brother is more intimate than that between a child and his father in Indian society. 59krtir acaryavilasavajrasya ratnadvlpanivasinah srimadagrabodhibhagineyasya prasiddhasiipa visvarupabhidhanasya conj. (A 114v6); 'phags pa mtshan yang dag par brjod pa'i rgya cher 'grel mtshan gsang sngags kyi don du rnam par ha ba zhes bya ba rgya gar gyi mkhan po sgeg pa'i rdo rje pa nyi ma can gyi rin po che gling na bzhugs pa dpal Idan byang chub mchog gi skal ba dang Idan gang gi mtshan dpal sna tshogs gzugs can du grags pa des mdzad pa rdzogs so (Tib. [226.1.6-8]). ^This expression is discussed below, following the citation of Taranatha's description of sGeg pa'i rdo rje. 61 T6h 2579-2582

19

Introduction

Author & Date

Taranatha.62 Lastly, the Tibetan refers to Vilasavajra as a pandita (mkhan po) rather than an acarya (slob dpon) as in the Sanskrit. Comparison of the Sanskrit and Tibetan colophons of the NMAA with Taranatha's description of the life of sGeg pa'i rdo rje is revealing. Taranatha's account is as follows: 63 King Devapala ruled for forty-eight years. After him, his son Rasapala ruled for twelve years. Since he did practically nothing new for the Buddhist Teaching, he is not counted among the seven Palas. During this period sGeg pa'i rdo rje (Vilasavajra), the acarya of Urgyan, spent ten years in £ri Nalanda, gave many explanations concerning the Mantrayana and composed the commentary on the Namasamglti. (V/2 lines of text omitted.) Among them, now about the acarya sGeg pa'i rdo rje (Vilasavajra). He was born in a place called Samia, was ordained in Urgyan and was a follower of the Vijnanamadhyamaka (rnam rig dbu ma pa) philosophy.

62 Sakurai (1987a, 88) states that Bu ston refers to 'Five dharmas of sGeg pa'i rdo rje' (sGeg pa'i rdo rje'i chos Inga), comprising the NMAA and the four works of Agrabodhi (rNal 'byor rgyud kyi rgya mtshor 'jug pa'i gru gzings, fol. 89b4 in The Collected Works of Bu ston pt. 11, T6h 5104). Though Sakurai accepts the identification of Agrabodhi with Vilasavajra, referring to the latter as the founder of the 'Jam dpal gsang Idan (Manjusri Guhyapanna) school, which takes its name from Agrabodhi's sadhana to Guhyapanna (Manjusrinamasamgitisadhana, Toh 2579), he comments that Agrabodhi's works all explain the Mayajaldbhisambodhikrama (dealt with in NMAA 4) in terms of the three samadhis, an exegetical category not found in the NMAA (Sakurai ibid. p.89). That the contents of works attributed to Agrabodhi and Vilasavajra significantly differ is supported by Davidson (1981, 8), who states that the Guhyapanna sadhana contains none of the concerns found in other works attributed to Vilasavajra. Davidson, however, speculates that Agrabodhi was later than Vilasavajra, and takes Agrabodhi's ritual approach to be based upon that of the NMAA. The Sanskrit colophon to the NMAA suggests that the reverse might be the case, ie. that Vilasavajra may have adopted elements in his ritual structure from the work of his uncle. According to their colophons the four works on the NS by Agrabodhi were translated into Tibetan by Smrtijnanakirti, the translator of the NMAA. The existence of an Agrabodhi who wrote on the NS from broadly the same ritual perspective as Vilasavajra, combined with the presence of the name Agrabodhi in the NMAA colophon, makes the identification of the two in the Tibetan more understandable. The mistranslation upon which it rests, however, is not one that a pandita such as Smrtijnanakirti is likely to have made and this, along with other mistranslations contained in the Tibetan translation of the NMAA (see Introduction to the Text, section 5), raise a question over the level of involvement he had in the translation. Perhaps the bulk of the translation was the work of his Tibetan disciples. Alternatively, the present translation could be the product of the revision of the Tibetan. 63This is a modified version of Chimpa & Chattopadhyaya's translation (HBI 271-272). For the Tibetan text see Schiefner (1868), Taranatha's rGya-gar-chos-'byun. St. Petersburg, 163. 22-164.19.

20

Author & Date

Introduction

After becoming learned in all the branches of knowledge, he meditated on Aryamanjusrlnamasamgiti in a small island called Madhima in Urgyan. When he was about to attain the siddhi of Arya Manjusri, the face of the picture of ManjuSri radiated bright light and kept the island illuminated for many days. Hence he was called Nyi ma dang 'dra ba ('Sun-Like'). A certain heretic felt the need for the five sense-organs of a Buddhist pandita as materials for his rituals. He came to kill this acarya. He [Vilasavajra] went on changing his own form into that of an elephant (glang po), horse, girl and boy. So he could not spot him and went away. Hence he was called sNa tshogs kyi gzugs (ViSvarupa, 'Possessing Various Forms'). During the latter part of his life, he extensively worked for the welfare of the living beings in Urgyan. After this, he attained the rainbow body, [ie.] the Vajra-body (rdo rje'i sku). His ordained name was dPal Idan byang chub mchog gi skal ba dang Idan pa (*Srimadagrabodhibhagin). His secret [ie. Tantric] name was sGeg pa'i rdo rje (Vilasavajra). Thus, the name of the author of the works composed by him, is differently mentioned as sGeg pa'i rdo rje (Vilasavajra), Nyi ma dang 'dra ba (*Suryasadr$a), sNa tshogs kyi gzugs (ViSvarupa) [and] dPal Idan byang chub mchog gi skal ba dang Idan pa (* £rimadagrabodhibhagin). The list of alternative names attributed to Vilasavajra by Taranatha is, with the addition of Nyi ma dang 'dra ba ('Sun-Like'), that of the Tibetan translation of the NMAA colophon. If it were not for the occurrence of Nyi ma dang 'dra ba in the concluding list of Vilasavajra's names, this name makes more sense in the context of the story given to explain it, as one subsequently given to the island rather than to Vilasavajra. In the Tibetan translation of the NMAA colophon the parallel phrase, nyi ma can, is descriptive of the island Ratnadvipa (it is in apposition to rin po che gling, not to sgeg pa 'i rdo rje). Comparison with the Skt. colophon shows a progression from the Skt.'s "who dwells in Ratnadvipa" (ratnadvipanivasinah) to the Tibetan translation's "dweller on the jewel island, 'Endowed-with-sun'" (nyi ma can gyi rin po che gling na bzhugs pa), then to Taranatha's description of sGeg pa'i rdo rje meditating on an island called Madhima, after which he became known as 'Sun-like'. These differences could be variously accounted for. The Tibetan translation of the colophon could, with nyi ma can, be preserving a word lost in the Skt. or it could be an addition, its source perhaps being a version of the story current at the time of the translation of the

21

Introduction

Author & Date

NMAA into Tibetan (or at the time of its revision) later reported by Taranatha. The place-name Ratnadvipa is lost in Taranatha's rGya gar chos 'byung, where the island is called Madhima, and 'Sun-like' has become a name of Vilasavajra. Taranatha may have received a confused version of an early legend surrounding the description ratnadvipanivasinah or the legend could be one that took as its source the Tibetan translation of the colophon. This latter explanation seems most persuasive in the case of the name *£rTmadagrabodhibhagin, which Taranatha states was Vilasavajra's monastic name. The Tibetan for the name is identical with that in the translation of the NMAA colophon, which as has been seen is based on a misconstrual of the Sanskrit. Thus Taranatha is offering - or transmitting - an account of a name of Vilasavajra that only exists as a product of mistranslation. However, Taranatha may provide a partial solution to a separate problem, namely where Vilasavajra lived. Both Taranatha and the Blue Annals describe him as living in Uddiyana (Urgyan)64 but this conflicts with the statement in the colophon to the commentary on the Guhyagarbhatantra that he was an acarya from Nalanda65 as well as with the colophon to the NMAA, which has him as an inhabitant of Ratnadvipa.66 Taranatha's statement that Vilasavajra spent ten years at Nalanda

^Taranatha's HBI does not contain the account of Buddhajnana being a pupil of Vilasavajra's in Uddiyana. 65Reported by Davidson, 1981, 7, note 18. Sanje Dorje (ed.) Commentaries on the Guhyagarbha Tantra and Other Rare Nyingmapa Texts from the Library of Dudjom Rinpoche, New Delhi, 1974, 222.5. 66Bu ston (Blue Annals I 367) gives Ratnadvipa (nor bu gling) as the birth place of sGeg pa'i rdo rje, whereas Taranatha has sam sa(Schiefher 1868, p. 164.6). Chimpa and Chattopadhyaya, however, read Sisa (HBI 272); and Dhih (Vol. 3,118-119), in an account of the life of Vilasavajra that is in part a retelling of Taranatha, has Samsar village for his birth-place, which with no supporting argument is equated with Khaipkhar, present day Kamgada. The geographical location of Ratnadvipa is uncertain. Davidson wonders whether it may be the vihara of Ratnagiri in Orissa and, taking the NMAA colophon to identify Agrabodhi and Vilasavajra (following the Tibetan), suggests that the conflict between the alternative locations of Ratnadvipa and Nalanda could be the product of confusion in the NMAA colophon resulting from a later acarya called Agrabodhi from Ratnadvipa considering himself, or being considered, as an incarnation of the Nalanda scholar Vilasavajra, ie. he suggests that the NMAA colophon gives information about Agrabodhi

22

Introduction

Author & Date

during which time he wrote the NS commentary is not improbable, at least in general terms. Teachers and scholars of this period often moved around from place to place and it would not be unlikely for someone of Vilasavajra's standing to spend a lengthy time at an important centre such as Nalanda and then (perhaps) return to his home territory, whether Uddiyana or Ratnadvipa. Both the internal and external evidence concerning Vilasavajra point to ascribing him a date of the latter part of the eighth century. Counter indications are Taranatha if his dating is followed Vilasavajra would have to be placed at least half a century later - and the two texts attributed to him dealing with Hevajra and Vajravarahl. If the link between Vilasavajra and rMa Rin chen mchog is accepted, it is possible to give a more definite date for the time that he was active, ie. c. 779 CE. As for so many of the figures of this period the bulk of the evidence remains circumstantial, making precise chronology almost impossible.

rather than Vilasavajra. This explanation is not sustainable given that the Skt. NMAA colophon does not identify Agrabodhi with Vilasavajra. A Ratnadvipa occurs in ch. 7 of the Saddharmapundanka Sutra where it is the goal of the journey through the forest in the parable of the conjured city.

23

Structure & Contents

Introduction

5. Structure and Contents of the NMAA 5.1 Introduction The NMAA comprises fourteen chapters,67 one on each section of the NS as enumerated by Vilasavajra after the opening stanzas of NMAA chapter one: Ch. Ch. Ch. Ch.

1 2 3 4

Ch. 5 Ch. 6

Ch. 7 Ch. 8 Ch. 9 Ch. 10 Ch. 11

On 'The Request for Instruction' (adhyesana). On 'The Reply' (prativacanam). On 'The Survey of the Six Families' (satkulavalokanam). On 'The Method of Awakening according to the Mayajala (mayajalabhisambodhikramah). On 'The Vajradhatu-Mahamandala of Bodhicittavajra' (bodhicittavajrasya vajradhatumahamandalam). On 'The Praise [of the Knowledge-Being ManjuSri] via the Perfectly Pure Awareness of the Dharmadhatu' (suvis" uddhadharmadhatujnanasvabhavena stutih). 68 On 'The Praise [of the Knowledge-Being ManjuSri] via the Mirror-like Awareness' (adarSajnanasvabhavena stutih). On 'The Praise [of the Knowledge-Being ManjuSri] via the Discriminating Awareness' (pratyaveksanajnanamukhena stutih). On 'The Praise [of the Knowledge-Being ManjuSri] via the Awareness of Equality' (samatajnanasvabhavena stutih). On 'The Praise [of the Knowledge-Being Manjus'ri] via the Awareness of the Performance of Action' (krtyanusthanajnanadvarena stutih). On 'The Praise [of the Knowledge-Being Manjus'ri that is of the nature] of the Five Awarenesses via the five Tathagatas' (pancatathagatamukhena pancaj nanastutih) .69

67The titles are taken from the individual chapter colophons, the bracketed additions being justified by other material from the NMAA, as described in the following two notes (for the Skt. chapter colophons see Appendix 111). 68It becomes clear that the 'Praise' (stuti) is praise 'of the Knowledge-Being Manjus'ri' in the opening lines of chapters 7-10. Thus: idanim pratyaveksanajnanamukhena bhagavatah sarvatathdgatajndnakdyasya manjusrijndnasattvasya stutim aha (chapter 8); and, idanim samatdjndnamukhena bhagavatah sarvatathdgatajndnakdyasya manjusrijndnasattvasya stutim adhikrtydha (chapter 9). 69The colophon to chapter 11 reads, "Here ends the chapter on 'The Concluding Praise', the eleventh in the commentary on the Aryandmasamgiti [called] 'An Explanation of the Meaning of the Name-Mantras'." (aryanamasamgltitlkayam namamantrarthavalokinyam upasamharastutyadhikara ek§das"amah parisamaptah) and the Skt. is supported by the Tibetan translation (Tib. [221.3.8]). It is not clear whether this should be regarded as corrupt, influenced by the colophon of chapter 14 (aryanamasamgltitlkayam namamantrarthavalokinyam upasamharadhikaraS caturdaSamah samaptah) or whether Vilasavajra is stating that these five NS verses of homage are to be regarded as concluding the NS proper. In order to preserve a distinction between chapters 11 & 14 I have followed the NS title given in the introductory section of chapter 1, namely, pancatathagatamukhena pancajndnastutih (Text 1.39) with the additional, 'of the Knowledge-Being Manjus'ri that is of the nature', following the opening of chapter eleven itself: "Now, the five Tathagatas, Mahavairocana etc., praised the Fortunate

24

Introduction

Structure & Contents

Ch. 12 Ch. 13

On 'The Benefit' (anusamsa). On 'The Arrangement of Mantras of the Wisdom-Wheel' (prajnacakramantravinyasah) .70

Ch. 14

On 'The Conclusion' (upasamharah).

The NMAA generally proceeds by commenting on the NS word by word, from the beginning and through to the end of each chapter. Exceptions are an introductory section at the beginning of the first chapter, which consists of opening stanzas, the division of the NS into topics and sections, and a statement of the NS's subject matter and purpose; and chapter 4, which consists ofasddhana built around two key verses of the NS (26-7) and which lays down the central structure of the NMAA's ritual content, the Vajradhatumandala. Chapters 4-10 are concerned in varying degrees with the elaboration of this ritual structure, which with the doctrinal context provide the unifying factors for the whole text. Apart from some initial statements in chapter one's introductory section, the doctrinal content is developed piece-meal, material being introduced when a word from the NS provides an opportunity. The NMAA is also something of a general Buddhist hand-book in that it gives glosses, explanations and etymological analyses (nirvacana) of words and terms familiar to Mainstream and Mahayana Buddhism. These features make it hard to summarise the contents of chapters individually. A further consequence of the nature of the NMAA's composition is that its doctrine and ritual organisation are not immediately apparent. This is not to say that Vilasa-

One - the Embodiment of Knowledge, who dwells in the Vajradhatumahamandala - with five verses whose nature is Awareness." (adhuna mahavairocanadayah panca tathagatah pancabhir jnanatmikabhir gathabhir bhagavantam jnanamurtim vajradhatumahamandalantarvartinam samstuvanti sma). 70The title given to the 13th section of the NS in chapter one - mantradvarena stutih, "The Praise via Mantras" (Text 1.42) - differs from that implied in the colophon of chapter 13, namely prajnacakramantravinyasah. The latter is an amplification of the topic of the section, mantravinydsah, given earlier in chapter one (Text 1.17-18), the meaning of which is not very clear as it stands. NS sections 12-14, like 6-11, are all called 'praises' (stutih) - presumably of Manjus"rijnanasattva - in their enumeration in chapter one. In the corresponding chapter colophons the word stuti is not used, leaving just the chapters concerned with the Five Awarenesses (ch. 6-10) and the 'concluding' homage (ch. 11) as 'praises'.

25

Structure & Contents

Introduction

vajra is not explicit or transparent in his exegesis. To the contrary, increasing familiarity with the NMAA shows a writer with a grasp of his subject and the ability to convey it clearly and economically. It is, rather, that the style of the genre prevents very much systematic development of material. My aim in what follows is twofold: to give an outline of the NMAA's ritual categories and overall framework together with some comment on the doctrinal perspective; and to introduce the contents of chapters 1-5. To an extent these concerns overlap, since by the end of chapter 5 the ritual framework of the NMAA is established, and thus to give an account of it also goes some way towards introducing the contents of the first five chapters.71 Some preliminary comment is required on the nature of the NS and how Vilasavajra divides it into sections. 5.2 The Namasamgiti and its divisions

As stated before, the NS itself is a short work. It consists of 162 verses in anustubh metre followed by a prose section describing the qualities of the NS and the benefits accruing to the person who recites and meditates on it (anusamsa). There is then a short passage containing mantras, followed by five concluding verses (upasamharah). The first 25 verses act as an introduction to the main part of the text, giving it the appearance of scripture: Vajradhara accompanied by a retinue of fierce Vajrapanis asks the Buddha £akyamuni to teach the Namasamgiti.12 £akyamuni replies and, being pleased with Vajradhara's words, agrees to his request. The core verses (26-162) consist for the most part of a series of predicates in the nominative

71 A full study of the contents of chapters 1-5 is beyond the scope of the present work, as is exploration of the extensive NS literature available in Tibetan translation. 72The NS does not open with the usual evam may a srutam ... ("Thus have I heard ...") of Buddhist Sutras and Tantras. However, among Tantras at least, this opening is not invariable. The Vajrabhairavatantra, like the NS, starts simply with atha (see Siklos, 1990). In the 14th century Bu ston classified the NS as a Tantra (Toh 360) placing it at the head of the rgyud (tantra) section of the Tanjur immediately preceding the Kalacakratantra. Bu ston's designation of the NS and Kalacakratantra as Non-dual Tantras was not accepted by Tsong kha pa who, not admitting the category, took them as Mother Tantras of the Anuttara class.

26

Structure & Contents

Introduction

case whose unstated subject is assumed to be ManjuSri. These are the 'Names' of the Namasamgiti. The concluding verses depict Vajradhara and his retinue rejoicing in what they have heard and further praising the NS. In many, though not all, of the extant Skt. MSS the NS is divided into the sections enumerated in the NMAA, some of which came to circulate independently in collections of stutis and stotras. The following table indicates the relation of the structure of the NS to the NS sections enumerated in the NMAA. Structure of the Namasamgiti

Divisions in NS MSS & NMAA

Frame Text: \ Opening Verses: (1-25)

\ a) Vajradhara's request (1-16)

Adhyesana (1-16)

j b) Sakyamuni's reply (17-22)

Prativacanam (17-22)

';

Satkulavalokanam (23-24)

\ c) Introduction to core text (23-25)

Mayajalabhisarnbodhikramah (25-27) Core Text:

\ a) Initial gat ha & homage (26-27)

(26-162)

Vajradhatumahamandalam (28-4 1 ) Suvi^uddhadharmadhatujfianam (42-66c) b) The 'Names * (28-157)

AdarSajnanam (66d-76) PratyaveksanajMnam (77-118) Samatajnanam (1 19-142) Krtyanusthanajnanam (143-157)

c) Homage (158- 162)

Pancatathagatajnanastutih (158-162)

Frame Text: Concluding Sections:

AnuSamsa (in 6 parts)

a) Praise & Benefits of the NS b) Mantras

Mantravinyasah

c) Concluding verses (163-167)

Upasamharah (163-167)

( d) Colophon

Table 1. The NS and its Divisions in NS Manuscripts and the NMAA.73

73NS verse numbers are given in brackets. Section titles are those of the NS MSS, taken from NS.Dav., allowing comparison with those of chapter 1 of the NMAA. They have been slightly modified, however. The NS gives the title compounding it with the number of verses of each section. Thus 'prativacanagathah sat' (NS.Dav. 51, 9) is here "prativacanam\

27

Introduction

Structure & Contents

It is not clear at what point this division of the NS into sections was made. 74 That it had in part to do with the influence of exegesis is suggested by the names of many of the sections as well as the unusual transition between sections 6 and 7, which falls at the end of the third pdda of a verse (NS 66c). It is not impossible that the division is Vilasavajra's. 75 Vilasavajra's summary of the NS differentiates NS 1-162 from the rest, stating that the NS consists of 162 verses plus 150 verse-equivalents for the anusamsa

74As usual with Buddhist scripture there is the question of whether it is possible to isolate different strata in its composition. Davidson argues that the 'frame text', that is, the opening and closing verses along with the prose section on the benefits of the NS and the section of mantras, probably represent a later stratum in the development of the text. Thus the core verses (26-162) constitute its initial form. He sees these verses as representing, "a basic meditative form complete with devotional homage in the final five verses (158-162)" and as being, "the instructions of a vajracarya" (Davidson 1981, 3). Given the nature of the core verses I find this interpretation puzzling. They contain no instructions as such but are descriptions of different embodiments and functions of wisdom; and it is not immediately apparent that they can be described as 'a basic meditative form' unless this points to their potential for contemplative recitation. Early commentators on the NS such as Vilasavajra and Manjusnmitra developed visualisation sddhanas centred on the NS verses but these are works of exegesis. The core text possesses a unity of style and organisation of content that distinguishes it from the rest of the work, but it does not necessarily follow that it represents a different stratum of textual composition. Analysis of the content of the NS should be considered before any conclusions are drawn. 75Except for 'Request', 'Reply', 'Benefit' and 'Conclusion' none of the NS division titles found in the Skt. MSS make immediate sense in relation to their content. The section called "The Arrangement of Mantras" (mantravinyasah), for example, is puzzling since although it contains mantras, they are not arranged in any way. An expanded title, "The Arrangement of Mantras on the Wisdom-wheel" (prajnacakramantravinyasah), in the colophon of NMAA chapter 13 makes good sense within the context of the NMAA's sddhana where the mantras are visualised on a prajndcakra in the heart of the Adibuddha. Again, section 4, entitled "The Method of Awakening According to the Mayajala" contains nothing that could be described as a 'method' (krama). A commentary such as NMAA 4 offers this. Though the points of the text divisions found in the NS MSS are identical with those given by Vilasavajra there are some minor differences in the two sets of titles. One is that sections 6-10 are not called 'praises' (stuti) in the NS MSS so, for example, instead of 'The Praise via the Awareness of Equality' the NS has 'Twenty-four verses via the Awareness of Equality' (samatajnanagathaS caturvimsatih). As seen above, the title for section 13 (mantravinyasah) is expanded in the NMAA. Despite these differences it is possible that the present section divisions of the NS derive from Vilasavajra's NMAA enumeration. Another possibility is that they originated with Manjusnmitra, but it is unlikely that the titles, at least, are his. Vilasavajra and the NS MSS give NS 23-4 the title 'The Survey of the Six Families' and the NMAA duly itemises six families, whereas ManjuSrimitra enumerates at least nine families in his Ndmasamgitivrtti (see Davidson, 1981, 21-22, note 62, where the passage is partly translated). If the present title was accepted at the time Manjusnmitra wrote his Vrtti it is unlikely that he would not comment in accord with it. However, on this issue of the origin of the NS divisions and their titles further study of other early NS commentaries extant in Tibetan translation is needed.

28

Introduction

Structure & Contents

section.76 He is not unaware that the mantra passage and the concluding verses follow the anusamsd since he has just enumerated them as the final two sections of the NS, and the NMAA devotes a separate chapter to each. In assigning 150 verses to the anusamsd Vilasavajra must be referring to the whole of what follows verses 1-162. 77 The NS is in several ways a puzzling text. It is not immediately clear, for instance, who or what the Names of the NS name; the 'core text' (NS 26-162) refers to ManjuSri just once (NS 157d); the identity of the speaker of the pivotal verses 26 and 27 is unclear and their sense obscure, as is that of tad yathd, which opens the following verse (NS 28) that introduces the Names. Also the relation of the NS to the Mdydjdlatantra of which it claims to be a part is not apparent. Full discussion of these problems lies beyond the scope of this introduction, although Vilasavajra's solution to some of them will become clear in what follows. 78 The NMAA's

76tad ekatra dvasastyuttaraSatagathabhir adhyardhaSatenanus'arnsagranthena stutih samagra (Text 1.43--4); "So [altogether, the entire praise [of the Knowledge-Being ManjuSri] has [a total of] one hundred and sixty-two verses, [plus the equivalent of a further] one hundred and fifty verses for the [prose] Benefit [Section]." ^Differentiation of the anusamsd, mantravinyasa and upasamhdra from what precedes is also found in Maiijus'riniitra's NS recitation sddhana ('Jam-dpal-gyi mtshan gdon-pa'i man-ngag', *Manjusrindmasamgityupadesd: Toh 2555), translated from the Tibetan by Davidson (1981, 45-7). He states that there are a total of 262 verses in the NS of which 150 belong to the anusamsd. Though the arithmetic does not work out, the figure of 262 for the total again suggests a division of the NS into 162 verses plus the remainder (which could either be 100 verse-equivalents, keeping Manjus'rimitra's total or 150 verse-equivalents, keeping his figure for the anusamsd, but giving a total of 312 verses for the whole NS). Davidson argues (1981, 47, note 149) that since ManjuSrimitra and Vilasavajra treat the anusamsd, mantravinyasa and upasamhara sections as being different from the remainder of the NS, this suggests that they arose as a unit, later than the earlier verses, supporting his hypothesis concerning the stratification and development of the text. However, his hypothesis is that NS 26-162 form the earliest strata, not NS 1-162. In present-day Nepal, the whole text of the NS including the colophon, but excepting the anusamsd, is chanted. Bajracarya Dibyabajra informed me (Jan. 1992) that the anusamsd used to be chanted (though it was not clear whether this was within his lifetime) but more recently it was felt that there was no need, as the anusamsd described the benefits of chanting the NS and was thus not strictly part of it. The full benefits could still be obtained without chanting the anusamsd. Indeed, modem Nepalese editions of the NS omit it. 78My paper, 'Manjus'ri and 'The Chanting of Names' (Ndmasamgiti): Wisdom and its Embodiment in an Indian Mahayana Buddhist Text.' given at the The Eighteenth Symposium on Indian Religions, Oxford, 1993, examines some of these problems.

29

Introduction

Structure & Contents

interpretation frequently coincides with what independent study might suggest; it appears that Vilasavajra is indeed bringing out what is implicit in the NS rather than imposing a pre-existing point of view. 5.3 Manjusrijnanasattva and the letter A Commenting on the term manjusrjjnanasattva in NS 10, the NMAA reads, He is [described as] jnanasattva since he dwells in the heart, [that is, the consciousness] of all the Tathagatas. The Knowledge-Being Manjus"ri is not the Bodhisattva who is the master of the ten stages (bhumi). Rather, he is Non-dual Awareness, the Perfection of Wisdom itself.79 Identified thus as the Non-dual Awareness (advayajnana) that constitutes enlightenment, the figure of Manjusrijnanasattva is central to Vilasavajra's NS exegesis. Since advayajnana is that which makes one a Buddha, ManjuSrijfianasattva can be described as being at the heart of all the Tathagatas. This understanding has as its ritual reflex the placing of ManjuSrijnanasattva at the heart of Mahavairocana, the Buddha at the centre of the Vajradhatumandala, the NMAA's principal ritual structure. 80 ManjuSrijnanasattva therefore has these twin aspects: as Non-dual Awareness; and as deity, that Awareness embodied. Although Vilasavajra does not analyse the word jnanasattva it should probably be taken as a karmadharaya ('a being who is Awareness') and this analysis of jnanasattva referring to something that is both 'Awareness' and 'a being', makes it easier to see how the term is used sometimes to denote wisdom and sometimes that wisdom's embodiment. The prominence Vilasavajra gives to ManjuSrijiianasattva takes its basis in the NS which in its opening verses, anusamsa and colophon refers to itself as, "the Namasamgiti of Manjusrijnanasattva". 81 Manjusrijrianasattva's identification with

79jnanasattva iti / sarvatathagatahrdayaviharitvat /... nayam daSabhumlSvaro bodhsattvah kim tarhy advayajnanam prajnaparamita saiva manju&ijnanasattvah // (Text 1.215-218) 80The actual arrangement in the sadhana of NMAA 4 is even more complex (see section 5.8 below) since Manjusrijnanasattva is at the hub of the Wisdom-wheel at the heart of the Adibuddha who is at the heart of Mahavairocana. Nevertheless, Manjus'rijnanasattva can still be described as in the heart of Mahavairocana. 81 bhagavan...manjuSrijnanasattvasya...namasamgItim...dharayisyamy (NS 10-14); sarvatathagatajnanakayasya mafijusYIjnanasattvasyavenikaparis'uddha namasamgitih (anusamsa: NS.Dav. 61,

30

Structure & Contents

Introduction

jndna is made in the opening verses where he is described as 'the Knowledge-Body' (jnanakaya) and 'the Embodiment of Knowledge' (jnanamurti),82 as well.as in the anusamsd where he is twice qualified as 'the Knowledge-Body of all the Tathagatas (sarvatathagatajnanakaya)'. For his ritual function, there is a recommendation in the anusamsd that the practitioner should meditate on ManjuSrijnanasattva's form.83 Vilasavajra uses these NS references in the NMAA. At the end of the visualisation of oneself as Manjus'rljnanasattva in chapter 4 the anusamsa recommendation to meditate on his form is cited (see Text 4.193 ff); the colophon is incorporated at the beginning of NMAA on NS 1 (Text 1.72-74); and at the beginning of chapter 4 one of the two anusamsd passages that refer to Manjus'rljnanasattva as 'the Knowledge-Body of all the Tathagatas' is incorporated into the text (NS.Dav. 65, 34; Text 4.2-4). This last passage provides unequivocal textual support for Vilasavajra's identification of Manjus'rljnanasattva with jndna. If the term jnanakaya ('Knowledge-Body'), which must surely be understood, as it is by Vilasavajra,84 to be a karmadharaya, if, that is, it is taken to mean 'the body that is Knowledge', then to say that Marijus"ri is the jnanakaya of all the Tathagatas is to say that he is the Knowledge or Awareness that underlies or lies within all Tathagatas. Passages from the NS anusamsd are used here, as elsewhere,85 to legitimate key notions and structures in the NMAA. The same function is performed by citation of

21-22); bhagavato mafiju£rijnanasattvasya sarvatathagatajfianakayasya jnanamurter advayaparamartham namasamgltim (anusamsa: NS.Dav. 65, 3-4); bhagavato manjusnjnanasattvasya paramartha namasamgitih (colophon: NS.Dav. 69, 9-10; see Text 1.73-4 for the NS colophon incorporated into the NMAA). There is just one further reference to ManjuSrijnanasattva by name in the NS (see note 83) making a total of five. 82In NS 10: bhagavan jnanakayasya mahosnisasya gispateh / manjusnjnanasattvasya jnanamurteh svayambhuvah. 83bhagavato manjusrijfianasattvasya rupam alambayann (anusamsa; NS.Dav. 66, 23; and see Text 4.193 for its citation by the NMAA). 84kimvis~istasya jnanakayasyeti jfianam eva kayo jnanakayas tasya (Text 1.203-204). 85It is cited, for instance, in the discussion of the NS's subject matter (Text 1.56) and as part of the description of its purpose an anusamsd passage is incorporated (Text 1.63-5).

31

Introduction

Structure & Contents

other scriptures or authoritative sastra. After identifying ManjuSrijnanasattva with the Non-dual Awareness at the hearts of all the Tathagatas, for instance, the NMAA continues, "It is for this reason that Dignaga said, "The Tathagata is the Perfection of Wisdom, that is to say, Non-dual Awareness""86 In NMAA on NS 10 Manju^rijnanasattva is said to be of the nature of the five Awarenesses (panca jnanani) through a correlation with the five epithets given to him in the verse.87 The five Awarenesses constitute another key interpretative category in Vilasavajra's NS exegesis, ranging in application from the overall structuring of the NS (as in the titles given to the NS sections) to the interpretation of individual Names. These Awarenesses can be seen as aspects or functions of Non-dual Awareness. 88 The NMAA later glosses NS 25's 'the verse of the Lord of Speech' (gatham ... giram patch) as the verse of ManjuSrijnanasattva (Text 3.5-6). This is the pivotal verse NS 26, "A A I I U U E AI O AU AM AH. I, the Awakened One, the Embodiment of Knowledge, am in the heart of the Buddhas of the three times".89 Vilasavajra therefore takes the T of this verse to be ManjuSrijnanasattva; this interpretation makes good sense of the NS and is consistent with the identification of ManjuSrijnanasattva with the Non-dual Awareness definitive of Buddhahood. As described previously, chapter 4 of the NMAA, which comprises the commentary on the verse just cited (NS 26) and the two verses enclosing it (NS 25 & NS 27), is a sadhana that has Manjusrljnanasattva as its central figure. His form and appearance is described in some detail (Text 4.172-180): he is six-faced and two armed; in each hand he holds the stem of a lotus and on each of the lotus flowers

86ata evaha dignagapadah / prajnaparamita jnanam advayam sa tathagatah / iti //(TexM.218220) 87pancajnanatmako bhagavan manju&ijnanasattvah (Text 1.214). 88For discussion of their role in the NMAA see section 5.6 below. 89a aiiuueaioauamah sthito hrdi / jnanamurtir aham buddho buddhanam tryadhvavartinam //(NS26; TexM. 127-8)

32

Structure & Contents

Introduction

there rests a volume of the Perfection of Wisdom; at his heart is a moon-disc on which is a letter A. In the language of ritual whatever is placed at the heart of a deity symbolises its nature or the source of its identity. ManjuSrijnanasattva is at the heart of the Tathagatas and the letter A is at ManjuSrijnanasattva's heart. The NMAA takes the NS 28's 'born from the letter A' (akarasambhavah) as qualifying Manjus"rijrianasattva,90 suggesting it should be understood as symbolising his source or origin. Commenting on the same verse Vilasavajra states that the letter A stands for 'The Limit of Reality' (bhutakoti),91 a synonym for sunyatd ('Emptiness'), the ultimate nature of things. So the letter A represents reality, in the sense of the truth about things, and Manjus"rijnanasattva represents awareness of that reality at the purest and most basic level. He depends on the letter A - is 'born' from it - since awareness of reality depends on that reality. In the mechanics of visualisation the deity is transformed out of its seed syllable (bijaksara). Hence, this is a transformation or, better, a transposition, from truth ('A') to realisation of truth and its embodiment (Manj us"rij nanasattva). In a longer passage in chapter 4 the 'A' at Manjus"rijnanasattva's heart is described as, "the essence of the Perfection of Wisdom (prajnaparamita), the cause of the arising of the Awareness of the Omniscient Ones, the origin of all £ravakas and

90That akarasambhavah does qualify 'ManjusYijnanasattva' is made clear at the end of chapter five (see Text 269-272). Earlier in the chapter Vilasavajra comments on the epithet as follows: "He is 'born from the letter A' (akarasambhavah > akarajanitah) because the Fortunate One, the Knowledge-Being [Manju&i] has as his nature the pure Knowledge-Body of the Tathagatas." (akarasambhava iti akarajanitah / bhagavatojnanasattvasyapari^uddhatathagatajnanakayasvabhavatvat Text 5.5-6). That the jnanakaya is the nature of \hejnanasattva can be taken as a reiteration of the identification of Manjusnjnanasattva with 'the Knowledge-Body of all the Tathagatas' of the anusamsa. There is also the implication that the jnanakaya is in some sense more basic than the jnanasattva, perhaps because it is in a sense less embodied. 91paramaksara iry ... naksaratity aksarah/ bhutakotirupenavicalitatvat / paramaS casav aksaraS ceti paramaksara iti // (Text 5.10-12). "It [ie. the letter A] is "the supreme syllable". The word 'syllable' means 'that which does not perish'. This is because it is without movement as a result of having the Limit of Reality as its nature. [Hence the word] 'paramaksarah' means 'the supreme syllable', [that is to say, 'the supreme imperishable reality']."

33

Structure & Contents

Introduction

Pratyekabuddhas, the Provisions of Merit and Wisdom of all the Mahabodhisattvas, the letter of ultimate reality, the cause of all letters";92 elsewhere the letter A is said to be of the nature of the Dharma-Sphere (dharmadhatu).93 This confirms the view that Vilasavajra saw it as standing for the true nature of things whose realisation Manjus'rijnanasattva embodies and also suggests that this nature is understood from a Vijnanavada perspective; dharmadhatu is a Vijnanavada term, as is bhutakoti. Though the letter A is identified as 'the essence of the Perfection of Wisdom', Perfection of Wisdom is here interpreted in Vijnanavada terms as is illustrated by the citation, quoted above, of the Vijnanavada commentator Dignaga, who identifies it with Non-dual Awareness. 5.4 The 'Names' of the Namasamglti In the introductory section of chapter 1 Vilasavajra discusses the NS in general terms under a set of four headings: connection (sambandha), subject matter (abhidheya), purpose (prayojana) and the purpose underlying that purpose (tatprayojana).94 The NS's subject matter is said to be Non-dual Awareness (advayajnana), its purpose the attainment of Non-dual Awareness and the purpose of this attainment is to become a Tathagata. Under the heading of 'subject matter' the title Namasamglti is discussed. 'Names' (nama-) are explained as twofold; as texts and as "all things moving and unmoving". As texts they can be mundane, that is, non-Buddhist or supramundane, that is, Buddhist in which case they are the "Yoga, Kriya and Carya Tantras, the [twelve] categories of sacred utterance (pravacana), the Sutranta, the Abhidharma and the Vinaya".95 92prajnaparamitasvabhavam sarvajnajnanodayakaranam sarasravakapratyekabuddhanam utpattibhutam sarvamahabodhisattvanam punyajnanasambharabhutam paramarthaksaram sarvaksaranam karanabhutam [akararn vinyaset] (Text 4.177-180) 93tatra mahaprano hy akarah sa canutpadasvabhavas tasya dharmadhatusvabhavatvad (Text 5.2324). "Here 'the great breath' is the sound A; and that [sound A] has 'non-production' as its nature. [This is] because it has the Dharmadhatu as its nature." 94See7>xrl.47-67. 95namani yogakriyacaryatantrapravacanasutrantabhidharrnavinayalaukikalokottarani sarvastha varajangamani ca (Text 1.58-60). That ndmani, when qualified as -laukikalokottarani, should be under-

34

Structure & Contents

Introduction

This analysis is such that between them the two categories of Names cover all objects of experience. How would the chanting (samgiti) of such Names serve the purpose of the attaining of Non-dual Awareness; are they not Manju£ri's Names? Vilasavajra's statement should be interpreted in the light of two further comments he makes. The first follows the initial definition of the Names: The 'Chanting' is of those Names metaphorically, that is, not as they ultimately are, but conventionally, in accordance with the principle that all this is mere names until [one has reached] the upper limit of existence, [that is, Enlightenment].96 The second, found at the end of chapter 5, makes it clear that the Names are in fact those of ManjusYi, at least those of ManjuSri understood as Manjus'rijnanasattva: It should be understood that 'Manju&i, the best of the glorious', the Knowledge-Being who dwells in the heart of all the Tathagatas, is qualified [in the Ndmasamgiti] by phrases, whose words are Namemantras ..., the first of which is "That is to say, [he is] the Fortunate One, the Awakened One, the Fully Awakened One, born from the letter A ",97 The Names of the NS are therefore both those of Manjus"rijnanasattva and of all objects of experience whether mundane or supramundane.98 The distinction between a conventional (samvrti) and ultimate (paramartha) understanding of Names may help elucidate Vilasavajra's interpretation. According to the Yogacara perspective all objects of experience are the result of an erroneous partition of consciousness, which is by nature non-dual. Consciousness that has realised its true (non-dual) nature possesses Non-dual Awareness hypostasised as Manjus'rijnanasattva. From the

stood as meaning '[collections ofj words', ie. 'texts', is not only suggested by the context but follows Vasubandhu's reading ofnama as 'word' in his discussion of namakaya ("the collection of words") as a cittaviprayuktadharma (see AKBh.Prw. I: 250ff). It is likely that Vilasavajra had this in mind since he was familiar with the Abhidharmakosabhasya as evidenced by the citations from it in chapter 6 of the NMAA. 96tesarn namnam samgftir iti / gaunya sthitya na paramarthatah samvrtya tu / namamatram idam sarvam a bhavagraparicchedam iti nyayat // (Text \ .60-61) 97tad yatha bhagavan buddhah sambuddho 'karasambhava ity evamadibhir namamantraksarapadair ... visesyo jnanasattvah sarvatathagatahrdayaviharl manjusnh snmatam vara iti drastavyah (Text 5.269-272). 98 What is meant by describing the 'Names' as 'Name-mantras' is discussed below (see section 5.5).

35

Structure & Contents

Introduction

perspective of ultimate truth everything that can be named can be said to be a name of ManjuSrijnanasattva since, ultimately, everything is self-aware non-dual consciousness." From the perspective of conventional truth, however, Non-dual Awareness can take any number of forms some of which - the Buddhist path, scripture, Bodhisattvas etc - aid in the process of its own realisation, and these are objects of experience that have various names. Depending on the perspective the Names are thus names of ManjuSrijnanasattva or names of texts and things. This interpretation is borne out in the NS, where the Names do not consist in a straightforward list of epithets and attributes of ManjuSri as Bodhisattva or Buddha. Whoever the Names name is variously identified as Vajrabhairava (NS 66), Samantabhadra (NS 115) and Vajrasattva (NS 71), as well as an Arhat (NS 52), a Pratyekabuddha (NS 51) and a purified Bhiksu (NS 52). While these do not make sense as alternative names for ManjuSri, they do make sense as names of forms in which ManjuSri asjndna can appear, that is, at the level of conventional truth. 100 Their true identity, what ultimately underlies their nature, has to be seen as ManjusYi[jnanasattvaj. 101

possible difficulty results here from the difference between citta andjnana. It is justifiable from a Vijnanavada perspective to say that all names ultimately name citta since all that is perceived is a product of its partition into a subject-object modality, but can it equally be said that all names name [advayajjndnal Ifjnana is the product of citta becoming aware of its true nature, only those names that denote objects (ie. aspects of citta) that possess or denote (some degree of) insight into that nature can be said to nameynamz. If this is correct, Vilasavajra's statement that the 'Names' include all that moves and that does not move goes beyond what he can legitimately say as a Vijnanavadin. 100Alex Wayman (1985, 9) takes an opposite view, viz. that the 'Names' are expressive of the ultimate truth about Manjus'ri, and that the glosses and further identifications made by commentators represent the level of conventional truth. Some of the verses of the NS might suggest this interpretation, but it is untenable for many. Is a purified Bhiksu or a Pratyekabuddha ManjuSri's ultimate nature? Surely it is the other way round. Wayman's analysis is based on his reading the full title of the NS as containing the single word paramdrthandmasamgitih and taking paramartha as qualifying ndma. This commits him to explaining the 'Names' as ultimate. In fact the title contains the two words paramartha ndmasamgltih as given in the colophon of all the MSS that have been edited (see NS.Dav. 69, 9-10) and it is clearly the NS, rather than nama, that is qualified as paramartha. The single word paramdrthandmasamgitih does occur in the anusamsd where, based on the reading in the colophon as well as on the content of the text, it makes most sense to lake paramartha as qualifying ndmasamgiti, rather than ndma. 101 This interpretation tallies with the fact \hz\jndnasattva is not found as a 'Name' in the NS core verses, which one would not expect if ManjuSnjnanasattva is the figure named. 'Manjus"ri' on the

36

Introduction

Structure & Contents

This makes sense of the short statement under the heading 'connection' (sambandha) in the introductory section of chapter 1, namely that, "the connection between these two [ie. between 'Name' and the subject-matter] is defined as the relation between name and named or between means and goal." 102 Hence if the Names name the subject matter, Non-dual Awareness, they also name ManjuSrijnanasartva; but not only do they name it (or him), they are also the means to its (or his) realisation, the purpose or goal of the NS. One way of describing the central concern of the NMAA is as an explanation of how the Names can function as a soteriological means (upaya) and this involves a transition from the sphere of doctrine to that of method. 5.5 The 'Name-mantras' of the NMAA. The above gives an account of how the NMAA understands the Names of the NS in doctrinal terms. Transposed into the language of ritual the Names become mantras, the 'Name-mantras' of the passage cited in the preceding section and the title of the commentary. The 'Name-mantras' are sometimes, but not always, identified as the mantras of particular deities - in chapter 5, for example, they become the mantras of the deities of the Vajradhatumandala. The NMAA is structured so that all the Names become 'Name-mantras' and have a place in the Vajradhatumandala that has at its heart Marijus"rijnanasattva and the letter A. 103 The construal of the first two words in the title of the NMAA (NamamantrarthavalokinT) is not immediately apparent. Though taking them as a dvandva produces good sense, 'An Explanation of the Meaning of the Names and Mantras', Vilasavajra's discussion of the number of Name-mantras and their distribution

other hand - understood as a Bodhisattva - could be a 'Name', and is so found (NS 157). He also appears as a deity in the Vajradhatumandala of NMAA 5, where he is identified with one of the 'Names' of NS 38. 102anayor abhidhanabhidheyalaksanaupayopeyalaksano va sambandhah // (Text 1.62) 103por a description of this structure see section 5.7.

37

Structure & Contents

Introduction

among the deities of the Vajradhatumandala at the end of chapter 5 shows they must be understood as a karmadharaya, ie. as 'Name-mantras'. Vilasavajra states that the Knowledge-Being ManjuSri "is qualified [in the Namasamgiti] by phrases, the words of which are Name-mantras" (namamantraksarapadair ... viSesyo). 104 Though ndmamantrd- could theoretically be read as a dvandva here, since there are no mantras in the NS verses containing Names it must be taken as a karmadharaya, 'Name-mantras' (ie. Names that are mantras). 105 The compound 'Name-mantras' never occurs alone but in the context of two compounds: ndmamantraksarapada (as above) or ndmamantraksara. The meaning of these compounds requires some discussion since they are central to an understanding of the role of the Names in the NMAA's ritual structure. At the end of chapter 5 they are used on a number of different occasions. 106 An interpretation that takes aksara as 'syllable' and aksarapada as 'unit (or group) of syllables', reading 'syllables of the Name-mantras' for ndmamantraksara and 'groups of syllables that are the Name-mantras' for ndmamantraksarapada, is satisfactory in some of these contexts. In the description of the visualisation it makes sense to say that the mandate-deities transform into the 'syllables of the Namemantras' (note 106: 'ii'), and also that ManjuSrijnanasartva is qualified in the NS by

104This passage is quoted more fully in the preceding section. See note 97 for the Skt. text. 105This assumes that ndmamantra acts as a unit within the compound ndmamantraksarapada. That a division such as ndma-mantrdksara-pada can be discounted becomes clear in the following argument. 106They are used as follows: i. ManjusrI is said to be qualified with ndmamantraksarapadas (the example cited above) that are predicates, without verbs (Text 5.269-272); ii. the mandala-deities [previously individually identified as being denoted by particular NS 'Names'] are visualised as going forth and acting for the benefit of living beings. On their return they become ndmamantrdksaras, which are placed on the moon-disc seats of the Vajradhatumandala (Text 5.275); iii. The practitioner is recommended to recite the ndmamantraksara padas in order to gain the qualities mentioned in the anusamsd section of the NS (Text 5.284-5); iv. Combined together the ndmamantrdksarapadas are said to have eighty-six Names (Text 5.292-3); v. The ndmamantrdksarapadas are said to be connected to the deities 'in the relation of qualifier and qualified' (Text 5.297); vi. It is stated that tradition has it that there are eighty-six ndmamantrdksaras (Text 5.298).

38

Introduction

Structure & Contents

'groups of syllables that are the Name-mantras' (note 106: T). However, to say that the combination of'groups of syllables that are the Name-mantras' results in eightysix Names (note 106: 'iv') is problematic since the number of Names and 'groups of syllables that are the Name-mantras' are the same according to this analysis (each syllable-group being one Name or Name-mantra). There is no point in 'combining' the syllable-groups. That this reading of the meaning of aksara and pada is unsatisfactory is confirmed when it implies that Vilasavajra states that the number of 'syllables of the Name-mantras' is eighty-six (note 106: 'vi'), a number which is far too small for a syllable count. 107 Also, the number of Names has already been stated as eighty-six (note 106: 'iv'). 108 To create consistent sense pada has to be taken as meaning 'phrase', a meaning it bears in Vasubandhu's Abhidharmakosabhdsya, 109 and aksara as 'word', giving a karmadharaya, 'words that are Name-mantras', for ndmamantraksara, and a bahuvrihi, 'phrases whose words are Name-mantras', for namamantraksarapada. The two difficulties above are then resolved. If each NS word is a Name it makes sense to say that there are eighty-six 'Name-mantra words' (not syllables) (note 106: 'vi'), and if each deity of the mandala is qualified by one 'phrase whose words are Name-mantras' - which may contain one or more words; there are more Names than deities - then it also makes sense to say that these phrases (pada) when combined give a count of eighty-six Names (note 106: 'iv'). Thus, on this interpretation, the number of namamantraksarapada equals the number of deities and the number of namamantraksara equals the number of Names.

107A tempting solution was to treat the Skt. as corrupt here and follow the Tibetan translation, which appeared to read namamantraksarapada (mtshan gyi sngags kyi yi ge'i tshig rnams Tib.[194.5.8-195. Llf). 108 This count is accurate if each word of NS 28-41 is taken to be a Name. 109"7Wa or 'phrase' is understood as vdkya, a discourse, a phrase allowing the development necessary for a complete sentence, for example the stanza, 'Impermanent are the samskaras ...' and the rest." (AKBhTVw. I: 250)

39

Structure & Contents

Introduction

The conclusions of other chapters of the NMAA confirm this analysis of the compounds. Chapter 5 does not give a figure for the number of 'phrases whose words are Name-mantras' (namamantraksarapada), though in the body of the chapter the identifications between Names and deities are, for the most part, clearly stated. In chapter 6 Vilasavajra states that according to tradition (ity amnayah) there are thirtysix 'phrases whose words are Name-mantras', though these are not identified, or correlated with deities in the comment on the NS verses. It is also stated that the number of 'words that are Name-mantras' (namamantraksara) is one hundred and eighty-one. Chapter 7 also distinguishes between the numbers of nama­ mantraksarapada and namamantraksara, identifying NS Names as deities. In chapters 8-10, the last three containing Names, only the number of Name-mantra words (namamantraksara) is stated and Names are not correlated with deities in the commentary on individual NS verses. 110 To summarise, within the ritual context of the NMAA the Names of the NS are interpreted by Vilasavajra as mantras, referred to as 'Name-mantras'. The Namemantras are identified with mandala-deities in some chapters, each deity being allocated one or more Name-mantra. At the conclusion of each chapter dealing with the NS Names (chapters 6-10) the number of Name-mantras understood to be contained in the corresponding NS chapter is stated, and on two occasions (chapters 6 & 7) a number is given for the number of groups they are divided into, when they are so divided. It is these Name-mantras that Vilasavajra sets himself to explain in the NMAA and which give it its title, "An Explanation of the Meaning of the NameMantras". 111

110In chapter 10 the word namamantrapada is used for the identification of the number of Namemantras. This may reflect the fact that in this chapter the number of namamantraksara and nama­ mantraksarapada are in effect identical since the former are not allocated to deities, ie. nama­ mantraksarapada means 'phrases whose (single) words are Name-mantras'. 11 ^or discussion of the overall structure of the NMAA and the way in which the Name-mantras are treated within it see section 5.7 below.

40

Structure & Contents

Introduction

5.6 The Five Awarenesses (panca jnanani) The group of five Awarenesses (jnana) is a key interpretive category of the NMAA and one used throughout the text and applied at a number of levels. In the introduction at the beginning of the first chapter they are listed as the topics of five sections of the NS (nos. 6-10) as divided by Vilasavajra and correlated with the five Tathagatas as shown in the following table. 112 Tathagata Vairocana Aksobhya Amitabha Ratnasambhava Amoghasiddhi

Awareness (jnana) su vi Suddhadharmadhal ujnana _______ _ adars"ajnana pratyaveksanaj nana samatajfiana krry anusthanaj nana

| English translation ! Awareness of the Perfectly Pure DharmaJ Sphere I Mirror-like Awareness 1 Discriminating Awareness { Awareness of Equality \ Awareness of the Performance of Action

Table 2. The Five Tathagatas and Five Awarenesses As described earlier, Vilasavajra also states that ManjuSrijnanasattva has the five jnanas as his nature. 113 This follows their correlation with the five epithets given to Manjus'rijnanasattva in NS 10. Since Manjus'rijnanasattva is identified as advayajnana, 114 this suggests that the jnanas are to be seen as advayajnana's principal aspects or manifestations. In ritual terms this is reflected in Mafijus'rijnanasattva having the five Tathagatas as primary emanations. It now becomes clear what is intended by the identification of the five jnanas as the topics of sections 6-10 of the NS. They are the principal aspects, or modes of appearance, of Manjus'rijnanasattva seen as advayajnana, and thus the Names of these sections are Names of Manjusrijfianasattva appearing as one or other of the jnanas, which can in turn be hypostasised as the Tathagatas. Section 5 of the NS

112 See Text 1.13-17. The five Awarenesses and their correlation with the five Tathagatas became familiar currency of Tantric exegesis. Historically, the set of five Tathagatas almost certainly preceded their identification as reflexes of the five jnanas, various sets of five Buddhas are found in the Yoga Tantras as well as in the earlier Vairocanabhisambodhisutra and Amoghapasakalpa. See Yoritomi, 1990, pp. 693-716. 113pancajnanatmako bhagavan manjus"rijnanasattvah (Text 1.214). 114This identity is established in NMAA on NS 10 (Text 1.215-218). See note 79 above.

41

Structure & Contents

Introduction

('The Vajradhatumahamandala') contains Names but its topic is said to be bodhicitta rather than one of the jndnas. It is nevertheless so organised that each Name falls under the aegis of one the jndnas, except for those at the beginning of the chapter that are said to apply directly to Manjus~rijfianasattva and the letter A in his heart. Vilasavajra explains the Names of NS section 5 as being distributed between the jndnas; those of sections 6-10 are apportioned one section to onejndna. In terms of the mandala that is developed throughout the NMAA this is represented by the Vajradhatumandala surrounded by circles of Name-mantras embodying the different jndnas. 115 Four of the five jndnas are found in the Yogacara tradition and one, the Awareness of the Perfectly Pure Dharma-Sphere, appears to be a product of Tantric literature. 116 Yogacara works describe the four jndnas as arising on Enlightenment (mahabodhi) as a result of the purifying transformation (paravrtti) of the eight consciousnesses (vijnana). In the course of chapter 5's correlation of Names with deities and jndnas Vilasavajra shows familiarity with these origins: each of the four jndnas is described as having as its nature the transformation of one or more of the consciousnesses. The following table gives the correlation between consciousnesses and the four jndnas in the NMAA. It corresponds to the Vijfianavada correlation as seen, for instance, in Xuanzang's VijnMaSi (p. 684).

115The detail of this is examined in the following section. 116For a Yogacara discussion of the four jndnas see Xuanzang's Vijnaptimatratasiddhi (VijnMaSi 681-692). The five jndnas are listed in the Dharmasamgraha (DhSam 94) and comprise the fifth section of the Mahdvyutpatti (MVy 110-114). Prof. A. Sanderson has pointed out in his work on Advayavajra's Saptdksarasddhana that a likely source of the suvisuddhadharmadhdtujndna is Vasubandhu's Trimsikd 30ab: sa evdndsravo dhdtur acintyah kusalo dhruvah, "This [Mahabodhi] is the Pure Sphere, inconceivable, good, eternal". Xuanzang discusses this verse and the pure dhdtu, identified as the dharmadhdtu, immediately following his discussion of the four jndnas (VijnMaSi 693ff.). Also, La Vallee Poussin notes that the Buddhabhumisutra teaches that the Buddha-stage (buddhabhumi) consists of five dharmas, the pure dharmadhdtu and the fourjnanas (ibid. 694).

42

Structure & Contents

Introduction

Consciousness caksurupadi vasana klistamanas vikalpamanas caksuradi vijnana

Resultant Awareness adarsajnana samatajnana

= alayavijnana = klistamanovijnana = mano vijnana = panca vijnana

pratyaveksanajnana krty anusthanaj nana

Skt. text

5.81 5.104-5 5.131 5.157

Table 3. The Four jndnas and the Eight Consciousnesses (vijnana). Elsewhere in chapter 5, Names, sets of four deities and these four jndnas are correlated. Thus, the four Family-Mothers (kulamatr) are said to be 'Named' by the epithets of NS 29 and their activity linked to the four jndnas (Text 5.23-35). The following table gives the correspondences. The Family-Mothers are also said to be transformations of the syllables 'A' etc.. Family Seed Awareness Mother \ adarsajnana The great breath, a._non-production. \ Sattvavajri j A ! samatajnana A ! j Ratnavairi speech Free from „ „. ,. ^ ,.,.....—,„„..„„.,.. „„—«, „„.„„.... ,„.„„..„.„>?.„-.....,„„,——1,«,., ,„»,„„—.„.,.,.^—,. . „,„ ,„,„„. .™,™,—,L,,. -,.„„„„, ...„,.,. ..„......utterance. „ ._»™,,,,,-,~,..r.

NS Name(s)

Foremost cause_of all expression. _ [ Dharrnavajn (AM Very radiant as all speech.____ j Karmavajr! j AH

j pratyaveksanajnana [ krty anusthanaj nana

Table 4. The Four Family-Mothers and Four Awarenesses. In the comment on NS 36 the fourpitjddevis, said to be denoted by the Names of the verse, are described as different types of delight that arise with the fo\njndnas. ]l1 Correlation with four rather than five jndnas partly reflects their grouping in the Yogacara tradition, but also results from the fifth jnana being identified with the central deity of the mandala and the other four being associated with sets of four deities occupying the cardinal directions. The Dharma-Sphere Awareness in these circumstances is considered to incorporate the others. When the four jndnas appear in Yogacara contexts, the Mirror-like Awareness (adarsajnana) is considered primary in that it contains the seeds (bija) of the other three. This sense is retained in the NMAA. Following the passage on the Family-Mothers Vilasavajra says that the enlightened mind is of the nature of the Mirror-like Awareness since it is the cause of

117ata£ ca guhyapujacatustayatmakam yathakramam adars"adijn§nanispattau svavimoksamukhavisuddhibalenanabhogabhirativi^esotpadakatvat // (Text 5.178-9)

43

Introduction

Structure & Contents

the other jnanas arising. 118 In respect of Manjusrijnanasattva, however, all five jndnas may be considered equal aspects of 'his' advayajnana. There is an ascending hierarchy of importance such that ddarsajndna, suvisuddhadharmadhdtujndna and advayajnana in turn contain the other jnanas, which, depending on context, may or may not be considered as on a level with each other. 119 On other occasions Vilasavajra invokes just one of the jndnas in his comment; the link between the mantra-family called lokdloka and Aksobhya is explained as a result of him having the Mirror-like Awareness as his nature (Text 3.16-18) and the word 'wisdom' (prajfia) in the Name 'Great bearer of the weapon of wisdom', mahdprajndyudhadharah (NS 34a), is glossed as meaning the Discriminating Awareness (Text 5.117-8). 12° One further important use of the five jndnas in the NMAA occurs in chapter 4, where they are described as forms of Perfect Under­ standing (abhisambodhi) and used to structure part of the preliminaries of the sddhana. 121 When the Name 'Having the five Awarenesses as his nature', pancajndndtmakah (NS 59b), is glossed in chapter 6 it seems that Vilasavajra has nothing more to add since the comment is brief, simply listing the five jndnas and analysing the compound as a bahuvrihi. 122 5.7 The Ritual Structure of the NMAA The ritual structure of the NMAA is based on a modified version of the thirtyseven deity Vajradhatu mandala with Mahavairocana at the centre, as found in the first chapter of the Sarvatathdgatatattvasamgraha. Chapter 4 of the NMAA initiates 118 "And hence the very mind that has the Dharmadhatu as its essence ... is of the nature of the Mirror-like Awareness, because it is radiant through the absence of adventitious impurities and because it is the sign, [that is to say cause,] of the arising of other Awarenesses." (tataS ca yad eva cittam dharmadhatusvabhavam prakrtya ... tad evagantukamalabhavaprabhasvaratvaj jnanantarodayanimittatvac cadarsajMnatrnakam Text 5.36-39) ll9 Text 4.29-59 and 5.36-43 are the longest passages dealing with the five Awarenesses in the first five chapters. 120See also 5.211: agradhir iti suvisuddhadharmadhatujnanam. 121 See section 5.8 below. 122pancajnanatmaka iti / adarsasarmtapratyaveksarialqtyanustha^asuvisuddhadharaiadharujnananlty etani pancajnanani / tany evatma svabhavo yasya sa pancajnanatmakah //(D.76v5-77rl)

44

Introduction

Structure & Contents

a sadhana that is finally completed at the end of chapter 10, in the course of which all the Name-mantras of the NS are assigned a place in or around the Vajradhatu mandala. The bulk of the sadhana material falls in chapters 4 and 5, especially the former, which is wholly devoted to the sadhana and which is in some respects complete in itself. The remaining chapters contain considerably less material directly related to the sadhana, this being confined to their opening and closing lines. Chapter 4 essentially consists in the visualisation of the mandala-palace with oneself as the central deity. The thrones and seats of the mandala-deities are left unoccupied. The central deity, Mahavairocana, is visualised as having the Adibuddha seated on a moon-disc at his heart; in the Adibuddha's heart is a wisdom-wheel (prajnacakra), on which are placed various mantras from the NS. At the hub of the wisdom-wheel is ManjuSrijnanasattva, seated on a lotus throne; on a moon-disc in his heart is the letter A. The central figure(s) emanate light-rays that act for the benefit of living beings throughout space and these rays, after absorbing the jnana of the Tathagatas, are drawn back to unite with the letter A in the heart of ManjuSrijnanasattva. Six ancillary visualisations follow that repeat exactly the same pattern as above, but with Mahavairocana replaced in turn by the Tathagatas that head the six mantrafamilies enumerated in chapter 3; at the hub of the wisdom-wheel a deity embodying one of the six mantras from NS 27 (and found on the spokes of the wisdom-wheel in the initial visualisation) replaces ManjuSrijnanasattva. These visualisations could be seen to represent an initial refraction of the central and unitary advayajndna into six primary components. Chapter 5 populates the thrones and seats of the Vajradhatu mandala left empty in the previous chapter. The bulk of the chapter is concerned with identifying NS Names with the individual deities; this is performed within the context of a complex elaboration of doctrinal correspondences (viSuddhi) that has the function of showing

45

Introduction

Structure & Contents

how the mandala-deities that are emanations of Mafijiis'rijnanasattva are also reflexes ofadvayajnana. When the NS verses have been glossed and the identifications made, the mandala-deities are visualised as going forth and acting for the benefit of beings throughout the universe. On their return they become the words of the Name-mantras with which they have been identified and take their places on the seats visualised in chapter 4. The practitioner, maintaining a sense of identity with Manju£rijnanasattva, then recites the NS verses, the performance of which is described as leading to the amassing of hosts of qualities and, eventually, Buddhahood. The chapter concludes with a statement of the number and distribution of Name-mantra words. None of the remaining chapters that deal with the Names (chapters 6-10) contains the detailed identification of Name and deity found in chapter 5. Their concluding sections, however, do follow a similar pattern. The various figures or forms denoted by the Names are visualised as radiating outwards into the universe, acting for the welfare of creatures, becoming on their return the words of the Name-mantras. The verses of the NS are then recited and the results of such practice - the obtaining of Buddha-qualities and eventual enlightenment - are restated. The words of the Namemantras for these chapters are not placed on moon-disc seats in their own mandala at the conclusion of the visualisation, but are formed into circles that are added, at each successive stage of the sadhana, to the Vajradhatu mandala of chapters 4 and 5. In the initial visualisation of the mandala (in chapter 4) it is described as having an inner and outer area. The Name-mantras of chapter 6 are placed in a circle in the outer area, described as the second circle (cakra) of Mahavairocana. The Namemantras of the remaining four chapters occupy three further cakras that encircle the outside of the main mandala. Those of chapters 7 and 8 together occupy the third cakra; those of chapter 9 the fourth; and those of chapter 10 the fifth and final cakra:

46

Introduction

NMAA Chapter 6 ... „__ 8 9

10

Structure & Contents

Awareness (jnana)

Position in mandala

su vi £uddhadh armadhat uj nana adaiiajnana pratyaveksanajnana samatajnana krty anusthanaj nana

_

2nd cakra 3rd cakra 3rd cakra 4th cakra 5th cakra

Table 5. Correspondence of NMAA Chapters, Awarenesses and cakras. Vilasavajra does not explain why the Name-mantras of chapter 6 are placed inside the Vajradhatu mandala and those of chapters 7-10 outside, but since the Names of chapter 6 are interpreted as those of the Awareness of the Perfectly Pure Dharma-Sphere associated with Mahavairocana, and given that Mahavairocana is the central deity of the mandala their position makes sense. That Vilasavajra had this in mind is suggested by the statement that this chapter has thirty-six 'phrases whose words are Name-mantras' (namamanrraksarapada), which, with Mahavairocana at the centre gives thirty-seven, the standard number for the deities of the Vajradhatu mandala. The Dharma-Sphere Awareness is also often seen as containing, or as being the essence of, the other fourjndnas and so it is not inappropriate that it should be distinguished from the others. Why the Name-mantras of chapters 6 and 7 should occupy the same cakra rather than have their own is, however, not obvious. It perhaps reflects their association with the Vajra and Lotus (padma) families, which historically appear before the Gem (ratna: chapter 9) and Action (karma: chapter 10) families; their status could thus be equal to each other and higher than that of the other two. Or it may have been dictated by a prior division of the anusamsd into five sections that are interpreted by Vilasavajra as relating to the five cakras of the Vajradhatu mandala. However, it is uncertain that the anusamsd divisions were not Vilasavajra's own. As noted earlier, extant NS manuscripts that have the anusamsd divide it into six rather than five sections. Although chapters 6-10 do not make the identification of the Names with deities in the way that chapter 5 does (for example, "'Great bearer of the weapon of

47

Introduction

Structure & Contents

wisdom' is [the Name of] Vajradharma"), some NS Names are already deity names. This is particularly so in chapter 7, which is associated with Aksobhya's Mirror-like Awareness and the wrathful forms of the Vajra family. Here Vilasavajra identifies many NS Names as being those of Krodhas ('Wraths') by stating the Tantra in which they are to be found. It is also explained, before the concluding section of chapter 8, that ManjuSrijnanasatrva can appear in a range of forms in accordance with the needs of living beings (for example as ajar of plenty to the poor). Thus not all the Names of the NS are, or have to be, explained as names of deities. Another important feature of the ritual structure of the NMAA is Vilasavajra's use of six Buddha families, enumerated in chapter 3. They are identified with six mantra families that Vilasavajra interprets as being described in NS 23-24. The following table lists the six Tathagatas, their family name, their mantra-family name and associated jnan a. Tathagata

Bodhicirtavajra Vairocana Aksobhya Amitabha Ratnasambhava Amoghasiddhi

Family name

Mantra Family mahamudra (cakra) kuiatraya vajra __ ______ lokaloka lokalokottara padma ratna mahosnisa karma 123 mantra vi dyadhara

Awareness (jnana) dharmadhatujnana adarsajnana pratyaveksanajnana samatajnana krty anusthanaj nana

Table 6. The Six Buddha Families of the NMAA. The six Tathagatas of this scheme do not have an equal status. Just as Vairocana (or Mahavairocana) subsumes Aksobhya, Amitabha, Ratnasambhava, and Amogha­ siddhi so Bodhicittavajra, a hypostasisation of bodhicitta, subsumes the other five. Manjus"rijnanasattva is the source or nature of all six. The steps in this proliferation of figures, each incorporating the previous ones, appears to be a translation into ritual terms of successive attempts to describe liberative realisation in the most fundamental

123 The karma family isnot found in the STTS, which has four kulas: buddha-, padma-, vajraand ratna-. Yoritomi (1990) notes that it is referred to in the STTS commentaries of Buddhaguhya and Anandagarbha (the latter is certainly later than Vilasavajra) and also in the version of the Vajrasekharatantra translated into Chinese by Vajrabodhi (Ryakushutsunenju-kyo: Taisho 866). He also notes that the Vajrasekharatantra contains a reference to a sixihkula, the Vajradharakula.

48

Structure & Contents

Introduction

terms. 124 Thus, Mahavairocana (dharmadhatujnana) -*• Bodhicittavajra (bodhicitta) —*• Manjus'rijnanasattva (advayajnana). The sense of progressive incorporation is reflected in the increasing number of heads possessed by the three figures at the centre of the mandala. Mahavairocana has four, the Adibuddha five, and Manjus'ri­ jnanasattva six. Mahavairocana has one head for each of the four jnanas; the other two each have the heads of the previous figure plus one of their own. Identification of Bodhicittavajra with the Adibuddha is not, and cannot be made since Bodhi­ cittavajra as a Tathagata has the Adibuddha in his heart in one of the ancillary visualisations of chapter 4. Such an identification would, however, make sense and also explain the title of chapter 5 of the NMAA, 'On the Vajradhatumahamandala of Bodhicittavajra'; this title is puzzling since the Vajradhatu mandala is described in chapter 4 as having Mahavairocana at its centre. If Bodhicittavajra is the Adibuddha and the Adibuddha is interpreted as the inner aspect of the central deity then in chapter 5, as the inner aspect, Bodhicittavajra emanates the inner (first) cakra of the main Vajradhatu mandala and then in chapter 6, as the outer aspect of the central deity, Mahavairocana emanates its outer (second) cakra. Creating a sadhana that binds each of the core verses of the NS into a single framework and interprets them in a meaningful way is an intrinsically difficult task given which the resultant ritual structure of the NMAA and the multilevel level Russian doll-like visualisation of the central deity is a remarkable accomplishment. However, Vilasavajra's use of a six family and seven Tathagata (counting Manju&ijnanasattva as a Tathagata) system does not make the task any easier. The basic Yoga Tantra format of central deity with major secondary deities in the four quarters is put under considerable strain and at places cracks begin to show, as can be seen in the figure of Bodhicittavajra. In relation to Manju^rijnanasattva he is of equal

124The process of hierarchisation here gives the appearance of being doctrine-led rather than the rationalisation of a deity's pre-existing popularity by devotees and exegetes (as may be the case, for example, with some of the Yoganiruttaratantra deities).

49

Structure & Contents

Introduction

stature to the five Tathagatas and embodies one of the mantrarajas of NS 27. However, in relation to Bodhicittavajra the five Tathagatas are of lower status. 125 Bodhicittavajra would be a good candidate for the role of Adibuddha as described above, but his function as Tathagata makes this impossible, and his role as the head of a family (kula) of deities is largely theoretical. 126 Another area of strain shows in the ritual function of the Adibuddha; he appears almost redundant and could be removed with minimal loss since the role of being the underlying nature of all the other deities is fulfilled by MafijuSrijnanasattva. Occasional 'blips' ensue on a practical level, for instance the five Tathagatas emanated in chapter 5 have only four thrones (those visualised in chapter 4). 127 5.8 Chapter four: 'The Method of Awakening according to the Mdydjdla'. As described earlier, chapter 4 establishes the ritual context of the NMAA by initiating the sadhana that runs through the subsequent six chapters. The practitioner visualises the mandala-palace with him or herself as the central deity Mahavairocana who contains both the Adibuddha and Manjusrijnanasattva. Light is emanated out into the universe, promoting the welfare of others and, having absorbed ihejndna of the Tathagatas, is drawn back again to deepen the practitioner's realisation. The stages of the sadhana are summarised below.

125Bodhicittavajra's mahamudrd family is explained as 'foremost' (NS 24) for the reason Vairocana and the other Tathagatas all have [the mudrd of] bodhicitta as their nature (bodhicittavajrakulam ata evagryam vairocanadlnam bodhicitta svabhavat vat / tasya kulam mahamudrakulam ity arthah Text 3.18-20). 126The first chapter of the Guhyasamdjatantra is an important source for the function of Bodhicittavajra, where he is described as containing the five Tathagatas in his heart (GuSama 3, 1114). However, Bodhicittavajra appears to have been a transitional figure, his role and that of the Adibuddha being taken over by Vajradhara. As described, pressure for such a merging of roles is present in the NMAA. 127See section 5.9 for a possible solution to this apparant discrepancy.

50

Introduction

Structure & Contents

I. PRELIMINARIES Generation of the bodhicitta. i. Generation of the resolve to attain Perfect Enlightenment (samyaksambodhau cittam) via the five Buddha-family Vows. ii. Generation of a mind of Enlightenment that 'sets out' (prasthanabodhicitta) via the development of five-aspected Perfect Understanding (abhisambodhi), that is, the generation of the Five Awarenesses (panca jnanani)

II. GENERATION OF SELF AS DEITY 1. Generation of the Vajra-ground (vajrabhumi). i. Consciousness is visualised to have the form of the letter 'A', which is like a magical creation, located in empty space and naturally radiant. ii. A radiant fiery HUM, which emits coloured rays of light is visualised in the heart. Tathagatas at the ends of the light-rays purify all beings in all world spheres, after which the rays return into the HUM. iii. The HUM transforms into a six-faced five-pronged Knowledge-Vajra (jnanavajra) which pervades the six Buddha fields and which is encircled by flames. iv. In the centre of the jnanavajra a white BHRUM is visualised that transforms into a stupa (caitya) made of the four jewels, which is the size of the trichiliocosm. v. In the centre (garbha) of the caitya an inverted triangle (dharmodaya) is visualised, transformed out of a syllable HRIH. vi. Above the dharmodaya is visualised a double vajra (viSvavajra) generated out of a letter A. vi. With the visvavajra a white Vajra-ground (vajrabhumi) is empowered. 2. The Generation of the Mandala (as residence). On the Vajra-ground the temple (kutagara) is visualised, transformed out of the syllable BHRUM. It is square with four gates, adorned and embellished, with a circular inner area called the Vajradhatu mandala in which is visualised the throne of

51

Introduction

Structure & Contents

the central deity and those of the Buddhas of the four cardinal directions (Aksobhya etc.). Moon-disc seats are provided for the mandala-deities, each generated of the letter A. 3. Generation of Mahavairocana and the Adibuddha. i. As a transformation of the syllable AH one visualises oneself as Maha­ vairocana - white in colour, four-headed and seated on the central lion throne. ii. A moon-disc is visualised in Mahavairocana's heart on which is a syllable DHlH. This transforms into the Adibuddha - five-faced and eight-armed, with each of his four right hands holding a sword and each of his left hands holding a volume of the Perfection of Wisdom in 100,000 lines. 4. Generation of the wisdom-wheel (prajnacakra) and Manju£rijnanasattva. i. At the heart of the Adibuddha a prajnacakra, produced from the letter A, is visualised. The wheel has six spokes and two inner and two outer bands (mekhala). On the bands are placed mantras from the mantra section of the NS, the twelve vowels of NS 26 - identified as the twelve Tathagata bhumis - and the consonants of the Sanskrit alphabet. On the spokes are six mantras derived from the six epithets of the homage of NS 27. The following figure shows the arrangement: On the bands (mekhala): 1. om sarvadharmabhavasvabhavavi^uddhavajra a a am ah 2. a a i I u u e ai o au am ah 3. ah ah sarvatathagatahrdaya hara hara om hum hrih bhagavan jnanamurte vaglsVara mahavaca sarvadhannagaganamalasuparis\iddhadharmadhatujnanagarbha ah 4. ka kha ga gha na ... ha On the spokes (ara): i. om vajratflcsnaya namah; ii om duhkhacchedaya namah; iii. om prajnajnanamurtaye namah; iv. om jnanakayaya namah; v. om vaglsVaraya namah; vi. om arapacanaya namah.

Fig. 1. The Arrangement of Mantras on the Wisdom-wheel (prajnacakra). 128

128The mantra of the first band is described as the root (mula) mantra of Manju^rijnanasattva; that of the third as his garland (mala) mantra (Text 4.158; 4.167).

52

Introduction

Structure & Contents

ii. Transformed from a syllable A at the hub of the prajnacakra one visualises oneself as Manjus~rijfianasattva, seated on a lotus throne, six-faced, two-armed, immersed in samadhi, with the Sentiment (rasa) of Tranquility, each hand holding a lotus on which there rests a volume of the Perfection of Wisdom. On a moon-disc in Manjuirijnanasattva's heart one visualises a letter A 5. Emanation and Reabsorption of Light-Rays. i. The mantras of fas prajnacakra become radiant, making Mahavairocana shine with light illuminating the Sana world. ii. The light rays then go to all the Buddha fields that make up the universe where they achieve the welfare of beings. iii. They next illuminate the assembly of Tathagatas, Buddhas and Bodhisattvas and, entering their heads, they encircle the vajras in their hearts and absorb WisdomAwareness (praj naj nana). iv. Returning to Mahavairocana, the light rays re-enter his body and, identified as the essencelessness of dharmas, unite with the letter A at Mafijus~rijnanasattva's heart. v. By thus meditating on ManjuSrijnanasattva, the moon-disc and the letter A at his heart, it is stated that one will see Manju^ri in his apparitional form (nirmanakaya) accompanied by Buddhas and Bodhisattvas and that, through the power of meditation, Buddhahood and Vajradharahood will be attained in this life. 6. Visualization of the six prajnacakras of the Buddha families. The Buddhas of the six families now in turn take the central position of the mandala replacing Mahavairocana, and six visualization sequences are performed that are structurally identical with what has preceded. At the heart of each figure is visualized the Adibuddha, and at the heart of the Adibuddha a prajnacakra. The number of spokes on the prajnacakra corresponds to the number of syllables of the mantra of the deity that is to reside at the heart of the prajnacakra. The mantra is then

53

Structure & Contents

Introduction

visualised one syllable per spoke on the wheel and the corresponding deity placed at its hub. The six deities derive from the epithets of NS 27 and their mantras are those that occupied one spoke each of the original prajndcah'a. The process of emanation and reabsorption of light rays proceeds as before. The following table summarizes the visualisations of the six prajnacakras giving them in the order in which they are visualised. For each prajnacakra, the bija that is visualised on a moon-disc at the heart of the innermost deity is also used to generate both the central Tathagata and the innermost deity. The innermost figures are all single-headed and two-armed with the right hand holding a sword and the left an emblem of the presiding Tathagata. Tathagata

bija

Amitabha Aksobhya Vairocana

Hrih Hum Ah

Amoghasiddhi

Ah

Ratnasambhava Om Bodhicittavajra A

Prajnacakra

mantra / no. of spokes 8 Om vajratlksnaya namah Om duhkhacchedaya namah 8 10 Orn prajnajnanamurtaye namah ?8 Oinjnanakayaya namah Om vaglSvaraya namah Om arapacanaya namah

8 9

Innermost Deity Colour Name

Object lotus Vajratlksna red Vajrakhadga dark blue vajra wheel Prajnajnana white Jnanakaya

green

VaglSvara Arapacana

yellow white

double vajra jewel book

Table 7. The Six Ancillary Visualisations of NMAA Chapter 4. Remarks on Some Features of the Sadhana The preliminaries to a sadhana, which create the ethical and metaphysical context for the subsequent visualisation, are often rather elaborate, involving descriptions of suitable times and places for practice, preliminary purificatory ablutions, a (seven­ fold) service (puja) before a visualised assembly of Buddhas and Bodhisattvas, the generation of the Brahmaviharas and a meditation on emptiness. The present sddhana's preliminaries, however, centred on the generation of the bodhicitta through the family vows and the development of 'Perfect Understanding that is of five forms (pancakarabhisambodhih)', are comparatively simple. The 'five forms' of Perfect Understanding are the Five Awarenesses the gener­ ation of which produces, according to Vilasavajra, the 'setting out' mind of

54

Structure & Contents

Introduction

enlightenment (prasthanabodhicitta). 129 This use of the Five Awarenesses in the preliminaries is noteworthy as in later sadhanas they became associated instead with the different stages in the evocation of the central deity. 130 In the NMAA sadhana the Awarenesses are developed in the following order: i. adarsajnana; ii. samata°\ iii. pratyaveksana°\ iv. suvisuddhadharmadhatu°\ v. ki'tyanusthdna0 . The first three are generated by reflection on some conceptual or mantric correlate of the respective Awareness. There are no instructions on how to develop the fourth, the Awareness of the Completely Pure Dharma-Sphere. It is a culmination of the other three, characterised as a state free from any dichotomy of subjects and objects that is radiant by nature. 131 The previous resolution to act for the welfare of beings causes the practitioner to rouse from this non-discursive state and the final Awareness is developed out of compassion for living beings. As the Awarenesses are developed the nature of the compassion associated with each changes. There is a progression from compassion that takes living beings as its object (adarsajnana), to that which takes dharmas as its object (samatajnana), to objectless compassion (pratyaveksanajnana), and finally to objectless great compassion (suviSuddhadharmadhatujnana & krtyanusthanajnana). The subtlest level of compassion - objectless great compassion - is thus attained by the fourth Awareness in order of development, the Awareness of the Completely Pure Dharma-Sphere, and maintained into the fifth Awareness.

129For the Skt. see Text. 4.29-56. 130For mKhas grub rje's account of the development of the five abhisambodhisin his rGyud sde spyi see Lessing and Wayman, 1968, 28-35. For the five ('Realisations') as a meansof structuring the generation of the central deity see Beyer, 1973, 11 Iff; R. Davidson, 1992; A. Sanderson, Advayavajra's Saptdksarasadhana. The identification of the five abhisambodhis with the five Awarenesses and their use both in the preliminaries and the organisation of the generation of the deity can be traced to the key passage in the Sarvatathagatatattvasamgraha that describes the Enlightenment (abhisambodhi) of Siddhartha (as Sarvarthasiddhi) in terms of an accelerated traversal of the Bodhisattva path achieved by means of meditation on a series of mantras (for the Skt. text of this passage see I. Yamada, 1981, 7-10. For a translation, see D. Snellgrove, 1987, 240-243.) 131 sakalagrahyagrahakavinirmuktam prakrtiprabhasvaram ... suviSuddhadharmadhatusunyatajnanam [bhavayet] (Text 4.51-53). Thus defined, there appears to be little to differentiate this Awareness from the Non-dual Awareness (advayajnana) of Manjusnjnanasattva.

55

Structure & Contents

Introduction

The differentiation of the Awareness of the Performance of Action (krtyanusthanajnana) from the other jnanas as something concerned with enlightened activity and which emerges out of a non-dual, non-discursive state (nirvikalpajnana), contrasts with other ways of viewing them, such as when the Mirror-like Awareness or the Awareness of the Perfectly Pure Dharma-Sphere is distinguished from the others. 132 The present distinction is rooted nonetheless in Yogacara understanding of krtyanusthanajndna where it is described as functioning for the benefit of beings by manifesting different Buddha-forms (nirmanakaya) (VijnMaSi 683). In later sadhanas the condition of a radiant non-discursive mind is more usually generated through reflection on a mantra that evokes (or summarises) the essential purity and emptiness of all things, including the sadhaka. This meditation on emptiness is taken to stand for achieving the Provision of Wisdom (jnanasambhara), the Provision of Merit (punyasambhara) having been symbolically gained through the performance of the proceeding part of the preliminaries. In the NMAA sadhana the summary statement upon which the practitioner is instructed to reflect in order to develop the Awareness of Equality - "Just as all phenomena are empty, without a self, so also am I empty, without a self' 133 - is close in sense to mantras used in later sadhanas such as, "OM all phenomena are by nature pure, I [too] am pure by nature". 134 The meditation on emptiness and the generation of the Awareness of Equality can be seen to share the same function and method. If the use of the five Awarenesses in the preliminaries does predate their association with the stages of generation of the main deity it is possible that their function to develop jndna, particularly the jndna that realises the emptiness of all entities, and part of their method, was transmuted into the meditation on emptiness.

132See section 5.6 above. 133yatha sarvadharmah sunya anatmanas tathaham api sunyo 'natma (Text 4.41). 134The other mantra often used - om sunyatajnanavajrasvabhavatmako 'ham - is found in the anonymous sadhana of Aryanamasamgiti in the Sadhanamala (sadhana no. 82; SaMa 159, 7-8). See Beyer, 1973, 33-36 for discussion of these mantras and how they are glossed in the Tibetan tradition.

56

Structure & Contents

Introduction

The unelaborated nature of the preliminaries and the role of the Five Awarenesses suggest that the NMAA sadhana is a comparatively primitive and early exemplar of the genre. Other features support this conclusion. There is no division of the sadhana into a 'Generation-phase' (utpattikrama) and 'Completion-phase' (nispannakrama) 135 and the whole of its contents would be classified as belonging to the 'Generationphase' if such a division were applied. Also, there is no distinction between 'Pledgedeities' (samayasattva) and 'Knowledge-deities' (jnanasattva), though in the NMAA Manjusri is 'the jnanasattva ManjuSri' and performs the same function as 'Knowledge-deities', namely standing for the true nature of the visualised deities. However, Manjus"ri is 'the' Knowledge-deity in the NMAA rather than one of many, and there is no merging of separately visualised Knowledge-deities with Pledgedeities to mark the point at which the sadhaka enters a state of full ritual identity with the jnanasattva of the deity whose sadhana is being performed. Nonetheless, the role of Manju&ri as jnanasattva in the NS and commentaries such as the NMAA could have contributed to the development of the more technical use of jnanasattva as Knowledge-deity. 136

135The 'generation-phase' of a sadhana encompasses the preliminaries and the visualisation of the central deity, often with a consort and retinue. This stage is distinguished from the 'completion phase', which is concerned with various yogas performed by the practitioner-deity in order to 'complete' the attainment of enlightenment. 136I do not know of any study on the samayasattva-jndnasattva distinction. mKhas grub rje, in his rGyud sde spyi, has the following to say in the chapter on Yoga Tantra: One generates the Symbolic Being (samayasattva) and draws in the Knowledge Being (jnanasattva), then applies the seals of the four Seals, but not if there is only the Symbolic Being or only the Knowledge Being. The purpose of executing the seals of the four Seals is to merge and unify the Body, Speech, Mind and Acts of the Knowledge Being with the body, speech, mind and acts of the Symbolic Being. There would be no foundation for merger if either were present by itself. This is comparable to having both Self Generation and Generation in Front. (Lessing & Wayman, 1980 ed., 235) This suggests that the distinction was in use in the Yoga Tantras, though mKhas grub rje cites the explanatory tantras, the Paramadya and Vajrasekhara, rather than the root Yoga Tantra, the Sarvatathagatatattvasamgraha, as confirming the picture he gives of the relation between the seals (mudra) and the sattvas. Vilasavajra cites both the Paramadya Tantra and Vajrasekhara Tantra in the NMAA, which suggests he was familiar with their content. mKhas grub rje's citation of these works, however, does not indicate how full a treatment they give to the subject. The question of whether Vilasavajra was aware of the notion of drawing the jnanasattva into the samayasattva, but choose to

57

Introduction

Structure & Contents

5.9 The VajradhdtumahdmandaJa (ch.5): Summary of Doctrinal Correspondences. As described earlier, in NMAA chapter 5 Vilasavajra identifies the Names of the fifth section of the NS as those of the deities of the Vajradhatu mandala, and at the end of the chapter they take their places on the empty thrones and seats visualised in chapter 4. The identification of the Names as deities is accompanied by the enumeration of a series of correspondences with doctrinal categories designed to show how the mandala-deities are also manifestations of different aspects of the Non-dual Awareness (advayajnana) embodied by ManjuSrijfianasattva. The correlations established (vis"uddhi) are sometimes of considerable complexity as may be illustrated in the treatment of the Names of NS 32cd-35, which are identified as the names of sixteen samadhi deities. An underlying assumption, clear in the designation samadhi, is that all the mandala-deities are conceived of as states of mind, samddhis, of Manjus*rijnanasattva. The samadhi deities are dealt with by Vilasavajra in four groups of four, each of which is named by one of the NS verses (four deities are said to be named in NS 32cd). He approaches each group, ie. each verse, as follows. First, the enlightened mind is explained as having a set of four sunyatds as its nature. These are related to one of the four Liberation-Doors (vimoksamukha) identified as the four faces of Mahavairocana in chapter 4. The Names of the NS verse are next identified with the specific deities from the Vajradhatu mandala, usually with a reason for the Name being also a Name of Manjus"rijnanasattva. Each set of four sunyatds is then associated with one of the four Awarenesses (the five, minus the Awareness of the Perfectly Pure Dharma-Sphere), and each sunyatd is described as opposing certain errors and producing certain effects, the effects being generally linked with the name of the mandala-deity that

elaborate his sddhana differently, or whether his approach contributed to the development of that notion remains open to further investigation.

58

Introduction

Structure & Contents

corresponds to the sunyata. The treatment of the remaining verses, however, is rather less complicated. Though Vilasavajra states that there are eighty-six Name-mantra words, ie. eighty-six Names, to be distributed among the mandala-deities and allocates some of them, 137 he does not say how many mandala-deities there are altogether, though in the chapter they are specified by name. Thus, as well as the thirty-seven deities of the Vajradhatu mandala of the STTS, sixteen Bodhisattvas are enumerated (after the door-guardians), giving a total of fifty-three. However, Vilasavajra appears to identify Vajrasattva as a sixth Tathagata, 138 increasing the total to fifty-four. It is also a question whether the Tathagata Vairocana (identified in NMAA on NS 31: Text 5.60-1) is the central deity, namely, Mahavairocana. If not, then the total for the deities increases once more since the thirty-seven deity Vajradhatu mandala does not mention Vairocana separately from Mahavairocana. The issue is one of whether the Names of chapter 5 are those of all the mandala-deities, including the central deity, or just those of the retinue. The sadhana of chapter 4, in which the central deity is visualised and the seats of the retinue left empty would suggest the latter and this is supported by the statement in chapter 5 that the chapter's Namemantra words should be imagined as placed in order on the moon-discs arranged previously. 139 However, Vilasavajra later assigns seven of chapter 5's Names to Manju^rijnanasattva and the letter A in his heart. This suggests that the Names are not only those of the retinue. If it were assumed that every figure in Vilasavajra's ritual structure is 'Named' in chapter 5, including the central deity and the inner figures, then the Tathagata Vairocana can be taken as Mahavairocana and perhaps the Tathagata Vajrasattva can be identified as the Adibuddha. This proposal also has the

137See Text 5.293-298. 138 See Text 5.50-51 and note to the translation in chapter 5, NMAA on NS 30ab. 139 7Vx/ 5.276-277.

59

Structure & Contents

Introduction

merit of providing a solution to the difficulty of having six Tathagatas but only four thrones visualised for them in chapter 4. The following table summarises the correlations between the mandala-deities, the NS verse stated to name them (the Names themselves are not given) and the various doctrinal categories that they embody. NS Mandala deity Class of Defile­ Verse deity ment (klefc) 28 29

30 31 32

33

34

35

Manju&ljnanasattva The letter A Sattvavajri Ratnavajri Dharmavajri Karmavajri Vajrasattva Amilabha Aksobhya Vairocana Amoghasiddhi Ratnasambhava Vajrasattva Vajraraja Vajraraga

Emptiness (sunyata)

Liberation Awareness (jnana) Door (vimoksamukha)

adhyatma0 "bahirdha0 adhyatmabahirdha0 _ rnaha0 ^unyata0 paramartha0 samskrta0 asamskrta 0 atyanta0

s^inyata0

adarSajnana

animitta0

samatajMna

apranihita0

pratyaveksanajnana

Kulamatrs

Tathagatas raga raga dvesa moha krodha lobha Samadhis

Vajrasadhu Vajraratna Vajrasurya Vajrahctu Vajrahasa Vajradharma

anavaragra0 anavakara^^ prakrti 0 " anabhilaksana0 samskara 0 sarvadharma0 abhava 0 abhavasvabhavau

Vajratlksna Vajrahetu Vajrabhasa Vajravisva / Vajrakarma Vajraraksa Vajrayaksa Vajramusti

krtyanusthanajnana

Table 8. Correspondences of the Vajradhatumahamandala of NMAA 5 (part 1).

60

Structure & Contents

Introduction

NS Mandala deity Verse 36 Vajralasya Vajramala

Class of deity

37

Bahyapujadevis

38

39

40

41

Vajragita Vajranrtya Vajradhupa Vajrapuspa Vajraloka Vajragandha [Vaj£anku|a] jVajrajiasa] [Vajrasphota] Vajravesa Maitreya Manju&l Gandhahastin

Emptiness

Guhyapujadevls

(^unyata)

Dvarapalas

adhyatma0 bahirdha 0 adhyatmabahirdha0 rnaha0 ^unyata0 paramartha0 samskrta0 asamskrta 0 atyanla0 anavaragra0___ anavakara 0

Bodhisattvas

Jfianaketu Bhadrapala Sagaramati Aksayamati Pratibhanakuta Maha stamaprapta Sarvapayafijaha Sarva^okatamonirghatamati Jaliniprabha Candraprabha Amitaprabha Gaganaganja Sarvanivaranaviskambhin

Perfection (paramita)

dana° SIla°

Power (bala)

virya0 _____ ksanti 0 prajfia0 dhyana 0 pranidhi 0 upaya 0 ^raddha bala° virya smrti samadhi prajna

prakrti0 laksana0 sarvadharma0 abhava 0 abhavasvabhava0

Table 9. Correspondences of the Vajradhatumahamandala of NMAA 5 (part 2).

61

TRAN SLATION CHAPTERS 1-5 OF VILASAVAJRA'S NAMAMANTRARTHAVALOKINl, 'AN EXPLANATION OF THE MEANING OF THE NAME-MANTRAS'

Translation

Conventions

Conventions Used in the Translation. I have employed the following conventions in order to produce a translation that is both accurate and readable. Square Brackets Square brackets contain material not present in the Sanskrit text. This additional material generally has the purpose of making the sense of a passage clear and is often implicit in the Sanskrit. I have tried to translate so that if square brackets and their contents are removed what remains makes grammatical sense. Following this rule so as to leave punctuation intact produces some oddities of appearance, thus: "After the request for instruction, he[, that is, £akyamuni,j replies ..." (chapter 2: opening) To prevent this I have regularly placed the opening bracket after the comma: "After the request for instruction, he, [that is, £akyamuni,J replies ..." Square brackets also enclose inserted citations of NS verses. Whole verses of the NS are first cited in chapter 5, from NS 33 onwards. Before this Vilasavajra gives half or quarter verses, or just the opening words of a verse - the pratika. To make the translation more readable the NS verses are inserted where necessary, translated as understood by the commentary. Although Vilasavajra provides regular analyses of the syntactic structure of the NS where a sentence spans more than one verse, the commentarial practice of taking the root text one verse at a time can make it hard to sense how the NS itself is developing. For this reason I have appended the text and a parallel translation of the NS verses commented on in chapters 1-5 of the NMAA (Appendix VIII). Bold Face Bold face is used to indicate a citation of the root text, whether appearing in the text of the NMAA or inserted by the editor. This may be a verse, part of a verse,

63

Translation

Conventions

a word or part of a word, thus: "Mental delusion (mudhadhl-) means the delusion connected with thoughts {mudha^dhi-}. It, [namely, great delusion,] is the destroyer (sudanah > vinas~akah) of this ..." (NMAA on NS 3lab)

Glosses Vilasavajra often uses synonyms to gloss words of the NS. If these have no particular force in the original it is hard to reproduce them in translation without them becoming empty or trite and they are better left untranslated. In such cases I have adopted the following notation to indicate how a word has been glossed: (x > y) means 'x is glossed as y'. Thus, "[£akyamuni] withdrew ... his long (ayatam > dirgham), full (sphitam > bahalataram) ... tongue." (NMAA on NS 17) 'ayatam > dirgham' indicates that the word 'long' is the translation of ayatam, and that ayatam is glossed by the synonym dirgham. Without this convention the passage would have to be translated by something such as, "[£akyamuni] withdrew ... his long, namely, lengthy; full, namely, thick ... tongue." An arrow sign facing the other way, (x < y), means 'x is the gloss of y'.

Grammatical Analyses Idiomatic grammatical analyses, such as those of bahuvrihi or karmadharaya compounds, are almost impossible to translate. Also, since the word glossed has to be translated according to the gloss, attempting to give a translation often produces an apparently redundant sentence, thus: "Lord of the thunderbolt means lord of the thunderbolt, namely, of the vajra" (kuliSo vajras tasya iSvarah kuli£e$varah; Text 1.101) The difficulties mount when there is a sequence of analyses for a long compound. Occasionally, these problems can be overcome if the head-word is translatable by an English compound word. Alternatively, Sanskrit can be imported into the translation,

64

Translation

Conventions

though this is not always satisfactory and the translation quickly becomes overburdened with untranslated words. I have generally indicated the presence of a grammatical analysis by placing the word analysed within braces, {}, and using the notation evolved by Michael Coulson (1976) to indicate the particular analysis. Although this results in Sanskrit being present in the translation, the braces can be passed over without loss of sense while the information contained in the original is preserved. Coulson's notation makes use of punctuation marks to distinguish different types of compound, as follows: dvandva tatpurusa

Indicated by a semicolon, eg. {as~va;gajah}. Indicated by a hyphen, the case relation being given by a number above the hyphen, eg. {kuliSa^isVarah} for a genitive tatpurusa.

karmadharaya

Indicated by a colon, eg. {maha:bhavadrisambhetta}.

bahuvnhi

Indicated by an underscore beneath the relationship mark for the two elements, eg. {mahabalazTparakramah} for a bahuvnhi based on a locative tatpurusa; {mahairupah} for a bahuvnhi based on a karmadharaya. (In the first example the tatpurusa case number is placed beside the hyphen since typographical limitations prevent it being placed above.)

Parentheses Except for their use to enclose glosses (see above), parentheses, (), are used to indicate the Sanskrit or English translation of the preceding word or phrase.

65

Translation

Chapter I

TRANSLATION [Chapter 1. On 'The Request for Instruction'.] Having bowed to the noble one, 1 Manjusri, moon of knowledge and protector in the three times, and likewise to his Namasamgiti2 of profound and lofty character; having studied the Yoga, Carya and Kriya Tantras; also the system of the Perfection of Wisdom; the Surra and Abhidharma Collections, [and] the explanation of others,3 [that is, of other Acaryas]; the Birth Stories and the Sayings;4 [and] also the teaching of the Hymnographer,5 [that is to say, Matrceta,] the whole Vijnanavada, and likewise the Madhyamaka, and also worldly treatises; 6 and keeping in mind the teaching received from the succession of my teachers,7 I, who have been [so] requested, will now produce a commentary on this [Namasamgiti], which deals with that which is profound and lofty, with a mind that is tender with compassion for slow-witted beings.

MSS start with an additional salutation. A has om namas te varadavajrdya ("OM, Obeis­ ance to you, wish-granting vajra"), and the MSS of 5 have om manjughosasvamine ("OM, Obeisance to the Lord who is Manjughosa"). Tib. has a translator's homage to Tara. I take all of these to be additions to the text of the NMAA (see textual note on 1.0). A's homage alludes to NS 158a, namas te varadavajrdgrya. 2Titles of texts are not translated. ^anyanibandhanam. There is some doubt over this reading. See textual note 1.4. 4itivrttam (for itivrttakam). This is an incorrect back-formation from the Pali itivuttaka (Skt. ityuktaka), "So it has been said". As one of the twelve types of Buddhist literature (pravacanani) it is enumerated in NMAA 6 (on NS 55ab). Nine pravacana, excluding itivrttakam, are listed elsewhere (DhSam 62). Both lists also include 'birth stories' (jatakam). Itivuttaka is also the name of a particular work in the Pali Khuddaka Nikdya. The twelve pravacana are sometimes distributed among the three pitaka. (Eg., see mKhas grub rje's rGyud sde spyi in Lessing and Wayman, 1968, 53-7). ^stotrakara. La Vallee Poussin identifies the 'Hymnographer' as the famous Matrceta (AKBh.P0M. II, 205, note 2; DC, 230 note 3). I use 'teaching' to translate matam rather than the more literal 'thought' or 'doctrine', which seem inappropriate given that Matrceta's renown was due to his devotional compositions. For matam Tib. has gzhung, which is more often used to translate granthah, 'literary work', 'original composition' (MVy 1428, 1455, 1466). 6laukika. Works on non-Buddhist subjects such as grammar, metrics, poetics; also Hindu sectarian texts. 7The oral tradition as well as the textual tradition is thus referred to.

66

Chapter 1

Translation

Now I will state the divisions] of the body of this Ndmasamgiti. Thus: the Request for Instruction; the Reply; the Survey of the Six Families; the Method of Awakening according to the Maydjala; the Bodhicitta - as [the nature of] £rivajrasattva; 8 the Awareness of the Perfectly Pure Dharma-Sphere9 (suviSuddhadharmadhatujnanam) - as [the nature of] Mahavairocana; the Mirror-like Awareness (adarSajnanam) - as [the nature of] Aksobhya; the Discriminating Awareness (pratyaveksanajnanam) - as [the nature of] Amitabha; the Awareness of Equality (samatajnanam) - as [the nature of] Ratnasambhava; the Awareness of the Performance of Action (krtyanusthanajfianam) - as [the nature of] Amoghasiddhi; the Praise [which is of the nature] of the [five] Awarenesses - via the five Tathagatas; the Benefit; 10 the Arrangement of Mantras; and the Conclusion. These are the fourteen topics of the body [of the text]. Of these [topics], the Request for Instruction, in sixteen verses [NS 1-16], starts with [the words], Then Vajradhara ... [and continues] until, ... stood before [him] with his body bowed [in respect]. The Reply, in six verses [NS 17-

^srivajrasattvadvarena. When making identifications or correlations, Vilasavajra uses -dvdrena, -mukhena, -riipena, -svarupena, -svabhavena and -atmana interchangeably as synonyms. As well as in the present enumeration of NS topics, they are used in the division of NS verses immediately following (Text 1.27-43). Elsewhere they are used to establish further correlations with the five Awarenesses (Text 1.204-213; NMAA 11 [A.102v5-103rl]) and the five Tathagatas (Text 1.191197); and in chapter 5 they are used to give correspondences between 'Names' and mandala-deities (Text 5.23-31). The sense of these terms can often be translated by 'as'. The phrase 'x as y' can mean either 'x [embodied] as y' or 'x as [the nature of] y'. These meanings can be identical since, for example, jnana is both (doctrinally) the nature of, and (ritually) embodied as, the Tathagatas. The sense of instrumentality (present in -dvarena, -mukhena: lit. 'by means of) is sometimes required and then 'via' is a better translation. ^'Dharma-Sphere' (dharmadhatu). An important term in the NMAA, dharmadhdtu is employed by Vilasavajra in its Vijnanavada usage to denote the ultimate non-dual nature of things. The Abhidharma sense of dharmadhdtu as the object of the mind (manas) is in the background but since all is mind from a Vijnanavada perspective, everything is comprised by the dharmadhdtu. "Awareness of the Perfectly Pure Dharma-Sphere' stands for the dharmadhatu as realised by an enlightened mind. It is one of the set of five Awarenesses (panca jnanani), a key interpretive category of the NMAA (see Introduction}. wanusamsd. Not listed in MW (but s.v. anu Jsams, 'to praise'), the Buddhist sense of the nominal form is 'advantage', 'benefit' (BHSD 34).

67

Chapter 1

Translation

22], starts with, Then Sakyamuni, the Fortunate One ... [and continues] until, '... [listen] to it well.'

*O Fortunate One, [so be it.' replied

Vajrapani]. The Survey of the Six Families, in two verses [NS 23-24], starts with, Then Sakyamuni, the Fortunate One, [having surveyed] the entire ... [and continues] until, ... the great Mahosnisa family. The Method of Awakening According to the Mayajala is alluded to in three verses [NS 25-27], starting with, [Sakyamuni spoke] this [verse] endowed with the six [kingly] mantras ... [and continuing] until, Obeisance to you ... Arapacana. The teaching of Bodhicittavajra via the Great Mandala of the Vajradhatu, in fourteen verses [NS 28-41], starts with, That is to say... [and continues] until, ... master of the path of the Great Way. The praise [of the Knowledge-Being ManjuSri] via the Awareness of the Perfectly Pure Dharrna-Sphere, in twenty-five verses less a quarter [NS 42-66c], starts with, Mahavairocana, the Enlightened One ... [and continues] until, ... he is Vajranku&a, with a great noose. The praise [of the Knowledge-Being ManjuSri] via the Mirror-like Awareness, in ten verses plus a quarter [NS 66d-76], starts with, Producing terror as Vajrabhairava ... [and continues] until, ... he is speech, the best of those with a voice. The praise [of the Knowledge-Being MafijuSri] via the Discriminating Awareness, in forty-two verses [NS 77-118], starts with, Being suchness, selflessness ... [and continues] until, ... very radiant with the rays of knowledge. The praise [of the Knowledge-Being ManjuSri] via the Awareness of Equality, in twenty-four verses [NS 119-142], starts with, Supreme accomplisher of desired objects ... [and continues] until, ... he is Ratnaketu, the great jewel. 11 The praise [of the Knowledge-Being ManjuSri] via

H The Skt. reads "Ratnaketu, the great sage" (ratnaketur mahamunih Text 1.34-5). The NS has "Ratnaketu, the great jewel" (ratnaketur mahamanih NS.Dav. 59, v,142d). The commentary of NMAA 9 confirms the epithet as mahamanih (B.68r2, B.68r3), however, showing mahamunih to be corrupt.

68

Translation

Chapter 1

the Awareness of the Performance of Action, in fifteen verses [NS 143-157], starts with, To be understood by all the Buddhas ... [and continues] until, ... Manju&ri, best of the glorious. The praise [of the Knowledge-Being Manjusri, which is of the nature] of the five Awarenesses, via the five Tathagatas, in five verses [NS 158-162], starts with, Obeisance to you, foremost wishgranting thunderbolt ,.." 12 [and continues] until, ... obeisance to you, Knowledge-Body. The praise [of the Knowledge-Being ManjuSri via] 'The Benefit', starts with, This, O Vajrapani, O Vajradhara, is that ... [and continues] until, ... [he is will be] a king of the Dharma The praise [of the Knowledge-Being Manjusri] via 'The [Arrangement of] Mantras', starts with, OM - [PURE VAJRA WHOSE ESSENCE IS THE NON-BEING] OF ALL DHARMAS ... [and continues] until, ... WOMB OF THE DHARMASPHERE AWARENESS [...] - AH. The praise via 'The Conclusion', starts with, Then Vajradhara ... [and continues] until, ... taught by all the Perfectly Enlightened Ones [NS 163-7]. So [altogether, the entire praise [of the Knowledge-Being ManjuSri] has [a total of] one hundred and sixty-two verses, [plus the equivalent of a further] one hundred and fifty verses 13 for the [prose] Benefit [Section]. 14 And this is a resume of the Namasamgiti of the Knowledge-Being ManjuSri, the Fortunate One, the KnowledgeBody of all the Tathagatas. Due to the uncertainty about the meaning of words, everyone always proceeds

12Davidson translates this as "Reverence to you, the giver of the best, the foremost vajrd' taking varadavajragrya as two words (NS.Dav. 38, v.lSSab). I take them as compounded since the allusion to this verse in MS A's opening salutation, om namas te varadavajraya, indicates that varada qualifies vajra (unless varadavajraya is read as a corruption of the vocative varadavajragrya). Hgrantha. A unit of thirty-two syllables used to measure prose as well as verse. 14For discussion of the division of the NS by Vilasavajra see Introduction, section 5.2.

69

Chapter 1

Translation

[with the study of a text] on the basis of having determined its structure, 15 its subject matter, 16 its purpose, and the purpose of that [purpose]. 17 Hence the structure, and so forth, of this Namasamglti must be stated. To elaborate: if its structure and subject matter are not spoken of, then those who act on the basis of reason, assuming that it is unconnected and purposeless like the words of a madman, will not proceed to study. For this reason its structure and subject matter should certainly be stated. Yet those with faith proceed simply from the authority of the words of the Fortunate One. Since they progress from faith alone, they do not need to be told the subject matter and the rest. So here, [namely, in the Namasamglti], the subject-matter is Non-dual Awareness (advayajnana). And he, [that is, Sakyamuni,] speaks to this effect in the [Section on] Benefit: "[ ...] since it, [that is, the Namasamglti] illumines and focuses [awareness] on [Manjusri's] names for the sake of causing [pupils] to penetrate the non-dual realness of things." 18 The name 19 [of the text] is Namasamglti. [It is to be analysed as follows]: giti means 'chanting' (gana); samgiti means 'perfect chanting' (samyaggiti) [and] Namasamglti means 'The Perfect Chanting of Names' {nama^samgitih}. 'Names' [in the sense of texts] are mundane, [that is, non-Buddhist] and supramundane, [that

^sambandhah. Lit. 'connection', sambandha is used in more than one sense by Vilasavajra. Generally it is used to denote the grammatical structure of a passage under comment, ie. the connection between part of the text and what precedes or follows it. In the present list it refers to the connection bet ween the title (abhidhanam) of the NS and its subject matter (abhidheyam). ^abhidheya, lit. 'what is named'. The heading 'subject matter' includes consideration of the name or title (abhidhana) of the text since an explanation of the title involves an explanation of what is referred to by the title and it is assumed that the title describes the subject matter. 17 tatprayojana. This refers to the purpose behind the purpose, or ultimate purpose. (See textual note 1.47-8.) Lopez, 1988, 212, note 1, cites a Tibetan discussion of these headings. 18This citation is from the end of the third part of the NS anusamsd (NS.Dav. 64, 17). A few lines earlier is the following passage: "It, [that is, the Namasamglti,] is the non-dual realness of things since it is separated from the characteristics of duality" (advayadharmata dvayadharmavigatatvat: NS.Dav. 64, 14). 19abhidhanam. This is included under the head of 'subject matter' (see note 16).

70

Translation

Chapter 1

is, Buddhist. The latter are] Yoga, Kriya and Carya Tantras; [the twelve categories of] enlightened utterance; 20 the Sutranta, the Abhidharma and the Vinaya. And ['Names' also denote] all things moving and unmoving.21 The 'Chanting' is of those Names, metaphorically, that is, not as they ultimately are, but conventionally, in accordance with the principle that all this is mere names up to, [that is, until one has reached,] the upper limit of existence. The connection (sambandhah) between these two, [that is to say, between the title and the subject-matter,] is defined as the relation between name and named22 or between means and goal.23 And as for the purpose [of the text], it is the attaining of the Non-dual Awareness by Bodhisattvas practicing the observances of the Way of Mantras, who realise [it] directly 'by means of concentration on the real and fervent application, through faith that is permeated with wisdom, [that is] spotless, absolute and that is penetrative of all dharmas\24 And so it should be added that the purpose of the purpose, [namely, Non-dual Awareness,] is [the attainment of] the final rank of the Tathagata, which arises from the limit that is the highest degree of meditation, [that is to say, from its culmination].

20pravacana. NMAA 6 (on NS 55ab) gives a twelvefold classification (see note 4). 21 The Abhidharma sense of ndma as a linguistic unit of greater length than a word is being exploited here, in addition to the usage of reference to 'things' in the sense of nameables. (See Introduction 5.4 for discussion of Vilasavajra's treatment of the 'Names' of the NS.) On the relation between names and their referents see P. Williams (1991, 198-9). 22Ie. the simple denotative relation. 23 Thus the [recitation of the] Names as means leads to the goal of Non-dual Awareness, the subject-matter of the NS. 24The single inverted commas in this sentence indicate material from the anusamsa of the NS that has been incorporated into the text. The original anusamsa. passage reads, adhimuktitattvamanaskdrdbhydm samantamukhavihdravihari sarvadharmaprativedhikayd paramaydndvilayd prajndnuviddhayd sraddhayd (NS.Dav. 65, 9-11). Excepting samantamukhavihdravihari, which is excluded, this is incorporated without change (Text 1.63-5). The phrase 'by Bodhisattvas practicing the observances of the Way of Mantras' (mantramukhacaryacarinam bodhisattvanam Text 1.63) also derives from the anusamsa (NS.Dav. 63, 4). Elsewhere in the anusamsa the nominative form mantramukhacarydcdri is found (NS.Dav. 65, 3). I take mantramukha as a synonym of mantranaya (ie. mukha as 'way', 'path', or 'method').

71

Chapter 1

Translation

And now the meaning of the words is explained.

[Then Vajradhara, the glorious, best tamer of those hard to tame, a hero [since he is] the conqueror of the three worlds, king of secrets, lord of the thunderbolt, (1)] [The word] 'then (atha)': it, [that is, the text] speaks about Vajradhara since he is topical, inasmuch as he is [the] one who seeks instruction. 25 The Fortunate One Vajradhara [is present] in order to request instruction in that 'Namasamgiti, possessing ultimate truth, of the Fortunate One, the Knowledge-Being Manjusri, spoken by the Fortunate One £akyamuni in the "Net of Samadhi" chapter occuring within the Mahayoga Tantra in sixteen thousand [lines called] the Aryamayajala' 26 for the sake of benefiting beings of weak [spiritual] capacities. Alternatively it, [that is, the Ndmasamgiti] says then [where] the word 'then' has a sense of succession: 'Then Vajradhara, accompanied by [Vajra-holding Krodhas27] such as Bhrkutitarariga, made obeisance to the protector, the Awakened One, the Fortunate One, the Tathagata, and, having joined his palms in respectful greeting, stood before

25The word atha is being taken here in the sense of indicating the introduction of a new topic (ie. atha adhikare). This contrasts with the second gloss below, "[Alternatively,] the word atha has a sense of succession" (athaSabda anantarye Text 1.75-6), ie. atha as meaning 'next'. 26This passage (Text 1.72-74) incorporates the NS colophon (marked in single inverted commas), though it differs slightly from the version in NS.Dav. (see textual note on 1.72). The full title of the NS is given as: manjusrijiidnasattvasya paramdrthd ndmasamgitih. The relationship of the NS and NMAA to the Mdydjdlatantra cycle remains a matter for further investigation. As well as the statement in the NS colophon, the opening verses of the NS (verse 13) describe it as being recited in the Mdydjdlatantra. Vilasavajra names the fourth section of the NS as 'The Method of Awakening according to the Mdydjdla' (mayajalabhisambodhikramah) and he opens the sddhana of chapter 4 by saying that the method "has come down through the line of my gurus, and is included within the Srimdydjdla [Tantra] in sixteen thousand [verses]". The version of the Mdydjdlatantra in the Kanjur (Toh 466), however, is not in sixteen thousand verses and though it contains a samddhi chapter, the NSis not found there (Davidson, 1981, 2). Nontheless, study of the contents of this work may cast light on the reasons for the statements in the NS and NMAA, as may exploration of the more extensive material on the Mdydjdla cycle in the rNying ma collection of Tantras (rNyingGB). 21krodha. Lit. *wrath', 'anger'. See NMAA on NS 5 below for a description of the appearance of these wrathful attendants of Vajradhara.

72

Translation

Chapter I

[&akyamuni] and spoke as follows'. This [also] gives the sentence structure [of verses 1-6 of the Ndmasamgiti]. [The name] Vajradhara means 'Vajra-bearer'. [His] profundity (gambhlrya) is referred to by this expression. The word glorious (Sriman) refers to one who possesses (-man) 'sn", that is to say, [one who possesses] 'goddesses'. And this alludes to his extensiveness (audarya). And so, what is being said [here]? It is that Vajradhara, the Fortunate One, has wisdom and skilful means as his nature.28 For this very reason it is said in the Srisamvara [Tantra], The glorious Vajrasattva (s~rivajrasattva), [that is, Vajradhara,] the Tathagata, is [present] in the whole sphere of space.29 It is !s"vara30 and so forth that are hard to tame [and they are so described] because they are tamed with difficulty. He is their tamer, that is to say, discipliner, because IsVara and so forth are to be disciplined by Vajradhara. Vajradhara is [described as] best because of his eminence. 31 [The phrase] the three worlds

28The word sriman literally means 'possessing glory' (sri with the possessive suffix -man). Vilasavajra's interpretation, 'possessing goddesses', associates the word with the deities of a mandala. Sri is identified asdevati, a word used in Buddhist contexts as equivalent to devata. This is seen, for instance, in sddhanas devoted to Tara where the term denotes the offering goddesses (pujadevl) (SaMa 140, 11; 180, 10; 185, 19; 195, 6; 199, 4). Thus to say that Vajradhara is glorious is to say that he is surrounded by a mandala of deities. The mandala deities represent the active, compassionate, allpervading aspect of the deity at the centre of the mandala and so the term sriman can also be associated with Vajradhara's skill in means (upaya). When considered alone without a retinue of mandala-deities Vajradhara symbolises both wisdom and skilful means; when he has a retinue the two qualities become separated, Vajradhara standing for wisdom and the mandala-deities for skilful means. As a result, Vilasavajra can state that 'Vajradhara' denotes profundity (gambhlrya). The quotation from the Samvaratantra that follows serves to illustrate the all pervasive aspect. Vajradhara is by implication identified with Vajrasattva of the quotation, and the sri of srivajrasattva is to be understood as meaning sriman. A parallel to the gloss of Srivajrasattva as Vajrasattva plus mandala-deities is seen in Bhavabhatta's Cakrasamvaravivrti where the name Sriheruka is glossed as Heruka plus his consort Vajra varahl: srivajravarahi / layd yukto herukah sriheruka iti / madhyapadalopi samdsah (CSViv folio 5rl). (I am grateful to Prof. A. Sanderson for this reference.) 29This quotation has been identified by K. Tanaka as coming from chapter 1 of the SBSYDJS and not from the Laghusamvaratantra (see Appendix 1 and Introduction note 20). 30The name of Siva. The "and so forth" refers to other Hindu deities who are to be brought into the 'discipline'. 31 The gloss is not as empty as it appears. There are other ways of construing parah ('best'), as 'full' (deriving from Jpr), for instance, or 'enemy'.

73

Translation

Chapter 1

refers to the earth, the atmosphere and the heavens. He is [described as] conqueror of the three worlds because it is his practice [or way] to conquer them. Alternatively, the three worlds are the three brothers Jayakara, Madhukara and Sarvarthasiddhi. He whose way it is to conquer these [brothers] is called conqueror of the three worlds. For this very reason Vajradhara, the Fortunate One, is a hero since he is not to be overcome by enemies. In the Sritattvasamgraha [Tantra] five secrets are mentioned. To be explicit: To the extent that living beings are converted [to Buddhism] 32 according to their individual natures, one who is pure should accomplish the purposes of [those] living beings using passion and so forth. With the Buddha's teaching as one's motive, [if] for the sake of the benefit of living beings one even kills all living beings, one is not stained by sin. Employing a Buddha's body,33 [if] for the sake of the benefit of living beings one even steals the wealth of others, one is not stained by sin. There is no pleasure equal to sexual desire. 34 Offering it, [that is, the pleasure,] to the Conquerors, one may resort to the wives of others [and provided it is] for the welfare of living beings one will gain merit.35

32Lit. "accept the discipline" (vinayam yanti Text 1.90). An alternative translation of this stanza is, "The more that beings accept the vinaya by their nature, the more, being pure, they are able to accomplish the purposes of beings through passion and so forth". In this case sucih has to be read as a plural. Though grammatically irregular, such usage is found in Buddhist metrical material. 33"Employing a Buddha's body" (buddhakayaprayogatah; sangs gyas sku dang sbyor ba las Tib. [185. 5.6/). It is unclear whether this means the taking on of such a body through visualisation or its actual attainment. ^raga. Though not prominent, use of sexuality as part of the path is present in the Yoga Tantras. It is present in the Sarvatathdgatatattvasamgraha although it was expurgated in the Chinese trans­ lation. Thus: "Throughout the triple-sphered [world] there is no evil like absence of passion. Therefore, you should not practice being indifferent to sexual desire" (viragasadrSam papam anyan nasti tridhatuke / tasmat kamaviragitvam na karyam bhavata punah // STTS 150, 12-13); '"For those whose nature is pure sexual desire is pure, but it is impure for non-Buddhist yogins. Observing the pledge of the pure-natured, one obtains success.' Thus said the Fortunate One, Avalokitesvara" (ragah suddhatmanam suddho hy asuddhas tlrthyayoginam / guddhatmanam tu samayam palayam siddhim apnuyad ity aha bhagavan avalokites~varah STTS 479, 2^). See also STTS 127, 3-10; 478-9. 35This could still fall within the limits of the Bodhisattva vinaya. Candragomin's 'Twenty Verses on the Bodhisattva vow' (*Bodhisattvasamvaravimsakd) reads (verse llc),"With mercy there is no [deed] without virtue" (Tatz 1985, 28, 35-6). For a parallel to these verses see SDPS 218, 1-15 (translation: p. 66-7).

74

Translation

Chapter I

And whatever action one performs, [if it is done] for the sake of the benefit of living beings, motivated by the Buddha's teaching, one will obtain much merit. King of secrets {guhya^rat} means king of those five secrets.36 Alternatively 'guhyarat' may be analysed as 'one who reigns because of the [five] secrets' {guhya^rat}. 37 He, [that is, Vajradhara,] is [also described as] the lord of the thunderbolt, namely, [the lord of] the vajra {kulisa^iSvarah}. With each [epithet so far discussed] should be construed the verb [of the sentence of the first six verses of the Namasamgiti which is structured as follows]: 'Vajradhara, the Fortunate One, [...] standing in front [of £akyamuni], joined his palms in respectful salutation and said the following', [what he said] being stated later [in verses 7-15]. [With eyes like an opened white lotus, with a face like a fullblown lotus, throwing the best of vajras upwards with his hand again and again ... (2)] [In the word] 'vibuddhafpundarikaksah]', 'pundarika' means 'white lotus' (sVetapadmam), 'vibuddha' means 'opened' (vikasitam), [and] 'aksi' means 'eye' (caksuh). He, [that is, Vajradhara,] is described as one with eyes (aksini > locane) that are like an opened white lotus: this is the [grammatical] analysis {vibuddha:pundarikam} {vibuddhapundarikaiaksah}. He is also described as one with a face (ananam > mukham) like a full-

36Wayman (1973, 37-8) cites a verse from the $ri-Paramadya Tantra that refers to five secrets: "The great weapon of the great lord who has the supreme success (siddhi) that is great, is said to be the five-pronged thunderbolt which is the great reality of the five secrets." Anandagarbha identifies these in his commentary, the Sriparamadyatlka. Wayman lists these: "(1) the bodhicitta, (2) understanding it, (3) its realisation, (4) its non-abandonment, and (5) the knowledge characterised by attainment." 37The term guhyarat is first glossed as meaning that Vajradhara is king, that is master of, the five secrets. The second gloss contrasts with this, saying he is king (over something else) as a result of the five secrets, ie. they are what give him his power. In this case -rat is being analysed etymologically, from the root Jraj, 'to reign'.

75

Chapter I

Translation

(pra-), or thoroughly (prakarsena-38 ) blown (-utphulla) lotus (kamalam > padmam) {protphullakamalaiananah}. The structure [of the first six verses of the Namasamgiti] is [this]: 'Grasping the middle of the [best of] vajra[s] with his own hand, [and] throwing [it] upwards again and again (muhur muhuh > punah punah), which is to say, [and] catching the vajra that has been thrown up, [...] he, [namely, Vajradhara,] spoke as follows'. [Accompanied by numberless Vajrapanis, with Bhrkutitaranga ('Waves of Frowns') at their head39 - heroes [since they are] tamers of the hard to tame, possessing hideous forms and [displaying] the heroic [Sentiment], ... (3)] [Now] the passage starting 'Accompanied by...'. The attendants of Srivajradhara are listed in the Vajrapdnyabhiseka Tantra as Bhrkutitaranga and so forth, the five hundred Krodhas.40 [This verse] connects [with the rest of the sentence] as follows: 'in the company of those [attendants], Vajradhara, the Fortunate One, bowed to the perfectly enlightened one, and spoke as follows: ...'. And those wrathful attendants, in respect of ultimate truth, are without number. Therefore he, [that is, £akyamuni,] said accompanied by numberless Vajrapanis. And also, like Vajradhara, they tame the hard-to-tame. This is the reason that he, [that is, &akyamuni,] said tamers of the hard-to-tame. [And] for this same reason they are heros, since they cannot be overcome by those [that are "hard-to-tame"]. Hence

^prakarsa, a grammatical term, indicates the sense of the verbal prefix pra- mpra-utphulla. (An alternative analysis of pra- is prathamye, "in the sense of 'first'"). ^bhrkutltarangapramukhaih ("with Bhrkutitaranga at their head"). Davidson and Wayman take this differently: "With endless Vajrapanis showing billows of angry brows" (Davidson 1981, 19); "was along with [retinue] lords having ripples of furried brow, and so on" (Wayman 1985, 57-8). That Bhrkutitaranga is a proper name is suggested by the termination -pramukhaih , which functions in the same way as -ddi in stating the opening term(s) of lists. 40For a discussion of the wrathful forms of Vajrapani see Davidson, 1981, 18, note 52.

76

Translation

Chapter 1

he, [that is, £akyamuni,] said heroes. What Sentiment41 do they possess? They possess the heroic Sentiment. What form do they have? They have a form that is hideous. 42 It is for this reason that they are described as possessing hideous forms and [displaying] the heroic [Sentiment].

[Throwing [their vajras] upwards with their hands, with the tips of their vajras flashing, supreme [since they are] agents for the [ultimate] purpose of the worlds through their wisdom, skill-inmeans, and great compassion, (4)] The word 'ullalayadbhih' means throwing upwards. How [is this done]? With their hands. [They are also described as] having the (kotayah > agrabhagah)

of their

vajras

{vajra^kotibhih}

tips

flashing

(praspuradiyajrakotibhih}. What nature do they, [that is, the attendants,] have? [They are described as] functioning as agents for the [ultimate] purpose of the worlds by means of wisdom, [skill-in-]means and great compassion. In this [description] wisdom (prajna) is for the purpose of realising [the nature of] all dharmas; skillin-means (upaya) is acting for the [ultimate] purpose of all beings by explaining all mundane and supramundane samadhis according to [the individual's] suitabilty; compassion [is the meaning of 'karuna'] because it, [that is, compassion,] suppresses (runaddhi) 'ka'. [The term] 'ka' is the name of pleasure. [This 'etymology' applies] since the compassionate one, being one who suffers the sufferings of others, subordinates his own pleasure. Great compassion

('mood' or 'sentiment'). One of the key categories of Indian aesthetics. Eight rasa (or nine, ifsanta is counted,) are usually enumerated: the erotic (smgara), the heroic (vira), the disgusting (bibhatsa), the angry (raudra), the mirthful (hasya), the terrific (bhayanaka), the compassionate (karuna), the wonderful (adbhuta) and the tranquil (santa). ^bibhatsa. Like vira, bibhatsa is a rasa. For a description of this form see commentary on NS 5 below.

77

Translation

Chapter I

{mahaikaruna} is [that compassion which is] by nature objectless. [The term] 'jagad' is plural43 ('worlds') [and refers to] the three-sphered [worlds]. 44 The [ultimate] purpose (arthah> prayojanam) is that of those [worlds] {jagad^arthah}. For this very reason, [namely, that they are agents of the worlds' welfare,] those Krodhas are described as supreme (parah > utkrstah) .

[With an excited and pleased attitude, joyful, possessing the body-form of Krodhas, protectors, accomplishing the business of the Buddhas, with their bodies bowed [in respect] - (5)] What is their nature? They have as their nature an attitude which is both excited and pleased: 'excited' in their bodies and 'pleased' in their minds. [A synonym for] 'aSayah' ('attitude') is 'abhiprayah'. They are, therefore, described as [being] with an excited and pleased attitude {hrstatustaras'ayaih}. It is for this reason they are joyful (muditah > harsitsah), since they are accomplishing the [ultimate] purposes of all living beings. They are described as possessing the body-form {vigraha^rupibhih} of Krodhas {krodha^vigraharupibhih}. What is that form [like]? It has very terrifying loud laughter; many heads, hands and feet; distorted eyes, erect hair, and tawny knitted eye-brows; a garland of skulls as an [head] adornment; snakes as ornaments, a lower garment [made] of a tiger-skin, and so forth.45 It says [they are] accomplishing the business (krtyani > kartavyani) of the Buddhas {buddha^krtyani}. For this very reason, they are protectors, that is to

cajagac cajagac ca jaganti. This parallels Panini's Astadhyayl where vrksas ca vrksas ca vrksas ca is said to explain the plural vrksah (AA 1.2.64). Vilasavajra specifies this here since the compounded stem form jagad is indeterminate as to number. (I am grateful to Prof. A. Sanderson for this reference.) ^traidhatukani. This adjectival form denotes the threefold division of loka ('world') into kamaloka ('the world of sensuous desire'), rupaloka ('the world of [pure] form') and drupyaloka ('the formless world'). 45 See textual note 1.144-6 for discussion of some difficulties in this passage.

78

Chapter 1

Translation

say, discipliners, since they perform the business of the Buddhas. [The attendants are further

described

as

being]

with

bodies

bowed

[in

respect]

{pranataivigrahaih}. The structure [of the first six verses of the Namasamgiti] is [to be understood as follows]: 'Vajradhara, the Fortunate One, [...] accompanied by those [Krodhas ...] whose bodies were bowed [in respect], uttered the following', [what he said being] stated later [in verses 7-15].

[Having bowed to the Lord, the perfectly enlightened one, the Fortunate One, the Tathagata, [he, ie. Vajradhara,] joining his palms [in respectful salutation], and standing before [Sakyamuni] said the following: (6)] What is he [Vajradhara] doing? Bowing [in respect]. To whom? To the Lord. He, [namely, &akyamuni,] is called Lord (nathah > svami)46 because he instructs protectorless creatures. How does it, [that is, the Namasamgiti,] also describe the Lord? As the perfectly enlightened one (sambuddham > samyaksambudddham); not as a £ravaka or Pratyekabuddha. [And the text] qualifies the Lord yet further as the Fortunate One (bhagavan). It so describes him because he is endowed with qualities (bhaga).47 Alternatively he is [so-called] because of he routs (bhanga) the four Maras. Accordingly there is this verse: He is called 'bhagavan' in this world because he destroys (bhagna) those things not conducive to enlightenment, [namely] the activity of the defilements, [re]birth, and also the obstacles of the [subtle] defilements and knowables (jneya).48 46There is a play here between the two senses ofndtha as 'protector' and 'lord'. 47bhaga. Lit. 'good fortune'. ^klesakanna tatha janma klesajneyavrti tathd / yena vaipaksika bhagnas teneha bhagavan smrtah // (nyon mongs las dang de bzhin skye / / de bzhin nyon mongs shes bya'i sgrib / / mi

79

Chapter I

Translation

According to this etymology he is the 'bhagavan' because he destroys (bhagnavan) the defilements and so forth. Alternatively, [returning to the first gloss, he is called 'bhagavan'] because he is endowed with the qualities (bhaga) such as sovereignty. And there is the following verse to this effect: There is the tradition that [the word] bhaga names the following six things: complete sovereignty, beauty, fame, glory, knowledge and activity.49 How is the Lord further described? As the Tathagata. Tathagata means 'one who has attained (gata) suchness (tathata)',50 that is to say, emptiness. Alternatively, he is [called] Tathagata since just as those [past] Tathagatas proceeded, [present ones] proceed, [and future ones] will proceed, so (tatha) he proceeds (gacchati).51 [The word] Tathagata is connected with, [that is, is the object of,] the preceding [phrase] having bowed to. While standing before [£§kyamuni], [and] joining his palms [in respectful salutation], Vajradhara, the Fortunate One, said the following: ["For my benefit, for my sake, out of favour to me, O lord, so that I may obtain [the method called] the Awakening according to the Mayajala, (7)] [Next there is] the passage beginning [with the words] for my benefit. [Here]

mthun phyogs chos gang yin pa / / 'joms pas de 'dir bcom Idan 'dod / Tib.fl86.3.8-186.4.1J ) The term dvrti is here synonymous with dvarana, which would be the normal word in a prose context. Both derive from a AT (sgrib pa is often used for dvarana: MVy 814, 815, 1383, 6512). The terms klesdvarana and jneydvarana have Yogacara origins: the former can be overcome by &ravakas but the latter only by followers of the Mahayana. The latent traces of the klesas can only be overcome by the practice of the bhdvandmdrga. 49These two glosses are also cited by Haribhadra who attributes the second to the Buddhabhumisdstra (AAA 1. 2-3: see ASPP 272, 5-11). I am grateful to Prof. A Sanderson for this reference. See testimonia to Text 1.159-164 for further citations. (See also textual note 1.159, 1.164.) 50The gloss of tatha as tathata exemplifies the commentarial device of glossing a word as its abstract. ^tathdgata is here read as meaning 'so-going', ie. 'proceeding in the same way'.

80

Translation

Chapter I

my benefit means 'what is good for me' {mat^hitam}; [and] since it will spiritually transform [Vajradhara and others] in the future, benefit refers to what is advantageous in the long run. [The verse continues,] 'for my sake, that is to say, for my [ultimate] purpose, out of favour (anukampaya > anugrahaya) to me (me > mama), O lord, may the Fortunate One teach52 so that (yatha), namely, in such a way that (yena prakarena), I may obtain - [obtain] what, did he, [that is, Sakyamuni,] say? - [obtain] the Awakening according to the Maydjala, that is to say, the method called 'the Awakening according to the Mayajala". The syntactic connection with the verb 'teach' that is stated subsequently [in verse 9] is as follows, 'may the Fortunate One [teach] in such a way that I may obtain that [method]'.

[For the benefit of all living beings sunk in the mire of ignorance, whose minds are agitated by the defilements, for the sake of obtaining the highest fruit, (8)] Also [the text says] 'For the benefit of all living beings sunk in the mire of ignorance'. In this expression it is ignorance (ajnanam > avidya) that is compared to a mire since it a place of attachment and because it is difficult to get out of. [Vajradhara's request for instruction is] for the benefit of living beings who are sunk in that mire {ajnanapanka^magnanam}. How also are those beings described? As ones whose minds are agitated by the defilements. The defilements are six in number. And accordingly, in the Abhidharma he, [namely, £akyamuni,] says,53

52An alternative and more natural object for the verb prakdsayatu ('may he teach') is namasamgitim (NS 11). Vilasavajra reads bhagavan in NS 10 rather than bhagavaj-, ie. reading a vocative and not a compound with jndnakayasya, and starts a new sentence with NS 10. The accusative ndmasamgitim becomes the object of dhdrayisydmi in NS 14. 53I take the unstated subject of the verb aha to be £akyamuni, ie. that Vilasavajra follows the view maintained by the Vaibhasikas (but opposed by the Sautrantikas) that the Abhidharmapitaka was the

81

Translation

Chapter 1

[The defilements are] pride, view, doubt, lust, anger and mental confusion. 54 [It is] living beings that have minds that are agitated by these [six] defilements [that] are described [as above in the text]{kle^vyakula-}{klesavyakulaicetasam}. The syntactic connection [with the subsequently stated verb 'teach'] is as follows,

'May the Fortunate One teach for the sake of obtaining the highest fruit'. The highest fruit is Buddhahood. [Vajradhara requests instruction] for the sake of (-praptaye > -artham) obtaining (praptih > pratilambhah) that [fruit].

[May the Fortunate One, the perfectly enlightened one, the teacher, the teacher of the worlds, he who knows the essence of the great pledge[-being], he who knows the disposition of [others'] spiritual capacities, [and] who is supreme, teach. (9)] He, [that is, Vajradhara,] said, 'May the Fortunate One [...] teach. [This should be understood to mean: 'May the Fortunate One who is] the perfectly enlightened one [teach]', that is to say, [teach] as the Eternal One, [namely, as Vairocana]; '[May the Fortunate One who is] the teacher, so-called because he instructs the world, [teach]', that is to say, [teach] as Aksobhya; '[May the Fortunate One who is] the teacher of the worlds {jagadguruh >jagatam guruh} [teach]', that is to say, [teach] as Ratnasambhava; '[May the Fortunate One] who know s

word of the Buddha. ^^mdnadrgvicikitsds ca ragapratighamudhayah. This is the common Abhidharma list of the basic defilements. Vasubandhu refers to them as anusayas in the Abhidharmakosa (AK 5:1) enum­ erating the same items though with two slight differences in terminology and in a differe nt order: mulam bhavasydnusaydh sad rdgah pratighas tathd / mdno 'vidyd ca drstis ca vicikits d ca. Later in the same chapter he has, kdme 'kusalamuldni ragapratighamudhayah (AK 5: 20ab). Here, the second pdda corresponds to the second pdda of Vilasavajra's citation. Perhaps both Vasub andhu and Vilasavajra had the same Abhidharma source in mind. Vasubandhu's Trimsikd 1 ld-12a reads, [klesd] ragapratighamudhayah // mdnadrgvicikitsds ca. This preserves the order where rdgah is placed first by splitting the list across two verses and so preserving the metre. Vilasavajra possib ly took the list from the Trimsikd, reversing the order to preserve metre (Vilasavajra was familiar with the Trimsikd; he cites it in NMAA 6). The list is found in the Dharmasamgraha, which follows the order of AK 5: 1: sat klesdh / rdgah pratighah mdnah avidyd kudrstih vicikitsd ceti (DhSam 67).

82

Chapter 1

Translation

the

essence

{tattva^jnah}

of

the

great

pledge[-being]

{mahasamaya^tattvajnah} [teach]', that is to say, [teach] as Amitabha - [here] pledge[-being] is the Fortunate One, Srivajrasattva; '[May the Fortunate One] who knows (-vit>vetti) the disposition (aSayah > abhiprayah) of [others'] spiritual capacities {indriya£a§ayavit} [teach]', that is to say, [teach] as Amoghasiddhi - [here] spiritual capacities are [threefold, namely,] weak, medium or superior.55 So, the meaning [of the passage] is, 'May the Fortunate One who has the five Tathagatas as his nature [and] who is supreme (parah > utkrstah) teach'.56

[O Fortunate One, [the Ndmasamgiti] of the Knowledge-Being Manju&ri, the Knowledge-Body, he who is endowed with a great coronal protuberance, the lord of speech, the Embodiment of Knowledge, the self-existing one, (10)] Next is the passage beginning 'O Fortunate One, [the Ndmasamgiti of the Knowledge-Being Manjus~ri,] the Knowledge-Body ...'. Its connect­ ion with the verb 'preserve', which is stated subsequently [in verse 14] is, 'O Fortunate One, I shall preserve the Ndmasamgiti which is associated with the Knowledge-Being ManjuSri'. How is [ManjuSri] described? As the KnowledgeBody - which means 'that body which is knowledge' {jnana:kaya} - that is to say, as the Mirror-Awareness. [He is also described] as endowed with a great coronal protuberance {mahaiusnisa} - that is to say, as the Awareness of Equality. He possesses the state of having a great coronal protuberance because he

55 See Abhidharmakosa and Bhasya for a fivefold classification of spiritual capacity (mrdu, madya, adhimatra, adhimatratara, adhimatratama: AK 6: 43ab & bhasya). The list of five is, in effect, identical to the list of three, the NMAA's adhimdtrabeing subdivided into three. 56Sakyamuni is depicted as the embodiment of all five Tathagatas, rather than being identified with Vairocana (portrayed in turn as the embodiment of the remaining four Tathagatas). &akyamuni can thus become a sixth Buddha.

83

Translation

Chapter 1

has the consecrations of all the Tathagatas as his nature. [He is also described] as 'gispatih' - which means lord (patih > svami) of speech (giram > Sabdanam)" that is to say, as the Discriminating Awareness. [He is described] as the Embodiment of Knowledge, [namely, knowledge embodied] {jnana:murtih} that is to say, as the Awareness of the Performance of Action. [He is described] as 'svayambhuh' which means the self-existing one in the sense that he exists (bhu > bhavati) spontaneously (svayam), that is, by the power of his own meditation. Accordingly, in the Sriparamadya [Tantra] it is said: He will realise the Awareness that is Completely Good of his own accord (svayam). 57 Hence [the description of ManjuSri as] the self-existing one - that is to say, as the Awareness of the Perfectly Pure Dharma-Sphere. What does this [identification of epithets with the Awarenesses] mean? [It means] that the Fortunate One, the Knowledge-Being Manjus*ri has the five Awarenesses as his nature. [In the expression] the Knowledge-Being Manju£ri {manjuSri:jfianasattva} 'Manjus"ri'

means

'one possessing splendour (s"rl) which is sweet

(manjuh > komala)' {manjuis'rih}. He is a 'Knowledge-Being' since he dwells in the heart, [that is, the consciousness] of all the Tathagatas. The Knowledge-Being Manju^ri is not the Bodhisartva who is the master of the ten stages (bhumi). Rather, he is Non-dual Awareness (advayajnanam), the Perfection of Wisdom itself. It is for this reason that Dignaga said, The Tathagata is the Perfection of Wisdom, that is to say, Non-dual Awareness. 58

57This citation suggests that svayambhu is being analysed as deriving from svayam + Jbudh. As it stands, svayambhu is problematic since, according to the basic Buddhist doctrine of Dependent Origination (pratityasamutpada), nothing is self-existent. 58This is the opening two padas of Dignaga's Yogacara commentary on the Perfection of Wisdom, the Prajnaparamitapinddrtha (PPA lab).

84

Chapter I

Translation

[In other words, the passage is to be understood as saying,] 'I shall preserve the Ndmasamgiti which is related to the Knowledge-Being ManjuSri who embodies the five Awarenesses'.

[[That] supreme Ndmasamgiti, in which there is deep meaning, lofty meaning, [and] great meaning, which is without equal, auspicious, [and] beneficial in its beginning, middle and end, (11)] How is that [Ndmasamgiti] qualified? In the words that begin '[That supreme Ndmasamgiti] in which there is deep meaning'. [In the expression] in which there is deep meaning {gambhiraiartham} [the phrase] deep meaning indicates meaning relating to emptiness. It is for this reason that he, [that is, Vajradhara,] will preserve it, [that is, the Ndmasamgiti]. [In the expression] in which there is lofty meaning {udaraiartham} [the phrase] lofty meaning refers to extensive (vaipulya) meaning. [The Ndmasamgiti is further qualified as] 'mahartham', meaning in which there is great meaning {mahaiartham}. Moreover, to have great meaning is to possess meaning for all living beings, because it fulfils all their hopes. [It is also qualified as] without equal {aisamam}. Why is it without equal? Because it has the Dharma-Sphere as its nature. [It is called] auspicious (Siva) because it is that in which all elaborations (prapanca) cease. 59 [It is further described as] beneficial in its beginning, middle and end. It possesses three benefits, for the reason that [through it] one obtains delight, joy and calm60 [respectively] in the times of hearing,

59sivam iti sarvaprapancopasamatvat. This must be an allusion to Nagarjuna's Mulamadhyamakakarika (yah pratityasamutpadam prapancopas'amam Sivam: MMK 1.2cd). I have taken prapancopasamam as a locative bahuvrihi following Candrakirti's Prasannapada commentary on the MMK (MadhyS 4). ^The list of seven Auxiliaries of Enlightenment (bodhyanga) contains 'joy' (pnti) and 'calm' (prasYabdhi) as its third and fourth members.

85

Translation

Chapter 1

reflecting and meditating. 61 It is that [Namasamgiti ] that he, [that is, Vajradhara,] "will preserve". [The meaning of] the phrase supreme Namasamgiti is understood, [that is, there is no need for further comment]. [Which was proclaimed by past Buddhas; [and] on account of that, [it is that] which future [Buddhas] will proclaim again, and which the perfect Buddhas of the present proclaim again, (12) Which is recited in the great Tantra [called] the Mayajala by immeasurable joyful great mantra-bearing Vajradharas, (13) And I, of firm intention, will preserve that [Namasamgiti] until cessation. So that I may be, O lord, the preserver of the secrets of all the perfectly enlightened ones [may the Fortunate One teach]. (14)] How is the Namasamgiti further described? [With the words] starting '[which was proclaimed] by past [Buddhas]'. [It is the Namasamgiti] 'which was proclaimed by past Buddhas, and on account of that (hi > yasmad) [it is that] which those in the future who are perfectly enlightened will proclaim again; and it is also the Namasamgiti which the Buddhas of the present, the Fortunate Ones, proclaim again'. [Next is] the passage, [that is, verse 13,] beginning with [the word] 'mayajala'. [It qualifies the Namasamgiti as] that 'which is recited in the great Tantra [called] the Mayajala by immeasurable great Vajradharas, bearing great mantras, who are joyful'.

61 These constitute the Three Wisdoms (prajfia): srutamayi prajna, cintamayi prajna, bhavandmayi prajna (see AKBh on AK 6: 5).

86

Translation

Chapter 1

[The text continues, in verse 14,] 'And I will preserve that [Namasamgiti] until cessation (a niryanat > nirvanam yavat)' How is he, [that is, Vajradhara, described]? He is [described as] of firm (drdhah > avicalah) intention (aSayah > abhiprayah) {drdhaias'ayah}'. What this, [that is to say, verses 12-14,] 62 means is that the Namasamgiti is [a teaching] which is proclaimed by the Tathagatas of the three times. Therefore, this Dharma-teaching is taught as being [a unconditioned entity] like space. There is a quotation to this effect: Whether Tathagatas arise or whether Tathagatas do not arise the essential nature (dharmata) ofdharmas remains the same. 63 [Next comes the line] beginning 'So that I may be ...'. This is connected with the previously stated verb] 'teach' [of verse 9, as follows], 'So that I may be, O lord, the preserver of the secrets of all the perfectly enlightened ones, may he, [that is, £akyamuni] teach'. [The word] preserver (dhrk) here, is the correct reading. 64 [I will teach [it] according to the particular dispositions of beings, (15ab)] 'Having understood that [Namasamgiti] properly I will teach [it] according to the particular dispositions (-vis'esatari > -vis"esat) of all beings.' With what purpose [will he teach it]?

62 Tib. orders the text differently here, placing this paragraph and the subsequent quotation earlier. See textual note 1.247. 63This is a well-known quotation in Buddhist works, cited by Vasubandhu as coming from a sutra (AKBh on AK 3: 28ab). In the Mahayana revision of the Sarvastivada Abhidharma dharmata is an unconditioned entity (asamskrtadharma), as is space (akasa). Buddhas, on the other hand, are conditioned entities (sarnskrtadharma). (See testimonia to Text 1.254-5 for further citations of the verse and textual note 1.254. See AKBhPru. 412-3; 515, note 192, for Pali version.) ^This statement presupposes the existence of variant readings, though none are attested in the extant NS MSS edited (see NS.Dav. 50).

87

Translation

For the destruction of all the defilements [and]

Chapter I

for the

abandonment of all ignorance." (15cd) Here, by all the defilements are meant [both] the defilements (klesa) and the products of the defilements (upakleSa). 65 [Vajradhara will teach the Ndmasamgiti] for the sake of their destruction {a$esakle$a£nas"aya} [In the phrase] for the abandonment of all ignorance, [the words] all ignorance means ignorance with respect to all objects. It is for the sake of that abandonment that he will teach [the Namasamgiti].

[Having thus requested the Tathagata for instruction, Vajrapani, the lord of the Guhya[ka]s, 66 joined his palms in respectful salutation and stood before [him] with his body bowed [in respect]. (16)] [Having] thus begins [the next passage to be discussed]: 'Having thus (evam), that is, 'in this way' (anena prakarena) requested the Tathagata for instruction, the Fortunate One Vajrapani, [namely, Vajradhara,] the lord of the Guhya[ka]s (guhyendra) [...]'. He is called 'guhyendra' here because he is king (indrah > raja) of the Guhya[ka]s {guhya^indrah}, that is to say, [king of] the Yaksas dwelling in Adakavatl. 67 [The verse concludes, 'Vajradhara] joined his palms in respectful salutation [and] stood before [£akyamuni] with his body bowed

65Vasubandhu defines upaklesa as the offspring of klesa, giving the example of hatred (kle£a) engendering anger (upaklesa) (AKBh on AK 5: 46). 66The translation 'lord of the Guhya[ka]s' for guhyendra, rather than 'lord of the Guhyas' is suggested by Vilasavajra's gloss and the presence of the epithet guhyakddhipah in NS 22. 67This gloss gives the potentially identical epithets guhyendra and guhyardt, fromNS 1, different senses. Adakavatl (Icang lo can Tib.[187.4.4-5f) is the capital of the yaksas. The Mahdvyutpatti (MVy 4137) gives Icang lo can for both AtakavatI and AlakavatI, and Das (p. 399) identifies the former as the abode of Kuvera or Vaisravana and the latter as a place name. However, AlakavatI is probably a variant of AtakavatI, and Tib. 's use of the same word for Adakavatl reinforces Edgerton's suggestion that AtakavatI and Adakavatl are to be identified (BHSD 8). See Edgerton (ibid. ) for textual references to Adakavatl.

Translation

Chapter 1

(prahvakayah > pranatakayah) [in respect]. The request for instruction is an entreaty accompanied by a preparatory act (samskara). In this case the preparatory act is the respectful salutation and so forth. [Here ends] the chapter on 'The Request for Instruction', the first in the commentary on the Aryanamasamgiti [called] "An Explanation of the Meaning of the Name-Mantras". Sixteen verses [are commented on].

89

Chapter 2

Translation

[Chapter 2. On 'The Reply'.] [Then £akyamuni, the Fortunate One, the perfectly enlightened one, supreme among the two-footed, extended out of his mouth his long, full, beautiful tongue; (17)] Following the request for instruction, he, [that is, £akyamuni,] replies in the passage beginning, 'Then Sakyamuni'. [The word Sakyamuni means] 'sage (muni) of the Sakya clans'. He is called a sage because he possesses silence (mauneya) in [the spheres of] body, speech and mind.68 The meaning of Fortunate One and perfectly enlightened one is understood, [that is to say, there is no need for further comment]. 69 [Sakyamuni is described as] supreme (uttamah > Sresthah) among the two-footed (dvipada^uttamah), that is to say, [supreme among] gods and men. [Having first extended it, £akyamuni] withdrew (upasamhrtya < nirnamayya) into his mouth (mukhe < mukhat)70 his long (ayatam > dirgham), full (sphitam > bahalataram), beautiful - since it resembles a conch shell [with a spiral] that turns to the right - tongue. And it is [also] the store of all teachings (dharma).

[Displayed a smile to living beings purifying the three lower destinies, illuminating the three worlds [and] chastising the enemies, the four Maras, [18]] What does [£akymuni] do [next]? The passage starting, '[Displayed] a smile' explains. Its structure is as follows: '[he] displayed a smile to living beings

^kayavakcittamauneyayogan munih. In chapter 6 Vilasavajra uses the same explanation in his account of NS 42b (mahamauni mahamunih), "['With the great rank of a sage' (mahamauni).] A sage (munih) is [so-called] because he possesses silence in [the three spheres of] body, speech and mind." (kayavakcittamaunayogan munih: B.22v9) 69See above, NMAA 1 on NS 6, for Vilasavajra's treatment of bhagavdn. 70Vilasavajra reverses the literal meaning of the text.

90

Chapter 2

Translation

(lokanam > sattvanam), purifying the three lower destinies, that is to say, purifying [the realms of] hell beings, animals and ghosts, and [then] replied to Vajrapani'. There are three signs of the Tathagatas' being about to teach the Dharma, namely, a smile, a glance, and the emitting of rays of light. Sometimes one [of the signs occurs], sometimes two, sometimes [all] three occur together. For example, in the Srimayajalatantra two are mentioned, the smile and the glance. Sometimes all three [are found], as in the Perfection of Wisdom in Twenty-Five Thousand Verses [where] Maitreya's glance, [as well as] his smile and the emanation of light-rays from his face occur. Sometimes there is one [sign], as in this very Namasamgiti, where it says, "[£akyamuni] displayed a smile to living beings". And this smile is said to be of four kinds: 71 one that is accompanied by emanation from the major and minor limbs; one that is accompanied by emanation from hair pores; one that is accompanied by emanation of natural light; [and] one that is accompanied by emanation from the sense [organ] of the tongue. That [last] one among these is the 'tongue-miracle'. It is a prior-sign of the [Tathagatas'] being about to teach. It, [namely, the text,] demonstrates [that, namely, Sakyamuni's smile and its effects72] with the passage beginning 'illuminating the three worlds'. Its structure is as follows: '[he] displayed a smile illuminating, that is, causing a display of, the three worlds, namely, the earth, the atmosphere, and the heavens, and [then] replied to Vajrapani.'

71 /ac ca smitam caturvidham ('dzum pa de ni rnam pa bzhi ste Tib.[188.1.3]). Tib. supports the Skt., though what follows could be an enumeration of the different types of the third sign, that of the emanation of light. Given that 6akyamuni's smile in the NS is said to "illuminate the triple world", the smile and possible accompanying effects are seen as separate from the emanation of light as a sign in its own right. Vilasavajra appears to be suggesting that the present case is one of the 'tongue-miracle'. 72The intent of this passage is rather obscure and my addition is conjectural.

91

Translation

Chapter 2

With the words, "chastising the enemies, the four Maras', [the text now] states how that smile is further qualified. The four Maras are: the aggregates, the defilements, death, [and] the god.73 Those same [maras] are enemies {mara:arayah} and their chastising {marari^asanam} means their 'instruction' (anuSasanam) [or] 'punishment' (nigrahah).

[And replied to Vajrapani, of great strength, lord of the Guhya[ka]s, with a sweet sublime voice filling the three worlds: (19)] "Having displayed a smile", he, [that is, Sakyamuni,] replied to Vajrapani with a voice (gira > vaca) filling the three worlds. With what kind of [voice did he reply]? With a sweet one. Its sweetness is due to it being pleasurable to the ear. [His voice is also described as] sublime (brahml). This is because it is endowed with brahma-sound (brahmasvara). The word 'brahma in ' brahma-sound' is the name for that which is great, that is to say, for that which is of great quality. 74 As it is said, Sublime (brahmam) merit is produced in the many thousand assembly [of worlds]. One should understand the phrase, "thousand[worlds]", to refer [here] to a great trichiliocosm.75 The meaning [of the preceding quotation] is thus, 'in the assembly consisting of many great trichiliocosms'. And accordingly it is said, He pervades the great trichiliocosm with his voice. Just as those inside hear, so also do those outside.

l^devaputrah. The literal sense, 'son of a god', is not accurate since gods cannot literally have sons and Mara has no father. Thus devaputra is a synonym of deva 74This gloss is based on an etymology of Jbrh meaning 'to grow great' or 'to grow strong' (DhP 17.85). I am grateful to Prof. A. Sander son for this reference. l^trisahasramahasahasro lokadhatuh. Lit. 'the world system consisting of a triple thousand great thousand (worlds)'.

92

Chapter 2

Translation

Here, those 'inside' are [those] within this great trichiliocosm; those 'outside' are those that are excluded from [this] world-system. So brahma- is used because of [its denotation of the] greatness [of the sound]. [The expression] 'brahma-sound* [literally] means the sound of that {brahma^svarah}, [that is to say, the sound of brahma, in other words "great sound"]. Alternatively [the word] brahma refers to £akyamuni, the Fortunate One. [In this case] the 'sublime speech' (brahmi vak) is the speech of that Brahmana, [that is, of Sakyamuni]. 76 [The passage can be understood, therefore, to be saying that], 'with just that one sublime voice filling (vyapnuvantya), that is to say, pervading (apurayantya), the three worlds, he replied to Vajrapani'. And accordingly, it has been said: Just as there is but one voice, though with many languages, different in every country, possessed by the destroyer of birth, so, since it is unfaltering [and] clearly articulated, it is the cause of the awakening of all people.77 ["Well done! O Vajradhara, glorious one. Good for you, Vajrapani. Since you, who are so compassionate for the sake of the benefit of the world, (20) are eager to hear from me the Namasamgiti of the Knowledge-Body ManjuSri

that is of great

meaning, auspicious, [and] destructive of evil, (21) I will teach you [this] good thing, O lord of the Guhyakas. (22ab)] [Next is] the passage beginning 'Well done! O Vajradhara'. Its structure is as follows: 'Good for you (te > tava), Vajrapani ([vajrapanaye >] vajrapaneh). O

76Vilasavajra's glosses of the word 'sublime' used to describe Sakyamuni's voice (brahmya gira) are that it is, i. a voice possessing great qualities; ii. his own (Buddha-like) voice. 77For discussion of the view that the Buddha taught the whole of his teaching by uttering a single sound see E. Lamotte, The Teaching of Vimalakirti. (tr., S. Boin) 1976, 12, notes 2-5.

93

Translation

Chapter 2

Vajradhara, glorious one, since (yat > yasmat) you are eager to hear from me the Namasamgiti which is associated with the Knowledge-Body Manjusri, I will therefore (tat > tasmat) teach you (te > tava), O lord of the Guhyakas, [this] good thing'. 78 '"Why have you, [O Vajrapani,] asked me, [that is, why have you made this request]? [It is made] for the sake of the benefit of the world, that is to say, as a favour to the world, because you, [O Vajrapani,] are so compassionate. [And] how is this Ndmasamgiti which you are eager to hear further qualified? By the adjectives 'pavitram' and 'aghanaSinlm'. The first means auspicious (pavitram > mangalyam) and the second destructive of evil (aghanaSinim > papanaSinim)."

[Listen to it well with a one-pointed mind." "O Fortunate One, [so be it." replied Vajrapani.] (22cd)] Listen with a one-pointed mind. This means, 'Do not [listen] in the manner of a pot that is face-down, leaky or impure, but listen well'. 79 When the Fortunate One had spoken thus, Vajradhara said 'O Fortunate One', that is, 'O Sakyamuni, so be it.'; in other words, 'I will listen in the way [you have] just stated'. [Here ends] the chapter on 'The Reply', the second in the commentary on the Aryanamasamgiti [called] "An Explanation of the Meaning of the Name-Mantras". Six verses [are commented on].

78See textual note 2.50 on Vilasavajra's reading of NS 20-22. 79H. Guenther quotes and discusses a citation of this simile in the rDzogs-pa-chen-po klongchen-snin-thig-gi sngon-'gro'i khrid yig kun-bzang (Guenther, 1959: 38, note 25).

94

Chapter 3

Translation

[Chapter 3. On "The Survey of the Six Families'.] [Then Sakyamuni, the Fortunate One, having surveyed the entire great family of mantras - the family of mantra spell-holders, the triple family, ... (23)] Now immediately following the reply [to Vajradhara] he, [that is, Sakyamuni,] surveys the six [mantra] families in the passage beginning, Then Sakyamuni, the Fortunate One. Here, [the phrase] the entire great family of mantras is an introductory summarising statement80 referring [to the families] in general [terms only, without going into particulars]. The structure [of the whole passage, that is, NS 23-4], together with the [main] verb, ["uttered",] which is stated later [in verse 25], is as follows: 'Having surveyed those [families], he, [that is, £akyamuni,] uttered this verse having the characteristic of non-origination, pertaining to the Lord of Speech, namely, the Knowledge-Being ManjuSrT. The family of mantra spell-holders {mantravidyadhara^kulam}. [In this expression the term] 'vidyadhara' means 'spell (vidya)-holder (dhara)', [that is, something containing a spell] {vidya^dhara}. [The term] 'manrravidyadhara' denotes a spell-holder that is of the nature of a mantra! mantra:vidyadhara}. [It does] not [denote a spell-holder] that is of the nature of a dharani.^ The family of mantra

^grahanakavdkyam (sdud par byed pa'i tshig Tib. [188.3.2]). This term, used also in chapter 5 (Text 5.270) to describe the NS phrases containing the Name-mantras of Manjusrljnanasattva, seems to be a technical term having the sense of 'an utterance that grasps [the essence of something], ie. a summary statement. 81 Since 'spell' serves equally to translate both dharani and vidya I have left dharani untranslated in order to preserve their distinction. Though the grammatical analysis is clear, the sense of this passage is obscure: mantravidyddhara is analysed as a karmadharaya whose elements are mantra and vidyadhara. The option of taking it as a tatpurusa with a dvanda as the first element, which makes more immediate sense (ie. 'holder of mantras and vidyds') is not followed. Also, the function of the suffix -atmaka added to both mantra and to dharani is unclear (cf. NMAA 5.48-9, mahatmako rdgas ca mahdrdgah). Vilasavajra seems to be making a distinction between mantras and dhdranis as

95

Chapter 3

Translation

spell-holders is the Action family [of Amoghasiddhi]. The triple family [is so called] because it has [the triad of] body, speech and mind as its nature. It is the family of Vairocana.

[... the family of the seen and that which is beyond the seen, the great family of the sight of the worlds, the foremost family of the Great Seal, [and] the great family of the great coronal protuberance - (24)] The family of the seen and that which is beyond the seen {lokalokottara^kulam}.82 [The word] 'lokah' is to be derived from the root Jlok in its meaning of 'to see'. 83 [Furthermore it is to be understood] in the sense of 'what is seen' [rather than 'the seer']. [This family] is the lotus family [of Amitabha]. The great family of the sight of the worlds {loka^aloka}. This is Aksobhya['s family] since he has the Mirror-like Awareness as his nature.84 The great family of the sight of the worlds {lokaloka^kulam} is the vajra family.

different sorts of things that contain vidyas. The family under consideration is characterised as a holder of the former. 82 'that which is beyond the seen' (lokottara). The text does not gloss lokottara and thus its translation as "what is beyond the seen" is conjectural, based on the approach of the gloss on loka. A passage may be lost due to scribal misreading: the eye could easily jump a section if, as is likely, the gloss on lokottara started with some form of the word lokah. If a passage has been lost it was also missing in 776.'s exemplar since Tib. contains no extra material (Tib.[1883.6-7]). In his analysis of loka Vilasavajra appears to be making a link with the Discriminating Awareness (pratyaveksanajfiana) of Amitabha, head of the lotus family with which the lokalokottara family is identified. The root from which the word pratyaveksana derives also means 'to see' (Jiks). The term lokottara, often translated as 'supramundane' in distinction to 'mundane' (loka), was used by the eponymous Lokottaravadin School to refer to the Buddha's life and activities (as being non-ordinary). The Mahavastu describes Bodhisattvas as enlightened with regard to loka and lokottara (MV i.86.4: cited by BHSD 465). Whether Vilasavajra has this sense in mind is not clear. ^lokr darsane (= DhP 1.76). 84Vilasavajra appears to be linking aloka to adarsa and thus to adarsanajnana, the Mirror-like Awareness of Aksobhya. Both words are composed of the verbal prefix a and a root meaning 'to see' (a Jlokand a Jars). The Skt. gives a tatpurusa analysis of lokdhka, "lokaloka means 'sight of the worlds'". Tib. differs, taking it as 'illuminating the worlds' but this loses the link with adarsa (lokanam aloko lokalokah; 'jig rten mams la gsal bar byed pa ni 'jig rten snang byed Tib. [188.3.7]).

96

Translation

Chapter 3

The foremost Great Seal family. [This] is the family of [the Tathagata] Bodhicittavajra [and], therefore, it is foremost. [This is] because Vairocana and [the other Buddhas of the directions] have the bodhicitta[-SQa\] as their nature. 85 The Great Seal family is the family connected with that [seal] {mahamudra^kulam}. The great family of the great coronal protuberance (mahosnisakulam). All the Tathagatas possess a great coronal protuberance because of the bestowal on them of the consecration of Lordship of the Dharma throughout the triple-sphered world. 86 The family associated with that [state of having a great coronal protuber­ ance] 87 is [called] the great family of the great coronal protuberance. It is the jewel family. 88 [Here ends] the chapter on 'The Survey of the Six Families', the third in the commentary on the Aryanamasamgiti called, "An Explanation of the Meaning of the Name-Mantras". Two verses [are commented on].

85Thus, Bodhicittavajra's family is 'foremost' since bodhicitta is the essential mark (mudra) of all other families, and Bodhicittavajra is the hypostasisation of bodhicitta. 86An alternative translation of this passage is, "The condition of having a great coronal pro­ tuberance is a result of the bestowal of the consecration of Lordship of the Dharma throughout the triple sphered world of (ie. 'by'?) all the Tathagatas ". Cf. chapter 5 on NS 33 (Text 5.91-2): mahattvam cintdmaninipatayd sarvatathagatanam traidhdtukadharmardjydbhisekdtmakatvdt. This latter passage connects possessing the consecration of the Lordship of the Dharma with being like a wishfulfilling jewel (cintamaniriipataya), thereby providing a rationale for the Mahosnisa Family being the Jewel Family. 87Vilasavajra analyses mahosnisakulam as mahosmsatvakulam (mahosmsatvam / tasya kulam mahosnisakulam mahad Text 3.21-22). 88For a summary of the correlations between mantra-families, Tathagatas and family name see Table 6 (Introduction, section 5.7).

97

Translation

Chapter 4

[Chapter 4. On 'The Method of Awakening According to the Mayajala' .] Now I shall teach 'The Method of Awakening according to the Srimayajala [Tantra]\ which has come down through the line of my gurus, and is included within the Srimayajala [Tantra] in sixteen thousand [verses]. Now, at the start of this [method], the Bodhisattva who 'practices the observances of the Way of Mantras', [and] who desires to make manifest 'this crest-jewel called the Namasamgiti, which has non-duality as its ultimate nature, [and which is] associated with the Knowledge-Being ManjuSri, the Knowledge-Body of all the Tathagatas, [and] the Embodiment of Knowledge',89 [this Bodhisattva] who is intent on the protection of the whole mass of beings without remnant, who possesses the mind of aspiration90 for the sake of Great Awakening,91 who embodies both wisdom and means, having attracted beings by means of the four elements of attraction,92 should generate a resolve to attain the highest perfect enlightenment. And thus he, [that is, the Bodhisattva,] says, May all the Buddhas and Bodhisattvas bear witness to me. I, called so and so,

89The single inverted commas indicate material from the NS anusamsdthat has been incorporated into the text (Text 4. 2-4). In the Sanskrit the incorporated material is continuous: mantramukhacaryacari imam bhagavato manjusrijnanasattvasya sarvatathagatajnanakayasya jnanamurter advayaparamartham namasamgitim nama cudamanim (NS.Dav. 65, 3-5). See textual note 4.4 on the reading namasamgitim nama cudamanim. 9Qpranidhdnacitta. The 'mind of aspiration' and 'the mind that has set out' (prasthanacitta) constitute the twin aspects of the bodhicitta, the former being a prerequisite for the development of the latter. Thus, £antideva: tad bodhicittam dvividham vijnatavyam samasatah / bodhipranidhicittam ca bodhiprasthanam eva ca // (BCA 1.15) ^mahdbodhi. This term is used by the Vijnanavada to denote the Enlightenment resultant upon the purificatory transformation (paravrtti) of the eight consciousnesses (vijnana). (See VijnMaSi 684ff). 92samgrahavastu. These are the four means by which a Buddha or Bodhisattva draws beings to the Dharma. The Dharmasamgraha gives the standard list: catvdri samgrahavastuni / danam priyavacanam arthacaryd samdndrthatd ceti, "The four elements of attraction are generosity, friendly speech, acting for the benefit [of others], and impartiality" (DSam 19).They are acquired by the Bodhi­ sattva as part of his practice of the Perfection of Means (upayaparamita). See H. Dayal, 1932, 251-9.

98

Chapter 4

Translation

from this time onwards until I am seated on the terrace of enlightenment [make the following vows], "Just93 as the lords, [that is, the Buddhas,] of the three times94 have resolved to attain perfect enlightenment, [so] shall I raise the highest unsurpassed thought of enlightenment (bodhicitta). I firmly undertake the threefold training in morality, the amassing of good moral qualities, and the morality of serving the interests of beings.95 Starting from today I shall accept the Buddha, the Dharma, and the Sangha, the supreme excellent triple-jewel, [which is] the vow, arising from the union, [that is, the family,] of the Buddhas. In the excellence of the great family of the vajra, I shall take up the vajra, bell and gesture96 in their true nature; [and] I shall accept the teacher. In the excellent, great family of the jewel, in the delightful pledge,97 I shall offer four gifts six times a day.98 In the pure, great family of the lotus, which arises from the Great Awakening,99 I shall take up the true Dharma which is both exoteric and esoteric and has three paths. 100

93From this point onwards the citation is also found (again, as a citation) in the Sarvadurgatiparisodhana Tantra (SDPS, 146, 8-25). 94traiyadhvikd ndthdh ('lords of the three times'). The three times are the past, present and future (see the gloss below on NS 26's "Buddhas of the three times": Text 4.147). 95sattvarthakriyasilam. This may be an allusion to arthacarya, the third 'Means of Attraction' (samgrahavastu). Dayal (1932, 254) cites arthakriya as an alternative to arthacarya. ^mudra. This refers to hand gestures that are either drawn on a mandala (cf. SDPS 244, 22) or made by the practitioner in a ritual context. They may symbolise deities as well as accompany different ritual actions such as the invocation, worship and dismissal of deities. ^samaya ('pledge'). In translating samaya as 'pledge' rather than 'assembly' I follow Tib. and SDPS.Tib.' s dam tshig. 98 See chapter 5, comment on NS 36a, where are mentioned "the Four Gifts of worldly objects, fearlessness, the Dharma and consecration". 'Consecration' (abhiseka) appears to be a Tantric 'fourth' added on to the familiar tripartite classification ofddna. "mahdpadmakule suddhe mahdbodhisamudbhave (byang chub chen po las 'byung ba'i Tib.) (Text 4.22). My translation follows Tib.. Skorupski translates, "In the pure family of the great Padma, in which there arises the great Enlightenment" (SDPS 19, 1-2), following SDPS.Tib. (byang chub chen po 'byung ba la). This seems the less straightforward interpretation, though both are possible. ^saddharmam pratigrhndmi bdhyam guhyam triydnikam (Text 4.21). The 'three paths' are likely to be the ways of the &ravaka, Pratyekabuddha and Bodhisattva rather than the Hinayana, Mahayana and Vajrayana (the term Vajrayana is not common usage in the Yoga Tantras). Tib. takes bdhyam and guhyam to qualify triydnikam rather than saddharmam (phyi dang gsang ba'i theg pa sum) but this creates problems of interpretation. What are the exoteric and esoteric forms of the way of the &ravaka etc. (or indeed the Hinayana etc)?. SDPS.Tib. takes them as separate qualifiers of

99

Translation

Chapter 4

In the excellence of the great family of action, I shall take up in its true nature the vow that is all-inclusive, [namely,] the performance of devotion (piija), to the utmost of my ability. Having raised the highest supreme thought of enlightenment, [and] taking up the complete vow, [that is to say, all the vows,] for the sake of all beings, I will cause those to cross over who have not crossed over, I will free the not-free, I will encourage the discouraged, [and] I will place beings in Nirvana. 101 " Having generated the thought of enlightenment in accordance with the taking of the vows, he should develop the perfect understanding that is of five forms 102 by the following method: As for the first, the Mirror-like Awareness, he should develop [it] by means of reflection. Just as classes of entities, being associated in place and time, are reflected in the orb of a well-polished mirror, so, perceiving the mass of beings, differentiated by differences of destiny, region, and birth, in the consciousness of Great Awakening (mahabodhicitte), the outflow of the perfectly pure Dharma-Sphere, [and] having perceived them there within that very spotless Mirror-like Awareness, he should then generate compassion that takes beings as its object. He should behold all creatures in the Mirror-like Awareness as they really are, thinking, "Oh! Alas! Beings are protectorless, lack a refuge, have no ultimate goal, are suffering, sunk in the ocean of transmigration. For the sake of those beings, just as I shall free myself entirely, completely, in all respects, from all suffering, so likewise [shall I free] all beings".

saddharmam (phyi dang gsang ba theg pa sum: SDPS.Tib. 147, 20). 101 The Sarvadurgatiparisodhanatantra also closely parallels this paragraph: karundvasena sarvasattvottaranaya bodhicittam utpadayed atirnatdrandydmuktamacandydndsvastdsvdsanaydparinirvrtaparinirvdtandya sakalasattvadhdtoh samsdrasamudrdd uttdrandya ca (SDPS 142, 2-4). ^pancakarabhisambodhim. See Introduction 5.8 for discussion of the five jndnasas the five forms of abhisambodhi.

100

Chapter 4

Translation

Then he should apply himself zealously to [developing] the Awareness of Equality, by means of the sameness that is the selflessness of all phenomena. He should enter into the selflessness of all phenomena contemplating, "Just as all phenomena are empty, without a self, so also am I empty, without a self. Accordingly he, [that is, Sakyamuni,] has spoken in such a way as, "Just as [one understands] oneself [to be], so should one understand all beings; just as [one understands] all beings [to be], so should one understand all phenomena". Then, through the realisation that the nature of any phenomenon is its selflessness, he should generate compassion that has phenomena as its object by means of the Awareness of Equality. And then, uttering the mantra, THE SOUND A IS THE FIRST OF ALL DHARMAS SINCE IT IS UNARISEN FROM THE BEGINNING, 103 the mantrin should realise the meaning of the mantra entirely [and] in brief, having contemplated the essencelessness of phenomena along with [compassion] that is without an object. In this way the Discriminating Awareness [is developed]. Immediately following [this], he should realise the Awareness of the Emptiness that is the Perfectly Pure Dharma-Sphere, which [Awareness] is accompanied by objectless great compassion. [It] is completely free from all subjects and objects; 104 it is radiant by nature; 105 [and] it has a nature in which parts are extinguished.

mukham sarvadharmdndm ddyanutpannatvdt (Text 4.46). This mantra occurs twice in the Sarvadurgatiparisodana Tantra (SDPS 160, 15; 282, 13-14). In chapter 1, with the syllable om at its beginning, it precedes the generation of the main mandala. In chapter 3, following verses of praise, it accompanies an oblation to the guardians of the ten directions (daSadiglokapalah). According to Beyer (1973, 146), it is used as a general purpose mantra. Skorupski's translation, "The syllable A is the source of all the dharmas on account of their fundamental non-origination" (SDPS 26; 99), is inadequate since the reason in the ablative case (adyanutpannatvat) must qualify the subject ( 'the syllable A'), not the complement ('the source of all dharmas'). ^grdhyagrdhaka. Lit. 'that which is grasped and that which grasps'. This Yogacara term describes the world as experienced by the unenlightened mind, ie. composed of a duality of subjects (grahaka) and objects (grahya), the result of the erroneous partition of the non-dual flow of perceptions (vijnapti) (see MSam 2.3). ^^prakrtiprabhdsvara. The description of consciousness as 'radiant by nature' has its roots in

101

Translation

Chapter 4

And then the yogin - who is, [at this point, immersed in non-discursive Awareness, and so] separated from acting for the sake of all beings - aroused by virtue of his mental continuum, which has been affected by his earlier resolution, 106 should cause the Awareness of the Performance of Action to arise by means of objectless great compassion. It is by that [Awareness] that one accomplishes the [ultimate] interests of all beings, since that [Awareness] is the pure worldly knowledge that is attained in the wake of the non-discursive Awareness. 107 And that [Awareness] should be understood to be the mind (citta) that is the thought of enlightenment that has the characteristic of setting out (prasthanalaksana), [the mind] that has conventional and ultimate [truth] as its nature, [and is] by nature pure. 108 Then the yogin should imagine his consciousness to be the letter A, having a form like that of a magical creation, located in [empty] space, naturally radiant, like a star. Arising out of that [syllable A], he should visualise himself as the great syllable HUM, dark blue in colour, endowed with masses of flames of wrath, surrounded by many hundreds of thousands of rays of light. It, [that is, the syllable HUM] is allbeneficent, the supreme bliss. He should imagine that rays of many kinds, [that is, of many colours] are emitted from that syllable HUM and that all the Tathagatas

early Buddhism (cf. pabhassaram idam cittam: Anguttara 1.10). In Mahayana Buddhism both the Yogacara and Tathagatagarbha traditions adopted the term. See La Vallee Poussin, VijnMaSi 109-113, for discussion of its use. 106Ie. the resolve to attain Perfect Buddhahood (samyaksambodhi) as embodied in the Buddhafamilies' vows. ^ Q1nirvikalpajnana. Again, this is Yogacara terminology. In Xuanzang's account of the four jndnas it is associated with the Mirror-like Awareness (VijnMaSi 688); according to the Mahdyanasamgraha when perceptions are burnt in the fire of nirvikalpajnana the truth about things, the Perfected Nature (parinispannasvabhava) is revealed (MSam 3, 29). W^vaiyavadanikam 'pure'. Xuanzang distinguishes dharmas as sdmklesika ('defiled') or vaiyavaddnika ('pure'), the two terms being vrddhi forms of samklesa and vyavaddna. (See VijnMaSi 182: La Vallee Poussin reconstructs vyavaddnika from the Chinese).

102

Translation

Chapter 4

[manifest] from the ends of the rays; and that they, [that is, the light-rays, together] with all those Tathagatas, go to all the Buddha-fields [and] purify all the mass of beings in all the world-spheres without remainder. Having generated the thought of enlightenment and so forth [in those beings], 109 up until the revealing of the miracles of all the Buddhas, they return back again [and] enter that syllable HUM. He should [then] visualise a knowledge-vo/ra, which is the transformation of that [syllable HUM]; which possesses the six colours - dark blue, [yellow, red,] and the rest; which pervades the six Buddha-fields such as Abhirari; which is encircled by flames, [and] unapproachable by the enemies that are all the Maras; [and] which has six faces 110 [and] five prongs. Having visualised a white syllable BHRUM in the interior of the core, [that is, inside the central sphere] of that vajra t [and] having visualised a caitya through the transformation of that [syllable BHRUM], made of the four jewels, big as the trichiliocosm, [he should visualise] a dharmodaya 111 by means of a syllable HRlH in the centre of that [caitya]. Above that [dharmodaya], through [the transformation of] a letter A, [he should visualise] a crossed-v#/ra [and]

^bodhicittotpddddikam krtvd. The Sanskrit appears to say that the light-rays with the Tathagatas are generating the bodhicitta and so forth in themselves. The causative karayitvd would give a better meaning: "Having caused the arising of the bodhicitta and so forth to occur, ...[in those beings]". It is likely that the 'mass of beings' are to be visualised as going through the stages of the Path rather than the light-ray Tathagatas. As Tathagatas they do not need to go through such stages (though they could be interpreted as displaying the stages of the Path as a show in order to convert and benefit beings). The function of the visualisation appears to be to create a pure (and next, a protected) realm within which the evocation of the mandala and central deity can proceed; visualising all beings as enlightened accords with such an aim. ^^sanmukham 'with six faces'. I am not certain to what this refers. Later in the sddhana lightrays emanate from the central deity, Mahavairocana, and illuminate the Saha world; then leaving 'the six vajra-faces' they pervade all Buddha fields (Text. 4.183-186). Could it be a triple vajra, ie. a visvavajra with a further vajra in the third dimension? This would result in a vajra with six faces, one in each of the cardinal directions plus the zenith and nadir (often vajras have makara faces out of whose mouths the prongs of the vajra emarge). Wayman (1980, 229, note 25) cites a passage by Padmavajra that mentions a four-faced vajra (rdo rje kha bzhi) that is visualised in the heart; but whether or not this is a visvavajra is not made explicit. Since it is encircled with flames and unapproachable by Maras, the function ofthejndnavajra must be to create a protected space - the size of the 'six Buddha-fields' - for the performance of the sddhana. 111 dharmodaya. Lit. 'the source of [all] dharmas', this is represented as a triangle.

103

Translation

Chapter 4

with that [crossed-vq/ra] he should empower a white ground constructed of vajras. Then one should imagine a temple [on that ground] with a roof of vajras, gems and jewels, transformed out of the syllable BHRUM. And how is that [temple] described? He, [that is, Sakyamuni,] has said, It has four corners, four gates [and] is adorned with four arches. It is decorated with bells, ornamental strips of cloth, banners and so forth; [also] with garlands and Yak-tail plumes. 112 At all the corners and in the joints of the projections of the doors it is inset with vo/ra-like jewels [each of] which has a sickle-moon joined with it. It is beautified with various colours, [and] adorned with strings and half-strings of pearls. He should depict the outer mandala with powders of five colours. Having entered its interior enclosure, which is circular, he should then put in place the second [mandala] which is called [the mandala of] Vajradhatu. It is encircled with garlands of vajras [and] adorned with eight pillars. It is depicted with the five powders and ornamented with half-strings of pearls and so forth. And whatever characteristic has been specified for the outer mandala, all of that he should apply fully [and] in particular to this [inner mandala] also. Only one thing he should not produce, [namely] the arch at the gate-joints; and therefore he should not create gates [or] arches for that [circular enclosure]. 113 And afterwards, he should arrange the thrones of the deities in the following way. Among those [thrones] he should place the throne of the principal [deity] in the centre, and the thrones of those of Aksobhya and the rest in the cardinal directions, starting with the east. And he who understands yoga should create those [thrones] as transformations of their own seed [syllables]. Here, the creation of the thrones [is as follows]: in the centre, a lion-throne; in the east, an

. Lit. 'belonging to the yak (camara)', an insignia of royalty. 113Parallel passages containing some whole lines and some part lines of the above are found in the STTS (63, 13-18; 111, 3-8; 140, 32-36) and the SDPS (160, 23-25; 190, 11-12). For example: carurasYam caturdvaram catustoranaSobhitam / catuhsutrasamayuktarn pattasragdamabhusitam // konabhagesu sarvesu dvaraniryuhasandhisu / khacitarn vajraratnais tu sutrayed bahyamandalam // tasya cakrapratikasam pravisyabhyantaram puram / vajrasutrapariksiptam astastambhopaSobhitam // (STTS 63, 13-18)

104

Translation

Chapter 4

elephant-throne; in the south, a horse-throne; in the west, a peacock-throne; [and] in the north, a garw^/a-throne. 114 [For the deities], from Sattvavajri up to VajraveSa, 115 he should visualise moondiscs [as seats, each] generated out of the letter A. And then, on the principal seat, [he should visualise] Mahavairocana, generated by means of the syllable AH. [Why has he four faces?] Because consciousness which is of the nature of the Dharma-Sphere since, by its nature, it lacks such forms as the grasped - is four-faced, since the four faces that are the freedoms (vimoksa) of emptiness and the rest, are the avenues through which all samadhis arise since their, [that is, the freedoms',] object is the Dharma-Sphere. He is white in colour because he has the Dharma-Sphere as his nature; he has braids of hair [stacked up on his head] as a crown 116 and is unadorned, because he is one whose mind is tranquil; [and] he makes the bodhyagri gesture 117 because he has both wisdom and means as his nature. 118

114See P. Pal (1988, S 19, S 24) for some murals depicting such thrones in Ladakh dating from the time of Rin chan bzang po (C. 11th.). 115These are the thirty-six deities of the Vajradhatumandala (see STTS 11-60; NispY 44-7). In the STTS, however, the sixteen samadhi deities (Vajrasattva, Vajraraja etc.) are enumerated before the four kulamatrs (Sattvavajri, Ratnavajri etc.). ^jatdmakuta. I read jatdmakuta as a karmadharaya rather than as a bahuvrihi of a dvandva, ie. as 'with a crown (makuta) of braided hair (jata)' rather than 'with braided hair (jata) and a crown (makuta)'. A bronze figure of Vairocana from Korea dated to the second half of the eighth century depicted with a simple hair crown and two hands in the bodhyagri mudrd (Snellgrove, 1978, p.235) supports this analysis, as does the present description of Mahavairocana as nirdbharana ("unadorned"), which suggests he is to be visualised with the appearance of a bhiksu rather than a cakravartin. However, the other analysis cannot be ruled out and in examples of figures with a jatdmakuta thus understood of Japanese origin the crown is placed in front of they aft* which rises up behind it (see B. Frank, 1991, 170, plate 88; 122, plate 54). I am grateful to Prof. A. Sanderson for pointing out this latter analysis of jatdmakuta as well as for the reference. lllbodhyagrimudrd. Some MSS read bodhyahgimudrd (see textual note 4.105), also found as the name of a mudrd. The Nispannayogdvali has bodhyahgl for Vairocana's mudrd in the Vajradhatumandala (NispY 44, 7) and De Mallmann (1964, 260) gives it as a synonym for the dharmacakramudrd. However, the STTS (89, 4-5) describes a bodhdgrl (emend to bodhyagri) mudrd as follows: vdmavajrdngulir grdhya (emend to grdhyd) daksinena samutthitd I bodhyagri ndma mudreyam buddhabodhipraddyikd II. This accurately describes the mudrd of some surviving Korean and Chinese bronzes of Vairocana where his right hand grasps the raised left forefinger, completely encircling it (see Snellgrove, 1978, pp. 235, 369). Y& changes the correlations of this passage. See textual note 4.102-5.

105

Chapter 4

Translation

Having generated the conviction that he is, himself, Mahavairocana, as described, through the yoga of the four Buddha-thrones, 119 he should visualise a moon-disc in his heart; [and] above that, transformed out of the syllable DHlH, [he should visualise]

the

Fortunate One, the Adibuddha, having

five

faces

(pancananah > pancamukhah), having five crests, that is to say, 'having five hairbraids'. It is through tying up those [hair-braids that] he has a head-dress of five hair-braids. 120 [The five faces are] endowed with five colours: dark blue in the east; yellow in the south; red in the west; green in the north; on the top, he has a white face, the face of the most excellent horse. 121 He is tranquil, with the ornaments of a youth, 122 possessing the erotic [sentiment, 123 and] wearing about himself a cloth of many colours. He has eight arms, holding at his heart with four hands, the one hundred thousand [verse] Perfection of Wisdom [scripture], divided into four parts; [and] carrying, in each of the other four hands, a sword of wisdom in the gesture of striking. [He should visualise] through the yoga of the four Buddha-thrones.

^^caturbuddhdsanayogena. This compound occurs again, also in the instrumental, after the visualisation of the Adibuddha at the end of the paragraph. Its sense is unclear. ^pancaclraka. For discussion of this epithet see De Mallmann (1964, 13-14). This and the two preceding passages in bold are citations of NS 93cd (pancananah panca&khah panacirakasekharah). Pancasikha is also the name of a gandharva - sometimes their king - who has been suggested as a possible antecedent or model for ManjuSri (M. Lalou 1930, 66-70). ^paramasva. The significance of this is unclear. Candrabhadrakirti, in his NS commentary the Aryamanjusrindmasamgitindmavrtti, calls the Vajradhatumandala the paramasvamandala (Wayman 1985, 29). Paramas'va is also the esoteric name of Hayagriva. I22kumarabharanopeta (byis pa'i rgyan gyis brgyan pa Tib.). This must allude to Manjusri's epithet htmarabhuta, 'in the form of a youth' or 'in the form of a prince 7 common in Mahayana sutra literature. (See Lamotte 1960, 1-3.) ^sasrngara (sgeg pa dang Idan pa Tib.). Though srhgdra can mean 'adornment' this is not its most usual sense and there are many other words commonly used for 'adornment' in a context such as this. Also, "possessing ornaments" is unlikely to follow "with the ornaments of a youth". Taking it as the name of a rasa is supported by Tib. (and sgeg mo is given in a Tibetan lexicon for Idsyd, the dancing girl whose dance depicts the emotion of love: Jaschke, p.l 14).

106

Translation

Chapter 4

[[Sakyamuni] spoke this verse of the Lord of Speech, which is endowed with six kingly mantras, 124 in which there is the source of the non-dual, [and] which has the characteristic of nonorigination: (25)] The structure [of the verse] is: '[Sakyamuni] spoke this verse of the Lord of Speech, which implicitly designates the twelve stages (bhumi) of the Tathagatas'. How is [the verse described]? As endowed (-samyuktam >-upetam) with six kingly mantras. The six mantras will be described later. What is the nature of [this verse]? It is one in which there is the source of the non-dual (advayodaya). Duality is [the world of] subjects and objects. That which does not have that [duality of subjects and objects] is 'the non-dual'. [The word] 'udaya' means 'that from which it arises', [namely, 'source']. [The compound word] 'advayodaya' means 'the source of the non-dual' {advaya^udaya}. He, [that is, £akyamuni,] spoke that verse in which there exists the source of the non-dual {advaya-udaya}. How also is [the verse] qualified? As one which has the characteristic of non-origination. This means that it has the Dharma-Sphere as its nature [and] that it has emptiness (s"unyata) as its nature.

["A A I I U U E AI O AU AM AH. I, the Awakened One, the Embodiment of Knowledge, am in the heart of the Buddhas of the three times. (26)] In the Srlvajramandalamkaratantra the twelve stages of the Tathagatas are mentioned: The [first] great stage is All-Round Radiance, and that is like the circle of the sun. And the second is Nectar Radiance, shining as brilliantly as the moon.

^^sanmantrardjdnasamyuktam. See textual note 4.117 for discussion of the Skt..

107

Chapter 4

Translation

And the third is Sky Radiance, established in the region of space. And Vajra Radiance, the fourth stage, is delightful. It is called 'Vajra' through its association with firm resolve. The fifth [stage] is called Jewel Radiance, which is established at consecration. 125 The sixth is called Lotus Radiance. It shines like a lotus; [and] since it is an inherently pure entity it is pure [and] unclinging. The seventh, called Radiance of Action, is said to cause Buddha activity. The eighth is called Incomparable [Radiance, since] it has no comparison. The ninth is called [Radiance] Without Equal since it penetrates all comparisons. The tenth is called Radiance of Wisdom, the unsurpassed Buddha-stage. The eleventh radiant great stage is [called] All-Knowing [Radiance]; and the twelfth, which is experienced individually, is [called] The Fulfiller of the Knowledge of Yogis. These twelve Tathagata stages should be understood as being transformed out of the twelve vowels, beginning with A. [The phrase] 'of the Buddhas of the three times' refers to Buddhas past, [present] and future. [The verse says] 'I [...] am in [their] heart' because he, [that is, the Knowledge-Being ManjuSri] dwells in the heart of all the Tathagatas. [He is] the Embodiment of Knowledge, because he has the Knowledge-Body as his nature; [and] the Awakened One, because he is enlightened as to things as they really are.

[OM - Obeisance to you, Fa/ra-Sharp, Destroyer of Suffering, Embodiment of the Knowledge of Wisdom, Knowledge-Body, Lord of Speech, Arapacana." (27)] Now one should contemplate a wisdom-wheel (prajnacakra) in the heart of that [visualised form of the] Adibuddha [in ones heart]. With what characteristics [should

125 In the system of the bhumis followed by the Bodhisattva, consecration (abhiseka) takes place in the tenth and final bhiimi where one is consecrated as a Buddha. However, an alternative to the standard name of dharmamegha ('Cloud of the Dharma') for this bhumi found in the Dasabhumikasutra is abhisekabhumi (DBS 55). It is unclear what the significance is of having abhiseka at this point in the Tathagata bhumis.

108

Translation

Chapter 4

one contemplate it]? He, [that is, Sakyamuni,] explains it as suggesting the law of dependent origination. And thus he says, The hub is sustained by the spokes, and the spokes depend on the hub. And by means of dependent origination, this wheel is set in motion. [The wheel] is generated from the letter A. It has six spokes with six colours, four bands (mekhala), [and] a symmetrical hub. Visualising a letter A in the middle of the hub of that [wheel], the mantras - beginning with the root (mula) mantra of the Fortunate One, the Knowledge-Being [Manjus'ri] - should be visualised as white in colour on the four bands. On the first of those bands [one should visualise]: OM PURE VAJRA WHOSE ESSENCE IS THE NON-BEING OF ALL DHARMAS A A AM AH 126 On the second band of that [wheel] one should visualise the twelve vowels that express the twelve stages of the Tathagatas. Then, beginning with the first spoke, one should arrange the six mantras in due order, according to the succession of the colours of the families as follows: OM - OBEISANCE, O VAJRA-SHARP, on the first; OM - OBEISANCE, O DESTROYER OF SUFFERING, on the second; OM - OBEISANCE, O EMBODIMENT OF THE WISDOM-KNOWLEDGE, on the third; OM OBEISANCE, O KNOWLEDGE-BODY, on the fourth; OM - OBEISANCE, O LORD OF SPEECH, on the fifth; [and] OM - OBEISANCE, O ARAPACANA, on the sixth. On the rim outside [those spokes] one should arrange the garland mantra of the Fortunate One, the Knowledge-Body of all the Tathagatas: AH AH - O HEART OF ALL THE TATHAGATAS, KARA KARA OM HUM HRIH, O FORTUNATE ONE, EMBODIMENT OF KNOWLEDGE,

w sarvadharmdbhdvasvabhdvavisuddhavajra a d am ah (Text 4.159-160). This mantra is from the mantravinydsa section of the NS.

109

Chapter 4

Translation

LORD OF SPEECH, OF GREAT SPEECH, WOMB OF THE DHARMASPHERE AWARENESS OF ALL DHARMAS THAT IS PERFECTLY PURE AND STAINLESS LIKE THE SKY - AH. 12? One should [mentally] deposit outside [that] on the second band [the consonants], beginning with the letter KA and ending with letter HA. This is the arrangement of the mantras of the [wisdom] wheel. [Next,] he should visualise himself as the Fortunate One, the Knowledge-Being [Manjusri], born from the letter A situated in the middle of that wheel, with six faces, radiant like the autumn moon, with the best of sapphires in his beautiful hair, 128 with a halo that has the brilliance of the orb of the newly risen sun, with all the Tathagatas as [head-]ornaments, immersed in samadhi, seated on a variagated lotus throne, with two books of the Prajnaparamita above blue lotuses held in his two hands, 129 [and] as in the Sentiment of Tranquillity. 130 Visualising a moon-disc in the heart of that [Knowledge-Being], above it, he should place the letter A, the essence of the Perfection of Wisdom, the cause of the arising of the Awareness of the Omniscient Ones, the origin of all Sravakas and Pratyekabuddhas, the accumulation of the merit

127a/z ah sarvatathdgatahrdaya hara hara om hum hrlh bhagavanjndnamurte vdgJsvara mahdvdca sarvadharmagagandmalasuparisuddhadharmadhdtujndnagarbha ah (Text 4.168-170). This mantra is also from the NS mantravinyasa section. The final compound can be variously construed. Davidson (1981, 44) has, "O embryo of the gnosis of the dharmadhdtu, being very pure and stainless like the spatial field of all dharmas—dh", taking garbha as 'embryo'. Wayman (1985, 115) has, "O womb of the pure Dharma-Sphere Wisdom where all natures are pure like the sky! AH". For garbha, 'womb' (ie. 'that which contains') is probably better than 'embryo' since Manjus"ri is the subject of the description. Wayman's translation is doubtful since it takes the first element of the compound as a bahuvrihi (ie. it would retranslate as gagandmalasarvadharmasuparisuddhadharmadhdtujndnagarbha). Using the notation adopted for showing grammatical analysis, the following shows how I have taken the compound: (sarvadharma&{gagandmalasuparisuddha:dharmadhdtujndna))garbha. ^indranlldgrasacciram (i ndra ni la mchog gi gtsug phud Tib.). I take cira to refer to head hair, as in Manjus'ri 's epithet pancaciraka, rather than clothing. This is supported by Tib. (gtsug phud), though it does not translate the compounded sat. 129Ie. each hand holds a lotus on which there rests a book of the Perfection of Wisdom. H°sdntarasa. The ninth rasa of Indian aesthetics.

110

Chapter 4

Translation

and wisdom of all the Mahabodhisattvas, the letter of ultimate reality, the cause of all letters. 131 Then, having imagined the syllables of the mantras on the [wisdom] wheel to be symmetrically placed, to have the colours described, to be radiant [and] capable of destroying the mountain of ignorance, he should visualise his own body as shining with those rays of light. The bright rays go out of the four faces [of Mahavairocana] and illuminate the Sana world. 132 Next, with six colours, they emerge from the six vajra faces, 133 go to the various Buddha fields, the collection of which makes up the whole of space, [and] accomplish the welfare of the mass of beings without remainder in those Buddha fields. Having illuminated the assembly of all the Tathagatas, Buddhas and Bodhisattvas, they enter their mouths [and] circle the vajra in their hearts from left to right, extracting their Wisdom-Knowledge (prajnajnana). They, [namely, the lightrays, then re-]enter one's own, [that is, Mahavairocana's four] mouths, [the rays being conceived] as identical with the essencelessness of all dharmas, [and] become united with the mantra-syllable ['A'] on the moon[-disc] in one's own heart. Thus, the Bodhisattva, delighting in uninterrupted meditation, practising the observances of the Way of Mantras, who meditates in the these three ways - [firstly] "taking the form of the Fortunate One, the Knowledge-Being Manjusrl as an object of meditation", that is, [ManjuSri] as the Knowledge-Body; [secondly] "reflecting on his form", that is, as the moon-disc; and [thirdly] "meditating on his form", that is, as the letter A - "will, before very long, come to see him, [that is, ManjuSri,] in his

131 This has the appearance of a passage incorporated from scripture, though I have not been able to trace it. ^sahalokadhatu. The name given to the world system in which we live (for textual refs see BHSD 588). 133On the six-faced vajra see note 110 above.

Ill

Translation

Chapter 4

Transformation Form-Body, 134 provided he follows the [rules of the] Vinaya. And he will see all the Buddhas and Bodhisattvas accompanying his Transformation Form-Body in the vault of the sky". 135 And therefore, through the power of meditation, one will attain Buddhahood [and] Vajradharahood here in [this] very life. In this way 'The Method of Awaken­ ing according to the Mayajala' is indicated. And now I will describe the six other wisdom-wheels according to the division of the families. Concerning these, one should firstly carry out the complete initial yoga, 136 [then] remove the Fortunate One Mahavairocana, established in the centre of his mandala, [and] in his place visualise Amitabha, red in colour, born from the syllable HRlH, dwelling in samddhi. Likewise in his heart one should visualise the Adibuddha; [and] in his [that is, in the Adibuddha's] heart, one should visualise an eight-spoked [wisdom-]wheel with four bands, occupied by the root mantra and so forth. And then, one should arrange on the eight spokes the eight seeds [of the syllables of the mantra] OM VAJRATlKSNAYA NAMAH. In the middle of the hub of the [wisdom-Jwheel, one should visualise

^^nirmdnantpakdya ('Transformation Form-Body'). In the Triple-body (trikaya) system of Yogacara Buddhology the Transformation-body nirmdna[rupa]kdya and the Body of Enjoyment (sambhogakaya) comprise the twofold Form-body (rupakaya), which is distinguished from the Dharma-body (dharmakaya). The emanation of nirmanakayas by Buddhas to act for the benefit of beings is associated with the Awareness of the Performance of Action (see VijnMaSi 691). (For discussion of the bodies of the Buddha and bibliography see La Vallee Poussin, VijnMaSi 762-813.) 135The quotations are from the NS anusamsa section (see NS.Dav. 66, 23-6, though Davidson's text differs slightly from that cited here). ^ddiyoga. This term is found in a tripartite classification of the generation-stage (utpattikrama) of sadhanas into three samadhis: i. adiyogo nama samddhih (the preliminaries and the complete generation of the deity); ii. mandalardjdgri nama samddhih (the deity radiating the retinue of the mandala from the womb of the consort); iii. karmardjdgri samddhih (the whole mandala employed for the benefit of all creatures). Beyer (1973, 117) states that the three samadhis have their origin in Guhyasamdja Tantra exegesis, being mentioned in the explanatory Tantra, the Vajramdld. However, they were also employed by Hevajratantra commentators, eg. Ratnakaras"anti in his Muktdvali nama Hevajrapanjikd (NAK MS no. 5.98; fol. 20r9-22r8: rf. from A. Sanderson, Advayavajra's Saptdksarasddhana). It appears that Vilasavajra uses the expression ddiyoga to cover the preliminaries and the generation of the mandala, but not the generation of the deity as included in the adiyogo nama samddhih.

112

Translation

Chapter 4

Vajratiksna, transformed out of the syllable HRlH, red in colour, possessing the ornaments of a youth, with a head-dress of five hair-braids, clad in a garment of many colours, holding a lotus in his left hand [and] bearing in his right hand a sword of wisdom. Visualising the syllable HRlH in his heart, above a moon-disc as before, he should meditate proceeding in accordance with the previously described sequence of emanating and withdrawing [light-rays]. Again, one should similarly visualise Aksobhya in the centre of the mandala, born from the syllable HUM, dark blue in colour, [and] displaying the earth touching gesture. Within his heart one should imagine the Adibuddha, and likewise, an eightspoked [wisdom-]wheel [in the Adibuddha's heart]. One should arrange the [root] mantras [and so forth] as before, and sit the eight seed[-syllable]s of [the mantra] OM DUHKHACCHEDAYA NAMAH on the spokes. In the middle [of the hub of the wisdom-wheel] one should visualise the Fortunate One Vajrakhadga, 137 born from the syllable HUM, deep blue in colour, with the form of a youth, adorned with all the ornaments of a youth, with five hairbraids, wearing about himself silk cloth, [holding] a vajra in his left [hand, and] bearing a sword in his right [hand]. Contemplating the syllable HUM in his heart, above a moon-disc, one should carry out in the same way [the process] of emanation and [withdrawal of light-rays]. Again, in the same way one should imagine Vairocana in the centre [of the mandala], transformed out of the syllable AH, white in colour, with [his hands in] the Bodhyagri gesture. 138 In his heart [one should visualise] the Adibuddha [and] likewise [in the Adibuddha's heart] a ten-spoked [wisdom-]wheel. Making an envelope with the [root] mantras [and so forth, on the bands of the wheel], one

vajrakhadga (jam dpal sdug bsngal gcod pa Tib.). Tib. retranslates into Sanskrit as ManjuSri-

duhkhaccheda.

l ^bodhyagnmudrd. See note 117 above.

113

Chapter 4

Translation

should arrange the seed-syllables of [the mantra] OM PRAJNAJNANAMURTAYE NAMAH on the ten spokes. In the middle of the [wisdom-]wheel one should visualise the Fortunate One Prajnajnana, transformed from the syllable AH, white in colour, with five hair braids, endowed with the ornaments of a youth, wearing beautiful clothes, holding a wheel with his left [hand, and] a sword of wisdom in his right, transformed from the syllable AH. Visualising a moon[-disc] in his heart [and] above that the syllable AH, one should carry out in the same way [as before the process] of emanation and [withdrawal of light-rays]. In the same way, one should contemplate Amoghasiddhi at the centre of the mandala, bom from the seed syllable AH, green in colour, endowed with the gesture of fearlessness. Likewise, one should imagine the Adibuddha in his heart [and] an eight-spoked wheel [in the Adibuddha's heart]. Making an envelope of the [root] mantra and so forth as before, one should arrange the eight seeds of [the mantra] OM JNANAKAYAYA NAMAH on the eight spokes. In the middle of the [wisdom-]wheel one should visualise the Fortunate One Jnanakaya, born from the syllable AH, green in colour, with the ornaments of a youth, with a head-dress of five hair-braids, holding a double vajra with his left [hand], [and] bearing a sword with his right. Visualising a syllable AH above a moon [disc] in his heart, one should meditate in the same way [as before], performing the [process of] emanation and withdrawal [of light-rays]. Once more, in the same way, one should meditate on Ratnasambhava at the centre of the mandala, born from the root syllable, [namely, OM,] yellow in colour, displaying the gesture of giving. Likewise, in his heart one should visualise the Adibuddha, [and] an eight-spoked wheel [in the Adibuddha's heart] with the root mantra and so forth [on its bands]. On the eight spokes of that [wisdom-wheel] one

114

Chapter 4

Translation

should arrange the seeds of [the mantra] OM VAGISVARAYA NAMAH. At the hub of the [wisdom-]wheel one should visualise VagiSvara, born from the seed syllable OM, yellow in colour, endowed with every ornament, with five hair braids, wearing around himself a many coloured cloth, holding a jewel in his left [hand], bearing the sword of wisdom in his right. Visualising in his heart the syllable OM above a moon-disc, one should carry out [the process] of emanation and [withdrawal of light-rays] in the same manner [as before]. Similarly, one should visualise Bodhicittavajra, transformed out of the syllable A, white in colour, possessing the Erotic Sentiment, crowned with the five Buddhas, seated in a state of Diamond-pride, adorned with every ornament, holding a vajra and bell, [placed] as before, at the centre of the mandala. Likewise, in his heart one should visualise the Fortunate One, the Adibuddha, [and] a nine-spoked [wisdom-]wheel inlaid with the root mantra and so forth [on its bands] in his [that is, the Adibuddha's,] heart. One should arrange the nine seed syllables of [the mantra] OM ARAPACANAYA NAMAH on the spokes. At the hub of the [wisdom-]wheel one should visualise the Fortunate One Arapacana, produced from the letter A, white in colour, adorned with the ornaments of a youth, with five hair-braids, wearing a many-coloured garment around himself, [holding] a book at his [left] side [and] a sword in his [right] hand. Visualising a letter A above a moon[-disc] in his heart, one should meditate as before, carrying out the [process of] emanation and withdrawal [of light-rays], and making firm the [deities'] dwelling place and so forth. In each case this should be added, [namely, the stabilising of the mandala visualisation]. 139

139The Skt. word order is puzzling here (see textual note 4.253) and it is unclear what is to be added 'in each case' (ie. to the instructions for each of the prajndcakrasl). I have taken it to be the requirement to stabilise the visualisation of the mandala (nilayadrdhibhava) as this is both the last and the only new instruction.

115

Translation

Chapter 4

Also, this is to be understood as the Sixfold Method of Awakening According to the Srimaydjdla, according to the division of the [Buddha] families. 140 There are, therefore, seven mandalas in the Aryanamasamgiti, and these should be known as the special instructions of the guru. In this way is 'The Method of Awakening According to the Mayajala [explained]. [Here ends] the chapter on 'The Method of Awakening According to the Mayajala', the fourth in the commentary on the Aryanamasamgiti called, 'An Explanation of the Meaning of the Name-Mantras'. Three verses [are commented on].

5.8).

140For a summary of the visualisations of the six prajnacakras see Table 7 (Introduction, section

116

Translation

Chapter 5

[Chapter 5. On 'The Vajradhatu-Mahamandala of Bodhicittavajra'.] [That is to say, 141

[the Knowledge-Being Manjusri is] the

Fortunate One, the Awakened One, the Fully Awakened One, born from the letter A. 142 The letter A is the head of all letters, the great good, the supreme syllable. (28)] Next, he, [that is, £akyamuni,] spoke about the emanation and [withdrawal of deities] associated with Bodhicittavajra via the great mandala of Vajradhatu. [In the passage] beginning That is to say, [the Knowledge-Being ManjuSri is] the Fortunate One, the meaning of [the Names] the Fortunate One, the Awakened One, the Fully Awakened One is understood, [that is to say, it needs no further comment]. He is born from the letter A (akarasambhavah > akarajanitah) because the Fortunate One, the Knowledge-Being [ManjuSri] has as his nature the pure Knowledge-Body of the Tathagatas. The letter A is above the moon-disc in his, [that is, the Knowledge-Being Manju^ri's] heart. He, [namely, Sakyamuni,] next states the attributes of that [letter]. It is at the head (agryah > agrebhavah) of all letters (sarvavarnanam > sarvaksaranam) {sarvavarna^agryah}. It is the head, that is to say, 'best', because it

^tadyatha. Vilasavajra makes no comment on this opening, which I take as a statement that the subsequent verses amplify the gatha of NS 26-7. 142I translate this half verse in accordance with Vilasavajra's statement later in the chapter (Text 5.294) that four Name-mantras (namamantrani) qualify Manjus'rijnanasattva. He also cites the whole of NS 28ab as the first of the phrases whose words are Name-mantras (namamantraksarapadani) (Text 5.269), and as each phrase denotes one figure it follows that all the Names of NS 28ab are Names of Manjus'rijnanasattva (see Introduction section 5.5 for discussion of the Name-mantras). Other readings are clearly possible. Thus Wayman (1985, 68): "Accordingly, is the Buddha, Bhagavat, the Sambuddha arisen from A" taking akarasambhavah as the predicate; and Davidson (1981, 22): "And in this way, the blessed one, the Buddha [Manjusri], the completely awakened, born from the syllable A, is the syllable A, the foremost of all phonemes, of great meaning, the supreme syllable" taking the two half verses together with akdrah as the predicate.

117

Translation

Chapter 5

is at their beginning. 143 It is the great good {maha:arthah} for the reason that it causes the production of benefit for the whole world. It is the supreme syllable (paramaksarah). By etymological analysis 144 syllable (aksarah) [is seen to] mean 'that which does not (na < a-) perish (ksarati < -ksarah)'. This is because it is without change, due to having the Limit of Reality145 as its nature. [Hence the word] 'paramaksarah' means the supreme syllable, [that is to say, 'the supreme unchangeable reality'] (parama:aksarah}. In this way what is really named146 by 'Knowledge-Being' and 'the letter A' is explained. And that same [nature], since it is completely free from all obstacles, should be understood to be the nature, one by one, of each of the manclala-deities arising in the consciousness of yogins. 147 Just as their samadhis, through indicating their nonapprehension (anupalambha) of error (viparyasa) and being opponents (pratipaksa) of [erroneous] perception, have the characteristic of the Liberation-Door148 of

143Cf. om akdro mukham sarvadharmdndm ddyanutpannatvdt ("The sound 'A' is the first of all dharmas since it is unarisen from the beginning."), the mantra used in chapter 4 as the basis for developing the Discriminating Awareness. See note 103 above. ^aparaparydya (mam grangs gzhan Tib.[191.3.4]). I take aparaparyaya to be used in the same way as paryaya, ie. to mean 'synonym' or, more literally, 'convertible term'. The present context suggests 'etymological analysis' since a desired meaning is being derived through a constructed etymology in the fashion of a nirvacana or nirukti. The Tibetan rnam grangs (paryaya) is used to refer to the component parts of words according to their etymological value (Das, 758). 145bhutakotih (yang dag pa'i mtha' Tib.[1913.4] ). Sometimes translated more literally as 'The Goal for Beings' bhutakotih is one of the synonyms for sunyata as understood by the Vijnanavada (see Vasubandhu's bhdsya on the Madhydntavibhdga 1.15). ^svarupdbhidhdna. Lit. 'true naming'. 147This sentence marks the beginning of the description and enumeration of the mandala-deities of the Vajradhatu mandala. However, the text is corrupt, making interpretation difficult (see textual note 5.14—15). I have taken the opening tad eva to refer to the previous sentence's svarupdbhidhdnam and thus the sentence as stating that the 'true naming', ie. true nature, of ManjusYijnanasattva and the mandala-deities is the same. An alternative is that tad eva should be taken as tad eva cittam, referring to the naturally radiant mind (prakrtiprabhasvam cittam) associated with Non-dual Awareness (advayajnanam) and generated by the practitioner in chapter 4 prior to the generation of the mandala (Text 4.60-61). In the course of the introductory sections to the mandala-deity identifications later in this chapter, the enlightened mind is referred to by tad eva (Text 5.72, 5.136), although these usages follow an earlier passage of the form, yad eva cittam ... tad eva (Text. 5.36-39). ^vimoksamukha. Chapter 4 identifies the four faces of Mahavairocana as the four vimoksamukhas (giving sunyata as the first), describing them as avenues for samadhis. Three vimoksamukhas are commonly found in sutra literature, ie. sunyata, animittaftdj, apranihitaftd] (DhSam 73, MVy

118

Translation

Chapter 5

Emptiness, in the same way, through being opponents of appearance, wrong desire and involvement in error, they [also] have as their nature the Liberation-Doors of the Signless, the Wishless [and] Non-Involvement149 respectively. And hence, in due order, he, [that is, £akyamuni,] spoke about the nature of the family-mothers, Sattvavajri, [Ratnavajri, Dharmavajri,] and [Karmavajri]. [[He is] the great breath, a non-production. Free from speech utterance. Foremost cause of all expression. Very radiant as all speech. (29) 150] The great breath, a non-production [is to be understood] as meaning Sattvavajri. Here the great breath is the sound 'A'; and that [sound 'A'] has nonproduction as its nature. [This is] because it has the Dharma-Sphere as its nature and is the cause of the Mirror-like Awareness.

1541-1544). The fourth, anabhisamskara, appears to be peculiar to the mantranaya. See note following. ^^anabhisamskara, 'Non-Involvement'. I take this to refer to a condition that allows the possibility of action that is free from karmic consequences. Edgerton notes that the term abhisamskara is used in Mahayana Sutra literature to mean both 'performance' (of action) and 'accumulation' (of merit or demerit) (BHSD 57). A possible translation of anabhisamskara would therefore be 'NonAccumulation'. YaSomitra uses abhisamskara to gloss the state of mental investigation (vitarka) that occurs between possessing the desire to act (chanda) and making the effort to do so (prayatna): "from desire [arises] investigation, a type of volition characterised by mental application" (chandad vitarkah cetanaviSeso 'bhisamskaralaksanah: AKBhVya 1129, 18-19). This suggests a translation of anabhi­ samskara that emphasises the mental aspect of action and I have therefore adopted the rendering 'NonInvolvement'. (Edgerton also gives 'mental preparation, application, determination' as a meaning for abhisamskara.) In the NMAA anabhisamskara is appropriately associated with the Awareness of the Perform­ ance of Action, the jnana concerned with enlightened activity in the world for the benefit of beings. Whether it is found as a vimoksamukha before this association with krtydnusthanajnana is a question for further research. The development of a fourth vimoksamukha may, like the fifth jnana, be an instance of the expansion of a pre-existing doctrinal category to meet the needs of a Tantric context. 150The translation follows Vilasavajra's interpretation of the verses containing the Names, ie. that each is comprised of a series of predicates (see Text 5.269-272). A more natural translation might be, "For the great breath is a non-production, free from the utterance of speech, foremost cause of all expression, very radiant in all speech", taking 'the great breath' (mahaprano) as the subject and the remainder as a series of predicates. So, Wayman (1985, 68): "The great inhalation is not a production, free from utterance by speech, chief cause of all speech, the clarification of all words".

119

Translation

Chapter 5

Free from speech utterance (vagudaharavarjitah) [is to be understood] as meaning Ratnavajri. He, [that is, ManjuSri,] is free from (varjitah > rahitah) that speech utterance {vag^udahara} since, as a result of the Awareness of Equality, [the faculty of] speech is not functioning. Foremost cause of all expression (sarvabhilapahetvagryah) [is to be understood] as meaning Dharmavajri. All expression {sarva^abhilapa} means the expression (abhilapah > abhilapanam) of all real entities (dharma), such as the five constituents (skandhas). The cause (hetuh > karanam) of that [expression] is the Discriminating Awareness. For this very reason it is foremost (agryah > pradhanam). Very radiant as all speech (sarvavaksuprabhasvarah) 151 [is to be under­ stood] as meaning Karmavajri. All speech {sarva:vak} means [the varieties of speech] as referred to in the passage that begins, "with the speech of the gods, with the speech of the nagas ,..". 152 He, [that is, Manjus"ri,] is for this reason very

^^sarvavaksuprabhasvarah There is a problem here of whether Vilasavajra takes this as one word or two (sarvavak suprabhasvarah). If one, unusually, he omits to state the grammatical relation between the two elements sarvavak and suprabhasvarah. The analysis he does give is of the type he employs when there are two words, or Names, that denote a deity, ie. 'Mafiju^ri is X, and so (ata eva) he is Y', suggesting "[Manjus'rijnanasattva is] all speech, [and so] very radiant" as the appropriate translation. At the end of the chapter Vilasavajra gives five as the number of Name-mantra words for the four Family-Mothers (kulamatr) denoted by NS 29 (mahaprano hy anutpado vagudaharavarjitah / sarvabhilapahetvagryah sarvavaksuprabhasvarah //) (see Introduction, section 5.5, on Name-mantras). If hi, a metrical filler, is discounted as a Name-mantra, there are five names when sarvavaksuprabha­ svarah is counted as one word. If it is two words, there are two Names giving total of six for the verse. A possible solution, keeping two Names for the final quarter verse, is not to count mahaprdnah as a Name. Vilasavajra identifies it as the letter A; the other five words (Names) could then be seen as qualifying it. However, this means that in NS 29a, Sattvavajrl would be denoted by anutpadah and Vilasavajra unequivocally cites the whole quarter verse as referring to her. I have therefore taken sarvavak and suprabhasvarah as one word in a karmadharaya relation, preserving the Name count of five. This interpretation is also in line with Vilasavajra's statement that the verses are made of predicates, ie. they are all Names, including 'the great breath' (see note preceding). The identification of mahaprdnah with the letter A is clarified later when the four Family-mothers are said to be trans­ formations of the letters A A AM AH, 'A' being the letter for Sattvavajri (Text 5.35-36). 152yar/zrnahamahamaharagah. The refrain, mahamahamaha-, in this and the following two NS

121

Chapter 5

Translation

[Next,] the passage beginning, The great festival of great passion. And here, at the beginning [of the verses that denote the Tathagatas], Amitabha is referred to [rather than Vairocana, the Eternal One]. This is in order to indicate the pre­ eminence of passion. For this reason he, [that is, Sakyamuni, putting passion first,] has declared: "Passion and hatred and delusion ..." There is said to be a great festival {maha:mahah} when the Tathagatas display worship that pervades the whole of space. The word festival, [therefore,] denotes worship. Great passion is passion that has a nature that is great. In this context, of the great festival of great passion {mahamaha:maharagah}, great passion is that which has the characteristic of ripening157 and liberating living beings. And that [great passion] is [of the nature of] the Fortunate One, the Glorious Vajrasattva. 158 [The truth of this is indicated] by reason of the words, "For great passion is the Fortunate One, Vajrasattva, the Tathagata." Producer of joy for all beings. [He is so described] since, for all beings

verses has a strong incantatory element that cannot be captured in translation. ^pacana (smin [cing grol ba'i mtshan nyid do] Tib.[191.5.5-6]}. The literal meaning of 'to cook' creates the link in the explanation of 'great passion': the heat of passion becomes the heat of the cooking that ripens one for liberation. Tib. confirms the Ski. reading which, given the common confusion of ca and va in Newari script, could have been interpreted as a corruption of pavana, 'purifying*, a reading that would have made almost as good sense. 158 Given the suggestion at the beginning of the passage (on NS 30) that the half-verse 'Named' Amitabha, it is unclear whether Vilasavajra is enumerating Vajrasattva as one of the Tathagatas of the mandala (as a samddhi deity Vajrasattva is identified with the Name mahakamah of NS 32cd). Perhaps Vajrasattva is being identified as the nature of Amitabha, though in the next paragraph the Name 'producer of joy for all beings' is specifically identified as Amitabha's. Thus it appears that NS 30a names Vajrasattva and NS 30b Amitabha, although this means that the general pattern of assigning a complete half-verse starting, "the great festival of great ...", to each Tathagata is not followed. However, the ritual structure of the NMAA is a sixfold one (six mantra families, six Wisdom-wheel visualisations) and if Vilasavajra needs Names for six Tathagatas a half-verse has to be divided. As the sixth Tathagata, Vajrasattva and his family are also associated with passion by Tsong kha pa in his sNags rim chen po (Peking blockprint 365b.3ff: ref. from Wayman, 1973, 47).

122

Translation

Chapter 5

who act out of passion, he causes [the arising of] joy (ratim > pritim) that has great bliss (mahasukha) as its nature. This is the nature of Amitabha. He, [that is, £akyamuni, then] spoke with regard to those who act in hatred:

[The great festival of great hatred, great enemy of all defilements. (30cd)] The great festival of great hatred. [Since it has been discussed above] the meaning of great festival is understood. Great hatred is Aksobhya and he is the great enemy of all defilements, that is to say, the great enemy of the defilements (kleSa), subtle defilements (upakleSa) and so forth. He, [that is, £akyamuni, next] spoke concerning those who act out of delusion:

[The great festival of great delusion, destroyer of stupefied mental delusion. (3lab)] The great festival of great delusion. [The meaning of] great festival is as before. Great delusion is Vairocana. Mental delusion (-dhfmoha-) means the delusion connected with thoughts {-dhi^moha-}. It, [namely, great delusion,] is the destroyer (sudanah > vinas"akah) of this stupefied mental delusion (mudha: dhlmoha-} {mudhadhimoha^nisudanah}. He, [that is, £akyamuni, next] spoke with concerning those who act from anger:

[The great festival of great anger, great enemy of great anger. (3 led)] The great festival of great anger. [The phrase] great festival has the same meaning [as before]. Great anger is Amoghasiddhi. Great enemy of great anger. [The words] great anger [are used] since it has [both] beings and

123

Translation

Chapter 5

non-beings as its object. 159 It is he, [that is, Amoghasiddhi,] who is the great enemy of this [great anger]. He, [that is, £akyamuni, then] spoke concerning those who act out of greed:

[The great festival of great greed, destroyer of all greed. (32ab)] The great festival of great greed. [The meaning of] great festival is [to be understood] as before. Great greed is Ratnasambhava. [He is] the destroyer of all greed because he intends the purification of all prosperity whether mundane or supramundane. 160 Since the four Liberation-Doors of Emptiness and so forth cause all samddhis, inasmuch as they, [that is, the Liberation-Doors,] are within the sphere of the Dharma-Sphere, that very [mind] has the Emptiness of the Internal (adhyatmaSunyata) 161 as its nature since it opposes errors concerning the self, [that is, what is subjective]. It has the Emptiness of the External (bahirdhas"unyata) as its nature

155'sattvdsattvavisayatvdt (sems can dang sems can ma yin pa'i yul gyi phyir Tib.[192.1.4J) Tib. supports the Skt. and takes sattvasattva as a dvandva. However, the sense is uncertain and it may be that both the Skt. and Tib. are corrupt. Can anger be 'great' because it extends beyond its usual object of (other) living beings to focus on all objects, thus becoming 'objectless'? The epithet 'great', eg. applied to 'compassion', is usually understood to indicate that it is based in realisation of the essencelessness of all dharmas and Awareness of the Dharma-Sphere ie. it is 'objectless' (see chapter 4, Text 4.51-3). l^sarvalobhanisudana iti sarvalaukikalokottarasampattiparisodhandrthatvdt. (chags pa thams cad sel bar byed pa ni 'jig rten dang 'jig rten las 'das pa thams cad kyi phun sum tshogs pa yongs su dag bar byed pa don nyid kyi phir ro Tib. [192.1.5-6]). The analysis and interpretation of this sentence is not straightforward. Only MS B attests sampatti but it is supported by Tib. (phun sum tshogs pa). Tib. takes sarva to qualify laukikalokottara and would translate as "[He is] 'the destroyer of all greed' because he intends the purification of the prosperity of all that is mundane and supramundane". I have taken both sarva ("all") and laukikalokottara ("mundane and supramundane") to qualify sampatti ("excellence", "prosperity", "welfare"). How the whole functions as an explanation of Ratnasambhava as "the destroyer of all greed" is uncertain. Does sampatti stand for all that is the usual object of greed? 161 This is the first of sixteen Emptinesses (s~unyata) that are correlated with the Names (and deities of the Vajradhatu mandala) of NS 32cd-35. The list corresponds to that found in the Astadasasdhasrikd Prajnaparamita (T 220, t.VII, k. 500, p.604a!7; rf. from Lamotte 1976, 2028) and in Vasubandhu's Madhyantavibhagabhasya (MAVibhBh on MAVibh 1:17; see Nagao, 24, 15-21), a work with which Vilasavajra was familiar (see Appendix I). In the MAVibhBh Vasubandhu gives some explanation of the sunyatas and relates them to MAVibh 1:17-20. See Introduction, section 5.9 for discussion and tabulation of the correlations. On the sixteen sunyatas see Lamotte's essay (1976, 1995-2043); also his translation of the Mahdprajndpdramitdsdstra (ibid. 2044-2151).

124

Translation

Chapter 5

subjective]. It has the Emptiness of the External (bahirdhaSunyata) as its nature because of the disappearance of the mental perception of the external, [that is, what is outward, as ultimately real]. It has the Emptiness of the Internal and the External (adhyatmabahirdhaSunyata) as its nature since it has the non-apprehension [as real] of the body as its nature, which is the substrate of that [internal and external perception]. It has Great Emptiness (mahaSunyata) as its characteristic because it is free from clinging to the [great and extensive] world of inanimate objects. 162 So it, [that is, the non-dual mind,] has as its nature the four samadhis - [those embodied as] Vajrasartva, [Vajraraja, Vajraraga] and [Vajrasadhu] - whose substrate is the Mirror-like Awareness [of Aksobhya]. Therefore he, [that is, Sakyamuni,] said,

Great object of desire. Great happiness. Great delight. Great joy. (32cd) Here great object of desire (mahakamah) is Vajrasartva. [The word] 'kamah' means 'that which is desired', [that is, it is to be understood as 'object of desire' rather than 'desire']. This means that he, [that is, Vajrasattva as a samadhi,] is desirable. Great happiness is Vajraraja, great delight is Vajraraga, [and] great joy is Vajrasadhu. Therefore, these [deities] are ordered in this way since the attainment of the Mirror-like Awareness - that has the nature of a reversion from (paravrtti) from impressions such as [those produced by] eye and form 163 - is the mastery of the [non-apprehensions] beginning with the non-apprehension [as real] of what is

162Vilasavajra follows Vasubandhu in relating adhydtmabahirdhasunyata to the body and mahasunyata to the inanimate world (bhajanaloka). See MAVibhBh on MAVibh 1:17. 163The 'reversion' is the purificatory transformation of the dlayavijnana ('Recepticle-consciousness'), which, according to the Vijnanavada, results in the arising of the Mirror-like Awareness. See Introduction, section 5.6.

125

Chapter 5

Translation

subjective, [that is, the mastery of the four Emptinesses]; and since, individually being the opponents of [different] errors, they cause: the arising of bodhicitta [via Vajrasattva]; attraction [via Vajraraja]; gratification [via Vajraraga]; and goodness [via Vajrasadhu]. In the same way, that very [mind] is [of the nature of] the Emptiness of Empti­ ness (SunyataSunyata) 164 since there are no appearances (nimitta) even in the aware­ ness of Emptiness. It is [of the nature of] the Emptiness of the Ultimate [truth] (paramarthaSunyata) because of non-affliction by appearances in Ultimate [truth]. It is [of the nature of] the Emptiness of the Conditioned (samskrtaSunyata) due to the lack of grasping of appearances in the Merit-Accumulation, which is conditioned. It is [of the nature of] the Emptiness of the Unconditioned (asamskrta£unyata) because, in the unconditioned also it does not take hold of any appearance in the Accumulation of Wisdom. So he, [that is, £akyamuni,] said,

With great appearance, [and] a distinguished body. Great colour, with great beauty. With a great name, great generous one. With a great extensive mandala. (33) Here, with great appearance {mahairupah}, [and] a distinguished body means Vajraratna. Appearance (rupam) means 'that which is represented (rupyate)', in other words, 'what is visible (vilokyate)'. It is great because, since he is like a wish-fulfilling jewel, he has as his nature the consecration of the Lordship of the Dharma 165 of all the Tathagatas throughout the triple-sphered [world]. And for this very reason, as he illuminates immeasurable [numbers of] Buddha-fields, he has a distinguished body.

164Vilasavajra links these four sunyatds with the second Liberation-Door, the Signless (animittavimoksamukha). ^dharmarajyabhiseka. This consecration is mentioned in chapter 3 in connection with the mahosnisakulam, identified by Vilasavajra as the ratna family of Ratnasambhava (see note 86).

126

Translation

Chapter 5

Great colour, with great beauty means Vajrasurya. 166 [He has the name great colour {maha:varnah} since he is endowed with six colours. 167 He has great beauty168 {mahaiyapuh} because he has the Awareness - consisting of Hearing, Thinking and Realising - of the Dharma-Sphere as his halo. With a great name, great generous one means Vajraketu. With a great name {mahainama} means possessing great renown. Why? On account of his practicing the Perfection of Giving to the limit. And hence, as he gives everything away, he is a great generous one {maharudara}. With a great extensive mandala means Vajrahasa (Diamond Laughter). [The word] 'mandah' means 'that which adorns', [in other words, 'an ornament1 ,] in accordance with the verbal root mand, 169 'to adorn'. Mandala means 'that which * * • » receives (grhnati > lati > -la) an ornament (manda-)'. It is extensive (vipulam > vistirnam) since it pervades all world-systems. Having a mandala [that consists] of laughter, that is, a nature (saram > [mandam]) 170 [so consisting], which is great [and] extensive, he is [said to have] a great extensive mandala

166The name found in Vajradhatumandala of the STTS and NispY is Vajratejas rather than Vajra­ surya, though the STTS gives Vajrasurya as an alternative (see STTS 36, 13). 167The explanation of mahavarnah makes the karmadharaya analysis puzzling; a bahuvrihi, "Of great colour", would be more appropriate. Wayman (1985, 70) reports that Smrtijfianaklrti identifies the 'six colours' as the six colours of sunlight. An alternative would be the colours of the six Buddhafamilies. 168 7Y£. has lus bong che ("of great body size", "of great body bulk") for mahavapuh Though vapus does mean 'form', 'figure', it more especially indicates a beautiful or handsome form, ie. 'beauty' (in a parallel way to rupin). This reading accords better with Vilasavajra's explanation of the Name in terms of a halo. I69madi bhusayam (= DhP 10.49). ^°sdram ('nature'). See textual note 5.102 for a discussion of some problems associated with this passage. Vilasavajra may be alluding here to a different gloss of mandala where manda- is glossed as saram ('heart', 'essence' or 'nature') rather than as 'adornment', and -la glossed as 'grasping' or 'enclosing'. An example is provided by Lessing and Wayman (1968, 270, note 1) who cite Padmavajra's STTS subcommentary, the Tantrdrthavataravyakhyana'. "''Mandala'' means an 'enclosing of essence' (manda-la) because of having the 'essence' in the sense of 'enclosing' it." (dkyil 'khor ni dkyil 'khor du Idan pas dkyil 'khor)" (T5h 2502; 96b-3). This analysis appears to have become the one most commonly followed by Buddhist Tantric commentators, though it is not certain whether it was current at the time of Vilasavajra. However, if Padmavajra was a disciple of Buddhaguhya he may have been a contemporary of Vilasavajra.

127

Translation

Chapter 5

{mahavipulaimandalah}. This is because he, [that is, Manjus"ri,] causes the illumination of the three worlds. 171 And so, [that mind] has as its nature the tetrad [of deities], Vajraratna, [Vajrasurya, Vajraketu] and [Vajrahasa] - the retinue of the Awareness of Equality. This is due to the lack of appearances as [previously] described - such as the [division into] subject [and object] - in the Awareness of Equality, which has a reversion from the defiled mind as its nature. And because, via the abandoning of their respective obstacles, they cause: [the bestowal of] consecration [as Vajraratna]; increase of radiance [as Vajrasurya]; the enjoining of the Perfection of Giving [as Vajraketu]; and attainment of joy [as Vajrahasa], this, beginning with Vajraratna, is their order. That same [mind] is [of the nature of] the Emptiness of the Endless (atyantaSimyata) because it does not cling to the desire for [achieving] the aims of endless beings. 172 It is [of the nature of] the Emptiness of that Without Beginning or End (anavaragraSunyata) because it is unattached to the desire, "I must not abandon the beginningless, endless world". It is [of the nature of] the Emptiness of NonDiscarding (anavakaras~unyata) because it is free from clinging to [the thought], "When I am in the condition of Extinction that is free from all attachment, I must not discard that which is good". It is [of the nature of] the Emptiness of the Innate (prakrtiSunyata) because it is free from the desire, "I must purify the predispositions (gotra) that are innate, [rather than acquired] in nature". 173

^^trailokyabhasakaratvat. This must allude to NS 18 where Sakyamuni's smile illuminates the three worlds (ie. the earth, the atmosphere and the heavens), smitam samdarsya lokanam apayatrayaSodhanam / trailokyabhasakaranam catiumarariSasanam // (NS 18) For comment on trailokyabhasakaranam see Text 2.23-25. 172Vilasavajra associates this and the following three sunyatas with the third Liberation-Door, the Wishless (apranihitavimoksamukha). 173The allusion here is probably to the Vijnanavada classification ofbijas contained in the alayavijndna (see VijnMaSi 562): "gotra is twofold, innate and acquired" (gotram dvividham prakrtistharn

128

Translation

Chapter 5

So he, [that is, £akyamuni,] said,

Great bearer of the weapon of wisdom. [Being] the great goad of the defilements, foremost. Of great fame, of great renown. Possessing great light, of great splendour. (34) Here, Vajradharma is the great bearer of the weapon of wisdom {maria:prajnayudhadharo}. The wisdom is the Discriminating Awareness. Since he has that Awareness as his nature, he is the bearer of that very [wisdom], which is a weapon (ayudham > Sastram). The great goad of the defilements {maha:kle£ahku$ah}, foremost means Vajratiksna. The defilements (kle£a) are passion and so forth [and it is] these 174 he is the goad of {klesa^ankuSah}. He is like a goad 175 since he has attained control over the defilements. This is why he is foremost (agranih > Sresthah). Of great fame, of great renown means Vajrahetu. 176 He is of great fame {maha:ya£ah}, because he, [that is, ManjuSri,] has as his nature the lordship 177 of all the mandala deities taught in the Tantra178 that are the means of ripening beings. For this very reason he is of great renown {mahaikirtih}. [And his renown is] great in that it pervades the world systems completely, without remainder.

samudanitam ca, BoBh 3.1). In other contexts gotra means 'lineage' or 'family', eg. &ravakagotra, Pratyekabuddhagotra, Tathagatagotra (VijnMaSi 103). I74tesdm ankusabhutatvat klesdnkusah (Text 5.121). Here, ankusabhutatvat has to be taken as a sapeksasamasa, tesdm qualifying ankusa instead of ankusabhutatvat. ^5ankusa, 'a goad'. This especially denotes the hooked implement used by elephant drivers, which is an instrument of control. I76vqjrahetu (rdo rje rgyu Tib.). In reading vajrahetu I follow two Skt. MSS and Tib.. Vajraketu, already identified with NS 33c, is the reading of the remaining MSS. 1 ncakravartisvabhavatvat. There is a play on the word cakra ('wheel') here: a mandala is shaped like a wheel - the word cakra is often used as a synonym for mandala - and Vilasavajra's word for 'lord' (cakravartin) means 'wheel turner'. 178It is unclear whether it is Tantra in general or the Maydjala, the Tantra from which the NS is purportedly taken, that is referred to here.

129

Translation

Chapter 5

Possessing great light, of great splendour means Vajrabhasa. 179 [In the phrase] possessing great light (jyotih > dlptih){maha:jyotih}, [the light is described as great] in that it diffuses throughout the triple-sphered [universe]. And therefore he is of great splendour {mahaidyutih} since he shines forth in the mental continua of all beings at the time of [mantra] recitation as the silence that is the avoidance of the defects of [pronounced] speech, whether [they be] of quick, medium or slow [speech]. And therefore, [that same mind] has as its nature [the tetrad of deities], Vajradharma, [Vajratiksna, Vajrahetu] and [Vajrabhasa], whose substrate is the Discrim­ inating Awareness - which [in turn] has as its nature a reversion from the [falsely] discriminating mind. This is because of the discrimination of the truth concerning all phenomena - due to the separation from the errors as previously stated - in the Discriminating Awareness. And because, as those [deities], it causes: the naturally pure dharmata (as Vajradharma); Wisdom-Knowledge (as Vajratiksna); non-turning back (as Vajrahetu); and the understanding of the intended meanings of all [speech] (Vajrabhasa); this is the order of these [four deities] starting with Vajradharma. That very [mind] is [of the nature of] the Emptiness of Marks (laksanas"unyata) because of non-involvement with respect to the Form Body (rupakaya) that consists of the [thirty-two major] marks and [eighty] minor marks. 180 It is [of the nature of] the Emptiness of All Dharmas (sarvadharmas'unyata) because of non-involvement with respect to the Body of Dharmas (dharmakaya), which has as its nature such [dharmas] as powers and confidences. It is [of the nature ofj the Emptiness of NonBeing (abhavaSunyata) because of the lack of attribution [of ultimate existence] to

179See textual note 5.127 for discussion of the possibility of reading Vajrabhasa (Diamond Light) rather than Vajrabhasa (Diamond Speech). 180This and this following three sunyatas are associated with the fourth Liberation-Door, NonInvolvement (anabhisamskaravimoksamukha).

130

Translation

Chapter 5

persons and things. And it is [of the nature of] the Emptiness of Non-Being and Essence (abhavasvabhavaSunyata) because of non-involvement in reviling the nature of Emptiness as it has been explained. 181 And therefore he, [that is, Sakyamuni,] declared,

Great creation-holder, [and so] wise. Great adept by means of the utility of creation. Great one, fond of the pleasure in creation. Great knower of the net of creation. (35) Here, great creation-holder {maha:mayadharah}, 182 [and so] wise means VajraviSva. [By] creation (maya) is [meant] the creation (nirmana) of varied, extensive and lofty worship for all the Tathagatas. One who possesses this is called a creation-holder {maya^dharah}. For this very reason he is wise (vidvan > panditah), since by this act of worship he acquires the Provision of Merit (punyasambhara). Great adept by means of the utility of creation is Vajraraksa {maha:mayarthasadhakah}. An adept (sadhakah) by means of the utility of creation (mayarthah > nirmanarthah) is one who expertly guides (sadhayati), that is, urges on, lazy beings through that [power of creation]. Great one, fond of the pleasure in creation means Vajrayaksa {maha:mayaratiratah}. He is fond of (ratah > saktah) the pleasure in creation {maya^ratih}{mayarati^ratah}, that is, [fond] of the pleasure in the creation of worship for the Tathagatas.

^yathoktasunyatalaksanapavadanabhisamskaranat (ji skad gsungs pa'i stong pa nyid kyi mtshan nyid la skur pa 'debs pa dang mngon par 'du mi byed pa'i phyir na Tib.[193.1.1]), "because of non-involvement in reviling the nature of Emptiness as it has been explained". I am not certain of the sense here. The 'nature of Emptiness as explained' might be reviled if, for instance, it is perceived as being nihilist. The translation follows Tib.; an alternative, "because of non-involvement hi reviling, which is empty by nature as has been stated" is more difficult to ascribe doctrinal sense. 182 Wayman (1985, 70) and Davidson (1981, 23), following NS.Tib. (mkhas pa sgyu 'phrul chen po 'chang Mukherji23) take maha to qualify maya. Thus, Wayman: "holding the great illusion".

131

Translation

Chapter 5

Great knower of the net of creation {maha:mayendrajalikah} is Vajramusti (Diamond Fist). The net of creation, that is to say, the assemblage of creations, refers to the unity of the Diamonds of Body, Speech and Mind. A knower of the net of creation is one who knows that 183 [unity]. It should be understood that the absence ofvrddhi here, [that is, the word 'indrajalikah' is used rather than the grammatically correct form, 'aindrajalikah',] is due to the text being scriptural. And so, there being the Awareness of the Performance of Action that has a reversion from consciousnesses such as [those arising in conjunction with the] eye, [ear and nose,] as its nature, this [very mind] is of the nature of the tetrad - Vajrakarma, 184 [Vajraraksa, Vajrayaksa,] and [Vajramusti] - that has the Awareness of the Performance of Action as its substate. Because, through the diminution of the [previously] stated misapprehensions, they cause: the worship of all the Sugatas [as VajraviSva]; protection [as Vajraraksa]; the pacification of obstacles [as Vajrayaksa]; [and the arising of the] Awareness of Emptiness with respect to the body and so forth [as Vajramusti]; the order of these [deities] that start with VajraviSva is this. Hence, since it is the obstacle to envy, immoral conduct, impatience and laziness, that very [mind] has the Perfections of Generosity, Moral Conduct, Patience and Energy as its nature.

vetti mayendrajalikah (Text 5.153). Taking -ika in the sense of "one who knows ..."can be derived from Panini (AA 4.2.59-60): tad adhite tad veda I kratukthadisutrantat thak, '[When the meaning is] "one who studies X" or "one who knows X", thak [is the affix] after (i) [names of] sacrifices, (ii) [the word] uktha etc., [and a compound] whose final member is [the word] sutra\ The meaning 'one who knows the net of creation' is justified by treating the expression "[the word] uktha etc." as indicating an akrtiganah, ie. a list that gives examples of words, rather than all the words, subject to a rule. Another Buddhist example of this treatment of-ika is Yasomitra's explanation of the term Vaibhasika as 'one who knows the Vibhasa' (see AKBhVya on AK 1.3). I am indebted to Prof. A. Sanderson for this note. 184Vajrakanna is the name for which Vajravisva is given as an alternative in the STTS (35, 1; 36, 13).

132

Translation

Chapter 5

Hence he, [that is, £akyamuni,J said,

Great lord of gifts, the best. Great possessor of moral conduct, [and so] foremost. Great bearer of patience, unshakable. Great strength of vigour. (36) Here, great lord of gifts, the best (maha:danapatih Sresthah} means Vajralasya (Lady of Diamond Play). Lord of gifts {danaSpatih} means lord of 'that which is given', 185 [given] out of a desire to confer benefit through worship. He is the best since he is the most eminent in the whole world, inasmuch as he bestows the Four Gifts of worldly objects, fearlessness, the Dharma and consecration. Vajramala (Lady with a Diamond Garland) is [the deity denoted by] Great possessor of moral conduct, [and so] foremost {maha:s~iladharo 'granih}. Because of [ManjuSri's] concentration of body, speech and mind [the word] 'slla' (moral conduct) [is used]. [It is to be understood in the sense of 'concentration'], since 'to be concentrated' 186 is a meaning of the verbal root siL It, [that is, moral conduct,] is characterised by abstention from the ten evil [acts] and has the code of precepts as its nature. The possessor of moral conduct {slla^dharah} is for that very reason foremost, that is, is one who goes in front. Great bearer of patience, unshakable (maha:ksantidharo dhirah} means Vajragita (Lady of Diamond Song). Bearer of patience - that is to say, [bearer of] tolerance (ksamanam), [or] forbearance (titiksa) - means 'one possessing that [patience]' {ksanti^dharah}. And hence, he is unshakable (dhirah) since he has firmness (dhairya) as a support even when among great sufferings.

185Thus danapatih is glossed as 'lord of gifts' rather than 'lord of [the act of] giving'. l * 6sila samadhau (= DhP 25.16).

133

Translation

Chapter 5

Great strength of vigour (mahaviryaparakramah) means Vajranrtya (Lady of Diamond Dance). Vigour (virya) means effort (utsaha) with respect to [developing] what is good. Vigour for what is evil and so forth, however, because of being lazy, is nothing but laziness; it is not vigour. He, [that is, Manjus~ri,] is the [great] strength (parakramah) of that [vigour] {vlrya%>arakramah} since he has completed the roots of merit [that are the basis] of all worldly and super-worldly [achievements]. 187 And so it, [that is, the non-dual mind,] has the tetrad of secret worship [goddesses, namely, Vajralasya, Vajramala, Vajragita and Vajranrtya,] as its nature, because, when the [four] Awarenesses - the Mirror-like [Awareness] and so forth arise in order, they give rise to special kinds of spontaneous 188 aesthetic delight by virtue of the purity of their respective Doors of Liberation. 189 Because that same [enlightened mind] is free from stupidity, 190 from distraction, 191 from incorrect resolve, and from the power of obstacles, he, [that is, £akyamuni,J declared,

^sarvalaukikalokottarakusalamulaparipuranat ('jig rten dang 'jig rten las 'das pa thams cad kyi dge ba'i rtsa ba las yongs su rdzogs par byed pa'o Tib. [193.3.5J) I follow Tib.'s reading of sarva qualifying laukikalokottara rather than kusalamula (jig rten dang 'jig rten las 'das pa thams cad). The latter alternative,".. .since he has completed all worldly and superworldly roots of merit" is unsatisfactory since 'roots of merit' are not so distinguished. Merit can, however, be directed to either worldly or supermundane purposes. The kusalamula are variously listed, eg., alobha, advesa, amoha (DhSam 138); bodhicittotpdda, dsayavisuddhi, ahamkdramamakdraparitydga (DhSam 15). ^anabhoga (Ihun gyis grub pa Tib. [193.3.6]).

189The enlightened mind is described as manifesting goddesses that embody forms of spontaneous delight that result from the purity of the vimoksamukhas (sunyatd, animitta, apranihita and anabhisamskdra) on the attainment of the four Awarenesses (jfiana). I have taken the compound ending -nispattau as a locative absolute (Text 5.178-179). I90dausprajna (shes rab 'chal ba Tib.[193.3.6]). This term is usually taken as an adjective meaning 'unwise' as is its non-vrddhi form dusprajna Here, being the opposite ofprajnd, it must be understood in a nominal sense, ie. as dausprajnatva. ^viksepah. The term aviksepah is classified as a supplementary positive mental event (ku&lacaitta) in Yogacara Abhidharma (see VijnMaSi 338). Here viksepa is seen as the opponent to the concentration of the Perfection of Meditation (dhyanaparamita).

134

Translation

Chapter 5

Dweller in concentration in great meditation. Possessor of a body of great wisdom. Great power, great [skill in] means. Resolve, ocean of knowledge. (37) It should be understood that in this verse [the parts dealing with] meditation and wisdom are reversed in order; 192 and so the meaning is as follows. Possessor of a body of great wisdom {mahaprajnas~arira2xihrk} means Vajradhupa (Lady of Diamond Incense). Great wisdom {maha:prajna} is [to be understood as] a body {mahaprajiia:£arira}. He is 'one who possesses that [body]'. Dweller in concentration in great meditation is Vajrapuspa (Lady of the Diamond Flower) {mahadhyana^samadhisthah}. [ManjuSri is a] dweller in concentration {samadhi^sthah}, that is to say, in one-pointedness of mind, in great meditation {maha:dhyana}. [It is called] great because it has the Four Meditation States as its nature. He is in the state of samadhi in those Four Meditation States. Resolve (pranidhih) 193 is Vajradipa (Lady of Diamond Light). [The word] 'pranidhih' means 'that which is vowed', [that is, a 'vow' or 'resolve', rather than 'one who vows']. [He is so-described] because through the power of resolve Bodhisattvas are produced.

192Though Vilasavajra thus states that the order of enunciation (pathakrama) is not that of the meaning (arthakrama) it is not certain how this should be understood. He might be stating that, in terms of meaning, the pdda on wisdom should precede that on meditation, and that he will comment on them in this order, inverting their NS order. However, the standard list of paramitas always has dhydnapdramitd before prajnapdramitd and in the following gloss there is no reason given for treating prajnd first. An alternative explanation is that Vilasavajra's NS exemplar had the pddas reversed, ie. with the pdda on prajnd coming first, and though he follows the order before him he notes that the usual order has been inverted. The problem with this account is that the extant MSS have the preceding citation of NS 37 in the usual order. It would have to be argued that later scribes restored or changed the order of the pddas in the citation. 193Again (see note preceding) Vilasavajra appears to take epithets pranidhih and mahopdyah in reverse order. It also reverses the usual order of the paramitas.

135

Translation

Chapter 5

Great [skill in] means {maha:upayah} means Vajragandha (Lady of Diamond Scent). He, [the Knowledge-Being ManjuSri,] is so-called since, through the power of his [skill in] means, he fulfils the aims of the whole mass of beings without exception, according to their needs. Therefore, since it is characterised by the Perfections of Wisdom, Meditation, Resolve and [Skill in] Means, it, [namely, that same non-dual mind,] has the tetrad of the external offering [goddesses - Vajradhupa, Vajrapuspa, Vajradipa and Vajra­ gandha - ] as its nature. Since power (bala) and knowledge (jriana), 194 being the cause of joy on the Path of Enjoyment (sambhogikamarga), are introductions to the Dharma-Sphere, these two alone, power and knowledge, are separated [from the other Perfections]. 195 'Power' refers to the Five [Powers] of faith, [energy, mindfulness, concentration] and [wisdom]. 196 'Knowledge' [is to be understood] within [the context of, either] the Path of Vision or the Path of Development: with respect to the Path of Vision it refers to the [seven] Auxiliaries to Enlightenment (bodhyahga); 197 with respect to the Path of Development it refers to the eight Auxiliaries of the Noble Ones (aryastanga). 198

194Vilasavajra thus treats the paramitas in the following order: ddna, sila, virya, ksdnti, prajnd, dhydna, pranidhi, updya, bala, jndna. The usual order of enumeration is ddna, sila, virya, ksdnti, dhydna, prajnd, updya, pranidhi, bala, jndna (DhSam 17-18). 195As introductions to the Dharma-Sphere rather than as embodiments of it Vilasavajra is placing bala and jndna on a lower level than the other paramitas. This allows him to use mahdbalah of the NS to stand for the gate guardians, who are usually regarded as of a lower status than the other mandala deities. 196The standard list of the five Powers (bala) is, sraddhd ('faith'), virya ('energy'), smrti ('mindfulness'), samddhi ('concentration') and prajnd ('wisdom') (DhSam 47). It corresponds also to the five indriyas. 197The seven, also referred to as sambodhyanga are, smrti ('mindfulness'), dharmapravicaya ('discrimination of realities'), virya ('effort'), prf//('joy'), prasrabdhi ('calm'), samddhi ('concentrat­ ion'), and upeksd ('equanimity') (DhSam 49; MVy 988-99). 198The eight principal components of the Path (marga) for Mainstream Buddhism: samyagdrsti ('perfect view'), samyaksamkalpa ('perfect thought'), samyagvdc( 'perfect speech'), samyakkarmdnta ('perfect action'), samyagdjiva ('perfect livelihood'), samyagvydydma ('perfect effort'), samyaksmrti ('perfect mindfulness'), and samyaksamddhi ('perfect concentration') (MVy 997-1004).

136

Chapter 5

Translation

And therefore, because it has [the powers of] faith, energy, mindfulness and concentration as its character, it, [that is, the non-dual mind,] is of the nature of the four gate-guardians [of the mandala] 199 since they cause full confidence in the Mahayana, exertion, uninterruptedness of those [previous results], and complete satisfaction [respectively]. That same [mind], in relation to the purity of [its power of] wisdom alone, has sixteen samddhis - starting with [that embodied as] Maitreya200 - as its nature, [ordered] according to the divisions of [the sixteen] Emptinesses that have been previously stated. Therefore he, [that is, Sakyamuni,] declared, Great one, made of friendliness, [and so] immeasurable. Of great compassion, supreme wisdom. Great wise one, [and hence] very intelligent. Possessed of great [skill in] means, [and so] very learned. 201 (38) Here great one, made of friendliness, [and so] immeasurable means Maitreya. [The term 'maitri' (friendliness) is derived from the word 'mitram' ('friend') as follows:] 'maitram' means 'that which is connected with a friend

199The epithet 'great strength' (mahabalah) in NS 37 stands for the four guardians of the gates (dvarapalah), the first four balas in the list of five being taken as the bases for the deities. The four guardians of the Vajradhatumandala are usually Vajrankusa (East), Vajrapasa (South), Vajrasphota (West) and Vaj raves'a (North). Vilasavajra must have had these in mind given that in chapter 4 he enumerates the mandate-deities as starting with Sattvavajri and finishing with Vajravesa (Text 4.9899). 200Maitreya is the first of a set of sixteen Bodhisattvas that are enumerated in the commentary to the following four verses. Identifying the Bodhisattvas as samddhis continues Vilasavajra's exegesis of the mandala as being composed of Mahavairocana and attendant deities understood at a deeper level as the non-dual enlightened mind and its various samddhis. 20 ^mahdkrti ( 'greatly learned'). Davidson and Wayman read mahdkrtih ('action', 'activity') and translate accordingly: "with profound performance" (Davidson 1981, 23); "great deed" (Wayman 1985, 72). See textual note 5.205.

137

Translation

Chapter 5

(mitram)'; then, it has the suffix -f [added] at the end [to give 'maitri']. 202 Therefore203 he, [that is, Manjus"ri,] is made of friendliness (maitrfmayah) - that is, he is of the nature of friendliness - towards all living beings. Being a great one, made of friendliness {maha:maitrimayah} he is immeasurable, since, for the whole [immeasurable] mass of living beings, he possesses the love [one feels] for an only son. Of great compassion, supreme wisdom means Manjusri. Great com­ passion is [compassion that is] objectless. Possessing that [sort of compassion] he is of great compassion. Supreme wisdom refers to the Awareness of the Perfectly Pure Dharma-Sphere.204 And therefore it has been said, The Fortunate One has loftiness, [that is, compassion,] and depth, [that is, wisdom,] as his nature. Great wise one, [and so] very intelligent means Gandhahastin.205 He is wise, since he possesses great wisdom, [that is, wisdom] that has the discriminating comprehension of dharmas as its nature. Being the great wise one {maha:prajnah}, he is very intelligent.

2Q2pascdn mbantam ("then, it has the suffix -i [added] at the end"). This passage is concerned with deriving maitri ('friendliness') from maitram, which is derived in turn frommitram ('a friend') as a vrddhi form in the most general way by means of a genitive of connection: mitram asydstlti maitram (Text 5.206). 203ta/a/? (de ni Tih[193.5.4J). The force of the word 'therefore' is unclear. Tib.'s "As for that [friendliness]" may be better. 20*agradhir iti suvisuddhadharmadhdtujndnam. Vilasavajra takes agradhih as a karmadharaya as opposed to the more idiomatic adjectival bahuvrihi ("possessing supreme wisdom"). Given that the second word of the other three quarter-verses is an adjective, the most natural translation of agradhih would be to also take it in an adjectival sense. The exegesis of this pdda also breaks the pattern of the other three, which take the concatenative form "[He is] X, and therefore (ata eva) Y". This would be inappropriate here as it would make wisdom dependent upon compassion. Making agradhih a substantive, however, identifies Manjusrijiianasattva as wisdom while keeping compassion as an attribute. The traditional ascendance of wisdom is all the more important here given that the deity denoted by the quarter-verse is the Bodhisattva Manjusri. Thus, Manjusnjiianasattva as Non-dual Awareness is here depicted as taking the form of the Bodhisattva Manjusri. 205The name Gandhahastin denotes an elephant in must (BHSD 210).

138

Chapter 5

Translation

Possessed of great [skill in] means, [and so] very learned means Jnanaketu. 206 Being possessed of great [skill in] means {mahaiupayah} 207 he is very learned (mahakrti > mahapanditah), since he explains the Dharma clearly, on account of knowing the dispositions and propensities of all living beings.

Great one, endowed with the strength of magical power. Possessed of great impetus, [and thus] with great speed. Possessed of great magical power, named 'Great Lord'. Strong in great powers. (39) Here great one, endowed with the strength of magical power is Bhadrapala {rddhi^bala} {rddhibala^upetah} {maha:rddhibalopetah}. Possessed of great impetus, [and thus] with great speed means Sagaramati. He is possessed of great impetus{mahaivegah}, that is, [of great] power of progress, since he appears in many world spheres due to being motivated to attend and worship all the Tathagatas. For this very reason he is said to be with great speed. Possessed of great magical power, named * Great Lord' means Aksayamati. He is possessed of great magical power because he has the Provision of Merit belonging to wordly [beings], £ravakas and Pratyekabuddhas. He is named * Great Lord* as he has completed the Provision of Knowledge belonging to all the Tathagatas. Strong in great powers means Pratibhanakutah. He is strong, that is, exerts vigour, in the great powers {maha:bala}{mahabala^7parakramah}, that is, with respect to the Ten Powers of the Tathagata, which will be [fully] enumerated later208

206Jnanaketu is elsewhere the name of a samddhi (MVy 559). 2Q^mahopdyah. In NS 37c (mahabalo mahopayah) mahopayah is analysed as a karmadharaya ('great [skill in] means'). The present bahuvrlhi analysis provides a contrast. 208 The dasa tathdgatabaldni are listed in NMAA chapter 6.

139

Translation

Chapter 5

- [and] the first of which is, "The Knowledge Of What Is Possible And What Is Im­ possible".

Great shatterer of the mountain of existence. Bearer of the great vajra, solid. Great fierce one, greatly ferocious. Creating fear in 'Great Fear'. (40) Here, great shatterer of the mountain of existence means Mahasthamaprapta {maha:bhavadrisambhetta}. Existence (bhavah) refers to the five constituents [that are the basis] for clinging. Those same [constituents] are a mountain (adri){bhava:adri} since they are, [individually,] rocks (Saila) of ignorance.209 Because he has obtained correct and rightly oriented Awareness he is a shatterer (sambhetta) of that [mountain of ignorance] {bhavadri^ambhetta}, that is, his practice is [its] destruction. Bearer of the great vajra {mahavajra^harah}, solid means Sarvapayanjaha. The great vajra denotes Awareness that is completely beneficent (samantabhadrajnana). Because he, [that is, ManjuSri,] is firm and since he has a material form he is [described as] solid. Great fierce one, greatly ferocious means SarvaSokatamonirghatamati. He is a great fierce one{maha:kruro} 210 since he displays a very wrathful form to beings that are hard to control. Greatly ferocious: though peaceful, ManjuSri, the Fortunate One, also displays ferocity through a desire to discipline living beings.

evadrir ajfidnasailatvat. Alternatively: "Those same [constituents] are a 'mountain' (adri) as they are a mountain (saila) of ignorance". Both translations are possible, with 'constituents' taken either individually or collectively, since sailah can mean 'mountain' or 'rock'. Tib.[194.2.2] takes it as 'rock' (brag). 210The translation, "great fierce one" follows Vilasavajra's karmadharaya analysis. More idiom­ atic would be "greatly fierce". The translation ofmahdraudra should perhaps follow the same pattern as mahakrurah, but the fact that Vilasavajra does not give a grammatical analysis suggests he took it in its natural idiomatic sense, "very ferocious".

140

Translation

Chapter 5

Creating fear in * Great Fear* {mahabhayabhayam^karah} means Jalinlprabha. * Great Fear* is [the name of] Samkara, [namely, £iva,] in his Mahabhairava ("Greatly Terrifying") aspect. He, [that is, Manjusri,] produces fear even in him. Master of the great spells, [and so] protector. Master of the great mantra, [and so] guru. Established on the path of the Great Way. Master of the path of the Great Way. (41) Here master of the great spells (mahavidyottamah), [and so] protector is Candraprabha. The great spells {mahaividyah} are the four dhdranis that are characterised by text, meaning, mantra and seal. He is their master {mahavidya^uttamah} For this very reason he is a protector from [birth in] all the bad destinies since he completely purifies the behaviour of body, speech and mind of all beings. Master of the great mantra (mahamantrottamah), [and so] guru means Amitaprabha. The great mantra {maha:mantrah} is that which has the Great Seal as its nature. He is its master (uttamah > s"resthah) {mahamantra^uttamah}. For this very reason he is a guru since, knowing the dispositions and propensities of all living beings, [he gives] the special teaching of the spiritual realisations. Established (arudhah > sthitah) on the path (nayah > margah) of the Great Way means Gaganaganja {mahayana%iayah}{mahayananaya^ariidhah}. The mean­ ing of the Great Way will be explained next. Master of the path of the Great Way is Sarvanivaranaviskambhin. The Great Way is, [in its ultimate nature, to be understood as] the non-objectifying Awareness (jfiana) which comes about from realising the inexpressible [common] nature of Samsara and Nirvana. Alternatively, the support nature of the Great Way

141

Translation

Chapter 5

can be summarised as a threefold supremacy: of provisions, method, and of scope.211 If that is so, [that is, if the Great Way is seen in these latter terms], in what sense is it 'Great'? The Buddhas, the Fortunate Ones, benefit all living beings. So it is they who are 'Great' [and] it is their 'Way', which they [gain] for the sake of perfection and cessation. Therefore, the Great Way is the Way of those associated with it; in other words, Great Way means 'The Way of the Great {maha^yana}'. Alternatively, it is [called] the Great Way by virtue of being great in seven ways: because of the greatness of its basis, practice, wisdom, energy, skill in means, attainment and [enlightened] activity. [From this perspective] Great Way means 'The Way which is Great {maha:yana}'. [It is so described] since these [factors] do not exist in any other Way. This, in brief, is the threefold [meaning of the expression] Great Way. It should be understood as having meanings which explain [i.] the motive (hetu) [for following it, namely, the attainment of non-objectifying Awareness], [ii.] those to whom it applies (gocara), [namely, Buddhas,] and [iii.] its nature (svabhava), [namely, its sevenfold greatness].212 This is the case,213 [namely, that ManjuSri is the master of the path of the Great Way'] because, by depending on his [skill in] means alone, the Fortunate One, the Knowledge-Being [Manjusri], the Knowledge-Body of all the Tathagatas, has all the worldly and superworldly samadhis as his nature.

21 ^or sambharopdya- ('provisions, method') Tib. reads 'khor ba'i gnas thabs dang {samsaropaya}. See textual note 5.258. 212It is unclear here whether the threefold meaning of Mahayana includes its initial definition as non-objectifying Awareness. I have translated as though it does. If it does not, there should surely be a correspondence with the threefold supremacy (of provisions, method, and of scope). Yet after the initial definition, only two further explanations of 'Mahayana' are given, one of its scope (gocara) and one of its 'greatnesses'. For discussion of Tib. see latter part of textual note 5.258. 2l3 tad evam (de nyid Tib.[194.4.2]). The sense of this passage (Text 5.267-8) is uncertain. It is unclear what tad evam refers to ( Tib. appears to have read tad eva). Provisionally, I take it to be to previous explanations of the term Mahayana.

142

Translation

Chapter 5

It should be understood that 'ManjuSri, the best of the glorious',214 the Knowledge-Being who dwells in the heart of all the Tathagatas, is qualified [in the Ndmasamgiti] by phrases, the words of which are Name-mantras,215 the first of which is "Namely, [he is] the Fortunate One, the Awakened One, the Fully Awakened One, born from the letter A". 216 [These phrases] are summary utterances217 characterised as [being comprised of a series of] predicates, [and] lacking verbs. And these manciala-deities - Vajrasattva and [the rest of the deities of the Vajradhatumahamandala] - by virtue of their union with the Great Seal (mahamudra), go forth endowed with their own [distinguishing] colours and implements. 218 And having issued forth they perform the activities of the Buddhas for [the benefit] of all living beings in all world spheres. Returning [to the mandala], they become words that are the Name-mantras, coloured with the hues of their own families. One should imagine [them] placed in order on the moon-disc seats of the great Vajradhatu mandala that were arranged earlier.219 [Then] generating the conviction that one possesses the appearance, indicated earlier,220 of the Knowledge-Being Manjus'ri, the Knowledge-Body of all the Tathagatas, who dwells at the centre of the circle of those [mandala deities], 'a son of a [good] family or a daughter of a [good] family', a Bodhisattva 'practising the observances of the Way of Mantras should quietly recite

anjusnh srimatam varah (= NS 157d). This is an allusion to the final pdda of the NS that contains the Names, the only one to contain the Name Manjusri. ^^namamantraksarapadaih ('by phrases, the words of which are Name-mantras'): Text 5.270. For discussion of this compound see Introduction, section 5.5. 216This is NS 28ab, which contains the first Names of the NS and is the first verse commented on in NMAA chapter 5. 21 ''grahanakavdkya. This term is used at the beginning of chapter 3 (Text 3.5; comment on NS 23). See note 80. 218For reproductions of some Japanese illustrations of the different Vajradhatu mandala-deities see Lokesh Chandra, 1987 (Introduction, p.80-87). 219See chapter 4: Text 4.98-99. 220See chapter 4: Text 4.172-176.

143

Translation

Chapter 5

three times the crest-jewel called the Namasamgiti, which has non-duality as its ultimate meaning, of this same Fortunate One, the Embodiment of Knowledge. Or one who wants to read should perform [the recitation] using a book'.221 The pure one, becoming silent [and] reciting soundlessly - so that no-one can hear - [and] correctly, the phrases [of the Namasamgiti], whose words are the Name-mantras, will soon obtain the throngs of qualities enumerated in the [section in the Namasamgiti on] Benefit as well as other [qualities]: The best of men,222 [he] who preserves the Namasamgiti of ultimate meaning [and] who has properly collected the Provisions of Merit and Knowledge, will soon acquire all Buddha-qualities223 and fully awaken to unsurpassed perfect enlightenment. Teaching the unsurpassed Dharma to all living beings, [and] not entering [the state] of Final Extinction for many aeons,224 he will be a Dharma king with a drum of the true Dharma [resounding] in the ten directions. And combined together, the phrases [of the Namasamgiti], whose words are the

221 The passage starting, "a son of a [good] family ..." incorporates material, marked by single inverted commas, from two parts of the NS anusamsa. The Skt of the NMAA, with the material incorporated underlined, is: tasyaiva bhagavato jnanamurter advayaparamartham namasamgitim nama cudamanim yah kaS cit kulaputro va kuladuhita va mantramukhacaryacari bodhisattvas trihkrtva kanthagatam avartayisyati / pustakagatam va pathamanah pravartayigyati (Text 5.279-83). The NS passages, again with the material incorporated into the NMAA underlined, are: punar aparam vajrapane vajradhara yah kai cit kulaputro va kuladuhita va mantramukha­ caryacari imam bhagavato manju&ijnanasattvasya sarvatathagatajnanakayasya jnanamurter advayaparamartham namasamgitim nama cudamanim ... dharayisyati (NS.Dav. 65, 2—4). punar aparam vajrapane vajradhara ya imam namasamgitim nama cudamanim pratyaham akhandasamadanatas trihkrtva kanthagatam avartayisyati / pustakagatam va pathamanah pravartavisvati (NS.Dav. 66,21-23). In the translation I have taken mantramukhacaryacari'to qualify bodhisattvah whereas in the NS passage it qualifies kulaputro va kuladuhita va. This follows Vilasavajra who, in the opening of chapter 4 (Text 4.2-4), incorporates parts of this same anusamsa passage, starting with mantra­ mukhacaryacari, which he uses to qualify the word bodhisattvah. 222purusapumgavah. 77^.reads skyes bu gang zag {purusapudgalah}. See textual note 5.287. 22 *sarvabuddhagunan (sangs rgyas thams cad kyi yon tan mams Tib.[194.5.4J). Tib. translates, "... the qualities of all the Buddhas". 224aparinirvdnadharmd. I have taken dharma in the sense of 'characteristic'. Thus, lit., "one who does not have the characteristic of (attaining) Final Extinction ...". An alternative is to take pannirvana as an adjective ("extinguished"), "his Teaching unextinguished ...".An adjectival use would be less normal in Buddhist contexts, however.

144

Translation

Chapter 5

Name-mantras of [the deities of] this great mandala of Vajradhatu, contain [a total of] eighty-six names.225 Apportioning them, [that is, the names,] with respect to the mandala deities, they are connected [as follows]: four for the Knowledge-Being [ManjuSri] and the moon-disc [at his heart]; three for the letter A; five for Sattvavajri and [the remaining three family mothers]; eleven for the Eternal One [Vairocana] and [the other Tathagatas]. The remaining phrases, whose words are the Name-mantras, are connected in due order [with the deities] in the relation of qualifier and qualified. The first [mandala-deity] is Vajrasattva [and] the last Sarvanivaranaviskambhin. [Altogether] there are fourteen verses [of the Ndmasamgiti that contain the Names of the Vajradhatu mandala-deities]. According to tradition the total number of Namemantra words is eighty-six. [Here ends] the chapter on 'The Vajradhatu-Mahamandala of Bodhicittavajra', the fifth in the commentary on the Aryanamasamglti called, "An Explanation of the Meaning of the Name-Mantras".

225For discussion of the Name-mantras see Introduction, section 5.5.

145

Materials

Introduction to the text

INTRODUCTION TO THE TEXT 1. Materials For the edition of the Namamantrarthavalokini I had access to the following ten Nepalese manuscripts: ] MS belonging to the University of Cambridge (Bendall Add. 1708). PalmA: leaf, Newari script, 115 folios (26 missing), dated Samvat 57? (= c. 1450 CE). MS belonging to P. R. Vajracarya, Kathmandu (NGMPP Reel No. E B: 360/16). Paper, Newari script (first 13 folios in Devanagari = BI), 86 folios, undated. MS belonging to P. R. Vajracarya, Kathmandu (NGMPP Reel No. E C: 419/17). Paper, Devanagari script, 80 folios, dated Samvat 920 (= 1800 CE). MS belonging to Dharmavajracarya, Kathmandu (NGMPP Reel No. E D: 1369/3). Paper, Devanagari script, 194 folios, undated. MS belonging to P.B. Kamsakara, Kathmandu (NGMPP Reel No. E E: 1057/13). Paper, Devanagari script, 72 folios, dated Samvat 1021 (= 1901 CE). MS belonging to the National Archives, Kathmandu (NAK MS. No. 5F: 6863; NGMPP Reel No. A 133/12). Paper, Newari script, 135 folios (1 missing), undated. MS belonging to the University of Kyoto (Goshima & Noguchi 58). Paper, G: Devanagari script, 123 folios, undated. MS belonging to the National Archives, Kathmandu (NAK MS. No. 5H: 138; NGMPP Reel No. B 108/24). Paper, Devanagari script, 77 folios, undated. MS belonging to Dharma Ratna Vajracarya, Kathmandu (NGMPP Reel J: No. E 1920/2). Paper, Devanagari script, 121 folios (4 missing including the final folio). MS belonging to the National Archives, Kathmandu (NAK MS. No. 3K: 570; NGMPP Reel No. B 109/13). Paper, Devanagari script, 102 folios, dated Samvat 1954 (= 1897 CE?). I also consulted two xylograph editions of the Tibetan translation of the NMAA: Tib.D. sDe dge edition of the bsTan 'gyur. rGyud 'grel, Khu. 27vl-115v3 (T6h. 2533). Copy in the possession of the Bodleian Library, University of Oxford. Tib.P. Peking edition of the bsTan 'gyur. rGyud 'grel, Si. 31r8 - 135rl, in Peking Tibetan Tripitaka, ed., D. T. Suzuki, Tokyo: Tibetan Tripitaka Research Institute, 1956. Vol. 74, pp. 184.4.8-226.2.1. (Ota 3365)

] I worked with printouts from microfilms of all of these MSS. In the case of MS A I also consulted the original in the University Library, Cambridge.

146

Introduction to the text

Materials

A number of NMAA manuscripts have been microfilmed more than once by the NGMPP.2 Also, some NMAA manuscripts microfilmed by the NGMPP have been microfilmed or listed separately by Takaoka and Tsukamoto. 3 The following table gives duplications and correspondences and lists the extant copies of the NMAA that I am aware of. (Sigla) MA) ... 2. (B)

Manuscript Bendall Add. 1708

NGMPP E 360/16

3. (C)

NGMPP E 419/17

,4._(D) 5.(E) 6. (F) 7. (G) 8. (H) 9. (J) 10. (K)

NGMPP E 1369/3 NGMPP E 1057/13

.^.,.. ......... ................ 12. 13. 14. 15 16?

Duplications & Correspondences = E 41 8/6 (lacks fol. 1-13) = E 387/19 plus E 388/1 (lacks fol. 1-13) = E404/18(=fol. 1-13 only) = £370/17 = E 397/7 = Takaoka CH 380

"NGMPP A m/n Goshima NGMPP NGMPP NGMPP

& Noguchi 58 B 108/24 11920/2 B 109/13

——— .... ...... ......... ......... ................ ,.„,„..-„........ = Bir 172 = Bir Library Catalogue tr 590 = BSP vol. 7, pt.2, 77 '

"JASWR MBB-T-77 { IASWR MBB-1-117

[Jakaoka KA 50

Takaoka CA 43-3

! St. Petersburg

\ Baroda 43

|

Table 10: Duplications of microfilms and correspondences with catalogue numbers. I have not been able to consult manuscripts 11-16 in the above list. Nos. 11-14 (lASWR and Takaoka) are microfilm copies of manuscripts in private ownership in Nepal. Of these, nos. 11 and 12, microfilmed by the IASWR, do not appear to

2One of the difficulties in identifying copies of the NMAA microfilmed by the NGMPP is that they are found under a variety of titles in the NGMPP card catalogues in Kathmandu and Berlin. The only fully correct title is that for E 1057/13 (MS E): Aryandmasamgititikd (Namamantrarthavalokini). Other titles include Namasamgititika (E 404/18; E 1369/3; B 109/13), Aryandmasamgititikd (E 387/19 + E 388/1), and Paramdrthandmasamgititikd (B 108/24). Mantrdrthdvahkini is erroneously used as a title on two occasions (E 360/16; E 397/7). This is probably the result of taking Ndma in Namamantra- to mean 'called'. Mantrdrthdvahkini is found twice more as part of a longer title, eg. Namasamgititika (Mantrdrthdvahkini) (E 1920/2). On three occasions what was titled Ndmasamgiti turned out to be the NMAA (E 418/6; E 419/17; A 133/12). 30f the following, Tsukamoto (p.209) lists numbers 1, 5 (as Takaoka CH 380), 7, 9 (as Bir 72 = BSP tr590(2-77) ) and 11-15. No.12. is incorrectly listed as MBB H-l 17 rather than MBB 1-117. For the IASWR MSS Tsukamoto refers to the Title List (IASWR list). However, this does not give any details of MBB-I-77 (nos. MBB-I-77 to MBB-I-81 are unlisted). For these one must consult the descriptive catalogue by M. B. Bajracharya and C. S. George (IASWR). (But see note following.)

147

Introduction to the text

Materials

correspond with any of the NGMPP manuscripts.4 No. 13 (Takaoka KA 50) is almost certainly an apograph of my MS C. 5 No.14 (Takaoka CA 43-3), which is titled 'Namasamgiti Tika" in Takaoka's catalogue, is identified as the NMAA by Tsukamoto (p. 209). 6 These four microfilmed MSS (11-14) are on paper. Tsukamoto also identifies Takaoka CH 380 as the NMAA (again, titled 'Ndmasamgiti Tikd" by Takaoka). As a result it became clear that this was the same manuscript as NGMPP 1057/13 (= my MS E). No 15 was consulted by Minaev for his 1887 edition of the Ndmasamgiti. According to his description of MSS it then belonged to the Library of the Imperial University of St. Petersburg. 7 He gives no catalogue or accession number, and the manuscript is described as 'new'. Whether this refers to its age or its being uncatalogued is unclear. 8 It was not (subsequently) listed by Mironov and I have not been able to ascertain whether it is still held in the University Library. The identity of MS 16, owned by the Oriental Institute, Baroda, remains to be ascertained. The Baroda catalogue gives ''Namasamgititika' as a title.9

4The IASWR, however, can find no trace of the microfilms of either of these MSS. The fiche entitled Namamantrartavalokim (Aryandmasangititika) (MBB-I-117) contained, on inspection, the Durgatiparisodhanatantra and no fiche for MBB-1-77 have been prepared. 5 Sakurai uses Takaoka KA 50 in his edition of NMAA adhikaras 3^4 (= his MS Tj). See the note on his edition in Appendix VI. 6Sakurai consults this MS in his edition of the Vairocanabhisambodhi quotation in NMAA 6 (Sakurai 1989, p. 515-6). The variants in the apparatus suggest that it is a descendent of my hyparcherype e, possibly another apograph of C. 7See, Minaev, I. P. (1887) Buddhizm. Izledovaniya i Materialui. St. Petersburg. Vol. II, p. xii. This work also contains Minaev's edition of the Mahavyutpatti (For that of the NS see pp. 137-159.) Vol. I, which comprises a collection of essays by Minaev, contains (p. 226) a citation from chapter 6 of the NMAA and has been translated into French: see Assier de Pompignan, R. H. (trans.) (1894) Recherche'S sur le Bouddhisme, Paris. Davidson (1981, 49) states that Minaev had intended to translate the Ndmasamgiti using Vilasavajra's commentary. 8Other details given by Minaev are that it is an undated MS, 130 folios in length with 8-9 lines per page, in Nepalese script. I am indebted to Dr. G. A. Evison of the Indian Institute Library, Oxford, for arranging for the Russian of the description of manuscripts to be translated for me. 9See Baroda, p. 1462, Serial No 43, Accession No 13297.

148

Introduction to the text

Description of Manuscripts

2. Description of the Manuscripts 2.1 Introduction All the manuscripts are made of paper and in good condition with the exception of the palm-leaf MS A, which is damaged in parts. Four are dated (ACEK), the earliest being from the fifteenth century CE (A); one has its colophon missing (J); the remainder are undated (BDFGH). Of the ten, BG are the only completely uncorrected manuscripts. FDJEK have a very few corrections; and AHC are more heavily corrected. At points these last three are extensively corrected, sometimes by more than one hand. (More specific information on corrections is given below.) The presence of manuscripts with such corrections suggests that the present text was not copied only to gain religious merit, as Cuppers has suggested was the case for Buddhist Sutra texts in Nepal, 10 but that it was also a commentary used by working scholars. 2.2 Presence of Figure Numerals At the end of some of the (fourteen) chapters of the text there are numerals in figures. Bendall (p. 203) noticed this in MS A, citing the text at the end of chapters 6-8, and stating that the figures represented the number of sections in each chapter. This is not quite correct. What the figures represent is the number of Name-mantras (namamantrah) u in the corresponding chapter of the NS according to Vilasavajra's exegesis. 12 The figure is placed after the total of Name-mantras given in words, which whole in turn is placed either immediately before or after the colophon for the chapter. This number in figures is given only at the end of chapters that deal with the

10See Cuppers, (1990) The IXth Chapter of the Samddhirdjasutra: a text-critical contribution to the study ofMahdydna Sutras. Stuttgart, p. XIX.

11 In fact, Vilasavajra states that the count of the ndmamantrah for each chapter is one given by tradition (ity amnayah). See Introduction, section 5.5, for discussion of the ndmamantrah. 12Each Name-mantra is usually treated separately; to that extent each could be said to have a section devoted to it, giving Bendall's statement a certain truth.

149

Introduction to the text

Description of Manuscripts

Name-mantras of the NS, that is, chapters 5-10. All the MSS have these figures, with some slight variation between MSS (though only MS A has '// 86 // at the end of chapter 5). Before the totals (in words and figures) for the Name-mantras, the number of NS verses on which each NMAA chapter has commented is given in words. DJECGK (archetype 6) also follow these with the number in figure form. However, ABFH have no figure numbers here suggesting, through the application of stemmatic analysis, that they were not present in the archetype. For chapters 5-8, DJECGK also have the number of NS verses in words (given incorrectly as a whole number for chapters 6 and 7) after the totals for the Name-mantras, ie. they give the number in words twice. When this happens, the number in figures follows the second instance of the number in words. Manuscripts DEJBiGK have Namasamgiti verse numbers embedded in the text of the NMAA. They are placed, in figures, at the point where the commentary on each verse is completed; and in each chapter the numbering starts afresh from one (since the chapter divisions of the NMAA and NS correspond). The numbering in E is only partial: it omits some of the numbers in chapters 1 and 2, has only a few for chapter 6, and none for the rest. C has the numbers in the margin throughout, with insertion marks in the text to indicate their position. I take these NS numbers to be an addition to the text. They are only found in MSS from one part of the stemma (5): ABHF do not have them, so they should be absent from the archetype (£1) unless they have been omitted independently, which seems unlikely. However, their presence certainly makes reference to the commentary on particular NS verses easier, and this may have been the motive for adding them. The partial numbering of E could originate in various ways: it, or one of the manuscripts it descends from, could have omitted some of the numbers present in its exemplar, though this seems rather unlikely if the numbering was already

150

Introduction to the text

Description of Manuscripts

incorporated into the text. The alternative is that E, or one of its antecedents, had as its exemplar a manuscript such as C, that had marginal numbers, and that these were only partially incorporated. Perhaps some of them were obscure, or the scribe had doubts as to whether they should be part of the text. In C we possibly witness the transitional state between manuscripts with and without numbering. 2.3 Lacunae The existence of lacunae in the MSS and the way in which they are marked also reflects the stemma groupings of the MSS as ascertained through examination of variants. Thus, A has no lacunae marked; B indicates them only very occasionally, leaving empty space in the text to mark them. (B has only one lacuna of over two syllables: on fol. 22v.) F contains many lacunae, especially in the first half of the text, which are marked by empty spaces as in B. H also has many lacunae similarly marked, though some are also indicated by means of dashes that fill the space of the lacuna. In these cases, however, many of the dashes are part of laconose corrections inserted into the empty spaces of the lacunae of the uncorrected manuscript. The remaining six manuscripts (DJECKG) all contain lacunae that are generally marked by dashes. CK contain some lacunae marked by both dashes and empty space, usually one or two dashes followed by a longer empty space. This method of marking appears to be used when the scribe does not know the precise length of the lacuna. This is borne out where the text of lacunae marked wholly by dashes is attested by other manuscripts; in such cases the number of dashes often exactly corresponds to the number of missing aksaras. K, which is an apograph of C, generally preserves C's distictions between the different types of lacunae. But G, an apograph of K, often omits the space where K (and C before it) have lacunae that are marked by dashes followed by empty space. The number of dashes no longer gives an indication of the number of missing aksaras. This pattern is not universal for G: fol. 86r has a dash followed by a gap and

151

Description of Manuscripts

Introduction to the text

there is an occasional instance of two dashes followed by a half dash space before the following syllable. 2.4 Individual Manuscripts MS A (6-7 lines per side) Although this dated palm-leaf manuscript, the oldest of those extant, has been described at some length by Cecil Bendall, 13 his remarks can be enlarged upon and require some correction. Parts of this manuscript are in poor condition; some folios are damaged at the margins or corners with resulting loss of text and folio numbers (especially fol. 3855). There is some insect damage in the form of small irregular holes and though much of this is in the margins some has intruded into the text (eg. fol. 36-40, 52, 61). On a few folios the text is a little faded, and there is occasional text loss through detachment of a surface layer of the palm leaf (eg. fol. 107, 108, 114). Of the 115 folios there are 26 missing. 14 The script of MS A is 'hooked Newari', a form of Newari script distinguished by a curve or hook on the top of each letter, and common in Nepalese manuscripts of the thirteenth to fifteenth centuries CE (see Bendall, 1883, xxi ff). At the beginning of fol. 54r there is a change in hand, which continues until the end of the manuscript. 15 This change is also reflected in changes in the script, as well as in the numerals used in the pagination. 16 The text of chapters 1-5, as well as that collated 13 See Bendall, C. (1883) Catalogue of the Buddhist Sanskrit Manuscripts in the University Library, Cambridge. 203-4. 14Fol. 4, 5, 8, 9, 19-22, 26, 42, 43, 47, 49, 50, 59, 63, 78, 80, 91, 94, 97, 99, 109, 110, 112, 113 are missing. This list differs slightly from that given by Bendall since he does not list fol. 78 as missing. However, the folio microfilmed before fol. 10 has lost its number as a result of damage to its margin. That it is out of place is clear from the script, which is that of A2. Examination of its content indicates that it is fol. 92. Since Bendall counts fol. 92 as missing, the total number of missing folios remains the same, ie. 26. The microfilming process has made some folios (6r, 72v, 92r, lllr, 114r) appear dark causing partial illegibility. 15In fol. 84-6 the script slants forwards in contrast to its usual upright form. The change is not abrupt but occurs gradually over the course of fol. 84r. It is unclear whether this might be a different hand. Fol. 87r returns to the upright script of fol. 54-83. 16Bendall does not mention the change in hand. The examples of the script given at the end of the catalogue are from the second hand (see the "Table of Letters" and the "Table of Figure Numerals"). In this hand the aksaras for ta and bha illustrate forms found in early Bengali script. (Eg. see the

152

Introduction to the text

Description of Manuscripts

from chapter 6 for the purpose of establishing the stemma, is all of the first hand (ie. it falls within fol. 1-53). Whether the postion of the manuscript in the stemma changes with the change in hand remains to be ascertained. MS A is dated in the final colophon with words standing for figures. Unfortunately, the folio containing the colophon is damaged and some - perhaps two syllables - of the relevant text is lost and two further syllables are hard to read. Bendall reads the date as N.S. 577: "If the first member of the partly mutilated compound word expressing the figures of the year stands for asva, the date is N.S. 5 (isu) 7 (mahidhara) 7 (asVa) or A.D. 1457" (Bendall, 1883, 203). I am not convinced that asva can be read. What Bendall takes as an initial a is not clear, and what he reads as sva is probably better taken as ndha. I can read very little of the first part of the passage with any certainty; ie. no more than: [**]**ndha*mahidharesuganite samvat. However, the error resulting from the illegibility of this first word falls within a margin of only ten years, since it represents the final number of the date. An argument might be made for a different reading of the date on the grounds that mahi and dhara can also be used individually as word numerals. 17 If this were the case here the date could be N.S. 5 (isu) 1 (dhara) 1 (mahi) or 139ICE, which is still consistent with the period within which the hooked form of the script is known to have been used. But such a proposal leaves the role of the previous syllable unaccounted for (*andha*), a syllable that has every indication of being compounded with mahidharesuganite (ie. it has no visarga or anusvara). There are only a very few marginal corrections in the first portion of the text (Ai, ie. fol. 1-53), largely in a second hand. The number of corrections increases

facsimile of MS Add. 1699, I, fol. 5, which is in an early Bengali hand, at the end of Bendalrs catalogue. The forms of the syllables ta and bha are seen clearly here in the word amitdbha in line one.) 17 See Sarma, K.V. & Subbarayappa, B.V. (1985) Indian Astromony, A Source-Book. Bombay: Nehru Centre, appendix 5, 'Bhutasankhya—Word Numerals'.

153

Introduction to the text

Description of Manuscripts

markedly in the second portion (A2). With the exception of the few corrections to A] in the hand of Afs scribe, all the corrections use the script of A2, in a smaller size than that of the text. Though it is not possible to say whether they are in the hand of the scribe of A2, the form of the script suggests that they were made at a time close to that of the manuscript itself, since usage of hooked Newari had ceased by the sixteenth century CE. MS B (28.8 x 12.6 cms., 9 lines per side) This is, for the most part, a clear and legible manuscript mostly written in Newari script. However, fol. 1-13 are in Devanagari script, with fol. 14-86, the remaining 73, in Newari. The Newari folios are numbered one to seventy three in small figures in the bottom right hand corner of the verso. (The verso of the final folio, which contains the colophon on its recto, was not microfilmed but I assume it is numbered as the rest.) Analysis of the variant readings of the two portions shows that they occupy different places in the stemma. The Devanagari portion (Bi) is an apograph of MS C whereas the Newari portion is an independent witness of considerable value. It is not clear whether the pagination of the Newari folios is that of the original scribe. If not, then the original was unnumbered and probably included the opening folios, which were subsequently lost; if it were, then in B's exemplar those folios must have been damaged or already lost. In any case the first thirteen folios clearly replace a lost or damaged portion of the original or one of its precursors. Since BI is an apograph of C, and C is complete and in good condition, the readings of BI are not included in the edition of text. I have adopted a continuous pagination, ie. 1-86, for purposes of reference to the manuscript as a whole. MS B was microfilmed three rimes by the NGMPP (see Table 10 above). There are some errors in the filming where two folios have been turned at once: E 360/16 omits fol. 63v/64r and fol. 65v/66r; E 418/6 has fol. 65v/66r but still omits fol. 63v/64r. E 418/6 also omits the final folio. Both E 418/6 and E 387/19 plus E

154

Introduction to the text

Description of Manuscripts

388/1 omit the thirteen Devanagari folios. Although E 418/6 is by far the clearest film, most of the pagination numbers at the bottom right hand corners are not visible since they fall outside the frame of the photographs. Fol. 1-13 have also been microfilmed as a separate item by the NGMPP (E 404/18). B is an uncorrected manuscript with neither marginal nor interlinear corrections. MS C (22 x 10.6 cms., 12 lines per side) Written in Devanagari script this MS is dated "Nepala Sam 920" (= 1800 CE) in the colophon. Generally the text has just a few marginal corrections, usually of one or two syllables, in what is possibly the hand of the original scribe. There is one instance of a more extensive marginal correction: fol. 42r has three lines down the side and one along the top margin, in the hand of the original scribe. However, fol. l~4 are heavily corrected by at least two hands. MS D (24.7 x 8.3 cms, 5 lines per side) This is written in Newari script in a large clear hand. Fol. 1-30 contain passages that appear to have been intentionally marked over or highlighted, with the result that they are darkened on the microfilm. Generally these passages are still legible through the 'highlighting'. The marking bears no apparent relation to the content of the text. For the most part D is corrected only very occasionally, possibly by the original scribe. There is a flurry of corrections in chapter 6 in connection with a number of lacunae (see fol. 53v, 54, 55r). Here the corrections are either marginal or written to fill the lacunae. Otherwise the MS contains just two other marginal corrections. MS E (29.4 x 13.5 cms., 11/12 lines per side) Written in Devanagari script, dated "samvat 1021" (= 1901 CE), MS E is closely written and sometimes of poor legibility. It has a few small marginal corrections, mostly in what is probably the hand of the original scribe, though there are some very occasional interlinear corrections in a second hand.

155

Introduction to the text

Description of Manuscripts

MS F (18 x 10.5 cms., 9 lines per side) This is an undated manuscript - as with MS H, it has no final colophon - written in a Newari script often not easy to read. In particular, the vowels ale, and a I o are often hard to distinguish, as are sa and ga. The number of folios of F is incorrectly given as 85 in the title sheet at the opening to the NGMPP microfilm. The correct figure is 135, made up as follows: fol. 1-64 are numbered in the middle of the right hand side of the verso; fol. 65v is blank and unnumbered; fol. 66-8 are numbered 1-3 in the bottom right hand comer of the verso in a much smaller hand of poor legibility; fol. 69-134 are numbered 165 (number 8 is erroneously used twice), also in the bottom right hand corner of the verso, in the same smaller hand; and the text ends on fol. 135r. The reason for these changes is unclear since the manuscript is almost certainly in a single hand. Fol. 1-64 appear to have the original pagination. In any references to this manuscript, as well as in the table that gives the folio number for the beginning of each chapter in the individual manuscripts (see Appendix V), I have adopted a continuous pagination from 1 to 135. Like fol. 65v, fol. 42v is blank; fol. 43 is missing, one frame of the microfilm containing fol. 42v and fol. 44r. F has almost no corrections: in the hand of the original scribe there are four marginal insertions of one syllable and one of two syllables (19r, 21v, 85r, 117v, 84v); one marginal correction of one syllable (29r); and, possibly in a second hand, there is a one and a half line marginal insertion (4v). There are a small number of erasures, generally one to four syllables in length, making the erased text illegible. MS G (30.1 x 10.8 cms., 7 lines per side) This is a clear, legible, undated Devanagari manuscript. The punctuation is that of double dandas throughout. They appear lighter in shade than the body of the text in the microfilm, suggesting that they may be in a different ink. G has no corrections and is the most corrupt of the ten manuscripts.

156

Introduction to the text

Description of Manuscripts

MS H (32 x 12.5 cms., 9 lines per side) This is an undated manuscript - as with MS F, there is no colophon - written in Devanagari script, and of reasonable legibility. Fol. 39 has probably been rewritten; although it is in the hand of the orginal scribe the recto has eleven lines and the verso twelve, in contrast with the usual nine lines per side. On the verso the text finishes before the end of line twelve and is followed by connecting marks that run to the end of the line, indicating that there is no break with the text at the beginning of the next folio. From fol. 40 the verso is also numbered with a small figure in the bottom right hand corner. This second pagination starts at one (for fol. 40) and continues until the end of the manuscript. The main numbering also continues. The hand of the manuscript does not appear to change. H is the mostly heavily corrected of all the manuscripts, containing numerous marginal and interlinear corrections and corrections filling many of the lacunae marked in the original. There is one closely written nine-and-a-half line marginal correction occupying the upper and lower margins of fol. 75r. There are also some erasures. A few of the corrections are in the hand of the original scribe but most are in a second hand. MS J (27.9 x 11.4 cms., 7 lines per side) This is a clear and legible manuscript written in Devanagari script. Four folios are missing (fol. 1, 2, 29, 121) including the first and last. Any colophon is therefore lacking. The title - "aryanamasamgititlka (namamantrarthavalokimj" - has been written in Devanagari in a second hand in the margin above the first surviving leaf (fol. 3r). This manuscript is unusual in that it inserts all the verses of the Namasamgiti into the text, each generally preceded by a translation into Newari of the previous verse. These extra passages are distinguished from the rest of the text by being written in a slightly larger script and possibly in a different ink. On the microfilm they come out darker than the surrounding text. That these Sanskrit Namasamgiti verses are

157

Introduction to the text

Description of Manuscripts

insertions is also evident from the fact that they are present even when the text itself cites the whole Namasamgiti verse, resulting in a double citation. With the exception of one marginal correction in the hand of the scribe (on fol. 9r) it has no marginal or interlinear corrections. MS K (31 x 16 cms., 10 lines per side) This is a clear and and fairly legible Devanagari manuscript dated 'samvat 1954'. If this is the Vikrama Sam vat calender, it is equivalent to 1897 CE. It is almost uncorrected; it has a number of small interlinear corrections which, excepting two in a second hand, appear to be in the hand of the original scribe. The punctuation is with double dandas only, which may be in different ink to the rest of the text since they appear somewhat lighter on the microfilm. 2.5 Orthography The present manuscripts display a range of orthographic features, listed below, a number of which are peculiar to manuscripts of Nepalese origin. 18 Many of these occur frequently but they have not generally been incorporated into the apparatus since to do so would be to burden it unnecessarily. However, if there is some doubt concerning the sense or syntax of a passage then they are recorded; in some cases they are noted in the insignificant variants. 1) Homorganic nasal for anusvara. This feature occurs most notably as a final, particularly in MS A: eg. (1.1) dryamanjusriyan natvd rather than aryamanjusriyam natvd\ (1.2) ndmasamgitin gambhlroddra- rather than ndmasamgrtim gambhiroddra-\ (\.5)jdtakan cetivrttan ca rather thanjdtakam cetivntam ca. 2) Anusvara for homorganic nasal. This is found generally, though most widely in the more recent MSS: eg. (1.1) -mamjusriyam (CH) for -manjusriyam (ADEF); (1.2) gambhiro- (CD) for gambhiro- (AEFH); (1.17) pamca- (C) for panca-

18For discussion of the possible origins of some of these orthographic features see John Brough, (1954) 'The language of the Buddhist Sanskrit Texts.' BSOAS, 16, 25-75; and also Constantin Regamey, (1954) 'Randbemerkungen zur Sprache und Textuberliferung des Karandavyuha'/i5/a?/ca, Festschrift Friedrich Weller, Leipzig, pp. 514-527. (I am grateful to Kate Crosby whose unpublished English translation of Regamey's paper I have consulted.)

158

Introduction to the text

Description of Manuscripts

(ADEFH); (1.77) -amjaliputo (CH) for -anjaliputo (ADEF); (5.298) amkato (CDJ) for ankato (ABEFH); (5.300) namamamtra- (CDEH) for ndmamantrd- (ABFJ). 3) Anusvara in addition to nasal. Eg. Bj has (1.169) idamm for idam; (2.24) lokanamm for lokanam; (3.18) cagryamm for cdgryam; (4.25) bodhicittamm for bodhicittam; (4.141) nirupammd for nirupama; and also (1.155) anusasamnat for anusasanat; (2.4) -dnamntaram for -dnantaram; (2.58) -dmnvitah for -anvitah. 4) Gemination of consonants following the letter r. This feature is found to a greater or lesser extent in all the MSS. It is most common in those in Newari script. Eg. (1.1) aiyya- (D) -dharminim (CDE); -arddha- (codd.) as sarvavamndndm (D) varjito (5).

as against drya- (Aye); (1.2) -dharmmimm (Ay) as against (1.10) -drddrena (A) as against -drdrena (5H2); (1.44) against -ardha-\ (5.8) sarvvavarnndndm (BF) as against and sarvavarndndm (HJCE); (5.26) varjjito (By) as against

5) Parasitic superscript r with double consonants. Eg. (2.7) nirnamaiyyopa- (H) for nirnamayyopa-; (2.36) -maryydm (A) for -mayydm; (4.245) -nisarnnam (B) for 6) Confusion of sibilants. The most common form is between palatal and dental s, both s for s, and s for ^ being regularly attested: eg. (2.26) -sasanam (HCEF) for -sdsanam (AJD); (5.147) kuslda (5) for kuslda (ABF); (5.151) sakto (p) for jaJtto (AB); also (1.17) anusamsa (CAE) for anusamsa 19 (DH2); (2.12) -sodhanam (FHC) for -sodhanam (JDE); (2.15) rasmi- (FHC) for roj/wi- (IDE); (5.217) -dsaydnusaya- (|3) for -dsaydnusaya- (AB); and (2.27) anusdsanam (FHC) and anusdsanam (E) for anusdsanam (AJD). Less commonly, retroflex and palatal s are confused: eg. (5.85) -marsdt (Bye) for -marsat (A). 7) Identity of v and b. Since the Newari and Devanagari scripts of the present (and indeed all Nepalese Buddhist) MSS do not distinguish the letter b from the letter v, context has to determine which is intended. 8) tva used for ttva. This spelling is ubiquitous, with all the manuscripts almost invariably spelling words such as sattva, tattva, and mahattva as satva, tatva, mahatva: eg. (5.189) mahatvam (BFHECD) for mahattvam (J). In all cases I have regularised the spelling.

19Edgerton notes that Buddhist Hybrid MSS often have anusams- for anusams- (BHSD 34, s.v. anusamsa). As in the present case these are likely to be examples of sibilant confusion rather than instances of an alternative spelling.

159

Introduction to the text

Description of Manuscripts

9)y- used for y-. I have noticed only one instance of this usage, in the text of chapter six collated for establishing the stemma: (6.40) A has -jogena for -yogena (Bp). 10) / used for r. This occurs occasionally: eg. (3.11) vailocana- (E) for vairocana- (AyCC); (3.19) vailocana- (E) for vairocana- (Av£C); (5.229) virydlambho (5) for virydrambho (Ay). 11) Confusion of n and n. This is not very common and is mainly witnessed in the more recent manuscripts where n is found for n in instrumental singular forms such as (1.16) -rupena for -rupena. Occasionally it appears in ligatures: (1.60) gaunya (CDEFJ) for gaunya (AH2); (5.101) grhna- (Bv£C) for grhna- (E). More rarely n is found for n: (5.231) ghanah (CDF) for ghanah (ABHE). 12) Confusion ofsth and sth. As with confusion of n and «, this is most apparent in the more recent MSS: eg. B] has (1.66) naisthikam for naisthikam; (4.154) pratisthitah for pratisthitah\ as well as (4.57) prasthdna- for prasthdna-; and (4.111) avastham for avasthitam. 13) Use of the avagraha. The MSS do not show a consistent usage, the avagraha being often omitted. I have regularised the spelling so that readings such as sopi are transcribed as so 'pi. The avagraha is sometimes found - again the use is not consistent - within compounds, indicating loss of an initial a through sandhi. This usage is only recorded in the apparatus if there is some doubt about meaning. For example, at 2.62, A reads md'dhomukha-, whereas |3 has sadhomukha(misreading ma- as sd-); and at 4.54, |3 reads -kriydvirahito and A (erroneously) has -kriyd 'virahto. 14) Occasional irregularities of sandhi. If all MSS fail to show the correct sandhi this is noted in the apparatus, when corrected. If one or two MSS omit sandhi (and this occurs most commonly with final -ah and -h) this is not recorded. On occasion sandhi is omitted because of differing punctuation between MSS. Again, such omissions are not recorded, unless the punctuation adopted in the edition is attested by none of the MSS. Sandhi preceeding iti is usually preserved, though not consistently; differences between MSS are not generally recorded. 15) -m in pausa. The manuscripts show no consistency in their usage here, at some points using an anusvdra, at others a final m with a virdma. I have not recorded differences either in the apparatus or in the insignificant variants.

160

Introduction to the text

Description of Manuscripts

2.6 Other Features of the Script

Apart from these orthographic traits, the manuscripts share a number of other characteristics:- particular misreadings of aksaras due to their similarity in shape; 20 loss and addition of anusvara; and loss and addition of visarga. These, along with elisions of single aksaras, dittographs, haplographs, metatheses etc., I have recorded, if only in the insignificant variants. Recording errors such as these may not directly help restore the text; nonetheless they are useful for establishing the stemma, which in its turn aids the resolution of cruces; and if other manuscripts become available they are invaluable evidence in the work of investigating new stemmatic relationships. Where the manuscripts disagree over the presence of an anusvara or visarga it is not always a straightforward matter to decide whether one is dealing with an omission or an addition. Prose syntax is not always clear, particularly where abstract reasons are given in the form of a series of compounds. And although loss of an anusvara or visarga occurs most frequently particularly in the most recent manuscripts - on occasion their addition is seen: eg. (5.287) purusah (B) for purusa- (a); (5.290) anuttaram (H5) for anuttara- (ABF); (5.291) -dharmam (DH) for -dharma- (ABFCEJ). I have therefore preserved all the readings to allow for the possibility of future revision of the text. Where there is no doubt over the preferred reading the variants are placed with the insignificant variants. 2.7 Punctuation

At points the punctuation of the manuscripts varies considerably, and I have generally not included these since to do so would be to overburden the apparatus. Exceptions arise if there is an issue of sense, or when the punctuation I have adopted

20Aksaras easily confused in Newari script include sa I ma\ ta I bha; pa I ya; ca I va; na I ra; and ga I sa. Also the ligatures ddha and dva, and the vowels a I e, a I o, and u I u can be almost indistinguishable. An example of a misreading of m and bh in ligature is provided at Text 5.229 where B has viryarastd for vlryarambho (Ay).

161

Introduction to the text

Description of Manuscripts

differs from that of the available manuscripts and requires a change in sandhi. My underlying assumption is that punctuation is not part of the textual transmission. As a result I have felt free to regularise it where necessary so that it accords with syntax, for the most part using double dandas for sentence breaks and single dandas elsewhere. One 'punctuation' error that occurs on a number of occasions and should be noted is the misreading of a as a danda; so, for example, (5.197) B misreads -svabhdvatvat as -svabha / vatvat. Sometimes dandas can be inserted in the middle of word: eg. (4.41) BI has a / ndtmanas for andtmdnas. Generally, the punctuation in A indicates structure and semantics with by far the greatest accuracy, and it is thus when A is unavailable that I have had to do most regularisation.21

21 The punctuation generally deteriorates as one goes further along the transmission as reflected by the stemma. For example, in GK the distinction between single and double dandas, preserved in C, has been lost, all the punctuation being in double dandas.

162

Method of Editing

Introduction to the text

3. Method of Editing 3.1 Emendation There are a number of questions as to the degree to which an editor may legiti­ mately emend a text. This is not the place for a full discussion of them but I should clarify my position as regards the present edition. I take it to be the aim of an editor to establish the oldest form of the text that is accessible. I do not subscribe to a view of "minimal interference" if by this is meant an edition that preserves the idiosyncrasies of a particular scribe or script along with regional peculiarities of orthography. This would be to identify the text with what is likely to be a comparatively late layer of its transmission.22 The alternative, of attempting to unravel the different layers of textual transmission, can bristle with difficulties, sometimes to the extent that little progress is feasible. Nevertheless, this aim devolves upon the editor a duty to correct, emend and suggest possible solutions to cruces in the transmission as received. Three factors in particular can hinder the editor's task so construed: whether one is dealing with a single prototypical text or with a number of recensions; uncertainty concerning the language of composition of the text; and frequently, in the case of Buddhist texts, reliance on recent Nepalese manuscripts that are often very corrupt. The problem of multiple recensions is encountered in acute form in Buddhist Mahayana Sutras, which either evolved over a period of time or existed in different recensions from the beginning.23 This issue is less likely to be encountered in the

22In some cases, when for instance there survives only one manuscript of a text, there may be little alternative to adopting this approach to some degree. However, this can result in a transcription rather than an edition, with little or no attempt being made at fulfilling the critical obligations of an editor. Worse than this, in consideration of its use for further scholarship, is an edition that emends or adopts readings from other manuscripts without so stating. See, for example, Schopen's remarks on Dutt's edition of the Bhaisajyaguru Sutra (Schopen, 1977, 208-210). 23The various Chinese translations of Mahayana sutras are certainly evidence of different recensions. However, it does not necessarily follow, as is often assumed, that a later translation represents a later recension. Regamey, for instance, concludes that the Gilgit and Nepalese versions of the prose Kdrandavyuhasutra represent different recensions that must have existed from the earliest periods of the text.

163

Introduction to the text

Method of Editing

field of sastra since one is dealing with texts ascribed to a specific author and the difficulties often shift into the sphere of attribution. Varying degrees of opportunity for interpolation and expansion do exist, however, with the work of more renowned writers being especially susceptible.24 In the present case there is no reason to doubt that we are dealing with one text without recensions, and there is no sign of any significant interference of the sort just described; its style and content remain consistent. Secondly, uncertainty with respect to the language of the original inevitably makes the proposition of emendations uncertain. Material written in Prakrit or the (Buddhist Hybrid) Sanskrit of the Mahavastu may have suffered different types of Sanskritization - intentional, and that resulting from copyists 'corrections' - that can be extremely hard to disentangle. Again it is largely the texts of the (Mahayana) sutras and tantras that are implicated here. Vilasavajra's Ndmasamgititikd appears to have been written in chaste classical Sanskrit. Apart from the citations, the text of the present edition yields evidence only once of a non-classical form (the thematised form yoginah for yogi: see 4.60 and textual note). As for sandhi, although the readings of some of the more recent manuscripts (6) might suggest irregularities, those of the others, especially the palm-leaf (A), generally preserve it. The internal evidence - our author's careful commentarial style and range of learning - also suggests that as a scholar Vilasavajra was likely to have written in classical Sanskrit: one who cites Panini's Astadhyayi and Patanjali's Mahdbhdsya, as well as the Dhdtupdtha, is unlikely to have done otherwise. The present text is also relatively unaffected by the third obstructive factor, reliance on recent, often very corrupt, Nepalese manuscripts. Although we have only

24&antideva's Bodhicaryavatara is one text that has almost certainly suffered in this way: an early Tibetan translation preserved at Dun-huang contains 2lQy2 less verses than the later version that is found in the bKa' 'gyur, which latter corresponds with the Sanskrit text as we have it (see Saito, 1993).

164

Introduction to the text

Method of Editing

one palm-leaf manuscript of the NMAA (A), time and again it preserves good readings lost elsewhere, or corroborated only in the undated manuscript B, which stands out beside A in its value as a witness.25 If the edition had to be based on the more recent manuscripts (CEDJKG) it would be considerably less reliable.26 Hence, the task of editing the NMAA is more straightforward than is sometimes the case for Buddhist texts. Nonetheless, being a prose rather than a verse text, it lacks the regulation of metre and the punctuation that indicates each half-verse; it is thus more liable to suffer from scribal errors of omission. 3.2 Types of Emendation In the apparatus I distinguish three types of emendation using the abbreviations con:, em., and conj.. This attempts to differentiate levels of seriousness of emendation. These levels are not discrete categories so much as points on a continuum representing the degree of certainty with which an emendation can be maintained. The first level (corr.) is used for correcting a reading that would be fully attested by the manuscripts were it not that it contained an insignificant error, such as a misreading or elision of a single aksara. Its use signifies that I judge there exists a minimal degree of doubt as to its being the correct reading. The second level (em.) is used where the degree of doubt is greater but the emendation is not completely conjectural since the reading is in part attested and a path of corruption is discernible. This type of emendation typically occurs when a word is attested but the syntax of the passage requires a different inflection. The third level (conj.) is reserved for

25Features of its (Newari) script suggests that B is probably from the eighteenth century or earlier. Other valuable manuscripts are F and the contaminated H (both undated). 26CEKDJG are shown below (see section 4.2, Evidence for the Hyparchetypes) to descend from one hyparchetype (6). They contain considerable errors, particularly omissions, that it would not be possible to restore. Only CEK are dated, C (1800 CE) being approximately one hundred years earlier than EK. Examination of their script and physical condition (as far as can be made out from the microfilms) renders it unlikely that DJ are older than C. G, the remaining manuscript of this group, is an apograph of K, and must therefore have been copied this century.

165

Method of Editing

Introduction to the text

emendations that are diagnostic, and where there is little or no support from the manuscripts. Some emendations of this type are purely conjectural, representing a possible solution to a crux, as for example when sense or syntax indicates that the text contains an omission. Others have some support from the manuscripts, and sometimes it is possible to see how the present state of the transmission could have developed. Nonetheless, the degree of doubt for such emendations is high, and for the majority there is some comment in the textual notes. 3.3 Use of the Tibetan Translation In the preparation of the present edition I have made selective use of the Tibetan translation of the NMAA (Tib.), consulting it whenever there is a crux in the Sanskrit text or when the manuscripts present alternative readings that seem equally viable. I have not accorded it any privileged status but have treated its readings as those of a further witness, assessing them in that light. The evidence of Tib. is thus incorporated into the apparatus, where its alliances with the Sanskrit variants are stated and its readings cited, as appropriate. In the textual notes, when a wider context is needed for a proper assessment of its significance, Tib. is cited more extensively. My aim in using the Tibetan translation has been essentially one of improving the Sanskrit edition, and I have not therefore made an exhaustive comparison with the Sanskrit text; for example, I have not attempted to list every place where it differs from the Sanskrit. Despite this restricted use I have recorded sufficient readings to make an attempt at relating it to the stemma of the Sanskrit manuscripts worthwhile.27 Reading the Tibetan translation reveals a number of characteristics that are are relevant to an assessment of its value as a witness. Discussion of these is deferred to section 5 of the present chapter; it suffices for now to note that some of them occasional mistranslations, interpretive translations, and rearrangements of the Sanskrit - suggest that Tib. should be treated with caution. Also, errors shared with 27See section 4.4 below, The Tibetan Translation & the Stemma Codicum.

166

Introduction to the text

Method of Editing

some of the Sanskrit manuscripts point to the Tibetan translation being based, at least in part, upon a corrupt exemplar. The role of Tibetan translations in the editing of Buddhist texts (Tantric and otherwise) has been the subject of a continuing debate, triggered in large part by David Snellgrove's remarks in the introduction to his edition of the Hevajra Tantra. My own contribution to the discussion is best deferred until after my remarks on some of the characteristics of the Tibetan translation of the NMAA. 3.4 Positive Apparatus For the presentation of the variant readings I have adopted a positive apparatus, that is to say, the first variant in each entry comprises the accepted reading followed by the siglum (or sigla) of the manuscript(s) that attest it or, if it is an emendation, by the appropriate abbreviation. The advantage in this approach is that one can see at a glance which manuscripts support the preferred reading; there is no need to work them out by a process of deduction (ie. by seeing which are not mentioned as having variant readings). Furthermore, the ambiguities that can arise with a negative apparatus - when it may not be clear what a variant is a variant of - are eliminated; the readings that follow the initial accepted reading are the variants of that reading. The possibility of such ambiguity is perhaps most evident when recording variants in compounds. The methods of dealing with the problem used by those who adopt a negative apparatus placing the footnote mark within the compound after the word concerned, or citing more of the surrounding text - are unsatisfactory when confronted with a compound that has a number of elements each of which may have variants. In the present edition I have placed the footnote mark at the end of the word for which there are variants, whether that word is a compound word or not. This allows a smoother reading of the text than inserting them midway. If, as in a compound, it is only part of the whole word that has variants, the positive reading in the apparatus is

167

Method of Editing

Introduction to the text

cited without dashes or other marks that are usually indicative of it being part of a compound. Such marks are unnecessary since the purpose of the positive reading is to indicate the preferred reading where there are variants; reference to the edited text shows whether the whole or part of a word is referred to. Any of these marks appended to the positive reading could reinrroduce the possibility of ambiguity, of uncertainty as to what the variants are variants of. When dashes are used in variants following the positive reading, they refer back to the positive reading as their starting point. If more than one element in a word has variants then the entries for each are found at the same footnote in the apparatus but separated by a diamond (4). The present apparatus has three registers: a register of testimonia, not always present; a register of codices; and a register of variant readings. The register of codices notes any manuscripts that are lacking (and for which lines when necessary) and summarises the resultant stemma. When a manuscript noted as lacking is unavailable because of lost folios, its unavailablity is not also noted in each entry in the register of variants. Omissions of three or more lines are noted in the register of codices, but in such cases the omission is generally noted in the register of variants as well. 3.5 Insignificant Variants To reduce the burden on the apparatus I have removed insignificant variants, placing them at the end of the edition (after adhikara 5) where they are crossreferenced to the text by chapter and line number. Variants accounted insignificant include misreadings, elisions, haplographs and dittographs of single aksaras. However, if such an error is attested by all the available manuscripts it is included in the main apparatus, and the emendation noted. This is because I have followed a rule of including all emendations to the text, however small, in the register of variants. In general, I have erred on the side of caution in counting a variant as insignificant. If a preferred reading is only witnessed by one manuscript, though the others possess

168

Method of Editing

Introduction to the text

only insignificant variants as just defined it is still included in the apparatus. This is partly to show the slight support that the preferred text sometimes has, but there is also the consideration that whether or not a variant is insignificant depends on a correct understanding of the text. In passages where the sense is complex or doubtful, the presence or absence of an anusvdra or visarga can produce meanings of apparently equal viability; thus, what may be an insignificant variant in one part of the text may not be so elsewhere. Larger errors, such as the corruption of more than one aksara or the omission of a word or phrase, are also included as insignificant errors if they are the error of just one manuscript or of (the descendents of) one hyparchetype. Such errors are included only when stemmatic theory indicates that they could not be the reading of the archetype. 4. A Stemma Codicum for Vilasavajra's Namamantrarthavalokini 4.1 Introduction

For the purpose of constructing a stemma I initially collated readings of the (ten) MSS available to me for a passage of some ninety lines in the sixth chapter of the NMAA. In some of the manuscripts this passage contained a number of marked lacunae that I hoped would provide clear evidence concerning transmission of shared errors. I intended to include an edition and translation of the whole of chapter six in the present work but it subsequently became apparent that the chapter was too long for this to be possible. The text and variant readings of the collated passage are included at the end of the edition, after the insignificant variants. The variant readings from the passage provide strong evidence for the existence of hyparchetypes a, {3, y and 5. The relationship between the manuscripts of 5 is less clear. There is also evidence of contamination in some manuscripts, including those from 5, and it is sometimes not easy to distinguish instances of contamination from those of counter-examples to proposed sub-hyparchetypes. Nonetheless, K and G are

169

Introduction to the text

Stemma Codicum

clearly apographs of C. They have all the errors of C as well as some of their own. Further, G has all the errors of K and also some of its own, so in turn is an apograph of K. Since C is complete, in good physical condition and clearly microfilmed by the NGMPP I have not collated KG for chapters 1-5. The present edition of these chapters is thus based on the eight remaining manuscripts, ABCDEFHJ. Subsequent collation confirmed the existence of hyparchetypes a, (3, y and 6, and suggested the existence of £ and £, two sub-hyparchetypes of 5. As a result I have adopted them as hyparchetypes in the apparatus and textual notes, although their status as such remains rather tentative. Further collation also made it apparent that the first thirteen folios of B (= BI) - which replace a lost portion of the original manuscript - are also an apograph of C and so readings of BI are not included in the apparatus.28 Since K and G had previously been eliminated as apographs of C, the relation between BI and KG was deduced by examining a number of key passages in chapter 1 rather than by a full collation of variants. Nevertheless, the evidence is clear enough to show that BI and KG form separate transmissions from C. There follows a summary of the evidence for the stemma and then a discussion of contamination. All the variants in the proposed examples of contamination are given in the format used in the apparatus for the text. Thus the first reading is always the positive or preferred reading and is followed by the variants of the remaining manuscripts. A variant followed by a hyparchetype siglum means that the reading is attested by all the manuscripts available at that point which are the descendents of that hyparchetype. The hyparchetype sigla are not used in the apparatus to denote a reading inferable as that of the hyparchetype (as a result of examining the readings of its descendants). In other words, "indu y : indra 6" means that indu is attested by

28A close comparison of the readings of BI and C was possible since my initial draft of chapters 1-4 used manuscripts A and B, the only ones available to me at the time. Since chapter four ends on fol. 15r in B, I had collated all of the first thirteen folios, ie. all of BI.

170

Introduction to the text

Stemma Codicum

FH and indra by CDEJKG. 29 The existence of the hyparchetypes is not thereby presumed by the use of their sigla.

a

Fig. 2. Stemma Codicum of the Sanskrit Manuscripts of the NMAA. 4.2 Evidence for the hyparchetypes i) ABCDEFGHKJ (Q = all available MSS) ABCDEFGHKJ share significant errors indicating that they descend from a common archetype, (Q). For example, (4.201) they interpolate idam (CDJ have itam; E itah) before idanim. They read (4.207) astasu cdstau for (conj.) astdsv astau, (5.153) indrajdlikah for (conj.) mdyendrajdlikah, (5.154) indrajdlikah for (conj.) mayendrajalikas, (5.159-60) karmeti for (conj.) krama iti or karmakrama iti, and (5.287) pungalah (JDC have pumgalah) for (conj.) pumgavah. ii) ACDEFGHJK (a) and B descend independently from Q,. ACDEFGHJK (a) share errors not transmitted by B. For example they read (6.2) pravrttds for pravibhdvitds, (6.4) jdtasamgam for sangajdtam (conj. :

29In examples from chapters 1-5 the siglum 6 stands for manuscripts CDEJ since KG were only collated for the passage from chapter 6.

171

Introduction to the text

Stemma Codicum

sarig*i*jdtam B), (6.13) paramitd for na ddnapdramitd, and (5.287) paramdrthdfor paramdnha-. They omit (6.43) taj (Dc has ta\ and (5.239) mahdkrurah. ACDEFGHJK are not descendants of B since B has errors of its own. For example, it omits (6.10) ddnaparamita, (6.59) rupdrupyapratisamyuktdndm samskdrdndm hetau yad andsravam jndnam, (6.64) anvayajndne, (6.46) bhumau, (6.61) samyuktdndm, and (6.80) mdrgam. It reads (6.46) bhumi for vasitva, and (6.19) viryadi for dhiryddi. iip CDEFGHJK (B) and A descend independently from a (ACDEFGHJK). CDEFGHJK share errors not transmitted by A (or B). For example, they omit (6.10) paramitd, (6.76) jndnam / tat samudayajndnam, and (6.87-9) parijheyam iti jdndti samudayah prahdno na punah prahdtavyah / nirodhah sdksdtkrto na punah sdksdtkartavyah. They share a long interpolation (6.70): prajnd pari(parlGK}jneyam iti jdndti / samudaya prahlno na punah prahdtavya(-tavyah FH) nirodhasdksdt(-sdksd GK) tvatd(*tr*to F; vattd H) na punah sdksdtkartavyah (sdksdt om. HF; sdksdt H2) / samvrtijndnam ceti. They read (6.9) karmavibhakta for karmavibhakteh, (6.16) etena for ekena, (6.40) samddhdna for gunddhdra, (6.63) dvayajndnesu for anvayajndne, (6.63) jndnesu bhumau (GK havejndnisu bhumau) for jndnabhumau, and (6.71) duhkha for duhkhatah. Since A has errors of its own CDEFGHJK are not descendants of A. For example, A (6.35) reads verma- for dharma-, (6.54-5) omits prahdndya, (6.84) interpolates na punar bhdvayitavya iti. It (1.20) omits one and a half lines, atha -* prativacanam. iv^ FH (Y) and DJECKG (5) descend independently from B (CDEFGHJK) FH share errors not transmitted by DJECKG. For example, they omit two lines (6.61-3), rupdrupyapratisamyuktdndm samskdrdndm hetau -»• nirodhe yad andsravam. They share marked lacunae, reading (6.56-60) ucyate < -15 > pratisamyuktesu; and (6.74-6) samudayajndnam < ~48 > nirodhajndnam katamat. They read (6.38) jndnabhdvi< > for jndnasydvasthd, (6.41) sdmanyatvdd (against 6's sdmdnyatvdd) for sdmdnyalaksanatvdd, (6.51) < 3 >su for samskdresu, (6.52) < > (against 6's heto) for hetau, (6.67) < >hitdya for samavahitdn, (6.67) sammukhK > for sammukhibhutdn, (6.67) < > bhavanti paresdm for paresdm, (6.85) < > Y (vidha H2) for vidyd, and (6.83) patijndnam (against &sparijndnam) for parijndtam.

172

Introduction to the text

Stemma Codicum

DJECKG (6) share errors not transmitted by FH. By way of example DJECKG read (6.7) atiti for etfti, (6.38) srdvakdbhir for srdvakddibhir, (6.66) avihina for avihmam (H2 has avihina), (6.78) nisaranata for nihsaranatas ca, (6.80) mdrga for margato (GK have mdga), (6.85) upayasa for updddya, and (ibid.) tvana for jnanam (FH

v) F and H have their own errors, and so descend independently from v F has errors of its own not transmitted by H. For example, it omits five lines (6.43-8), [nairajtmyavisesat -* pratisthita iti // [dasa jhandni], and reads (6.23) -arthara for -drthah, (6.72) manasa for manasi, and (6.66) < >gamye for margasya (which is attested by AB against magamya by DJECKGH, the rest of p). H has errors of its own not transmitted by F. For example, it omits (6.27-8) dasaiva pdramitd nayo mdrga updyo yasydsau sa tathoktah. It has dittographies: (5.61) of vairocanam dhiydm mohah after dhiydm moho, (5.71-2) of sunyatddicaturvimoksamukhdndm dharmadhdtvdyatanamukhena sarvasamddhihetutvdt. It reads (6.1 ) -pravrtya for -pravrttds (attested by AB against -pravrttd by DJECKGF, the rest of p), (6.3) ddram for ddnam, (6.17) tvagund for 'ndhaganah (attested by AB against tvagand by DJECKGF, the rest of p), and (6.45) pramuditdhindm for pramudilddmdm.

vi) DJ (t ) and ECKG(B 1 )30 (e) descend independently from 5 DJ share errors not transmitted by ECKG. For example, DJ read (6.18) ihdksihin for ihd(ahd Y)ksihindn, (1 .64) prativedhikam yd for prativedhikayd, (1.73) bhagavdn sdkya for bhagavacchdkya, omit (1.110) protphullam (from the extant MSS, ie. those of p), read (1.131) rund - in contrast to FH's ruviti - for runaddhiti (EC). They read (5.85) paramdrthdt for pardmarsdt (EC having paramarsdt). There is no unambiguous support for ECKG deriving from a single hyparchetype in the portion of chapter six collated. That KG are direct descendents of C is clear (see below). Readings from chapters 1-4 also show that B] is an apograph of C (see below). However, evidence of shared errors in chapters 1-5 suggests that E and C, and thereby ECKGB], are related.

30Bi is placed in brackets since it is only extant for chapters 1-3 and part of 4.

173

Introduction to the text

Stemma Codicum

EC share errors not transmitted by DJ. For example, (1.79) EC read gdmbhirai for gdmbhirya, (1.147) -rupinir for -rupibhir, (1.148) -karar for -karair, (1.151) vaksyamdnamdna for vaksyamdnam, (1.191) sdsya- for sdsva-, (1.195) -sattvajna for -tattvajna, (1.251) adva- for tryadhva-, (4.122) ta yasya for yasya, and omit (5.62) mudha (reading dhlmohah for mudhadhimohah). They have close errors. For example (4.198) C reads jramani, and E vajramani, forjanmani; (5.68) E reads -moheti, and C -moha iti, for lobha iti. And both (5.68) read -moho for -lobho, and (5.138) -dropetdt for -dropandd. vii) D and J have their own errors, and so neither is the exemplar for the other. J has its own errors. For example J reads (6.17) samasta for samasto, (6.19) ddayah in contrast to the rest of (3 1 s ddaya for dddya (AB), yasya for yasydsau (H2 has yasya), (6.72) andsrava for andsravam. J interpolates (5.69), reading sarvamukhena sarvasamddhihetutvdt sarvalobhanisudana iti for sarvalobha­ nisudana iti, D has its own errors. For example D reads (6.7) sarvdm for sarvd-, (6.46) ddnddind (against the rest of S's ddnddindm) for ddnddi-, (6.77) pranimato for pranitato, (6.6) jndnodayakramena for kramena (but CGK have kramena jndnodaya), (5.45) am//i / (against ECJ's amita /) for amitd-, (5.71) dhdtvdyamana for dhdtvdyatana, (5.239-40) taryakdndm (in contrast to the rest of |3's taryantdndm) for durddntdndm. viii) E and CKG have errors of their own and so descend independently from £. E has errors of its own. For example it reads (6.7) kimapi (against the rest of (3's kipi) for kvipi, (6.8) tadabhirupe for tatpuruse, (6.36) -sthdnatato for -sthdnato, (6.42) dtmasatd for dtmasamatdm, and omits (6.84) me. In chapter four E omits one and a half lines (4.51-2). CKG have errors of their own, and C has no errors of its own that are not shared by KG. For example CKG read (6.6) kramena jndnodaya for kramena (D has jndnodayakramena), (6.47) pratitdt (against the rest of |3's pratisthitdt) for pratisthdndt, (6.53-4) yandsravam for yad andsravam, and (6.81) nairyatimanasi for nairydnikatas ca manasi. ix) KG are apographs of C. KG have all the errors of C and are never correct against C. They have errors of their own, and when C is in error their own errors can be seen to compound those of

174

Stemma Codicum

Introduction to the text

C. The additional errors of KG are minor, amounting to the loss, addition or alteration of one or two syllables, but they are sufficient in number and specific enough to make the relationship to C clear. For example, KG read (6.6) -vabhana- for -bhavana-, (6.8) alaki for aluki, (6.51) svaskaresu for samskaresu, (6.56) jnanas for jnanam, (6.59) skarana for samskaranam, (6.63) anvrya- for anvaya-, and (6.91) dam for idam. Error is compounded by KG, for example: (6.62) AB have (correctly) api, H2FCEDJ have atha, GK have ya atha; (6.80) AB have (correctly) mdrgato, CDEJ have marga, and KG mdga; (6.84) AB have (correctly) me bhdvita, CDJFH have ma bhdvita, and KG mama bhdvitadu; (6.68) A has (correctly) canasravan, CDEHJ have cdnydmscavdn, and KG have cdnydmscavdn. KG compound errors that C has alone. For example, (6.67) KG have sam // 1 // ndt (for sammukhibhutdn) where C has samindt, misreading C's T for a '!'. On the next line (6.68) KG have dhamam jindti for dharmdn jdndti, where C has dharmdm jdndti. But C places the anusvdra of dharmdm before the superscript V, which latter is above the vertical stroke of the 'o'. The result is that C appears to read dhamam jidndti, which could easily be copied as dhamam jindti. x) G is an apograph of K KG have almost identical readings, but G has all the errors of K plus some of its own suggesting that G is an apograph of K. For example G omits (6.49) duhkhajndnam, and reads (6.14) paramitdndm for pdramitdndm, (6.55) andrthavam for andsravam, (6.74) hetuta for hetutah. G compounds error. For example (6.2) ABE correctly have madhyamakd-, CCDFHJ have madhyamakd, CK have mddhyakdma and G has mddhekdma. xi) Bi is an apograph of C B] is the first 13 fol. of B. These folios contain chapters 1-4 and part of 5. The examples that follow are taken from chapter 1. B! has all the errors of C. For example, B^ alone read (1.4) pidyakam for pitakam, (1.10) karundrdrana for karundrdrena, (1.31) pdddna- for pddona-, (1.84) -pratidya- for -pratigha-. Bj also has its own errors. For example, it omits (1.175-6) mdydjdldbhisambodher iti / maydjdldbhisambodhir ndma kramas tasya yathdham Idbhl

175

Stemma Codicum

Introduction to the text

bhavdmi; and it reads (1.27) -vairocand for -vairocano, (1.60) paramdrthah samvrtydt for paramdrthatah samvrtya, (1.65) -vidvamyd for -viddhayd, (1.76) ndrtham for ndtham. There is a telling example of how B! (or one of its antecedents) has incorrectly incorporated a marginal correction found in C. At 1.93, C reads sarvasattvams tu na se pdpena lipyate. The right margin contains the words na sa dose • patha, and in the text two short horizontal lines above na and a short vertical stroke after pdpe indicate the syllables to be replaced by the 'correct' reading (patha). The 'corrected' reading is therefore sarvasattvams tu na sa dosena lipyate. The scribe of Bl (or one of its antecedents), however, did not notice the vertical stroke after pdpe and took the marginal correction to be an insertion. Also, he misunderstand the use of the term patha and took it too as part of the insertion. Thus he replaces C's na with na sa dose patha, giving the reading, sarvasattvdns tu na sa dose patha se pdpena lipyate\ A good illustration of the process of textual corruption. Fol. 1-4 of C are heavily corrected and in these folios Cc often follows 5, in contrast to Ca which follows Y- 31 For the text covered by C in these four folios B, follows Cc rather than Ca. For example B } CC read: (1.13) ddarsanajndnam bodhicittah where Ca has < ~7 >bodhicittah, (1.17) anusamsd mantravinydsah where Ca has an omission, (1.49) sambandhddiny where Ca has sambandhdny. xii) B, and K descend independently from C B! and K have their own errors indicating that they are independent copies of C. Bj incorporates - sometimes erroneously, as at 1.93 in the preceding note - all the corrections of C whereas K incorporates only some of them. This suggests:- i) that C was corrected more than once, and copied after one set of corrections, but before a second set, had been made; ii) that the copy leading to K was made before the one leading to B]. This partial incorporation of corrections by K in contast to B! also

31 See below for discussion on contamination in C.

176

Stemma Codicum

Introduction to the text

confirms the view, reached through consideration of the stemmatic affiliation of C's corrections, that C was corrected by more than one hand or one more than one occasion. 32 K follows Ca . For example: at 1.13, where KCa have < ~6 >bodhicittah whereas E } CC have ddarsanajndnam bodhicittah; at 1.83, where KCa have dharma whereas B 1? incorrectly treating Cc 's correction damyanta as an insertion, has dharma ddpyanta; and at 1.93, where KCa have sarvasattvdms tu na se papena lipyate whereas B, again incorrectly incorporating a marginal correction (see above), has sarvasattvdns tu na sa dose patha se papena lipyate. K incorporates some of C's corrections. For example, at 1.93, where Ca has sdrayan, E^CC have mdrayan, and K has mdsdrayan. (Here K incorrectly incorporates C's interlinear correction ma taking it as an interpolation before, rather than as a substitute for, sd). At 1.87-8, KB, both incorporate a longish marginal correction in C, silam asyeti trilokavijayi / ata eva viro bhagavdn vajradharah parair anabhibhavamyatvdt, but K adds its own errors, reading silan for silam, and -bhavatfyatvdt for -bhavamyatvdt. At 1.17, there is a more complex example. Ca has a lacuna of about seven aksaras for the reading anusamsd mantravinydsa. B^0 supply the correct reading but have -vinydsa for -vinydsa. K reads anusamsd mantra. K might have had the correction in its exemplar but have made an error of omission, dropping vlnydsa. Examination of C reveals that the lacuna is filled with the correction anusamsd mantra vi; the final part of the correction, nydsa, is placed in the margin. It is possible that vi (and nydsa) are later additions - in other words that C's lacuna was corrected twice; once with the addition of anusamsd mantra and then with the addition of vinydsa. So K's reading could be an accurate copy of its exemplar, since if it were to omit anything it would more likely to be the marginal nydsa, retaining anusamsd mantra vf. 33

32See discussion of contamination below. 33 This example, like that at 1.13 already cited, is relevant to discussion of contamination. (See following section.)

177

Introduction to the text

Stemma Codicum

In the above examples G follows K, and adds some minor errors of its own, which confirms the conclusion that it is an apograph of K. 4.3 Contamination A number of the manuscripts exhibit signs of contamination in chapters 1-5 and that part of chapter 6 collated for establishing a stemma. H and D each contain varying degrees of correction, these forming the source of their contamination. C, E and J show limited signs of contamination, and the evidence in the latter two is not decisive. C's first four folios contrast with the rest of the manuscript; the readings suggest that its exemplar may have changed from fol. 4v or 5r onwards. B shows no sign of contamination and, with the exception of one correction each, the same is true of A and F. Stemmatic theory can still usefully be applied in the evaluation of variants for this text after consideration of the level and nature of the contamination evidenced in these manuscripts. The contamination in HDC originates from corrections that can usually be identified as such; the negligible contamination in AF is of the same type; that in EJ, though incorporated into the text (except for the occasional correction in E), if it exists at all is of limited extent. In the following discussion of contamination in individual manuscripts, I give all the variants at each point in the form they are found in the apparatus, though for clarity of comparison I have removed any citations of the Tibetan translation. They are also prefaced by their text chapter and line number, and note is made of MSS lacking as a result of lost folios, eg. "(AB lacking)".34

34In the apparatus this information is obtained from the register of codices located above that of the variants.

178

Introduction to the text

Stemma Codicum

MS A shows no sign of contamination, with the exception of one correction in a second hand: 35 6.45

bhuminam ISvaro A2 : bhumi£varo ABDC : bhummamana< >s"vara CDEH2JKG : om. F : lac. H

The readings here are puzzling. The syntax of the passage makes it clear that bhummdm isvaro is the correct reading. According to the stemma bhumlsvaro should be the reading of Q. since it is attested by both AB as well as Dc . The correction of A must thus represent contamination from outside H. However, S's bhumlnamandsvara apparently has a remnant of the correct reading, implying that 5 is in its turn contaminated. Yet there is no other instance suggestive of contamination in 5. The problem would be solved if bhuminam isvaro is taken as the reading of O and both A and B allowed to have independently omitted the two syllables -nami- (ie. bhuml[ndm ijsvaro), enabling S's reading to be traced via ex. Such independently occurring omission seems unlikely, however. Readings of the later part of the manuscript - ie. chapters 7ff. - may cast more light. MS C does not generally show signs of contamination from outside |3. Two exceptions are the corrections damyanta (1.83) and pratlcchan (1.113). 1.83 1.113

damyanta Cc : dharmmataY •' dharmata £ECa (AB lacking) praticchan ECCC : pratitya celt. (AB lacking)

For the first four folios C may have had two exemplars, one related to y, the main exemplar used for the initial copy, and a second, related to 6, used to correct the first. Fol. 1-4, in contrast to the remainder of the manuscript, are heavily corrected, with some of the uncorrected readings showing agreement - often in error - with F and H (Y). The corrections are in at least two hands, which are not always easy to distinguish, for which reason I have adopted Cc for all corrections. In the instances cited above (1.83; 1.113) the hand is one almost certainly used only for a small 35A's position in the stemma and the question of contamination will have to be examined again for the later part of the manuscript where the hand changes and the number of corrections increases. See above, Description of Manuscripts, section 2.4.

179

Stemma Codicum

Introduction to the text

number of the corrections. Most of the corrections produce readings that agree with DEJ (5). Thus: 1.49

dheyamACav: dheyadikam DECC (BJ lacking)

1.52

avaSyam H2 : avaSyaA : s"yava £ECC : ava vCa (B lacking)

ivaksini £CC : iti aksim vCa : iti vaksin! E (AB lacking) 1.108 1.17-18 anuSamsa mantravinyasah ADE : < ~8 > Cay (anuSamsa mantra­ vinyasah suppl. H2; anuSamsa mantravinyasah suppl. Cc) (BJ lacking) 1.134 jagac ca jagac cajagac ca jaganti £ECC : jagac ca jaga Cay (AB lacking) arthakarair iti Cay : arthakaraih parair iti £ECC (AB lacking) 1.135 From fol. 5r onwards C agrees in significant error with EDJ rather than FH. (See, for example, 2.17, where C has pathyate as does EDJH2 as against y's pa< >.) It may be that at this point the scribe adopted as his main exemplar the manuscript he had been using for making corrections, abandoning the first. Two readings from fol. 1-4 suggest more specifically that C's initial exemplar may have been a manuscript between (3 and F: 1.7

lau(lo- DCc)kikam ca tatha Sastram guruparvakramagatam AEDH2CC : om. Y : < -18 >Ca (suppl. C6H2) (BJ lacking)

Here (1.7), C could not have F or H as its exemplar since it marks a lacuna for a half verse that they omit. 1.85-8

Silam asyeti -+ anabhibhavamyatvat om. FC (silam (y}asyeti -»• anabhibhavaniyatva suppl. F0 ; s"ilam asyeti trilokavijayi / ata eva viro bhagavan vajradharah parair anabhibhavaniyatvat suppl. Cc) 4 atha va jayakaramadhukarasarvarthasiddhikaras trayo bhrataras trilokah / tan vijetum Silam asyeti trilokavijayi om. EDJCCC (lacking AB)

C here (1.85-8) omits three lines also omitted by F but not by H, affiliating it with the line of transmission from |3 to F. F corrects the omission in the margin, as does C, but C's correction is not complete; it omits part of the passage that is also omitted, probably through a saut du meme au meme, by DJE. This example shows independently that C could have had as its exemplar neither H nor the corrected F. It also demonstrates how C was being corrected by 6 or one of its descendants. Affiliation with uncorrected F rather than H is also seen at 1.13.

180

Stemma Codicum

Introduction to the text

1.13

bodhicittam A : < -1 >bodhicittah FCa : < ~7 >bodhicittern H (adariajnanam bodhicittam H2) : adar&anajnanam bodhicittah ECC : om. D (BJ lacking)

The situation is even more complicated than I have suggested since C, uncorrected, does not always follow FH in fol. 1-4. The clearest example of this is at 1.10 where C does not follow FH in having a different half verse to the rest of the MSS. 1.10 1.56

sattvanam mandabuddhmam karunardrena cetasa A6H2 : sattvanam ca hitarthaya jnaninam moksapraptaye y (BJ lacking) dharmatavataranartharn 5 : dharniatadhvataranartham y : dharmatartha A (B lacking)

Furthermore, in fol. 4v, despite the examples to the contrary (see above, 1.134; 1.135), there are also readings where C follows 6 rather than y: 1.129 1.131

pravicayah conj. : pravicarya 5 : prasiddhaye y (AB lacking) runaddhiti £ : runa C : ruviti y (AB lacking)

It may be that fol. 4v is transitional, with both exemplars being used evenly. Alternatively, the change in exemplar may occur near the bottom of fol. 4r (where a change in ink, if not hand, is discernable) and the corrections to fol. 4v, fewer in number than in the preceding folios, form the tail end of a second and later strata of corrections. The adoption of differing quantities of the corrections by the various descendants of C, ie. KG and BI, supports such a theory. 36 MS D, like H, gives indications of contamination only where it has been corrected. In contrast to H, D has few corrections, possibly in the hand of the original scribe, their highest density occurring in the portion of the sixth chapter collated to establish the stemma. On a number of occassions Dc is correct against the rest of (3. 5.13

sartvadyaksarayoh Dc : sattvadyaksara D : sattvadya eyJ (AB lacking) 4 svarupabhidhanam Dc : paridhanam £ : < >petidhanam y : < >panidhanam C : pranidnanam E : om. B (A lacking)

36See discussion of variants at 1.13 & 1.17 in part xii ("B } and K descend independently from C"), section 4.2, above.

181

Introduction to the text

Stemma Codicum

Here (5.13) Dc provides a good reading in the absence of ones from AB. Else­ where Dc agrees with AB. For example: 6.41

samanyalaksanatvad ABDC : samanyatvad CDEGJK : samanyarvad FH

6.44

daSabhumiSvaro natho ADCms : daSabhumiSvarau natho B : da£abhumipada CDEGH2JK : om. F : lac. H bhummam TSvaro 'dhipatir A2 : bhumiSvaro 'dhipatir ABDC : bhummamana£vara jivapattir CDEH2JKG (-patir KG) : om. F : lac. H nairyanikataS ca manasi ABDC : nairyatimanasi CGK : nairyatamanasi E : nai< ~8 > FH : naiyatita manasi D : naiyatimanasi J : naiyatimanasi H2

6.45 6.81

These readings show that D's source of contamination could come from either a or B. The correction to D at 6.47, however, suggests the source may be the branch of the stemma containing B since Dc 's ucyacyate (= ucyate with dittography) is attested by B but not A: 6.47

natha ity ucyate tasv eva daSabhumisu B : natha ity ukte tasv eva dasabhumisu A : na< ~7 >da£abhumisu CDEGH2JK : napa ity ucyacyate tany eva dasabhumisu Dc : om. F : lac. H

MS E in general does not show signs of any contamination from outside (3. There are three instances possibly indicating the contrary. Firstly, (1.113) E is corrected in the margin to provide a good reading: 1.113

praticchan ECCC : pratitya FHDJCa (prati* Ea ) (AB lacking)

Secondly, at 5.19, E is the only MS from |3 not to contain a lengthy dittography: 5.19

samskaravimoksa BE : samskarapratipaksabhavena(-prati< ~4 > FH; bhavena suppl. *H2*) kramad animittapranihitanabhisarnskaravimoksa FHDJC (A lacking)

Thirdly, E is correct with AB against the rest of |3: 5.265

abhavadABE : abhavadyCC

Only the first of these examples is wholly convincing: the source of the correction must lie outside p. In the second case the dittography could have been omitted by the scribe through a saut du meme au meme (less likely, it could have been independently corrected); in the third, E's abhavdd could be the result of a scribal slip rather than a genuine inherited reading. Thus, it is possible that E

182

Introduction to the text

Stemma Codicum

uncorrected is uncontaminated, at least from any source outside (3 and its descendants. On a number of occasions E is correct against the rest of 6 and these are harder to assess. Some could be correct adventitiously as a result of an independent error by E. For example, 4.44 4.199 4.206 6.3

nairatmya AyE : nairarmye C£ (B lacking) mayajalaAByE : malajalaC : mala C cintayetAByE : vicintayet CC pratiAByE : pravi CDJGK

Others are not so easily explained as, for example, the retention of a syllable (6.39; 6.63) or vowel (4.67) or two words not attested by the rest of 6 (6.58): 6.39 6.63 4.67 6.58

urtarottarajnana AByE : uttarottajnana CDJGK anasravam ABE : asravam CDJGK : nasravam F : asravam H2 : om. H vikurvitani AE : dvivikurvitana C£ : < >kurvata(-to F)ni y : dvivikurvitani H2 (B lacking) anvayajnanam katamat ABDC : advayajnanam katamat E : JDa : < > HF : < - - - > CGK : om. H2

In this last example the reading of HF seen out of context is misleading since it is part of a larger lacuna. Thus (3 could have the correct reading, the larger lacuna ocurring as a separate error in y. However, a lacuna for anvayajnanam katamat should, according to the readings, be present in 5. These examples can be seen either as evidence of contamination or as providing counter-instances to the proposed structure of the stemma. They do not, I think, have sufficient weight to warrant changing the stemma at present; the evidence for the MSS of 8 sharing error is strong and it is not clear how the relationship of the MSS of 8 could be changed while accommodating the evidence of shared errors in DJ and EC reviewed earlier. Collation of further chapters of the NMAA may bring clarification.

183

Introduction to the text

Stemma Codicum

MS F shows no sign of contamination with the exception of a one and half line omission corrected in the margin, possibly in a second hand. 1.85-8

Sllam asyeti -* anabhibhavaniyarvat om. FC (Silam (y}asyeti -*• anabhibhavanlyatva suppl.

The corrected passage is otherwise fully attested only by H, though it is present in part in 5. The reasons for thinking that the correction is from a different exemplar than the original (ie. that F did not omit the passage and then correct it from the same exemplar) are speculative. C also omits this passage. Possibly C could have copied from F before it was corrected or from a descendant of F, copied before F was corrected; but other evidence suggests that an antecedent of F rather than F itself was C's exemplar here. (See example 1.7 in the discussion on MS C above.) If this is so then F's exemplar would have contained the omission. However, since H has the passage, a has to have had it. Thus if F is contaminated its source need be no further distant than a manuscript between a and F that still contained the passage. (For a discussion of C's correction of the omission, following 5, see above on 'MS C'.) MS H is the most heavily corrected of the manuscripts and is clearly contaminated. However, like D, the uncorrected text appears to be free from contamination, though not all the corrections are easy to identify as such, especially when they fill lacunae left as empty spaces in the uncorrected text. H is often 'corrected' so as to produce a more erroneous reading. H2 follows 5, for example, at: 1 . 125 6.56-8

svakaraih y : muhur muhuh svakaraih 6H2 (AB lacking) ucyate dharmajnanam // anv(adv- E)ayajnanam katamat / ruparupyapratisamyuktesu ABEDC : ucyate dharmajnanam ruparupyapratisamyuktesu DJCKGH2 : ucyate < -15 > pratisamyuktesu FH (ie. H2 omits anvayajnanam katamat, following 5, excepting E.)

The direction of contamination can often be seen, with H2 having readings only found in J. For example: 5.36 6 25

parinispanna BytD : parinispannatvad JH2 (A lacking) pariSuddhatvad AB : pariSuddhatva CDEGKFH : pariSuddhitva JH 2

184

Introduction to (he text

Stemma Codicum

6.42

nairatmya AB : nairatmya CDGEK : naigamya JH2 : naira F : nai*< > H

6.66

avihinam / yena AB : avihinam yad F : avihmam yada H : avihina para JH2

6.81

nairyanikataS ca manasi ABDC : nairyatimanasi CGK : nairyatamanasi E : naiyatitamanasi D : nai< ~8 > FH : naiyatimanasi J : naiyatimanasi H2

Occasionally H appears to be corrected from a source outside (3. For example, at 1.125 and 6.7, H2 is correct against (3. (In 6.7 a could have kvipi.) At 4.42 it is correct against a, and at 5.153 correct against H. 1.125 6.7 4.42 5.153

utksipadbhih H*2* : utksipyadbhih ECC (utksipya C) : utksipyabhih C : utksip< > FH (AB lacking) kvipi BH2 : ksipi A : kipi (3 (kimapi E) tatha caha H2 Tib. : om. AFHECDJ (B lacking) ekibhavah H2 : ekibhavam AB : ekibhava (3

The example at 4.42 is persuasive and in its light one is tempted to see BH2 's kvipi at 6.7 as evidence of contamination from the B branch of the stemma. The example at 5.153 is by itself insufficient to warrant a conclusion of contamination from outside the archetype - it could be the result of a (fortunate) scribal slip - and I have not noticed any other possible instance of such contamination. MS J has few readings suggestive of contamination. Since it is an uncorrected manuscript37 any contamination would have to be either inherited or the result of incorporation of corrections made to its exemplar. There are two potential examples: firstly, in 6.21-2, where J retains a syllable omitted by the rest of (3; secondly, in 5.239, where J does not witness an interpolation found in the rest of (3. 6.21-2 5.239

adhararupatvat ABJ : ararupatvat CDEFHKG tamonirghata B : tamonirghatana AJ : tamondhakaranirgha(-gho H)tana Dye

If independent scribal correction can be discounted (I am not convinced it can) the second example, since it retains the final -tana, suggests contamination from a or one

37J does have just one marginal correction. See above, Description of Manuscripts, section 2.4.

185

Stemma Codicum

Introduction to the text

of its descendents rather than from the branch of the stemma containing B. Other possible examples of contamination in J are less convincing. For example: 4.167 6.48

nemyamAJ : namyam CDy : navamyam E (lacking B) anvayaABJ : advaya CDEFHKG

Here the possibility of independent scribal correction is more likely. The case for contamination in J is not very strong. As with E, the results of further collation and editing may produce a clearer picture. 4.4 The Tibetan Translation & the Stemma Codicum Discussion of the relation between the Skt. MSS and Tib. can only be preliminary, since it draws upon a limited range of comparisons, although the implications of a number of these are nevertheless clear. Each of the examples summarised below is discussed more extensively within the corresponding textual note. i) Tib, has its own errors. For example, Tib. omits (5.243) samkara iti\ reads (5.258) 'khor ba'i gnas thabs {samsaropaya} for sambhdropaya; transcribes (4.207 & elsewhere, see textual note) te nama for nama. At 2.34, Tib. omits parsadi (though B is lost here). It is hardly surprising that Tib. has errors. There was at least one hundred years between the composition of the text and its translation. It would be unlikely that the exemplar(s) used by the translator would be an autograph manuscript; it could well contain corruptions. Errors could also enter at the time of translation or later, when the translation was revised. If some of these errors were derived from Tib.'s Skt. exemplar then, since they are Tib.'s (exemplar's) own errors, it follows that O is not in the direct line of descent from Tib.'s exemplar, otherwise the present MSS would also contain those errors. Unfortunately, the above examples are not sufficiently decisive to form a basis for such assumption; 'khor ba'i gnas thabs {samsaropaya} for sambhdropdya is perhaps the most likely candidate. Of course, Tib. could not be an apograph of any of the Skt. MSS since the oldest is dated c. 1450 CE, so it would

186

Introduction to the text

Stemma Codicum

be logically impossible for Tib. to have all the errors of, say, MS A as well as errors of its own. However, if Tib. and a shared errors and also contained errors of their own, then Tib. and a would share a hyparchetype that was an Indian manuscript predating the translation into Tibetan. ii) Tib, agrees in error with a At 5.90, B's vilokyata is to be preferred against a's vinasyata, which latter is supported by Tib.'s mi rtag fees pa 7 don to]. iii) Tib, agrees in error with (3 At 1.45, Tib. and (3 omit stutih samagrd // idani attested by A, thereby collapsing two sentences that make good sense into one without sense. (Unfortunately B is lost here.) At 2.40-1, Tib. again follows (3 in omitting three words: for A's tena mahattvad brahma / tasya svarah (em. : suro A) brahmasvaram, p has tena mahatrvad brahmasvarah. Tib. reads de bas na che ba'i phyir tshangs pa'i gsung ro. (B is lost here.) iv) Tib, is correct with A(B) against B For example, at 4.221 Tib. (byang chub mchog) supports AB's bodhyagriagainst P's bodhyangi-. There are also instances from chapters 1-4, where MS B is lost, when Tib. supports A against p. For example, Tib. supports: (1.67) A's iti against |3's ityddi; (1.183) A's klesdh sat (nyon mongs pa drug go} against P's klesd rdgddayah sat\ (4.6) A's mahdbodhikrt pranidhdnacittah (byang chub chen por smon pa'i sems} against (3's mahdbodhikrtapranidhdnacittah\ (4.96) A's tatrdsanavikalpah (de la gdan rnams brtag pa) against p's tatrdsanam vikaJpetah; and (4.96) A's madhye simhdsanam purve gajdsanam against P's madhye simhdsanam varam purve gajdsanam jneyam. At 4.105 Tib, (byang chub mchog) supports A's bodhyagriagainst P's bodhyangi-. The two examples cited where Tib. has the errors of P are striking. Both are errors unlikely to have been made independently and have therefore either been inherited from 776.'s Skt. exemplar or are the result of later contamination. If the former, then Tib. and P must descend from the same hyparchetype, since the

187

Slemma Codicum

Introduction to the text

possibility of Tib. descending from |3 is ruled out by the examples where Tib. is correct against (3 (ie. 0 has its own errors in relation to Tib.). The following figure shows this alternative: O

a

B

Tib. Fig. 3. Possible Relation of Tib. to the Stemma Codicum of the Sanskrit Manuscripts of the NMAA. 38 If Tib. and (3 do descend from the same hyparchetype, Tib. cannot be correct (unless through independent correction) against a; against either A or (3 when they are in agreement with B; or against a and B (ie. H). However there are such cases, where: v) Tib, is correct against a. For example: Tib. supports (2.41) at ha va (yang na), which gives good sense as the opening of a second gloss, against a's tatra; supports (3.4) ityadi / tatra (ces pa la sogs pa ste / de la), omitted by a. (For each of these B is lost.) At 5.199, Tib.'s drag pa chen po supports B's mahakrurah, omitted by a. vi) Tib, is correct against O. There is one instance where this may be the case: at 4.201 where Tib. supports idanlm {da ni} against ABy's idam idanlm (C? have itam iddnim; E has itah idanlm). These examples suggest, though perhaps not conclusively, that Tib. is contaminated; there is a slender possibility that they could be corrections made independently by the translator. It is not clear whether these examples represent the

38For the fiill stemma codicum of the Sanskrit Manuscripts see Fig. 2 (p. 171).

188

Stemma Codicum

Introduction to the text

contamination or whether it is seen in those where Tib. shares error with (3. Contamination could well have arisen at the time of Tib.'s revision, especially if it was 'corrected' against a different Skt. exemplar. In any case Tib. cannot be simply related to the stemma of the Skt. manuscripts.

5. The Tibetan Translation of Vilasavajra's Namamantrarthavalokini Title: Aryanamasamgititika Namamantrarthavalokini; 39 'phags pa mtshan yang dag par brjod pa'i rgya cher 'grel pa mtshan gsang sngags kyi don du rnam par Ita ba (Tib.P. 31r8-31vl [184.4.8-184.5.1]; Tib.D. 27vl-2). 5.7 The Translators According to the colophon of the blockprint edition(s) of the Tibetan translation of the NMAA the text was translated by Smrtijnanaklrti and revised by Phyag na rdo rje and kLog skya shes rab brtsegs.40 The Indian pandita Smrtijnanaklrti, who can probably be placed at the beginning of the eleventh century, is portrayed by the Tibetan histories as an important figure in the initiation of the 'later spread' (phyi dar) of Buddhism in Tibet.41 He worked in Eastern Tibet (Khams) where, apart from teaching, he was responsible for the translations of a number of texts, including the NMAA, some shorter works on the NS and his own NS commentary (Toh 2584). The Blue Annals discuss whether or not he predated Rin chen bzang po and conclude that he did.42 ^-avalokinl is my emendation. The transciption reads -avalokini. 40See Appendix III for text. Concerning kLog skya shes rab brtsegs, the Blue Annals (I 382) states, "The lo-tsa-ba kLog-skya Ses-rab-rtsegs heard the Tantras of the Samvara Cycle from Phammthin-pa". Pham-mthin-pa is earlier described as a pupil of Naropa (ibid. I 381). Bu ston's Chos 'byung adds: "Log-kya Shes-rab-tseg translated the (works of the) Cycles of Manjusnnamasangiti (mTshan-brjod)" (Obermiller, 1932, 220). If kLog skya shes rab brtsegs is a contemporary or near contemporary of Naropa (956-1040 CE) the revision of the Tibetan of the NMAA did not post-date the original translation by very long. (On the date of Naropa, see D. Snellgrove and T. Skorupski, The Cultural Heritage ofLadakh. Vol.11, p. 90, where they argue that 956-1040 CE is better than 10161100 CE, the figure given by H.V. Guenther in The Life and Teaching of Naropa, London, 1963.) 41 Some biographical information is found in Bu-ston's Chos 'byung (Obermiller, 1932, 214-5). See also Blue Annals 1 204-5, 346, 372, 395; II 417. 42Blue Annals I 204-5.

189

The Tibetan Translation

Introduction to the text

5.2 Xylograph Editions

In the preparation of the present edition I used two xylograph Tanjur editions of the Tibetan translation of the NMAA, the Derge and the Peking, one from each of the two Tanjur traditions.43 For those parts of the text that I consulted, however, the two editions were practically identical, showing only very minor variations in readings.44 My concern has not been to produce an edition of the Tibetan text but rather to consult it at points where the Sanskrit proved problematic. The following remarks concerning the translation and its relation to the Sanskrit transmission arise in this context. 5.3 Features of the Translation

The Tibetan translation of the NMAA displays a number of features that might be expected of any translation: occasional mistranslation and misinterpretation, as well as translation that is interpretive rather than strictly literal. The inherent difficulties of translating a Sanskrit commentary, such as dealing with glosses and grammatical analyses, sometimes produce Tibetan that is trite or empty of meaning. Part of the problem can be the comparatively limited vocabulary of Tibetan. Other features derive from the adoption of what must be a previously existing translation of the Ndmasamgiti for all citations of its verses. The most dramatic result of this is the rearrangement of the Skt. text of the NMAA wherever the order of comment on words in the NS differs from the order followed by the inherited NS translation. This generally arises in the following circumstances. The Skt. usually comments on a verse pdda by pdda, working from the first through to the last. If, however, the pre-existing NS translation has altered the order of the pddas, the

43The transmission of the Tanjur xylograph editions is less complicated than that of the Kanjur. All four - the Narthang, the Peking, the Derge, and the Cone - can be traced back to the edition of Bu ston. They divide into two families, the Narthang and the Peking editions on the one hand and the Derge and the Cone on the other. In the latter, the Cone edition is a direct copy of the Derge. 44Any variants in the passages I consult are noted in brackets after the preferred reading.

190

The Tibetan Translation

Introduction to the text

Tibetan translation of the NMAA rearranges the NMAA so that it follows the order of padas in the NS translation. This would of course make for better reading of the Tibetan, with the commentary taking the words of the verse, as translated, one by one from the beginning. Since the pre-existing NS translation sometimes differs from Vilasavajra in how it construes the Skt., this can produce a conflict in sense between verse citation and commentary. There is also at least one occasion where, faced with a clash between the inherited NS translation and the (Skt.) NMAA analysis, the Tibetan modifies the NMAA in order to make its analysis accord more with the inherited NS translation. There are also passages that are straightforwardly corrupt. Various junctures provide opportunities for corruption: the point of translation, revision, or the point of preparation of the blocks for the xylograph editions. Also, as seen above (section 4.4), Tib. is likely to have inherited corruptions from its exemplar(s). The following paragraphs contain examples and some further discussion of these features. i) Corruptions Some examples of small errors - of omission, addition and misreading - in Tib. that are not shared by any of the Skt. MSS are enumerated in section 4.4 (i) above. A extensive corruption is seen at the beginning of the passage in chapter 4 listing the attributes of Mahavairocana as visualised by the yogin: part of a sentence is misconstrued as a complete sentence, and its translation contains a misreading (caturtham for catur-) as well as an omission (see the textual note on 4.102-5). Chapter five has examples of additions where Tib. inserts citations of NS padas: thus Tib. has the half verses Blab, 3 led and 32ab where the Skt. (codd.) has 3 la, 31 c, and 32a respectively (see textual note on 5.56).45

45For discussion of instances of corruption that may have been inherited from Tib:s Skt. exemplar(s) see section 4.4 (iii) above.

191

The Tibetan Translation

Introduction to the text

ii) Mistranslation At

5.13,

Tib.

misinterprets the

dvandva jnanasattvadyaksarayoh

(jnanasattva;adyaksarayoh), taking it to be a karmadharaya (ye shes sems dpa' la sogs pa'i yi ge dag gi {jnanasattvadi:aksarayoh} Tib.[191.3.5]). The result, "of the two syllables jnanasattva, etc.", gives no sense in context. A lengthier mistranslation occurs in a passage that lists the attributes of Mahavairocana as visualized by the yogin, following each with an explanation of its significance. Tib. misreads the punctuation and so incorrectly links all but the last of the attributes and explanations (see textual note on 4.102-5).46 iii) Non-literal translation For example, at 4.174, ManjuSrijnanasattva is described with the bahuvrihi bdldrkamandalacchdyaprabhdmandalam, "with a halo that has the brilliance of the orb of the newly risen sun" (The Skt. is emended to -cchaya- from -cchaya-}. Tib. does not translate the element -cchaya- and uses the interpretive bskor ba rather than the possessive particle can giving, "encircled by a halo like the disc of the newly risen sun" (nyi ma 'char ka'i dkyil Itar 'od kyi dkyil 'khor gyis bskor ba Tib. [190.2.6-7J). (See textual note on 4.174.) Again, at 4.81, in a verse citation describing the decoration of the kutdgdra (ie. the mandala) Tib. translates the half-verse khacitam vajraratnais tu ardhacandrena samyutaih with rdo rje rin chen rnams dang ni / / zla gam (kham Tib.P.) rnams kyis yang dag brgyan (mnan Tib.P.) (Tib.[189.3.5]}. Though the Skt. is not without its problems - samyutaih is a conjectural emendation of codd.'s samyutam (see the textual note on 4.81) - Tib., perhaps constrained by the metre, deals with it interpretively, using the verb yang dag brgyan for both samyutaih and khacitam

46For discussion of a mistranslation that makes sense in context see textual note on 5.177. At 4.207, Tib. mistranscribes a mantra, adding an extra syllable, and omits the Skt. giving the number of syllables (again, see the textual note).

192

The Tibetan Translation

Introduction to the text

(yang dag preserves the verbal prefix sam-) and treating ardhacandrena as a plural, giving "...it is adorned with vo/ra-gems and half-moons".47 iv) Empty glosses At 1.110, the Skt., commenting on protphullakamalananah (NS 2b), glosses kamalah with padmam. Tib., which inherits pa dma for kamala, can but repeat itself, translating the gloss with pa dma ni pa dma nyid do (Tib. [186.1.3]}. A few lines earlier, at 1.106, commenting on the previous pada, vibuddhapundarlkaksah, the Skt. glosses pundankam with svetapadmam. Tib. has pa dma dkar po zhes pa ni dkar po'i pa dma'o (Tib.[186.1.2]}, "as for white lotus, [this refers to] a lotus that is white" Given that pa dma dkar po had been chosen to translate pundankam, the rendering of the gloss is bound to be empty. Nonetheless, pa dma dkar po is an obvious choice for a translation of pundankam even if it were not predetermined by the inherited translation: it is after all how the NMAA glosses it (and how it is translated in its three appearances in the MVy, at 1335, 1348, and 6147). Tib. has here fallen foul of a problem endemic to all translation of Skt. commentaries, namely, that if the root text is translated in accordance with the commentary, as it generally should be, it often renders translation of glosses empty. v) Adoption of a pre-existing Tibetan translation of the Namasamgiti For its citations of NS verses Tib. does not translate afresh but adopts what must have been a previously existing translation.48 That this is the case is most apparent where there is a conflict between the meaning of a NS verse as cited by Tib. and its analysis in the following comment. If Tib. translated the NS citations for itself such conflicts would be extremely unlikely, whereas in fact they occur frequently, particularly so in chapter 5 where the NMAA and the inherited translation often read

47 A further example of a non-literal translation is found at 5.175 (see textual note). 48 I generally refer to this translation as the 'inherited translation', which for the most part corresponds with the edition of the translation printed by Mukherji (1963) (= NS.Tib.). See section 5.5 below for discussion of its relation to the transmission of the NS in Tibetan translation.

193

The Tibetan Translation

Introduction to the text

the epithets starting mahd- differently. Before discussing possible sources of this inherited translation, there follow some examples of how it can differ from the analysis given by the NMAA, and a note of some of the consequences. Verse 40 of the Ndmasamgiti provides a typical example of what is commonly found in chapter 5. Here, the epithet mahdbhavddrisambhettd (NS 40a) is analysed by the Skt. as a karmadharaya with mahd- qualifying the rest of the compound: mahdms cdsau bhavddrisambhettd ca mahdbhavddrisambhettd (5.235-6). Thus the NMAA takes the compound to mean "great shatterer of the mountain of existence". However, in the opening citation of the verse (5.231 in the Skt.) Tib. reads otherwise: srid pa 'i ri bo chen po 'joms, "[he] shatters the great mountain of existence" (Tib.[194.1.8-194.2.1J). In the translation of the karmadharaya analysis this difference produces a confusing result, ie. chen po yang de yin la sridpa'i ri bo 'joms pa yang de nyidyin pas srid pa'i ri bo chen po 'joms zhes so (Tib. [194.2.3]). Here the analysis, that is, what corresponds to mahdms cdsau bhavddrisambhettd ca, is translated correctly but since the conflicting translation of the epithet is retained, its restatement at the end creates a nonsense ("Since he is great and one who shatters the mountain of existence, he shatters the great mountain of existence"). A similar conflict is seen in the treatment of the epithet mahdprajndyudhadharah (NS 34a), which is again taken by the Skt. as a karmadharaya with mahd- in apposition to the rest of the compound: mahdms cdsau prajndyudhadharas ca mahdprajndyudhadharah (5.119-120). This gives a meaning of, "Great bearer of the weapon of wisdom"; Tib., however, in its citation of the pdda has shes rab mtshon chen 'chang ba ste (Tib.[192.4.5]), "Bearing the great weapon of wisdom" and this time in the translation of the karmadharaya analysis it mistranslates the first part as well as restating the inherited translation of the epithet at the end: shes rab mtshon chen 'chang ba yang de yin la chen po yang de nyid pas shes rab mtshon chen 'chang ba'o(Tib.[l92A.I]).

194

The Tibetan Translation

Introduction to the text

A further example of the inherited NS translation conflicting with the interpretation of the NMAA and leading to mistranslation is found in the following verse (NS 35a). The Skt., following the familiar pattern, analyses mahdmdyddharah as a karmadharaya: mahdms cdsau mdyddharas ceti mahdmdyddharah (5.144-5). For this Tib. has chen po yang de yin la sgyu 'phrul yang de nyid yin pas sgyu 'phrul chen po 'chang ba'o (Tib.[193.1.3-4]). For the citation of the epithet Tib. takes mahd to qualify mdyd (sgyu 'phrul chen po 'chang {mahamayaAlharah}). It looks as if it may be trying to retain this in the translation of the karmadharaya, which retranslated into Skt. gives something like mahdms cdsau mdyd ca mahdmdyd­ dharah*9 Finally, I have noted one instance in the translation of these karmadharaya analyses where Tib. attempts to translate it all in accordance with the reading of the inherited translation. Thus in the analysis of the epithet mahdmdyendrajdlikah (taken by Tib. as sgyu 'phrul chen po mig 'phrul can) the Skt. reads mahdms cdsau mdyendrajdlikas ca mahdmdyendrajdlikah (5.153-4). 50 For this Tib. has sgyu 'phrul chen po yang de nyid yin la mig 'phrul can yang de nyid yin pas sgyu 'phrul chen po

mig

'phrul

can

no

{mahdmdyd

cdsdv

indrajdlikas

ca

mahdmdyendrajdlikah} Tib.[193.1.8-193.2.1]. Not only does the use of a pre­ existing translation of the NS by Tib. produce conflicts of meaning and passages that make little sense, it also exerts pressure on Tib. to translate in accordance with that translation.

49An example of limited vocabualary compounding the problems of adopting a pre-existing translation is seen in Tib. for NMAA on NS 31b (mudhadhimohasudanah) where gti mug is used to translate both mudha and moha. See textual note on 5.61. 50The Skt. here is emended. The available MSS read mahdms cdsdv indrajdlikas ca mahd­ mdyendrajdlikah, which must be corrupt. (See textual note on 5.154.) However, if 776.'s exemplar also read this, the path of further corruption to its present reading is easier to account for - all that is needed is for the lost 'mdyd' to be attached to 'mahd-.

195

The Tibetan Translation

Introduction to the text

At a number of points the NMAA either cites or incorporates passages from the anusamsd, the prose section praising the benefits of the NS.51 In translating these, Tib. does not follow the translation of the anusamsd as printed in the Derge or Narthang Kanjurs. I have not been able to ascertain whether there was an earlier translation that it did follow (see section 5.5 below). The same is true for the translation of the NS colophon where it is incorporated into chapter one (see textual note on 1.72). vi) Rearrangement of NMAA in accordance with the order of the inherited translation of the NS. As stated earlier, one of the most striking features of Tib., deriving from its adoption of the pre-existing translation of the NS, is the re-ordering of sections of the NMAA so that its order in commenting on the words or pddas of a NS verse accords with that of the words orpddas of the inherited translation. I will examine just one instance of this phenomenon, which should be sufficient to illustrate some of the frequent ramifications. Other instances, a number of which are discussed in the textual notes, are listed subsequently. NS 8 reads, ajndnapankamagndndm klesavydkulacetasdm / hitdya sarvasattvdndm anuttaraphaldptaye //. The inherited translation reverses the order of the first two padas, thus: nyon mongs pas ni sems dkrugs shing / mi shes 'dam du hying ba yi / sems can kun la sman pa dang / bla med 'bras bu thob by a 'i phyir // (Mukherji 5-6) Tib. has rearranged the passage on NS 8 so that the lines on klesavydkulacetasdm are translated first, followed by those on ajndnapankamagndndm, and finally by those on anuttaraphaldptaye. Hence punar api kimvisistds -> prakdsayatv iti 51 In chapters 1-5 the anusamsd is cited twice: a single line at 1.56, and a five line passage starting at 5.287 (see corresponding textual notes). It is incorporated (without attribution) on three further occasions: two and a half lines starting at 1.63; a little less at 4.2 and some four lines at 5.279ff. (again, see corresponding textual notes). In the last of the incorporated passages, material from two parts (cakras) of the anusamsd is combined.

196

The Tibetan Translation

Introduction to the text

sambandhah (1.182-7) is translated first, starting at Tib.[186.4.8]\ then ajndnapankamagndndm -* sattvds tesdm hitdya (1.180-2), from Tib.[186.5.2]; and anuttaraphalam -> tatprdptaya iti (1.187-8) stays last, at Tib.[186.5.4]). The rearrangement is unsuccessful, however, since the Skt. does not comment on the NS verses in discrete units that can be interchanged without interrupting the development of the interpretation. This is especially so in the introductory chapters preceding those that deal with the 'names' of Mafiju^ri. Unfortunately it is in these chapters that there is least coincidence between the order of the Skt. and that of the inherited translation of the NS verses, which results in the Tibetan in disruption of meaning. In the present instance, for example, Tib. opens the passage with sems can de rnams ji Ita bu zhe na, translating punar api kimvisistds te sattvd ity aha, but reference to the living beings (sems can de rnams; te sattvdh) only makes sense in context of the previous reference to them in the opening passage of the Skt., which Tib. has not translated. (Tib.'s omission of anything for punar api here could be seen as a remedy for the problem of putting first a sentence that starts, "Again ...") Later, Tib. includes the linking passage (sambandhah), anuttaraphalaprdptaye bhagavdn prakdsayatv iti sambandhah, at the end of the lines on klesavydkulacetasdm. This should properly have been included as the start of the comment on anuttaraphaldptaye; since Tib. omits it, it has to insert a citation of the last pdda before translating the commentary on it. These are not the only problems with this passage, but they are enough to show the tangle that can and often does result from rearranging the order of the Skt.. 52

52The Tibetan translation of the NMAA on NS 8 is as follows: sems can de rnams ji Ita bu zhe na / / nyon mongs pas ni sems dkrugs shing zhes pa ni nyon mongs pa drug go / / de skad mngon pa'i chos las gsungs te / nga rgyal Ita ba the tshom dang / / Mod chags zhe sdang gti mug go / / zhes gsungs so / nyon mongs pa de rnams kyis sems rnam par dkrugs pa'i sems can gang yin pa de rnams ni nyon mongs pas ni sems dkrugs shing ste / de'i phyir na de mams bla na med pa'i 'bras bu thob pa'i don du bcom Idan 'das kyis bshad du gsol zhes pa dang 'brel to (te Tib.D.) I I gzhan yang khyad par ci 'dra ba'i sems can zhe na / mi shes 'dam du bying ba yi / / sems can kun la sman pa dang / / zhes pa ste / de la mi shes pa ni ma rig pa ste / de nyid thogs pa'i gnas nyid kyi phyir ram / thar dka' bas na 'dam dang 'dra bas 'dam mo / / der bying ba'i sems can gang yin pa rnams la phan pa'i phyir

197

The Tibetan Translation

Introduction to the text

Tib. again rearranges the Skt. text so that it follows the order indicated by NS.Tib. in the passages on the following verses: NS Id (see textual note on 1.88); NS 4 (see textual notes on 1.125 & 1.129); NS 9b (Tib. reverses the order of the NMAA at 1.191fYon the epithets sastd andjagadguruh); NS 18 (see textual note 2.12).53 For the commentary on NS 6ab, Tib. changes the order of the Skt. but unusually - this is the only occasion I have noted - not so as to follow that of NS.Tib.. This suggests that at this point the inherited Tibetan translation followed by Smrtijnanakirti differs from that of the bilingual xylograph. (See textual note to 1.156.) In its citation of NS 38d, Tib., following NS.Tib., reverses the order of the two epithets but does not subsequently rearrange the order of the NMAA. Such preservation of the Skt. order of the NMAA in the face of a divergence in the inherited NS translation is uncharacteristic. (See textual note to 5.205.) 5.4 The Value of Tibetan Translations in Editing Buddhist Sanskrit Texts. As stated before (in section 3.3 above) I have treated the readings of the Tibetan translation of the NMAA as simply those of another witness and assessed them as such in the light of the evidence from the Skt. MSS. This might seem to state the obvious, but means that the readings of the Tibetan have not been accorded any privileged status, running contrary to the view, classically exemplified by David Snellgrove in the introduction to his edition of the Hevajra Tantra (HVT), that in the editing and translating of Buddhist Tantric texts the value of the Tibetan translation is equal to, if not greater than, that of the surviving Sanskrit. He says, for example, ...the general method has been to first ascertain the intended sense of the text and then edit the manuscripts accordingly. (HVT II, viii.)

bshad du gsol zhes pa'i tshig dang 'brel to / bla med 'bras bu thob bya'i phyir zhes ba ni bla na med pa'i 'bras bu ni sangs rgyas nyid do / / de

A word enclosed with angle brackets is an emendatory addition by the editor that is not attested by any of the manuscript readings available.

*

Following a word a superscript asterisk indicates that there is a note in the Textual Notes on the preceding word or passage.

[ ]

Square brackets enclose material additional to the edited text of the NMAA, eg. inserted verses of the NS and chapter headings.

t t

Obeli are used to mark a word or passage judged to be corrupt but which the editor is not able to emend.

Symbols and Abbreviations in the Apparatus and Textual Notes 1. Manuscript. Archetype and Hyparchetype Sigla A, B, C, D, E, F, H, J.

Upper case Roman letters are used as sigla for the Sanskrit MSS of the

NMAA. (See Introduction to the Text, Section 1) H, a, (3, Y, 5, e, t,:

Greek letters are used as sigla for the archetype (H) and hyparchetyes (a,

(3, y, 6, e, C) of the stemma codicum. Unless otherwise stated, they are used to denote the readings of all the extant descendants of the (hyp)archetye, rather than a reading inferred as being that of the (hyp)archetye itself (because, for example, it is attested by the majority of its descendants). 2. Abbreviations for Texts and Tibetan Translations NMAA

Vilasavajra's Namamantrarthavalokini. In plain text in the testimonia; elsewhere in italics. Figures give chapter and line number of the present edition, eg. 'NMAA 1, 202.' or 'NMAAfl, 202f denotes NMAA adhikara 1, line 202.

Tib.

The Tibetan translation of the NMAA. Unless otherwise specified, Tib. refers to the reading of both Tib.D. and Tib.P. Text references, following Tib. in brackets, are those of Tib.P., eg. Tib.fJ86.3.7J.

Tib.P.

Peking blockprint edition of Tib. (D. T. Suzuki, ed., Peking Tibetan Tripitaka, Tokyo:

Tibetan Tripitaka Research Institute, 1956, vol. 74, pp. 184.4.8-226.2.1) Tib.D. NS

Derge blockprint edition of Tib. Ndmasamgiti. NS 42 denotes Ndmasamgiti, verse 42 (verse numeration follows that of

NS.Dav.

NS.Dav.). Edited text of the NS as printed by Davidson (1981): in plain text in the testimonia, elsewhere in italics, eg. 'NS.Dav.[68, 2J denotes NS.Dav. page 68, line 2; 'NS.Dav.[68, v. 163a]' denotes NS.Dav. page 68, verse 163, pada a. (In the testimonia

207

Abbreviations

Text

references are given without brackets around the text reference. Eg. 'NS.Dav. 68, 2'; 'NS.Dav. 68, v.!63a'.) Variant readings recorded by Davidson are given with the source of the variant (as abbreviated by Davidson) indicated in superscript, eg. NS.Dav.Minaev denotes the reading of Minaev's edition of the NS. NS. Tib.

Tibetan translation of the NS according to the bilingual Peking xylograph edition as printed in Mukherji.

NS.Tib.D.N. Tibetan translation of the NS according to the Derge and Narthang xylograph editions as printed in Mukherji. Mukherji

Mukherji, Durga Das. (1963) (ed.) Aryamanjusrinamasanglti: Sanskrit & Tibetan Texts. Calcutta: University of Calcutta.

Sak.

Sakurai's edition of NMAA 3 & 4 as printed in Sakurai, 1988.

3. Symbols and Abbreviations A1

A as corrected by the original scribe.

A2

A as corrected by a later hand.

Aa

A before correction.

Ac

A after correction.

Aw£

Indicates a reading in A's margin.

corr.

correxi. 'I have corrected.'

em.

emendavi. 'I have emended.'

conj.

'I ha ve conj ectured.'

codd.

codices. All the manuscripts available at a place in the text.

cett.

ceteri. All the remaining manuscripts available at a place in the text.

om. A

omisit A. 'A has omitted.'

suppl. A

supplevit A. 'A has supplied.'

v.l.

varia lectio. 'Variant reading.'

lac. A

lacuna in A. Represent connecting marks in a MS, which indicate that there is not a break or lacuna in the text where there is a gap between aksaras, eg. 'u>»»pama A'. Parentheses generally contain additional or exceptional material. Uses: 1. Following a hyparcherype, giving - usually insignificant - exceptions to the reading record. Eg. "sarvamdadatvat codd. (sarvadadatvat BH)". If the exception involves a difference of one aksara only, the brackets may be placed within the reading, eg. "papadau tutsahah B : papady(-ody F)anutsahasya (3". 2. Within a reading of Tib., giving a variant. Eg. "pascan nibantam conj. : mjug ni nge'i (de'i Tib.D.) rkyen dang Idan pa'o Tib.[193.5.4T\ 3. Following corr., em., or conj., giving a reason for the reading proposed. This includes abbreviations of testimonia, Tib. with text reference and reading, the name of the

208

Text

Abbreviations

proposer of the emendation (if different from the editor). Eg. "kusidasattvani conj. (Tib.[193.1.5] sems can le lo can mams): kusidasattvam codd". 4. Following Tib. and containing a text reference and reading where Tib. supports an attested reading. Eg. "svasamadhinam F Tib.([191.3.6] rang gi ting nge 'dzin rnams) : svam samadhinam cett"

Angle brackets indicate a lacuna recorded in a manuscript, either empty (
), or

enclosing dashes () according to whether the manuscript has a blank space or has marked the number of syllables lost or illegible. For large lacunae the total number of lost or illegible syllables is indicated by a number within the angle brackets (). Where the number of such syllables is an estimate, eg. in a lacuna that is a blank space, the number is preceded by a tilda (< ~9 >). [ ]

1. Encloses source reference after a text abbreviation. Eg. Tib.[193.2.5], SDPS[14, 21]. 2. Encloses Tibetan that contextualises the relevant passage. Eg. "atitanagatapratyutpannanam conj. Tib. ([190.1.4] 'das pa dang ma 'ongs pa dang da Itar gyi [dus gsum gyi sangsrgyas rnams kyi'o]) : atltanagatanam codd".

{ }

Follows a Tibetan reading and contains a retranslation into Skt.. Eg. "krama iti em. : karmeti codd. : las kyi go rim shes bya'o {karmakrama iti} Tib.[193.2.5J". Hyphen. Uses: 1. At the beginning or end of a word or syllable, to indicate that the word or syllable is part of a larger word. Eg. "yathasvam (3 (-tvarn F) : yathasva B"; "nimittabhavat a (nimiha- F): nimittagrahabhavat B". 2. In the middle of a word at a breaks. A hyphen is never used in the first (positive) reading of the apparatus, which is fully stated, since the positive reading identifies which part of the edited text has variant readings.

*

Illegible syllables are indicated by asterisks on the line, eg. gu*rat.

* *

Uncertain letters or syllables are enclosed by superscript asterisks, eg. gu*hya*rat.

4

Separates entries of different sets of variants at a given footnote.

X->Y

Used inclusively to indicate a passage of text, 'from X to Y'. Eg. guhye — guhyarat, 'The passage starting guhye and ending guhyardt\

Abbreviations in the Register ofTestimonia unattrib.

Unattributed quotation,

attrib.

Attributed quotation.

209

Adhikdra 1

Text

TEXT [Adhikara I] 1 * aryamanjuSriyam naWa jnanendum2 * tryadhvatayinam3 * / tathasya4 namasamgitim gambhirodaradharminim // yogacaryakriyatantram5 tatha paramitanayam / sutrabhidharmapitakam6 vilokyanyanibandhanam7 * // 5

jatakam cetivrttam ca stotrakaramatam tatha / vijnanavadam akhilam tatha madhyamakam ca yat // laukikam8 ca tatha Sastram guruparvakramagatam / upadeSam ca samsmrtya prarthitena mayadhuna // vyakhyanam kriyate tasya gambhirodaravartanam9 /

10

sattvanam 10 mandabuddhmam karunardrena cetasa // idanim asya 11 namasamgiteh ^ariravyavasthapanam abhidhiyate // tad yatha

/

adhyesana

/

prativacanam

/

satkulavalokanam

/

Codd. a (A F H E C D); A, (3 (F H E C D); y (F H), 5 (E C D); e (E C), C (D). Lacking: B (fol. 1-13 lost); J (fol. 1-2 lost). Before the first line are found the following salutations: om namo manjughosasvamine 6 : om namas te varadavajraya A : < ~12 > y : rje btsun sgrol ma la phyag 'tshal lo Tib. [184.5.1] 2 jnanendum H*2* Tib.([184.5.2] ye shes zla) : jnanendu A^ 0&ane*du* C) 3 tryadhvatayinam con/. (Tib.) : tryadhvatayinam A : pravatayinim y (tryadhvatayinIm H2) : pradhvatayinam 5 (*dhvatayinam E; tpradhvatayinam C) 4 tathasya Ay : tathasmin 6 5 yogacarya A : yogasyartha F : yogasyartha H (yogacarya H2) : yogasyarthe 6 4 kriya< ~2 > vilokyanya- y (ie. tantram -»• pitakam lac. y; suppl. H2) 4 tantram AeH2 : tantre D : lac. y 6 pitakam EH2 : pitakam A : pidyakam C : padyakamD 7 anyanibandhanam ASH2 : anyatibandhanam F : anyapibandhanam H : rnam 'grel gzhan Tib. [184.5.3J 8 lau(lo- DCc)kikam ca tatha ^astram guruparvakramagatam AEDH2CC : om. y : < -18 >C (suppl. CCH2) 9 vartanam F : varta*am H (vata*am H2) : vartaram 5 : vattaram A 10 sattvanam mandabuddhmam karunardrena cetasa A6H 2 : sattvanam ca hitarthaya jnaninam moksapraptaye y 1 ] asya A : asyah (3 1

210

Adhikara 1

Text

mayajalabhisambodhikramah / bodhicittam 1 * s"rivajrasattvadvarena / suvis"uddhadharmadMtujnanarn mahavairocanasvabhavena / adars"ajnanam 15

aksobhyadvarena2 / pratyaveksanajnanam amitabhamukhena3 / samata/

jnanam ratnasambhavarupena

krtyanusthanajnanam amoghasiddhi-

svabhavena / pancatathagatamukhena jfianastutih / anuSamsa / mantravinyasah / upasamharaS ceti SarirarthaS4 caturdaSa // tatra atha vajradhara ityady5 arabhya yavat prahvakayasthito 'grata iti 20

sodas"abhir gathabhir adhyesana // atha6 s~akymunir bhagavan ityady arabhya yavat tat sadhu bhagavann7 iti sadbhir gathabhih prativacanam // atha Sakyamunir bhagavan sakalam8 ityady arabhya yavad mahosnisakulam9 mahad iti dvabhyam gathabhyam satkulavalokanam

//

imam san-

mantretyady arabhya yavad arapacanaya te nama iti tisrbhir gathabhir 25

mayajalabhisambodhikramah sucitah // tad yathetyady arabhya yavan mahayananayottama

iti

caturda^abhir

gathabhir

vajradhatu-

mahamandaladvarena bodhicittavajrasyabhidhanarn 10 // mahavairocano buddha 11

ityady

arabhya 12

yavad

vajrahkus'o

mahapa^a

iti

Codd. a (A F H E C D); A, B (F H E C D); Y (F H), 5 (E C D); e (E C), C (D). Lacking: B (fol. 1-13 lost); J(fol. 1-2 lost). bodhicittam -»• mahavairocanasvabhavena om. D 4 bodhicittam A Tib.D.([27v6] byang chub kyi sems) : < ~7 >bodhicittah FC : < ~7 >bodhicittem H (adarsajnanam bodhicittam H2) : adarsanajnanam bodhicittah ECC : om. D : byang chub kyi semsrdorje {bodhicittavajram} Tib.P.[184.5.6] 2 dvarena A : svabhavena B (om. E; dvarena H2) 3 mukhena A : svabhavena 6 (mukhena H2) 4 s"arirarthas" A : sarirarthaS 6 (SarirartheS D) 5 ityadi A : ityadav B 6 atha ->• prativacanam om. A 7 bhagavann corr. (NMAA 2, 64; NS.Dav. 51, \\22d) : bhagavan Y : bhagavann 5 (bhagavan D) : om. A 8 sakalam B : sakalamantram A 9 mahosnisa A : P usnisa (*u*snlsa F) 10 bodhicitta A Tib. : bodhicittacitta B (bodhicitta DH) 1J buddha A : buddho Y : buddha 6 12 arabhya A : arabhyah B 1

211

Adhikara I

Text

padonapancavimSatibhir 1 gathabhih suvisuddhadharmadhatujnanasva30

bhavena2 stutih // vajrabhairavabhlkara ityady arabhya yavad3 ghoso ghosavatam vara iti padonasardhadaSabhir gathabhir adarSajnanasvabhavena stutih // tathatabhutanairatmyetyady4 arabhya yavaj 5 jnanarcih suprabhasvara iti dvacatvarimsadbhir gathabhih pratyaveksanajnanamukhena stutih // istarthasadhakah6 para ityady arabhya yavad ratnaketur7 maha-

35

manir8 iti caturvimSatibhir9 gathabhih samatajfianasvabhavena stutih // sarvasambuddhaboddhavya10 ityady arabhya yavan 11 manjuSrih Srimatam vara iti pancadaSabhir gathabhih krtyanusthanajnanadvarena stutih // namas te varadavajragryetyady 12 arabhya yavaj 13 jnanakaya namo 'stu ta iti pancabhir gathabhih pancatathagatamukhena pancajnanastutih // iyam asau

40

vajrapane vajradhara 14 ityady arabhya yavad dharmaraja ity anusamsa 15 stutih // om sarvadharmetyady arabhya yavad dharmadhatujfianagarbha ah

32 cf. NMAA 8 [Cl: see, eg. B.42r8]: tathatdbhittanairdtmyabhutakotir anaksara iti 34 cf. NMAA 9 [B.60v5]; NS.Dav. 58, v.l 19a: istarthasadhakah parah 34 cf. NS.Dav. 59, v.!42d: ratnaketur mahdmanih; NMAA 9 [B.68r2, B.68r3] mahamanih Codd. a (A F H E C D); A, (3 (F H E C D); y (F H), 6 (E C D); 6 (E C), C (D). Lacking: B (fol. 1-13 lost); J (fol. 1-2 lost). 1 2 3 4 5 6 7 8 9 10 1J 12 13 14 15

padona A : padonna (3 (padona D) 4 vimsatibhir corr. : vimsati codd. jfiana A : om. P yavad corr. : yavat codd. nairatmyety conj. (NMAA 8; NS.Dav. Bi^ira alone^ Quad-) : neratmety A : naira tmya(-tmye H) ity 0 NS.Dav. Mukher^i; Vira : nairatmyam NS.Dav. [55, v.77a] yavaj corr. : yavat codd. sadhakah corr. (NS.Dav. [58, \.119a]) : sadhaka codd. ratnaketur Y : ratnaketuAS mahamanih conj. (NMAA 9; NS.Dav. [59, v. 142dJ) : mahamunir codd. vimsatibhir A : vimsati (3 buddhaboddhavya corr. : buddhabodhavya P : buddhabodhavya A yavan A : yavat P vajragryeryA : vajragrery P yavaj corr. : yavat codd. vajradhara A : om. P (vajradhara Ff) anusamsa corr. : anusamsa codd.

212

Text

Adhikara 1

iti mantradvarena stutih

//

atha vajradhara ityady 1 arabhya yavat

sarvasambuddhadeSita2 ity upasamharena stutih // tad ekatra dvasastyuttaraSatagathabhir adhyardhaSatenanus~amsagranthena stutih3 samagra4 / 45

idam5 * ca6 bhagavatah sarvatathagatajnanakayasya manjuSrijnanasattvasya namasamgiteh pindarthah // sarvatra vakyarthasam£ayena7 * sarvesam sambandhabhidheyaprayqjanatatprayojanadhyavasayapurvika8 * pravrttir iti sambandhadiny9 asya namasamgiter vacyani // tatha hi yadi sambandhabhidheyam 10 asya na

50

kathyeta tadonmattavakyavad 11 asambaddhatvam 12 anarthakyam ca sambhavayantah preksapurvakarino 13 na pravartante Srotum iti bandhabhidheyam 14 asya avaSyam 15 vacaniyam

//

sam­

^raddhavantas tu

42 cf. NS.Dav. 68, v.!63a: atha vajradharah 42-43 cf. NS.Dav. 69, v.!67d: samyaksambuddhadesitah Codd. a (A F H E C D J); A, (3 (F H E C D J); y (F H), 5 (E C D J); e (E C), C (D J). Lacking: B (fol. 1-13 lost); J, lines 42-51, (fol. 1-2 lost); J, fol. 3r, starts line 51: preksapurva .... 1 2 3 4 ^ 6 7 8

9 10 11 12 13 14

vajradhara ityady H2 (NS.Dav.[68, v. 163aJ) : vajradharetyady cett. sarva codd. Tib.([185.2.5-6] [rdzogs pa'i sangs rgyas] kun [gyi bshad]) NS.Tib. [Mukherji 102, 4] : samyakAtf.£>av./ y svakaraih y Tib. [186.2.2] : muhur muhuh svakaraih 6H2 prasphurad y '• prasphulla 6 (prasphullad D, praphulla J, praprasphulla C) prasphuranti conj. (Sanderson; Tib.[186.2.1] rab tu 'phro ba) : prasphullanam 5H2 (prapraphullanam J) : prasphurad eva< > y prasphurad y : prasphulla 5 (praphulla J) taihcodd. (om. H) kimatmakah 5 : kimatmaikagra y karah 6 : karair iti y pravicayah conj. (AKBh; Tib.[186.2.5] rab tu 'byed pa) : pravicarya 5 : prasiddhaye y prajna / upayo conj. (Tib.[186.2.5] shes rab bo / / thabs ni) : prajnopayayo 6 (prajnopayayau C) : prajnopayayoh y 4 's~esa 6 : ^ese y ('s"ese F) 4 laukika codd. (lokika C) : om. Tib. pradarSana yE : pradarsa CC yathabhavyataya £ : yathabhavataya CY sattvarthakaranam conj. (NMAAfl, 142]; Tib.[186.2.6] sems can rnams kyi don byed pa) : sattvakaranam codd. kamcorr. : ka codd. runaddhiti £ : runa C : niviti y sukha EJ : mukha cett.

219

Adhikara 1

Text

atmanah sukham avrnotiti karuna 1 / mahati casau karuna ceti mahakaruna / analambanatmika //jagac cajagac cajagac cajaganti traidhatukani / tesam 135

arthah prayojanam / ata aha prajnopayamahakarunajagadarthakarair iti 2 * / ata eva te krodhah para utkrstas te ata aha parair iti //

[hrstatustaSayair muditaih krodhavigraharupibhih

/

buddhakrtyakaraih nathaih sardham pranatavigrahaih // (5)] kimsvabhavah3 / hrstatustaSayasvabhavah / tatra4 hrstah5 kayais6 140

tusta£7 cittena / a£ayo 'bhiprayah8 / tataS cayam arthah / hrstatusta aSaya9 yesam te hrstarustaSaya ata aha hrstatustaSayair iti / ata10 eva mudita harsitah sarvasattvarthakaranad1 1 * / ata evaha muditair iti // vigrahasya rupam vigraharupam / krodhanam vigraharupam krodhavigraharupam / kim tad rupam / mahabhairavattahasanekaSirsakaracaranavikrtanayano-

145

rdhvakes^pingalabhrtbhangakapalamalalairikrrisarpabharanavyaghracamianivasanadikam 12 * / tad vidyate 13 yesam te krodhavigraharupinah / ata

Codd. a (A F H E C D J); A, 0 (F H E C D J); y (F H), 6 (E C D J); e (E C), C (D J). Lacking: A (fol. 4-5 lost), B (fol. 1-13 lost). karuna Y Tib. [186.2.4] : mahakaruna 6 arthakarair iti C*Y : arthakaraih parair iti CECC Tib. kimsvabhavah corr. : kimsvabhava 6 : kibhutava H : kibhubhava F tatra Y Tib.([186.2.8] de la) : tatah 6 (tata J) hrstah corr. : hrsta codd. kayais corr. : kayaih codd. tustasH2 : tustaceft. 'bhiprayah JE : abhipraya CDY : abhiprayah H2 hrstatusta asaya corr. : hrstatustasaya e : hrstatustasayo Jy : hrstatusta D ata eva mudita -> muditair iti om. H karanat conj. Tib.(fl86.3.1J [sems can gyi don] byed pa'i phyir ro) : karanani codd. (om.H) 12 mahabhairavattahasa conj. : mahabhairavo ttahasa Y : mahabhairava attahasa 6 (mahabhairava athasa C) 4 alamkrti conj. : alamkrtaco^. 13 tad vidyate Y : tad(ta C) vidyante 5

1 2 3 4 5 6 7 8 9 1° 1*

220

Adhikara I

Text

aha krodhavigraharupibhir iti // buddhanam krtyani buddhakrtyani krtyani kartavyani / ata aha buddhakrtyakarair iti / ata eva 1 nathah2 Sastaro3 buddhakrtyakaranat 150

/

ato nathair iti

// pranato vigraho yesam te

pranatavigrahah / atas taih / pranatavigrahaih sardham bhagavan vajradhara idam vaksyamanam aheti4 sambandhah5 //

[pranamya natham sambuddham bhagavantam tathagatam / kltanjaliputo bhutva idam aha sthito 'gratah // (6)] kim krtva / pranamya / kam / natham iti6 / anathanam lokanam 155

anusasanat nathah svami / tarn7 * // punar api kimviSistam natham8 ity aha sambuddham iti* samyaksambuddham9 na 10 Sravakam pratyekabuddharn 11 * // bhuyo 'pi natham vi^inasti bhagavantam iti /

Codd. a (A F H E C D J); A, (3 (F H E C D J); y (F H), 5 (E C D J); e (E C), C (D J). Lacking: B (fol. 1-13 lost); A, lines 147-9 (fol. 4-5 lost); A, fol. 6r, starts line 149: [bujddhakrtyakaranat .... 1 2 3 4 5 6 7 8 9 10 1]

ata eva 5 Tib.([186.3.4] de nyid kyi phyir na) : ata evahay nathah corr. : natha codd. (pm. H) s~astaro Jy : Sastaro £D aheti A : aha (3 (a< > F) sambandhah A Tib. ([186.3.6] 'brel lo) : om. (3 natham iti A : natham 5 : natham natham iti y tarn conj. : tarn natham A : natham (3 Tib.([l86.3.6] mgon po) natham A : om (3 buddham Ay : buddha 6 na Ay : om. 6 pratyeka A : adhya (3

221

Adhikdra 1

Text

bhagasampannatvad 1 bhagavan / caturmarabhahgad va / tatha caha / kleSakarma tatha janma kleSajneyavrti2 * tatha / yena3 vaipaksika4 bhagnas5 teneha bhagavan smrtah6 // iti //

160

anena7 nairuktena8 vidhina kleSadikam bhagnavan9 iti bhagavan 10

/

aiSvaryadigunayogad va / tatha caha / aisvaryasya samagrasya rupasya ya&asah Sriyah / jnanasyatha11 prayatnasyasannambhagaiti 12 smrtih13 * // iti // punar api kimbhutam natham ity 14 aha tathagatam iti / tathatam 15

165

159-60 cf. AAA 1.2-3, p.272, 5-6 (unattrib.): klesakarma tatha janma klesajneydvrti tatha /yena vaipaksika bhagnas teneha bhagavan smrtah // 163-4 cf. 1) AAA 1.2-3, p.272, 9-10 (attrib. BuBhuS): aisvaryasya samagrasya rupasya yasasah sriyah /jnanasyatha prayatnasya sanndm bhaga iti srutih; 2) AbhayaP lv6-7: aisvar­ yasya samagrasya rupasya yasasah sriyah /jnanasyatha prayatnasya sanndm bhaga iti srutir iti; 3) CSViv 6vl-2: aisvaryasya samagrasya rupasya yasasah sriyah /jnanasyatha prayatnasya sanndm bhaga iti smrtir iti bhagayogdd bhagavan aham; 4) SpPr 89, 2^4: aisvaryasya samagrasya jndnasya yasasah sriyah / vairdgasya ca moksasya sanndm bhaga iti smrtah // Codd. a (A F H E C D J); A, (3 (F H E C D J); Y (F H), 6 (E C D J); e (E C), C (D J). Lacking: B (fol. 1-13 lost). 1 2 3 4 5 6 7 8 9 10 1] 12 13 14 15

bhaga codd. (*gya A) avrti conj. (AAA) : avrtiscodd. : [nyonmongs shesbya'i] sgrib Tib.fl86.4JJ yena A : ye ca cett. vaipaksika A : vaipaksikam 6 : vaiparikay bhagnas A : bhagna (3 smrtah AH2 : smrtah (3 iti // anena em. : ity anena AH2 : ityane HE : ityene FCDJ (ityenai J) nairuktena Ae : tairuktena C • nairuktetad(stad H) y bhagnavan AJy : bhagnavann CDE bhaga van P : om. A jnanasyatha E AAA; AbhayP; CSViv : jMnasyartha (-ar*ma* A) cett. bhaga iti A AAA; AbhayP; CSViv, SpPr : iti bhaga(gah J) (3 smrtih AH2 CSViv : smrtah (3 SpPr : sYutih AAA AbhayaP ity Ay : iti ity 5 tathatam codd. (tathata A)

222

Adhikara I

Text

Sunyatam 1 gatah tathagatah / atha va yatha te tathagata gata gacchanti2 gamisyanti tatha gacchatiti tathagatah / tarn tathagatam pranamyeti purvena sambandhah / agratah3 sthito bhutva krtanjaliputah4 san bhagavan vajradhara idam aha //

170

[maddhitaya mamarthaya anukampaya me vibho / mayajalabhisambodher yatha labhl bhavamy aham // (7)] maddhitayetyadi / mahyam hitam maddhitam hitam5 ayatipathyam6 agamiparinamatvat7 // mamarthaya matprayqjanaya8 me mama he vibho anukampaya anugrahaya yatha yena prakarenaham9 labhi bhavami

175

/

kasyety aha / mayajalabhisambodher iti / mayajalabhisambodhir10 nama kramas tasya11 yathaham labhi bhavami tatha 12 bhagavan13 praka£ayatv iti vaksyamanena14 kriyapadena sambandhah //

Codd. a (A F H E C D J); A, p (F H E C D J); y (F H), 6 (E C D J); e (E C), C (D J). Lacking: B (fol. 1-13 lost). 1 2 3 4 5 6 7 8 9 1° 1J 12 13 14

sunyatam codd. (sunyata A) gacchanti A : om. cett. agratah corr. : agrata codd. putah AJ : puta P hitam A : him eF : hit JH : hin D ayati codd. (ayeti JH) agami AE : agami cett. matprayojanaya A5 : mahayojanayay prakarenaham corr. : prakarena aham codd. sambodhir A : sambodhi P tasya A : tasya P tatha 6 : tatatha A : tapoy bhagavan A6 : bhagaya y vaksyamanena J : vaksyamanena A : vaksamanena yeD

223

Adhikdra 1

Text

[ajnanapankamagnanam

kle&avyakulacetasam

/

hitaya sarvasattvanam anuttaraphalaptaye // (8)] ajnanapankamagnanam sarvasattvanam hitaya ca / tatrajnanam 1 avidya

180

saivasangasthanatvad2 duruttaratvac3 ca panka iva pankah4 / tasmin magna ye sattvas tesam hitaya // punar api kimvisistas te sattva ity aha kle£avyakulacetasah / kle£a sat5 * / tatha cahabhidharme / manadrgvicikitsaS6 ca ragapratighamudhayah / iti // 185

taih klesair vyakulani cetamsi yesam sattvanam te 7 kleSavyakulacetasah / atas tesam // anuttaraphalapraptaye bhagavan8 praka§ayatv iti sambandhah / anuttaraphalam buddhatvam tasya praptih pratilambhah / tadartham ta^)raptaya iti //

[praka&ayatu sambuddho bhagavan* gasta jagadguruh

/

mahasamayatattvajna indriya^ayavit parah // (9)]

190

prakaSayatu bhagavan ityaha9 / sambuddha iti SaSvatadvarena // jagadanuiasanac chastety aksobhyadvarena // jagatam gurur10 jagadgurur iti ratnasambhavadvarena // samayo 11 bhagavan £rivajrasattvah / maham^ casau samayas ca mahasamayah / tasya tattvam mahasamayatattvam /

Codd. a (A F H E C D J); A, P (F H E C D J); Y (F H), 5 (E C D J); e (E C), C (D J). Lacking: B (fol. 1-13 lost). 1 2 3 4 5 6 7 8 ^ 10 1*

tatraA : tatraY : tamtra(-tram C) 5 (tamca J) saivasanga A : saivasanga (3 duruttaratvac A : duttaratvac 6 : dubhrratvac H (duttaratvac H2) : dubhrratva F pankah A : panka (3 (pankas JH) kle^ah sat A Tib.([l 86.4.8] nyon mongs pa drug go) : kleSa ragadayah sat (3 vicikitsas A : vicikitsa P teA : om.. (3 bhagavan A : om. P ityaha A : ityadi P gurur Y : guruh A : guru 6 samayo AS : samaya Y

224

Adhikdra 1

Text

195

taj janatiti mahasamayatattvajna iti / amitabhadvarena // indriyani mrdumadhyadhimatrani 1 / tesam asayo2 'bhiprayah3 / tarn vettiti indriyaSayavid ity amoghasiddhidvarena // evam pancatathagatatmako bhagavan para utkrstah prakaSayatv ity arthah4 //

[bhagavan jnanakayasya mahosnisasya gi spat eh

/

manjuSrijnanasattvasya jnanamurteh svayambhuvah // (10)]

200

atha bhagavan5 * jnanakayasyetyadi 6 // he bhagavan manjuSrijnanasattvasya sambandhim 7 ya8 namasamgitis tarn caham dharayisyamiti vaksyamanena kriyapadena sambandhah // kimviSistasya jnanakayasyeti / jnanam eva kayo jnanakayas tasya jnanakayasyety9 adar^ajnanadvarena // 205

mahosmsasyeti / mahan 10 usmso 'syeti mahosmsas tasya11 mahosnisasyeti samatajnanadvarena // mahosnisarvam sarvatathagatabhisekatmakatvat // gispater 12 iti / giram sabdanam patih svami gispatis tasya gispateh pratyaveksanajnanadvarena // jnanamurter13 iti / jnanam eva murtir jnanamurtis

Codd. a (A F H E C D J); A, P (F H E C D J); Y (F H), 6 (E C D J); £ (E C), C (D J). Lacking: B (fol. 1-13 lost). 1 2 3 4 5

6 7 8 9 10 1J 12 13

matrani A : matratani 3 asayo (3 : asayad A 'bhiprayah AJ : bhipraya yeD (bhiprayas H2) ity arthah A : iti (3 : zhes pa dang 'brel to ('brel lo Tib.P[187.1.1J)Tib.D.[32r3J bhagavan J : bhagavan cett. : bcom Idan 'das kyi [ye shes sku] {bhagavaj-} Tib. [187.1.1] jnanakayasyetyadi conj. (Tib. [187.1.1] ye shes sku zhes bya bala sogs pas ni) : jfianasattvasyetyadi codd. sambandhini A : tambhavani (3 (sambhavam E) ya A6 : yo Y tasya jnanakayasyety A : tasyety 0 mahan A : mahamS casau P tasya mahosnisasyeti samatajnanadvarena om. F gispater A : gflipater (3 murterAH : murtir 5 (murtir C) : mu< >F 4 Unanamu]rter iti / jnanam eva murtir om. F

225

Adhikara 1

Text

tasya jnanamurter iti krtyanusthanajfianadvarena // svayambhuva 1 iti 210

svayam bhavatiti svayambhur bhavanabalad2 bhavatity arthah / tatha coktam Sriparamadye / samantabhadrajnanasya3 svayam evavabhotsyate / iti4* // atas tasya svayambhuva iti suviSuddriadharmadhatujnanadvarena // kim uktam bhavati / pancajnanatmako bhagavan manjuSrijnanasattvah /

215

manjuh5 komala £rir yasyasau manjusiih / jnanasattva iti / sarvatathagatahrdayaviharitvat jnanasattvah

/

/

manju£ri£ casau jnanasattva^ ceti 6 manjuSri-

nayam daSabhumiSvaro bodhisattvah7 kim tarhy8

advayajnanam prajnaparamita saiva maiijuSrijfianasattvah // ata evaha dignagapadah / prajnaparamita jnanamadvayam9 sa tathagatah / iti 10 //

220

tasya manjuSrijnanasattvasya pancajnanatmakasya sambandhinl 11 ya 12 namasamgitis tarn caham 13 dharayisyamiti 14 //

222 cf. NMAA 1, 202: namasamgitis tarn caham dharayisyamiti

Codd. a (A F H E C D J); A, (3 (F H E C D J); Y (F H), 6 (E C D J); e (E C), C (D J). Lacking: A, lines 222, (fol. 8-9 lost); B (fol. 1-13 lost); A, fol. 7v, ends line 222 ... tarn caham dharaya. 1 2 3 4 5 6 7 8 9 10 1] 12 13 14

bhuva A : bhuvar (3 (bhur H) bhavanabalad A6 : bhavanadbalad y samantabhadra ADC : Y£DC : om. J evavabhotsyate / iti conj. : evavabhotsyatheti A : evavabhautsatheti (3 : [kun tu bzang po ye shes kyi / / bdag] nyid mngon par byang chub bo(em. : po Tib.) Tib. [187.1.6] mafijuh corr. : mafiju codd. (manjum F) ceti A : ca P bodhisattvah -* manjusrijnanasattvah om. H (suppl H2) tarhiAEC : tahyCy : tajjeH2 advayam A : advaya P tathagatah / iti em. : tathagata iti A : tathagatam iti P sambandhinl A : sabandhani P ya AJy : yo £D tarn caham om. F dharayisyamiti corr. (NMAAfl, 202]; Tib.[187.2.3]bdag gis gzung(bzung Tib.P.) ngo shes pa) : dharayamiti P : dharaya< > A

226

Adhikdra 1

Text

[gambhirartham udarartham mahartham asamam Sivam / adimadhyantakalyamm namasamgitim uttamam // (11)] kimviSistam ity aha // gambhirartham ityadi* / gambhiras' casav arthaS

225

ca gambirarthah Sunyatarthah1 sa2 yasyam asti sa gambhirartha / atas tarn dharayisyamlti3 // udarartham iti / udaraS4 casav arthas" codarartho5 * vaipulyartha6 iti yavat / sa yasyam 7 vidyate sa udarartha / tarn ca // mahartham8 iti / mahan artho9 yasyam 10 sa mahartha / maharthatvam 230

punar niravaSesasattvarthata 11 sarvasaparipuranatvat 12 / 13 * // asamam iti / na vidyate sama yasya sasama 14* / asamatvam kutah / dharmadhatusvabhavatvat / 15 * // Sivam iti sarvaprapancopaiamatvat // adimadhyantakalyamm iti / sYutacintabhavanakalesu harsapriti-

227 cf. NMAA 1, 202: ndmasamgitis tarn caham dharayisyamlti

Codd. (3 (F H E C D J); Y (F H), 6 (E C D J); E (E C), C (D J). Lacking: A (fol. 8-9 lost), B (fol. 1-13 lost).

1 2

sunyatarthah ye : om. C sa 5 : se Y

12 13 14 15

sarvasaJC : sarvasa E : sarvasamDH2 : om. F tarn ca conj. (NMAA[1, 228]) : om. codd. sasama conj. : sa ity asamam codd. (om. F; ity asamam H 2) tarn ca conj. (NMAA[1, 228]} : om. codd.

dharayisyamlti conj. (NMAAfJ, 202]; r/fc.^/£7.2.5ybdag gis gzung zhes pa'o) : dharayamlti codd. 4 udara^ —>• mahartham iti om. E 5 udara^ casav artha^ codarartho conj. : udara < -14 > F : udarartham < ~8 >rtha H : udararthamrtha C : udarartham udarartha D : udarartham udarartha JH2 : om. E Tib. 6 vaipulyartha 5H2 (om. E) : masamam iti / na vidyate sama yasya ity asamam / asama< > kutah vaipulyartha Y (Y omii asamam -» kutah, line 231) 7 sa yasyam conj. : sa yasya F : sa yasya H : yasya sa 5 (om. E) 8 mahartham corr. : mahartham codd. (om. E) 9 artho Y£ : artha C 1 ° yasyam corr. : yasya codd. 11 niravas~esa corr. : niravaiSesa Y : niravasesasya 5 (niravas~evasya C) 4 sattvarthataJH2 : sattvathatah cett. (om. F) 4 [nirava^esajsattvarthata -» asamatvam kutah om. Y 3

227

Adhikdra J

Text

praSrabdhilabhat 1 trikalyanini2 / tarn dharayisyami 3 // namasamgitim 235

uttamam iti gatartham4 //

[yatltair bhasita buddhair bhasisyante hy anagatah / pratyutpannag ca sambuddha yam bhasante punah punah // mayajalamahatantre ya casmin sampragiyate / mahavajradharair hrstair ameyair mantradharibhih

//

aham cainam dharayisyamy a niryanad drdha&ayah /

240

yatha bhavamy aham natha sarvasambuddhaguhyadhrk

//

(12-14)] punar api kimviSistam5 ity aha / yatltair ityadi6 / atltair buddhair ya bhasita anagataS ca ye samyaksambuddha hi yasmad yam7 bhasisyante 245

punah 8 * / pratyutpannaS ca buddha bhagavanto bhasante yam9 nama­ samgitim 10 punar1 ] * apiti //

234 cf. NMAA 1, 202: namasamgllis Idm caham dhdrayisyamiti 243 cf. NMAA 1, 225: kimvisistdm iyddi. Codd. (3 (F H E C D J); y (F H), 6 (E C D J); e (E C), C (D J). Lacking: A (fol. 8-9 lost), B (fol. 1-13 lost). praSrabdhilabhat corr. : pra&bilabhat e : praSabitvabhat (-bhavat J) C : pras"arddhi bhavat y 2 trikalyanini 6 (trikalyam J) : trikatvoninl y 3 dharayisyami conj. (NMAAfl, 202] Tib.[187.2.8] bdag gis gzung ngo) : dharaya mi codd. 4 gatartham eD : gatarthahJH : gatarma F 5 kimvis~istam conj. (NMAAfl, 225]) : kimviSistam codd. 6 yatltair ityadi y : om. 6 7 yam 6 (om. J) : ya y 8 punah ye : punah punah C : om. Tib. 9 yam conj. : ya codd. 10 namasamgitim corr. : namasamglti codd. (namasamgitih H2 ) 1J punar codd. : yang dang yang du {punah punar} Tib.[187.3.2] 1

228

Adhikdra I

Text

mayajaletyadi 1 * / asmin mayajalamahatantre 'meyair2 mahavajradharair mahamantradharibhir3 hrstair ya sampragiyate tarn // aham cainam dharayisyamy a niryanad nirvanam yavat / kimbhuto4 250

drdhaSayah / drdho5 'vicala6 a£ayo 'bhiprayo yasyasau drdhaSaya iti // anenaitad uktam bhavati / tryadhvavartibhis7 tathagatair ya bhasita * namasamgitih8 / ato 'sya dharmaparyayasyakaSatulyatam9 * samdarSayati / tatha coktam / utpadad10 va tathagatanam anutpadad 11 va < tathagatanam > 12 * sthitaivaisa 13 dharmanam dharmateti pradaiiitam bhavatiti 14* //

255

yatha 15 bhavamy aham ityadi / he natha yatha sarvasambuddhaguhyadhrg bhavamy aham tatha prakaSayatv iti purvena sambandhah / dhrg ity ayam ^uddhapathah16 //

254-5 SekUdTl 71: utpadad va tathagatanam anutpadad va tathagatanam sthitaivaisa dharmanam dharmateti; AKBh 452, 9-10 (on AK 3, 28ab): utpadad va tathagatanam anutpadad va tathagatanam sthitaiveyam dharmata iti vacanat. Codd. P (F H E C D J); y (F H), 6 (E C D J); £ (E C), C (D J). Lacking: A (fol. 8-9 lost), B (fol. 1-13 lost). 1 ^ 3 4 5 6 7 8 9 10 11 12 13 14 15 16

mayajaletyadi (mayam- H) yC '• mayajalatyadi (ani E) e 'meyair corr. : ameyair codd. dharibhir corr. : dharibhih codd. kimbhuto J : kimbhuta cett. drdho H : drdha cett. 'vicala H : avicalacetf. tryadhva yC : adhva e namasamgitih y : namasamglti 6 tulyatam conj.(Sanderson) : tulyam DEy : tulya CJ : [nam mkha'] dang 'dra bar Tib.[187.3.3] utpadad corr. (SekUdTT; AKBh) : upadad y : upada 5 anutpadad y : anutpada 6 tathagatanam conj. (SekUdTi; AKBh) : om. codd. Tib.[187.3.3] sthitaivaisa J (cf. SekUdTi) : sthitivasa (-rsa H) y : sthitaivesam D : sthitaivaisam e bhavatiti conj. (Tib.[187.3.4] [bstan par] 'gyur ro zhes so) : bhavati codd. yatha y : tatha 5 suddhaHE : saddha CFC

229

Adhikara I

Text

[praka&ayisye sattvanam yatha&ayavi&esatah

/

(15ab)j

yathavat tarn aham adhigamya1 sarvasattvanam yatha£ayavisesat pra-

260

kaSayisye // kimartham ity aha /

a&esakle&anagaya aSesajnanahanaya iti

/ (15cd)

tatra asesaklesa iti2 kleSopaklesa grhyante tesam na£aya / aSesajnanahanaya iti a£esesu padarthesv ajnanam as"esajnanam aSesajnanahanis tadartham 265

prakaSayisye //

[evam adhyesya guhyendro vajrapanis tathagatam / krtanjaliputo bhutva prahvakayasthito 'gratah // (16)] evam ityadi / evam anena prakarena bhagavan guhyendro3 vajrapanis tathagatam adhyesya4* / tatra guhya yaksa adakavatlnivasinas5 tesam indro 270

raja / ata eva guhyendra ity ucyate // krtanjaliputo bhutva prahvakayah pranatakayo 'gratah6 sthita iti // samskarapurvika7 prarthana adhyesana / samskaro8 'tra pranamadi // aryanamasamgititikayam namamantrarthavalokinyam adhyesanaya prathamo 'dhikarah // gathah9 soda^a10 //

Codd. (3 (F H E C D J); Y (F H), 5 (E C D J); e (E C), C (D J). Lacking: A (fol. 8-9 lost), B (fol. 1-13 lost). 1 2 ^ 4 5 6 7 8 9 10

adhigamya Y£ : adhigatya C aSesaklesa iti YE : atesa C guhyendro yC '• guhyandro e adhyesya corr. : adhyesye codd. adakavatl5r/'6.(/7#7.4.4/lcanglocan) : adakavataY 'gratah corr. : grata codd. ('grata J) samskara 6 : satkara Y samskaro 6 (samskara J) : satkara Y (corr. H2) gathah corr. : gatha codd. sodaSa Y : sodaSah 6

230

Adhikdra 2

Text

[Adhikara 2]

[atha gakyamunir bhagavan* sambuddho dvipadottamah / niraamayy ayatam sphitam svajihvam svamukhac chubham // (17)] adhyesananantaram prativacanam aha / atha s"akyamunir ityadi / 5

Sakyakulanam munih kayavakcittamauneyayogan munih / bhagavan sambuddha iti gatarthah / dvipadottama iti dvipada devamanusyah 1 / tesam uttamah srestha2 ity arthah / svajihvam nirnamayyopasamhrtya3 sva­ mukha d iti4 svamukha5 ayatam iti dlrgham6 sphitam iti bahalataram Subham iti daksinavarta^ankhanibham / sarvadharmalayam ca //

10

[smitam samdargya lokanam apayatraya^odhanam trailokyabhasakaranam caturmararigasanam

//

/

(18)]

kdm krtva / smitam ityadi* / lokanam sattvanam apayatraya^odhanam narakatiryakpretasamsodhanam7 smitam dar^ayitva vajrapanim pratyabhasateti sambandhah // tathagatanam dharmades^nayas8 trilihgani9 / tad 15

yatha / smitam vyavalokitam raSmisamcodanam ca / kvacid ekam 10 kvacid dve 11 kvacit trini va yugapad 12 bhavanti / tad yatha / Srimayajala-

Codd. (3 (F H E C D J); y (F H), 6 (E C D J); e (E C), C (D J). Lacking: A (fol. 8-9 lost); B (fol. 1-13 lost). 1 2 3 4 5 6 1 8 9 10 11 12

manusyah J : manusya cett. (manusyam H) grestha JF : sresta cett. nirnamayyopa Dy : nirnamayyapa eJ svamukhad iti conj. : svamukhadibhih codd. svamukha y : svamukhe 6 dirgham J : dirgha cett. sam^odhanam 6 : sarnsadhatam y des"anayas y : des^nayas 6 triCC : triyE kvacid ekam H : kvacidd ekam cett. kvacid dve corr. : kvacid dvekam JF : kvacid ekam H : kvacidd ekam eD yugapad 6 : yutapa y

231

Adhikdra 2

Text

tantre dvayam 1 pathyate / smitam vyavalokitam2 ca / kvacit trim' yatha pancavimsatisahasryam prajnaparamitayam maitreyavyavalokanam smitam3 mukhena ra£mispharanam ceti / kvacid ekam yathatraiva namasamgitau 20

smitam samdarSya4 lokanam iti / tac ca smitam caturvidham 5 / arigapratyangaspharanena romakupaspharanena prakrtiprabhavispharanena jihvendriyarasmispharanena ca / tad atra jihvapratiharyam / deSanayah6 purvalingam7 // trailokyabhasakaranam8 ityadina darSayati bhur bhuvah svar iti trayanam lokanam abhasakaranam9 prakasanakaranam smitam dar£ayitva

25

vajrapanim pratyabhasateti sambandhah // punar api kimvisistam 10 smitam ity aha caturmarari£asanam iti / catvaro marah skandhaklesamrtyudevaputrah / ta evarayas11 tesam sasanam anusasanam nigrahah // [trilokam apurayantya brahmya madhuraya gira / pratyabhasata guhyendram vajrapanim mahabalam // (19)] smitam samdarSyeti 12 trilokam apurayantya vajrapanim gira 13 vaca pra­

30

tyabhasata14 / kimbhutaya / madhuraya / madhuryam 15 srutisukhatvat16 //

Codd. A, p (F H E C D J); y (F H), 6 (E C D J); e (E C), C (D J). Lacking: A, lines 17-25 (fol. 8-9 lost), B (fol. 1-13 lost); A, fol. lOr, starts line 25: sated sambandhah //. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

dvayam em. Tib.([188.1.1] gnyis) : 'dvayam codd. ('dvaya F) vyavalokitam codd. (-katamy; -kitam H2 ; -kanam C0) smitam conj. : smita codd. Tib.([1 SB. 1.2] 'dzum pa'i [zhal dang]) samdar^ya 6 : darsya C caturvidham 6 : caturvidhahy jihvapratiharyam / desanayah e : jihvapratiharya / deSanayah (dasa- H) y : jihvaprnayah J : jihva*pri*spharna desanayah D purvalingam corr. : purvalingaEY : purvam lingam (ligam D) C abhasakaranam em. Q^S.Dav.[50, v.18]; Tib.[187.5.4] ['jig rten gsum po] snang byed) : abhaskaranam codd. abhasakaranam em. (NS.Dav.f50, V.18J; Tib.[187.5.4] snang zhing gsalbar byed pa['i 'dzum ba bstan nas]) : abhaskaranam codd. (abhaskarana H) kimvisistam 6 (kimvisirsta C) : kimvisistah y ta evarayas Ay : trayavarayas 6 samdarsyeti em. : samdasyetiA : dar^ayati P giraeF : giramAC : §iroH pratyabhasata AT : pratyabhasatah eyD madhuryam AJ : madhuryam eyD (madhurya H) srutisukhatvat A : Sruti(ih E)sukhetva £ : srutimukhetva C : srutimukhatva y

232

Adhikdra 2

Text

brahmyeti 1 brahmasvaropetatvat2 / brahmasvara iti brahmasabdo mahato mahagunasyadhivacanam / yathoktam / brahmam3 punyam prasuyate4 'nekasahasraparsadi / iti5 * // 35

sahasraSabdena6 trisahasramahasahasro lokadhatur avagantavyah / anekasahasramayyam7 parsadity8 arthah // yatha9 coktam / lokadhatum 10 trisahasramahasahasram11 Sabdena spharatiti / yathabhyantarah Srnvanti tatha bahya apiti 12 / tatrabhyantara ye 'smims 13 trisahasramahasahasre 14 sthitah / bahya iti ye 15

40

lokadhatvantaritah16 / tena mahatrvad brahma / tasya17 svaro18 brahmasvarah19 * // atha va20* brahma21 bhagavan Sakyamunih / tasya brahmana iyam

Codd. a (A F H E C D J); A, p (F H E C D J); Y (F H), 5 (E C D J); e (E C), C (D J). Lacking: B (fol. 1-13 lost). 1 2 3 4 5 6 7 8 9 1° 11 12 13 14 15 16 17 18 19 20 21

brahmyeti em. : brahmeti AD : brahmeti CF : om. J : brahmati HE brahmasvaropetatvat om. J brahmam DH : brahmyam AeJF : prasuyate A : prasuyate iti D : iti eJ : < ~3 >Y : mahar*dya*m ca iti H2 : rab skye ba'o zhes Tib. [188.1.5] 'nekasahasra conj. : anekasahasram p : anekasahasrya A * parsadi / iti em. : parsaditi A : parsadam P sahasra A : sahasra (3 mayyam conj. : maryyam A : mayl P parsadity AJ : parsad ity ytD yatha A Tib.(JJ88.1.6J ji skad gsungs pa) : tatha P lokadhatum A : lokadhatu P mahasahasram A : mahasahasra P bahya apiti A : bahyapiti p (bahyaniti E) ye 'smirns corr. : yesmin A : yasmim p tri sahasra AH : trisahasre5F ye A : om. P lokadhatvantaritah A : lokadhatur ita P (lokadhatur iti E) brahma / tasya om. P svaro conj. : suro A : om. P brahmasvarah p : brahmasVaram A atha va conj. (Tib. [188.1.8] ywg na) : tatra codd. brahma P : brahma A

233

Adhikara 2

Text

brahmi 1 vak / taya brahmya2 ekayaiva3 gira4 trilokam5 apurayantya 6 vyapnuvantya7 vajrapanim pratyabhasata / tatha coktam / ekapi vani bhavasudanasya

45

nanabhidhanaih pratide£abhinnaih / yatha tatha8 sarvajanaprabodhahetur9 bhavaty askhalita10 vibhakteti 11 //

[sadhu vajradhara* Sriman* sadhu te vajrapanaye / yat* tvam jagaddhitarthaya mahakarunayanvitah // (20)

50

mahartham namasamgitim pavitram aghana&amm / manjusrijnanakayasya mattah srotum samudyatah // (21) tat sadhu desayamy esa aham te guhyakadhipa* / (22ab)j sadhu vajradharetyadi / te tava vajrapaneh sadhu / Sriman he vajra55

dhara yat yasmat tvam manjusrijnanakayasya sambandhini 12 ya namasamgitis tarn 13 mattah 14 Srotum samudyatas tat tasmat he guhyakadhipa te tava15 sadhu desayamiti sambandhah // kimartham tvayaham 16 prstah /

Codd. a (A F H E C D J); A, p (F H E C D J); Y (F H), 5 (E C D J); e (E C), C (D J). Lacking: B (fol. 1-13 lost). 1 2 3 4 H4 5 6 7 8 9 10 1] 12 13 14 15 16 ^

brahmi vak A : brahmi vak 0 brahmya em. : brahmya F : brahme H : brahma A : brahma 6 /jvci>» :. eka eva J : ekayava H : ekayeva F ekayaiva CK.ayaivii AeD oiram fi A : giram otra A p(3 gira trilokam A : trailokyam (3 apurayantya AE : apuryantya CC : apuryantyo Y vyapnuvantya A : vyapnuvantya 5 : vyapnuvantyo Y tatha A : yathaa ey : om. C scad) : sattva P + heturAY : hetuS (fl88.2.2Jiham hamscad) sarva AH2 Tib.(fl88.2.2Ji t >lit§Y : < >lit§ askhalita AH22 : astvalitaS vibhakteti AY : vibhaktah (-kta JE) iti 5 sambandhini AH2 : sambandhani P tarn AC : ta Y£ mattah A : mamtrahS : mantahY tava A : tatra P tvayaham A5 : bhuya(yo H)ham Y j*11

.A

.*__'

234

Adhikara 2

Text

jagaddhitarthayeti / jagadanugrahaya mahakarunayanvitah 1 san // kimvi£istarn namasamgitim Srotum udyata iti / pavitram iti mangalyam2 / 60

aghanasanim3 iti papanasanim4 //

[§rnu tvam ekagramanas tat sadhu bhagavann iti // (22cd)j sYnu tvam ekagramana iti 5 / anenaitad uktam bhavati / madhomukhaccliidraSucighatavat6 kim tarhi sadhu sYnu // bhagavataivam ukte sati vajradharah praha bhagavan7 Sakyamune8 evam astv iti9 / yathoktenaiva10 65

kramenaham srnuyam iti // aryanamasamgltitlkayam namamantrarthavalokinyam prativacanadhikaro dvitiyah // gathah sat1 1 II

Codd. a (A F H E C D J); A, (3 (F H E C D J); y (F H), 5 (E C D J); e (E C), C (D J). Lacking: B (fol. 1-13 lost). 1 2 3 4 5 6 7 8 9 10 11

karunayanvitah (3 (-amnvitah C) : karananvitah A mangalyam A : mangalyam (3 aghanasanlm AH2 : aghanaSinlm 6 : aghana&nam y papanasanim A : papana^inim ^ iti AH2 : om. P madho A : sadho (3 4 cchidra AH : cchida cett. bhagavan AH2 : bhagavan (3 Sakyamune AH2 : Sakyamunih 3 (-muni E) astv iti A6H2 : asthiti y yathoktenaiva Ay : yathoktanaiva J : ethoktanaiva eD gathah sat corr. : gathasatSH 2 : om. Ay

235

Adhikara 3

Text

[Adhikara 31

[atha Sakyamunir bhagavan* sakalam* mantrakulam mahad / mantravidyadharakulam vyavalokya kulatrayam //

(23)]

idanim prativacananantaram satkulavalokanam adhikrtyaha / atha Sakyamunir bhagavan1 < ityadi / tatra>2 * sakalam3 * mantrakulam mahad 5

iti grahanakavakyam etat samanyenopattam4 / tad5 vyavalokya giram pater manjusnjnanasattvasya6 sambandhini7 yanutpadadharmini gatha tarn imam bhasate smeti / vaksyamanakriyapadena saha8 sambandhah9 // mantravidyadharakulam iti

/

vidyam dharayatiti vidyadharah

/

mantratmakaS casau vidyadharah ca 10 mantravidyadharah / na 11 dharany10

atmakah12 / tasya kulam mantravidyadharakulam karmakulam ity arthah // kulatrayam iti kayavakcittasvabhavatvad vairocanakulam 13 * //

1 cf. NS.Dav. 51, v.23ab: atha sakymunir bhagavdm savakalam mantrakulam mahat

Codd. a (A F H E C D J); A, (3 (F H E C D J); y (F H), 5 (E C D J); e (E C), C (D J). Lacking: B (fol. 1-13 lost). 1 2 3 4 5 6 7 8 9 10 11 12 13

bhagavan codd. : bhagavamNS.Dav./^y, v.23] ityadi / tatra conj. (Tib.[188.3.2] ces pa la sogs pa ste / de la) : om. codd. sakalam A NS.Dav. [5J, v.23] : sakala (3 NS.Dav. Ne- A samanyenopattam AH2 : samanyatapattam e : samanyanapattat C : samanyanamattat F : samanyenomattat H tad AH2 : ta ye : na C jnanasattvasya Ay : hinasattvasya 5 sambandhini AH2 : sambandhani (3 saha A : om. (3 sambandhah A : sambamdharthah (3 caA : ceti (3 na AH2 : om. 0 dharanyatmakah A : dharanatmakah 0 vairocanakulam codd. : rnam par snang mdzad kyi rigs zhes pa'i don {vairocana­ kulam ity arthah} Tib.[188.3.5]

236

Text

Adhikara 3

[lokalokottarakulam lokalokakulam mahat / mahamudrakulam cagryam mahosnisakulam mahat // (24)] lokalokottarakulam 1 iti / lokr2 darSane3 / lokyata4 iti lokah / tasya 15

kulam lokalokottarakulam5 padmakulam ity arthah // lokalokakulam mahad iti / lokanam aloko lokalokah / aksobhyah / adarsajnanatmakatvat / tasya kulam lokalokakulam 6* vajrakulam7 ity arthah8 * // mahamudrakulam cagryam iti / bodhicittavajrakulam ata evagryam9 vairocanadlnam bodhicittasvabhavatvat / tasya kulam maha-

20

mudrakulam 10* ity arthah 11 * // mahosnisakulam mahad iti / sarvatathagatanam traidhatukadharmarajyabhisekapradanan12 mahosnisatvam / tasya kulam 13 mahosnisakulam 14 mahad15 ratnakulam ity arthah // aryanamasamgititikayam namamantrarthavalokinyam satkulavalokanadhikaras trtiyah // gathe 16 dve //

Codd. a (A F H E C D J); A, P (F H E C D J); Y (F H), 6 (E C D J); e (E C), C (D J). Lacking: B (fol. 1-13 lost). 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

lokaACF : loka CE : laukaH lokr H2 : lokya a darsane Aey : dar^ana C lokyata Ay : lokata 6 lokalokottara AEy : lokalokattara C : lokalokottara C mahad conj. (Tib.[188.3.8J chen) : om. codd. vajrakulam AyD (Tib.[188.3.8] rdo rje'i rigs) : cakrakulam eJ ity arthah H2 : iti a : om. Tib. evagryam A : evagryo P mahamudrakulam codd. : phyag rgya chen po'i rigs mchog {mahamudrakulam agryam} Tib.[188.4.1] ity arthah H2 : iti a : om. Tib. dharmarajya AH2 : dharmaraja P mahosnisatvam / tasya kulam AH2 : mahosnisa / tasya kulam y : om. 6 mahosnisakulam FtfS : om. AH mahad A5H2 (mahata C) : mahat aramH : mahamtamF gathe AH2 • gatha p

237

Text

Adhikara 4

[Adhikara 4] adhuna ^rimayajalabhisambodhikramo 1 guruparamparayata2 ucyate / sa ca SrimayajalasodaSasahasrantahpati3 / tatradau tavan mantramukhacaryacari imam bhagavato manjuSrijnanasattvasya sarvatathagatajnanakayasya jrianamurter advayaparamartham4 namasamgltim nama cudamanim* 5

saksatkartukamo bodhisattvo 'SesanavaSesasarvasattvadhatuparitranaSayo5 mahabodhikrt6* pranidhanacittah prajnopayatmakas caturbhih samgrahavastubhih sattvan samgrhya anuttarayam7 samyaksambodhau cittam u^)adayet8 // tatha caha / samanvaharantu mam sarvabuddhabodhisattva aham amukanama9

10

imam velam upadaya yavad a bodhimandanisadanat 10 / utpadayami paramam bodhicittam anuttaram / yatha traiyadhvika 11 * nathah 12 sambodhau krtani^cayah 13 //

2-4 cf. NS.Dav. 65, 3-5: mantramukhacaryacarl imam bhagavato manjusrljnanasattvasya sarvaiathagatajndnakdyasya jnanamurter advayaparamariham namasamgltim namacudamanim 12 cf. SDPS 146, 9: yatha traiyadhvikanathdh sambodhau krtaniscaydh.

Codd. a (A F H E C D J); A, (3 (F H E C D J); y (F H), 5 (E C D J); £ (E C), C (D J). Lacking: B (fol. 1-13 lost).

1 2 3 4 5 6 7 8 9 10 11 12 13

kramo A : kramah H2 : krama (3 guru A : Sriguru [3 sahasrantahpatl A : sahasryantahpatl y : sahasryantahpata 5 paramartham AH : paramartha 5F 'SesanavaSesa A : as"esavas"esa yC : a^esovatesa E 4 paritranaSayo A : paritrana nayo 6 : paritrananaya y krt A Tib.([l88.4.6-7] byang chub chen por smon pa'i sems) : krta p samgrhyanuttarayam con. : samgrhya anuttarayam codd. utpadayet A6H 2 : utpadayat y amukanama(-nama Ii) codd. nisadanatew. : nisadanat A : nivedanat P traiyadhvika codd. : traiyadhvika SDPS[146, 9] na\hahASDPS[146,9] : natha(3 niscayah A : niscaya p (-^caya E, -Scayo F)

238

Text

Adhikdra 4

trividham silaSiksam ca kusaladharmasamgraham 1 * / sattvarthakriya^ilam2 ca3 pratigrhnamy aham drdham // 15

buddham4 dharmam ca samgham ca triratnagram anuttaram / adyagrena grahlsyami5 samvaram buddhayogajam // vajram6 ghantam7 * ca mudram ca pratigrhnami tattvatah / acaryam ca grahlsyami8 mahavajrakuloccaye // caturdanam pradasyami satkrrva tu dine dine /

20

maharatnakule yogye9 samaye ca manorame // saddharmam pratigrhnami bahyam guhyam triyanikam / mahapadmakule Suddhe mahabodhisamudbhave 10 // samvaram sarvasamyuktam pratigrhnami tattvatah / pujakarma yatha^aktya mahakarmakuloccaye //

13 cf. SDPS 146, 10: trividham silasiksam ca kusaladharmasamgraham 16 cf. SDPS 146, 13: adyagrena grahlsyami samvaram buddhayogajam 17 cf. SDPS 146, 14: vajraghantam ca mudram ca pratigrhnami tattvatah 18 cf. SDPS 146, 15: acaryam ca grahlsyami mahavajrakuloccaye 20 cf. SDPS 146, 17: maharatnakule yoge samaye ca manorame 22 cf. SDPS 146, 19: mahapadmakule suddhe mahabodhisamudbhave.

Codd. a (A F H E C D J); A, P (F H E C D J); Y (F H), 6 (E C D J); £ (E C), C (D J). Lacking: B (fol. 1-13 lost). 1

kuSala conj. (SDPS[146, 10]\ Tib.[188.5.2] dge ba'i [chos ni sdud pa dang]) : kusa lam codd. 2 sattvartha Aye : sattvarthe C 3 ca P : om. A 4 buddham AEy : buddha CC 5 grahisyami H2 SDPS[146, 13] : grhisyami cett. 6 vajram y : vajra A6 SDPS[146, 14] 7 ghantam AJ : ghantamny : ghantaeD 8 grahisyami H2 SDPS[146, 15] : grhlsyami ce//.(-hisyami CD) 9 yogye A : yoge P SDPS [146, 17] (yogya SDPSACD, yoga SDPSE ) 10 samudbhave AH2 SDPS[146, 19] : samudbhavam P

239

Adhikara 4

Text

utpadayitva 1 paramam bodhicittam anuttaram /

25

grhitva2 samvaram krtsnam sarvasattvarthakaranat // atirnams3 tarayisyami amuktan mocayamy4 aham / ana£vastan aSvasayisyami5 sattvan6 sthapayisyami7 nirvrtav iti // samvaragrahanapurvakam bodhicittam utpadya8 pancakarabhisam30

bodhim bhavayed anena kramena // prathamam tavad adar£ajnanam pratibimbayogena bhavayet / yatha parisuddhadar£amandale9 taddeSakalasamnihitah padarthajatayah pratibhasante 10 tatha suvis"uddhadharmadhatunisyande 11 mahabodhicitte gatide^ayonibhedabhinnan 12 sattvadhatun 13 pa^yams 14 tatra tan sunirmala 15 adarSajnane drstva tadupari sattva-

35

lambanam 16 karunam utpadayet // aho batami sattva anatha a^arana aparayana duhkhitah samsararnavanimagnas 17 tesam satrvanam arthaya

25 cf. SDPS 146, 22: utpadayitva paramam bodhicittam anuttaram 26 cf. SDPS 146, 23: grhitam samvaram krtsnam sarvasattvarthakaranat 27 cf. SDPS 146, 24: atirnams tarayisyami amuktan mocayisydmy aham 28 cf. SDPS 146, 25: anasvastan dsvasayisydmi sattvan sthdpaydmi nirvrttdv iti.

Codd. a (A F H E C D J); A, 3 (F H E C D J); Y (F H), 6 (E C D J); e (E C), C (D J). Lacking: B (fol. 1-13 lost). 1 2 3 4 5 6 7 8 9 1° 1J 12 13 14 15 16 17

utpadayitva A SDPS[146, 22] : utpadayami (3 SDPSAC grhitva A : grhitam (3 SDPS[146, 23] atirnams ACJH : atirnas DF : atirnan E SDPS[146, 24] mocayamy codd. : mocayisyamy SDPS[146, 24] asVasayisyami ACH : ams"vasayisyamieF sattvan conj. (SDPS[146, 25J, Tib.[188.5.7] [mya ngan 'das la] sems can [dgod]) : sattva A : sarvasattva 3 (-sattvam J) sthapayisyami codd. SDPSB : sthapayami SDPS[146, 25] bodhicittam utpadya A : bodhicittotpadya p dar&mandale A : darsanamandale £ : dar&mamandala YC : dar^anamandalais H2 pratibhasante A : pratibhasante P nisyande A : nisyande P gatides~a AY : de^a e : daSa C * bhinnanAY : bhinnat 6 sattva codd. (sattvan H2 ) pasyams corr. : pa^yan codd. (pasyen E) sunirmala y • sunirmale A6 alambanam P : avalambanam A samsararnava A6 : samsararnave y

240

Text

Adhikara 4

yathaham atmanam sarvena sarvam sarvatha sarvaduhkhebhyo 1 mocayisyami2 tatha sarvasattvan3 apity adarsajnane4 sarvasattvan asamvrtya paSyet5 // 40

tatah samatajfianam adhimuncet sarvadharmanairatmyasamatayogena6 / yatha sarvadharmah 7 Sunya anatmanas tathaham api Sunyo 'natmeti 8 #

vibhavya sarvadharmanairatmyam praviSet // tatha caha9 / yathatmanam 10 tatha 11 prajanati sarvasattvan 12 / yatha sarvasattvams 13 tatha prajanati sarvadharman ityadi // tatah sarvadharmanairatmyasvabhavadhigamad 14 45

dharmavalambanam 15 karunam samatajnanenotpadayet / tatas cakaro mukham 16 sarvadharmanam adyanutpannatvad iti mantram uccarya mantrarmam17 bhavayen mantri nikhilena samasatah / vibhavya dharmanairatmyam analambanaya saha //

50

evampratyaveksanajnanam 18 //

Codd. a (A F H E C D J); A, p (F H E C D J); Y (F H), 6 (E C D J); £ (E C), C (D J). Lacking: B (fol. 1-13 lost). 1 2 3

sarvaduhkhebhyo AyC •' sarveduhkhebhyo £ mocayisyami A : mocayami |3 sarvasattvan AH2 : sarvasattvatvan CF : sarvasattvatan E : sarvasattvasyan J : sarvasattvasvan D 4 adarsajnane A . adarsajfiana P (adar^ana- E) 5 pasyet AF : paSyat 6H 6 nairatmya A : nairatmye P 7 sarvadharmah A : sarvadharma P (sarvadharma H) 8 gunyo 'natmeti corr. : Sunyonatmeti A : ^unyanatmeti P 9 tatha caha H2 Tib.([189.1.5] de skad kyang gsungs pa) : om. cett. 10 yathatmanam em. : yathatmana H2 : yatha atmana 6 : yatha atmana A : yatha a< > Y

11 tatha em. (Tib.[189.1.5] de 'drar) : tatha A : om. P 12 sarvasattvan em. (Tib.[189.1.5-6] sems can thams cad) : sarvasarvasattvan AH : sarvasattva 5F 13 sarvasattvams em. : sarvasattva codd. 14 nairatmya AyE : nairatmye CC 15 lambanam P : lambanim A 16 mukham AH : mukha 5F 17 mantrartham A : mantrasattvartham £F : mantrasattvartham J : matrasattvartha D : mantra< >rtham H 18 jnanam A : jnana P ('mitabha E)

241

Adhikara 4

Text

samanantaram 1 evanalambanamahakarunanugatam sakalagrahyagrahakavinirmuktam prakrtiprabhasvaram2 pratyastamitavayavasvabhavam suviSuddhadharmacMtuSunyatajnanam3 bhavayet // tataS ca sarvasattvarthakriyavirahito4* yogi purvapranidhanahitacittasam55

tanavaSad utthapito 'nalambanaya5 mahakarunaya krtyanusthanajnanam bhavayet6 / nirvikalpakajnanaprsthalabdha^uddhalaukikajnanatvat7 tasya tenaiva sakalasattvarthaparisamaptih8 / tac ca prasthanalaksanam bodhicittam samvrtiparamarthasvabhavam prakrtivaiyavadanikam cittam iti veditavyam // tato yogi9 * cittam mayopamakaram aka£ade£astham prakrtiprabha-

60

svaram tarakakaram vicintya10* tadudbhutam krodhajvalamalakulam 11 nilavamam anekara^mi^atasahasraparivrtam samantabhadram paramanandam atmanam brhad dhumkaram* cintayet12 // tasmad dhumkarad anekakararasmayo 13 vinihsrtas 14 tebhyaS ca ra^mimukhebhyah sarvatatha65

gatas tais ca sarvatathagataih sarvabuddhaksetrani gatvasesanavaiesesu 15

64 cf. NMAA 4, 185-6: sadbhyo vajramukhebhyah sadvarna vinihsrtya (vinihsrtya A : vinisrtya P) Codd. a (A F H E C D J); A, P (F H E C D J); y (F H), 5 (E C D J); e (E C), C (D J). Lacking: B (fol. 1-13 lost). ' 2 3 4 5 6 7 8 9 10 1! 12 13 14 15

samanantaram -* pratyastamitavayava[svabhavam] om. E prabhasvaram AH : prabhasvara CF : prabhasvara C : om. E dhatu 6 : dhatum Ay kriyavirahito P : kriya'virahto A utthapito 'nalambanaya corr. : utthapitah / analambanaya codd. bhavayet A : saksatkuryat (3 laukikaAEH2 : lokika CyC sattvartha 3 : sattvartham A * parisamaptih A : parisamapte P (parisamapteh H2) yogi conj. (Tib.fl89.2.4] rnal-'byor pas) : yoginasrokurvata(-to F)ni y : dvivikurvitani H2 pravi^antiti conj. (Sanderson) : pravi&tah A : pravisanta(-tas J) 5 : praviSantam y vamam Aye : varanam C marari AyE : malari C : nalari C vajravaratakasya A : vajrava(dha H)rata(ta E)ka P paravrttya corr. : paravrtya A : paravrtam P (ravrtam H)

7

catu AJ : catu Dye

8 9 1° 1J 12 13

tadupary corr. : tadupari codd. akarena 0 r/6.(/7£PJ. 2-37 yige a las) : ahkarena A mayim AH : may! 5F varnam em. : vamam codd. (varna F) cintayet A : vicintayet P tac Ay : tarn 5

243

Adhikara 4

Text

caturaSram caturdvaram catustoranabhusitam / ghantapattapatakadisrakcamaravibhusitam 1 // konesu caiva sarvesu dvaraniryuhasarndhisu /

80

khacitam vajraratnais tu ardhacandrena samyutaih2 * // harardhahararacitam vicitrair upaSobhitam3 / rajobhih pancavarnais ca samlikhed4 bahyamandalam // tasya cakrapratika£am pravisyabhyantaram5 puram6 / vajradhatvabhidhanakhyam dvitiyam vinyaset tatah //

85

vajramalapariksiptam astastambhopaSobhitam / rajobhih pancabhis caiva harardhadivibhusitam // mandalasya tu bahyasya yad uktam sarvalaksanam / sarvam tasyapi tat7 kuryat samasena visesatah // ekam eva na kuryad vai toranam8 dvarasamdhisu /

90

atas ca tasya dvarani toranani na kalpayet // iti /

Codd. a (A F H E C D J); A, P (F H E C D J); Y (F H), 5 (E C D J); £ (E C), C (D J). Lacking: B (fol. 1-13 lost). 1 2 3 4 5 6

vibhusitam P : bhusitam A samyutaih conj. (Sanderson) : samyutam codd. vicitrair upaSobhitam ADFE : vicitrarupaSobhitam HJ : vicitrair upaSobhitam C samlikhed JY : likhed va A : samlikhye CD : samlikhe E abhyantaram A : osvantaramS : asvantaramY puram A : punah P

8

toranam A : torana P

7

tat AY : yat 5

244



Text

Adhikara 4

asanam 1 ca nyaset paScad devatanam yathakramam2 / tatramadhye3 nyasec caiva pradhanasya4 tu asanam // purvadidiksu sarvasu aksobhyadmam ca vistaram5 / 95

svabijaparinispannarn kalpayet tac ca6 yogavit // tatrasanavikalpah7 / madhye simhasanam8 purve gajasanam9 daksine turagasanam pascime mayurasanam uttare garudasanam* iti // sattvavajryadyarabhya yavad vajrave&paryantanam 10 akaraparinispannani candramandalani 11 vicintayet /

100

tataS ca pradhanasana 12 ahkarena 13 parinispannam mahavairocanam / yato yad eva 14 cittam prakrtya grahakadyakaravirahad dharmadhatusvabhavam tad eva* caturmukham s"unyatadicaturvimoksamukhanam dharmadhator alambanatvena sarvasamadhiprasutihetutvat / Suklavamam 15 dharmadhatusvabhavatvat / jatamakutopetam 16 nirabharanam ca

Codd. a (A F H E C D J); A, (3 (F H E C D J); Y (F H), 5 (E C D J); £ (E C), C (D J). Lacking: B (fol. 1-13 lost). 1 2 3 4 5 6 7 8 9 10 1' 12 13 14 15 16

asanam Ay : asanai e : asane C kramam A . kramat P tatra p : tan A pradhanasya AyC : pradhanasye £ vistaram A : vistaram (3 tac ca P (taca F) : tasya A tatrasanavikalpah A : tatrasanam vikalpetah (3 : de la gdan rnams brtag pa Tib. [189.4.2] sirnhasanam A Tib. : simhasanam varam P gajasanam A Tib. : gajasanam jfieyam P vajraAyC : vajrame 4 paryantanam A : paryantam p candramandalani Ay : om. 6 asana corr. : asane codd. ahkarena Ay : akarena 6H2 yato yad eva A : yade< >va H : yatoya< >va F : yad eva 6 varnam AH : vama 5F makutopetam AJH : mukutopetani eDF

245

Adhikara 4

105

Santacittatvat / bodhyagrimudrayuktam 1 * prajnopayatmakarvat // tarn evarnbhutam mahavairocanam atmanam adhimucya 2 caturbuddhasanayogena taddhrdaye candramandalam tadupari dhihkarena parinispannam adibuddham bhagavantam / pancananam iti pancamukham / pancaSikham iti pancacirakam3 / tasyaiva4* bandhanat pancacirakaSekharam / panca-

110

varnopetam / purvena nilam daksinena pi tarn paScimena raktam uttarena haritam / murdhni5 paramaSvamukhavad avasthitam mukham ^vetam 6*

/ bantam 7 kumarabharanopetam 8 sa^rngaram 9 vicitra-

vastraparidhanam astabhujam ^atasahasrikam10 prajnaparamitam caturdha vibhajya caturbhih karair hrdi samdharyamanam 11 * / aparaiS caturbhih 1 15

karaih prajnakhadgam 12 praharanabhinayena dharayantam caturbuddhasanayogena13 vyavasthitam bhavayet //

105 cf. NMAA 4, 221-2: vairocanam sitavarnam bodyagrimudrayuktam dhkaraparinispannam (bodhyagri AB Tib.([190.5.5] byang chub mchog) : bodhyangi(-ga F) P)

Codd. a (A F H E C D J); A, p (F H E C D J); Y (F H), 5 (E C D J); £ (E C), C (D J). Lacking: B (fol. 1-13 lost). 1 2 3 4 ^ 6 7 8 9 10 11 12 13

bodhyagri A Tib.([189.4.6J byang chub mchog); NMAA[4, 221-2] : bodhyangl P 4 yuktam A : vutam P adhimucya A : adhimumcya P pancacirakam A : pancaclram P tasyaiva con/. : tad eva codd. murdhni P : murdhani A yasya tarn conj. (Sandersori) : om. codd. Tib. Santam Ay (^anintam H) : ^anta 6 abharanopetam A : obharanopetam H : obharanavarnopetam F : abharanavarno petamS sa^mgaram AF : saSrngara 6H ^atasahasrikam A : Satasahasra P samdharayamanam em. : samdharyamanam codd. (sadharyamanam D) khadgam em. : khadga codd. yogena A : om. P

246

Adhikara 4

Text

[imam sanmantrarajanasamyuktam* advayodayam

/

anutpadadharminim gatham bhasate sma giram patch // (25)] imam dvadaSatathagatabhumyupalaksitam 1 gatham2 giram pater 120

bhasate smeti sambandhah // kimbhutam sanmantrarajanasarnyuktam3 iti vaksyamanasanmantropetam // kimatmikam advayodayam 4 iti / dvayam grahyam5 grahakam ca / tan6 na vidyate7 yasya8 tad advayam9 / udety asmad 10 ity udayah11 / advayasyodayo 12 'dvayodayah 13 / sa vidyate yasyam 14 gathayam sadvayodaya 15 tarn 16 bhasate sma // punar api kim-

125

viSistam / anutpadadharminim17 iti dharmadhatusvabhavam 18 Simyatasvabhavam 19 ity arthah //

[a a i i u ii e ai o au am ah sthito hrdi / L







jnanamurtir aham buddho buddhanam tryadhvavartinam

//

(26)] §rivajramandalamkaratantre dvadas^atathagatabhumayah pathyante //

130

Codd. a (A F H E C D J); A, p (F H E C D J); y (F H), 6 (E C D J); £ (E C), C (D J). Lacking: B (fol. 1-13 lost). 1 2 3 4 5 6 7 8 9 10 1] 12

13 14 15 16 17 18 19

bhumy corr. : bhumi codd. (bhumim H) gatham AJF : gatha eD : gartha H rajanaA : rajanamSH : rajanahH2 : *g*ajatamF advayodayam AJ : advayodvayam yeD grahyam AH : grahya CFE : grahr C tan na AJ : tan na eD : taun na H : ta*tt*a F vidyate A5 : vidyeteti H : vayateti F yasya AyC '• ta yasya £ advayam A6 (advaya D) : aksayam y * aksayam /< ~8 >advayasyodayo F asmad AJ : asmabhideD : asmamH : lac. F udayah ACH : udayamh C : udayam E : lac. F odayo corr. : odayah H : odvayah A : adayah £F : om. C * advayasyo­ dayo om. C 'dvayodayah corr. : advayodayah Jy : advayodvayah £D : om. A yasyam AJH : yasya eDF sadvayodaya corr. : sa'dvayodaya A : sa'dvayodaya eH : sa advayodaya J : sa'dvayodvaya D : sa'dvayadvaya F tarn A : gatham 6F : gatham tarn H dharminim A : dharminim(-ni F) gatham (3 svabhavam AH : svabhava F : svabhava 6 svabhavam Ay : svabhavam 5

247

Text

Adhikara 4

samantaprabha mahabhumih 1 suryamandalavac ca sa / amrtaprabha dvitiya tu candravat suprabhasvara2 // gaganaprabha trtiya tu gaganabhumipratisthita3 / vajraprabha tu ya bhumis4 caturthi sa manorama // 135

drdhadhyasayayogena5 vajreti parikalpyate / pancami ratnaprabha nama abhiseke pratisthita // sasthi padmaprabha nama padmavat suvirajate / svabhavasuddhadharmarvat6 nirmala nisparigraha // saptami karmaprabha nama buddhakarmakari7 smrta /

140

astamy8 anupama nama upama tatra na vidyate // navami nirupama9 nama sarvopamaprativedhatah 10 / da£ami prajnaprabha nama buddhabhumir anuttara // ekadaii sarvajnata mahabhumih11 prabhasvara 12 / dvadasi pratyatmavedya 13 tu yogijnanaprapurika // iti //

145

eta dvadaSatathagatabhumayo 'karadidvada^asvaraih 14 parinispanna veditavyah //

Codd. a (A F H E C D J); A, P (F H E C D J); Y (F H), 5 (E C D J); e (E C), C (D J). Lacking: B (fol. 1-13 lost). 1 2 3 4 5 6 7 8 9 10 11 12 13 14

mahabhumih A : mahabhumi Y : mandalabhumi 5 suprabhasvara AH : suprabhasvaram eF : saprabhasvaram C pratisthita AeD : pratisthitah Y : pratisthitam J bhumi^ A : bhumi P drdhadhyasaya A : tadadhyaSaya EY : tadadhyasama CC svabhava AY : prabhava 6 karma A : dharma p astamy corr. : astami codd. nirupama AYC : nirupamae : dpe Idan {sopama} Tib.1191.1.2] prativedhatah A Tib.([19LL2] so sor rtogs) : prativodhakah P sarvajnata mahabhumih A : sarvajnaprahha nama bhumi P prabhasvara AY : prabhasvarah 6 vedya P : vaidya A 'karadi corr. : akaradi codd.

248

Adhikara 4

Text

buddhanam tryadhvavartinam ity atitanagatanam 1 buddhanam / hrdi2 sthito 'ham iti sarvatathagatahrdayaviharitvat / jnanamurtir iti jnanakayasvabhavatvat / buddha iti yathavasthitapadarthavabodhat3 //

[om vajratiksna duhkhaccheda prajnajnanamurtaye

150

/

jnanakaya vagi&vara arapacanaya te namah // (27)] idamm tasyadibuddhasya hrdaye prajnacakram vicintayet4 // kimviSistam* ity aha pratityasamutpadarthasucakam 5 / tatha caha / araih* samdharyate6 nabhir nabhau7 carah8 pratisthitah / pratityotpadayogac ca cakram etat pravartate // iti //

155

akarena parinispannam sadvarnopetam sarasatkam9 * caturmekhalasamyuktam suvibhaktanabhi // tasya nabhimadhye 10 akaram vicintya / bhagavato jnanasattvasya mulamantradayo mantraS catasrsu mekhalasu sitavarna eva11 bhavaniyas12 // tatraprathamamekhalayam 13 om sarvadharmabhava160

svabhavaviSuddhavajra14 a a am ah // tasyaiva dvitiyamekhalayam 15

Codd. a (A F H E C D J); A, p (F H E C D J); y (F H), 6 (E C D J); 6 (E C), C (D J). Lacking: B (fol. 1-13 lost). 1

2 3 4 5 6 7 8 9 1° 1! 12 13 14 1^

atitanagatapratyutpannanam con/. Tib.([190.1.4] 'das pa dang ma 'ongs pa dang da Itar gyi [dus gsum gyi sangs rgyas rnams kyi'o]) : atitanagatanam codd. hrdi A : hrdaye 6 : hrdaya Y yathavasthita A : yathasthitap 4 padarthavabodhat A : padadvabodhat 6 : padavabodhat Yy vaDoanat vicintayet AH : cintayet 5F samutpadartha AJE Tib.([190.1.6] rten cing 'brel par 'byung ba'i don) : sam utpa(-ttva H)rtha CDY * sucakam AY : sucikam 5 samdharyate Aye : samdharyati t, nabhau AJH2 : nabhi nabhau DYE carah AH2 : cara P sadvarnopetam sarasatkam conj. : sadvarnopetam arasatkam codd. tasya nabhi om. P eva AF : evarn 6H bhavaniyas A : bhavanaya(-am H) P prathama AY : om. 5 dharmabhava AY : dharma 6 4 visuddha AFe : suddha HC dvitiya AY : dvitiya 6

249

Text

Adhikara 4

dvadasatathagatabhumyabhidhayakan 1 dvadaSasvaran2 cintayet3 // tatah4 prathamaram arabhya5 yathavat mantrasatkavinyasam6 kulavarnakramena kuryat // tad yatha / prathame / om vajratiksnaya namah7 * / dvitiye / om duhkhacchedaya namah8 / trtiye / om prajfiajnanamurtaye namah / 165

caturthe / om jnanakayaya namah9 / pancame / om vagiSvaraya namah 10 / sasthe / om arapacanaya 11 namah12 / cakrasya13 bahirnemyam 14 bhagavatah sarvatathagatajnanakayasya malamantram nyaset / ah ah15 sarvatathagatahrdaya 16 hara hara om hum hrih bhagavan jnanamurte 17 vagTSvara mahavaca sarvadharmagaganamalasuparisuddhadharma-

170

dhatujnanagarbha ah // bahir dvitlyayam mekhalayam kakaradihakaraparyantan18 niveSayet // ity ayam cakrasya mantravinyasah //

168-9 NS.Dav. 68, 12-13: a ah sarvatathagatahrdayam hara hara / om hum hrlh bhagavan jnanamurttivdgisvara mahavaca', SaMa 160, 19-20: a ah sarvalathdgalahrdaya hara hara I om hum hrih bhagavan jnanamurttivagis\>ara mahdvdca; Tib.[190.2.4-5]: a ah om sar ba ta tha ga ta hr da ya / ha ra ha ra / om hum hrih bha ga wan (bam Tib.P.) / dznya na mu (mu Tib.P.) rtti / ba gi swa ra / ma ha wa (bd Tib.P.) tsa.

Codd. a (A F H E C D J); A, p (F H E C D J); Y (F H), 5 (E C D J); £ (E C), C (D J). Lacking: B (fol. 1-13 lost). 1 2 3 4 5 6 7 8 9 1° 1] 12 13 14 15

abhidhayakan A : abhidhayakam (3 svaran A : svaram (3 cintayet Ay : vicintayet 6 tatah ASH 2 : om. y prathamaram arabhya A : prathama< >m arabhya Y : prathamamaniparitye 5H2 vinyasam A : vinyasa (3 namah A : te namah 0 7ifc./790.2.3/ namah AY : te namah 6H2 Tib. namah conj. : te namah codd. Tib. namah conj. : te namah codd. Tib. arapacanaya AyC : alapamcanaya e namah conj. : te namah codd. Tib. cakrasya AY : cakra 6 nemyamAJ : namyam CDy : navamyam E ah ah A J : ah ah Y£D : a ah om Tib. [190.2.4] : a ah SaMa [160, 19J; NS.Dav.[68, 12] : a a om NS.Dav. Ne : am ah NS.Tib.[Mukherji 99, 2J: NS.Dav MukherJl' Vira : aaNS.Dav.B : om. NS.Dav. 7 16 hrdaya codd. Tib.[190.2.4]\ SaMd[160, 19J; NS.Dav. Minaev- Tib- B- : hrdayam NS.Dav. [68, 12] 17 jnanamurte codd. NS.Dav. Minaev (ajnana- C) : jnanamurti Tib.[190.2.4J; NS.Dav. [68, IB]; SaMa [160, 19-20] 18 paryantan AH2 : paryantam (3

250

Text

Adhikara 4

tasya cakrasya madhyavartinam akarajam bhagavantam sanmukham jnanasattvam SaracchaSarikaprabham 1 indranilagrasacciram 2 balarkamandalacchayaprabhamandalam3 * sarvatathagatamayabharanam4 samadhi175

samapannam vicitrapadmasanopavistam ubhayakarasaktanllotpalordhvasthitaprajnaparamitapustakadvayam5 Santarasopetam atmanam vicintya / taddhrdaye candramandalam vibhavya tadupari prajfiaparamitasvabhavam sarvajnajnanodayakaranarn6 sarvaSravakapratyekabuddhanam utpattibhutam7 sarvamahabodhisatrvanam8 punyajnanasambharabhutam9 paramartha-

180

ksaram sarvaksaranam karanabhutam akaram 10 vinyaset 11 // tataS 12 cakrasthitani 13 yani mantraksarani yathoktavarnopetani 14 suvibhaktany15 ajnana^ailavidhamanaksamani 16 gabhastimanti vicintya tai£ ca ra^mibhih svasariram uddyotamanam 17 vibhavayet //

Codd. a (A F H E C D J); A, P (F H E C D J); y (F H), 6 (E C D J); e (E C), C (D J). Lacking: B (fol. 1-13 lost). 1 2 3 4 5 6 7 8 9 10 1] 12 13 14 15 16

17

chasanka 6 : chasankam A : chaSakam H : cha^aka F indranllagra AJ (Tib.[190.2.6] i ndra ni la mchog) : indramalagra eD : mdra jalagra H : indratalagra F 4 sacciram AyJ (saclra H) : saccarid eD mandalacchaya em. : mandalacchaya codd. mayabharanam A : mayabharana P karasakta P : karesakta A 4 nilotpalordhva A : nilotpaladvaya P (nllotpara- C) 4 dvayam Aye : dvaya C jnanodayakaranam A : jnanodayakaram y : jnanadayakaram 5 utpattibhutam Ay : mur(mu C)ttparttibhutam CD : murttibhutam JE sarva Ay : sarve 6 sambharabhutam Ay : sambhutamS (sambharam C, sambhuatam C 2) akaram A : akara 6 : aikaramy vinyaset P : vinyasya A tala^AJy : ataS eD sthitani Ay : sitani 6 varnopetani Ay (vamopeta< > F) : mantro(tra C)petani SH2 suvibhaktany A : bhaksany P (< >bhaksany F) vidhamana Ay (vidharmana H) : dvidharmana(mata E) 5 4 ksamani A . ksa rani ey : ksatrani C uddyotamanam Ay : udyanam 6

251

Adhikara 4

Text

uddyotayanto 1 raSmayas caturbhyo2 mukhebhyo nirgatya sahaloka185

dhatum avabhasya / tatah sadbhyo3 vajramukhebhyah4 sadvarna5 vinihsrtya6 sarvakas~adhatusamavasaranani 7 nanabuddhaksetrani gatva tesu buddhaksetresv as"esanava£esasya8 sattvadhator artham krtva sarvatathagatabuddhabodhisattvaparsadam9 avabhasya tesam vaktrena pravisya taddhrdvajram 10 pradaksimkrtya tatprajnajnanam akrsya dharmanairatmya-

190

samataya svavaktraih11 pravisya svahrccandrasthitamantraksarena sahaikibhuyavasthita 12* cintayet13 // evam anavaratabhavanabhirato mantramukhacaryacari bodhisattvo bhagavato 14 manjusrijnanasattvasya rupam alambayann iti jfianakayena tadrupam anuvicintayann15 iti candramandalarupena / tadrupam anu-

195

dhyayann ity akararupena / evam tridhakarena bhavayan16 / tarn eva nirmanarupakayena 17 na cirad eva vinayavas*am upadaya draksyati 18 /

Codd. a (A F H E C D J); A, (3 (F H E C D J); y (F H), 5 (E C D J); £ (E C), C (D J). Lacking: B (fol. 1-13 lost). 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

uddyotayanto A : uddyotyate H : uddyotyatai F : uddyan(-ya*n* Qacatai £ : ud dyantacatai C caturbhyo A : catur 3 sadbhyo codd. (sadbhyas J; sadbhya*^* E) vajramukhebhyah A Tib.([1903.4] rdo rje'i sgo [drug] nas) : cakramukhebhyo y : cakrebhyo £D : cakrebhyah J varna AH : varno 5F vinihsrtya A : vinisrtya (3 akas"a A6 : akas"e Y a^esanava^esasya corr. : as"es"esanavas"esasya A : a^esanava^esa (3 parsadam P : parsad A vajramA : vajrim P 4 taddhrdvajram -» svavaktraih om. D svavaktraih AyC : svavaktreE : sva*jaihJ : om. D sthita iti conj. : sthitam A : sthitam (3 (sthita D) cintayet Ay Tib.(fl90.3.7] bsgom par bya'o) : vicintayet 6 (vicintat D) bhagavato AJy : bhagavanto £D anuvicintayann A : anuvicintayan F : anuvicintayenn E : anuvicintayet HCC bhavayan A : bhavayet (3 nirmana Ay : nirvana 5 draksyati ACD : draksati JEy

252

Adhikara 4

Text

gaganatalagatams ca sarvabuddhabodhisattvan nirmanarupakayasahagatan draksyatiti 1 // tataS ca bhavanabalad ihaiva janmani2 buddhatvam vajradharatvam prapnotity3 anena kramena mayajalabhisambodhikramah4 200

sucitah // idanim5 * kulabhedena caparani sat prajnacakrany apy ucyante // tatra purvam tavat sarvam adiyogam krtva bhagavantam mahavairocanam svamandalamadhye6 vyavasthitam7 apakrsya8 tatsthane 'mitabham9 rakta­ varnam hrihkarodbhavam samadhisthitam 10 vicintya taddhrdaya 11

205

adibuddham tathaiva dhyatva taddhrdi castaram cakram caturmekhalayutam 12 mulamantradibhih13 parivestitam 14 cintayet 15 // tataS caresv astasv astau 16* bijani om vajratiksnaya nama 17 * iti vinyaset // cakranabhimadhye vajratlksnam raktavarnam balabharananvitam 18 pancacirakasekharam vicitravastrasamvitam 19 hrihkaraparinatam vamahastena20

Codd. n (B A F H E C D J); B, a (A F H E C D J); A, p (F H E C D J); Y (F H), 5 (E C D J); e (E C), C (D J). Lacking: B, lines 197-99 (fol. 1-13 lost); B, fol. 14rl starts line 199, fprdjpnotity anena .... 1 2 3 4 5

6

7 8 9 10 1^ 12 13 14 15 16 17 18 19 20

draksyatiti con. : draksyati iti eD : draksati iti JY : draksyati A janmani AyC " jramani C : vajramani E prapnotity Y : prapnoti ity 6 : prayotity A : prapratlty B mayajala ABYE : malajala C : mala C idamm conj. (Tib. [190.4.3] da ni) : idam idanlm ABy : itam idanirn CC : itah idanim E

sva a : om. B

vyavasthitam A : vyavatthitam B : vyavasthitah 6 : vovyavasthitah Y apakrsya AB : avakrsya 6 'mitabham corr. : amitabham codd. samadhisthitam a : samavistham B taddhrdaya corr. : taddhrdaye codd. mekhala ABFe : mukhamla CH mantradibhih B : mantradi a parivestitam ABY : parivyasthitam 5 cintayet AByE : vicintayet CC astasv astau conj. : astasu castau codd. namaAB : namah 3 : lenamahTib.fi90.4.6] anvi tarn a : an vita B samvitarnBJY : samcitam A : samvirame : saviram D hastena AB : haste (3

253

Adhikara 4

Text

210

padmadharam daksinahastena 1 prajnakhadgadharam 2 vicintya taddhrdi purvavac 3 candramandalopari4 hrihkaram vibhavya purvoktaspharanasamharanakramena krtva bhavayed iti // punas tathaivaksobhyam5 mandalamadhye6 mlavamam humkarajanitam bhusparSamudrayuktam dhyatva taddhrdaye cadibuddham tathaiva

215

castaram7 cakram vibhavya purvavan mantravinyasam krtva om duhkhacchedaya nama8 ity astau bijani caresu niveSya madhye bhagavantam vajrakhadgam9 humkarodbhutam nilavarnam kumararupam sarvabalabharanavibhusitam 10 pancaciram cmapattaparidhanam vamena vajram daksinena khadgadharam 11 vibhavya taddhrdi candramandalopari humkaram dhyatva

220

tathaiva spharanadikam kuryad iti // punas tathaiva madhye vairocanam sitavarnam 12 bodhyagrimudrayuktam 13 * ahkaraparinispannam 14 vicintya taddhrdaya 15 adibuddharn 16 tathaiva da^aram 17 cakram mantrair vidarbhitam krtva om prajnajnana-

Codd. a (B A F H E C D J); B, a (A F H E C D J); A, (3 (F H E C D J); y (F H), 6 (E C D J); e (E C), C (D J). Lacking: A, line 223 (fol. 19-22 lost); A, fol. 18v, finishes line 223, ...cakram mantraifrj.

1 2 3 4 5 6 7 8 9 1° 1] 12 13 14 15 16 17

daksinahastena B : daksine a dharam AB : bhrtam (3 purvavac ABS : buddha y candra A6 : catu B : samjfia Y aksobhyam A : askobhyam B : aksobhya p mandalamadhye AB : mandalam tasya madhye(madhya F) (3 tathaiva castaram(-ra DE) a : tathaivastaram B nama AB : namah 3 vajrakhadgam codd. : 'jam dpal sdug bsngal gcod pa {mafijusriduhknacchedah} Tib. [190.5.3] sarvabalabharana a : sarvabharana B khadga ABS : khadgam Y sita a : sveta B 4 varnamABY : varna 5 4 bodhyagri AB Tib. ([190.5.5] byang chub mchog ) • bodhyangl(-ga F) (3 yuktam AB • yutam P ahkaraAB : ah (3 : yigea{akara} Tib.([190.5.4] taddhrdaya H : taddhrdaye cett. adibuddharn AJE : adibuddha B : adibuddhams daSaram A : das"ara P : daslra B

254

Adhikdra 4

Text

murtaye nama 1 iti bijaksarani daSasu caresu vinyasya2 cakramadhye 225

bhagavantam prajnajnanam Suklavarnam pancaciram kumarabharanopetam vicitrambaradharam vamena cakradharam daksinena prajnakhadgam ahkaraparinatam3 * vibhavya taddhrdi candram tadupary ahkaram4 dhyatva tathaiva spharanadikam kuryad iti // tathaivamoghasiddhim5 mandalamadhye haritavarnam abhayamudranv-

230

itam ahkarabvjodbhavam6* dhyatva taddhrdaya7 adibuddham tathaiva vicintya cakram astaram purvavan mantradividarbhitam8 krtva / tataS caresv astasu9 om jnanakayaya nama 10 iti bijani castau 11 vinyasya12 cakramadhye bhagavantam jnanakayam haritavarnam kumarabharanam pancaciraka^ekharam ahkaraparinispannam 13 vamena visVavajradharam 14 daksinena

235

khadgadharam vicintya taddhrdi candram tadupary ahkaram 15 dhyatva tathaiva spharanasarnharanam kurvan16 bhavayed iti //

Codd. n (B F H E C D J); B, 3 (F H E C D J); Y (F H), 5 (E C D J); e (E C), C (D J). Lacking: A (fol. 19-22 lost). 1 2 3 4 5 6 7 8 9 10 11 12 13 14

nama B : namah P vinyasya B : nivesya (3 ahkarap : ahkaraB : yigea{akara} Tib. [190. 5.6] ahkaram codd. : yi ge a {akara} Tib. [190.5. 7] siddham em. : siddhi BP ahkara B : akara (3 : yi ge sa bon a Tib. [190. 5. 8] taddhrdaya corr. : taddhrdaye codd. vidarbhitam B : vinyasam p astasu B : astasu P namaB : namah P bijani castau B : om. P vinyasya P : nityasya B ahkara BeJ : akara D Tib.([191.1.2] yi ge a) : *karaH : illeg. F vamena visvavajradharam codd. (< ~9 > F; va H; mena visvavajradharam suppl.

15 ahkaram Bye : trakaram C : yi ge a {akaram} Tib. [19 1.1. 2] 1 6 kurvan em. : kurvana B : krtva P

255

Adhikara 4

Text

punar api tathaiva ratnasambhavam pitavarnam varadamudropetam mulaksarasambhutam 1 mandalamadhye dhyatva taddhrdaya2 adibuddham tathaiva dhyatva3 cakram astaram mulamantradisamyutam tadastaresu4 om 240

vagiSvaraya nama5 iti bijani vinyasya6 cakranabhau vagiSvaram pltavarnam sarvabharanopetam pancacrram vicitravastraparidhanam omkarabijajatam7 vamena ratnadharam daksinena prajnakhadgadharam dhyatva taddhrdi candramandalopari omkaram vibhavya8 tathaiva spharanadikam kuryad iti // tathaiva bodhicittavajram9 Svetavamam Srrigararasopetarri pancabuddha-

245

mukutam vajragarvataya nisannam 10 * sarvalamkaravibhusitam 11 vajraghantadharinam akarena parinispannarn 12 tathaiva mandalamadhye vicintya taddhrdaye bhagavantam adibuddham tathaiva vibhavya13 taddhrdi cakram 14 navaram mulamantradikhacitam 15 om 16 arapacanaya nama 17 iti

Codd. O (B F H E C D J); B, (3 (F H E C D J); Y (F H), 6 (E C D J); e (E C), C (D J). Lacking: A (fol. 19-22 lost). 1 2 3 4 5 6 7 8 9 10 1] 12 13 14 15 16 17

mulaksara B : maunaksara 6 : maunaksara F : nanaksara H : yi ge om {om kara} Tib.[191.1.3] taddhrdaya corr. : taddhrdaye (3 : taddhrdaya B tathaiva dhyatva By : om. 6 tad Bye : sad C nama B : namah (5 vinyasya BDy : vinyasec J : vinyase e jatamBE : jatayCC vibhavya BJ : vibhavyas Dye vajram B : vajra 3 garvataya nisannam conj. : gar*vt*ayanisarnnam B : gardbhayadivarnam C : garbha adivamamE : garbhapavarnam J : garbhapadavarnam D : gardabhavamamH*2* : garbha< 2 >vama F : rdo rje snyems pas bzhugs pa Tib.fl91.L8J sarvalarnkara Bye : sarvakara C parinispannarn B : parinatam (3 vibhavya (3 Tib.([191.2.1] bsgoms te) : om. B cakram By : cakra 6 mula p Tib.([l91.2. /yrtsaba) : mala B 4 khacitamBy : kheditamS omp Tib. [191.2.1] : om. B namaco/r. : namah ey : te namah CH2 : tenama B Tib. [191.2.1]

256

Adhikdra 4

Text

nava bijaksarany 1 aresu2 vinyasya3 cakranabhau bhagavantam arapacanam 250

suklavarnam balabharanabhusitam4 pancaciram vicitrambaraparidhanam akaraparinispannam5 pustakankam 6 * khadgapanim vicintya taddhrccandropary7 akaram dhyatva purvavat spharanasamharananilayadrdhibhavadikam 8 krtva bhavayed9 iti sarvatra yojamyam* // ayam api 10 kulabhedena 11 £rimayajalabhisambodhikramah sodha 12

255

veditavyah 13 // ata evaryanamasamgitau 14 sapta cakrani bhavanti tani ca gurupade^ad veditavyani // etavanmayajalabhisambodhikramah // aryanamasamgititikayam15 namamantrarthavalokinyam mayajalabhisambodhikramasyadhikara^ caturthah // gathas tisrah //

Codd. Q (B F H E C D J); B, p (F H E C D J); Y (F H), 6 (E C D J); e (E C), C (D J). Lacking: A (fol. 19-22 lost). 1 2 3 4 5 6 7 8

9 10 11 12 13 14 15

bijaksarany codd. (bijaksara E) * [bijaksarajny aresu -» vicitra[mbaraparidha nam] om. E aresu B : caresu p (om. E) vinyasya BJy : vinyase CD : om. E bhusitam B6 : vibhusitam y parinispannam p : nispannam B pustakankam B : pustaka(-ke D)p : phyagna po ti Tib. [191.2.3] taddhrc B : taddhrdaye (3 samharana conj. (Tib.fl91.2.3Jbsduba dang) : samharanam (3 : samspharana B + nilayacon/.(riZ?./^/9;.2J-^/gnaspa[brtanpargyurpa]) : nilayam P : lilayaB bhavayed ByDc Tib.([191.2.4] bsgom par bya'o) : kuryad6 ayam api (3 Tib.([191.2.4] 'dir yang ) : ity ayam B kula p 776. (rigs) : asilaB sodha Y Tib. (drug tu) : prodha D : praudha eJ : om. B sodha veditavyah -» etavan mayajalabhisambodhikranah om. B evarya conj. (Tib.fl91.2.5] 'phags pa) : evayamcodd. aryanama B r^.(/79;.2.6/'phagspamtshan) : iti nama (3 (nama D)

257

Adhikara 5

Text

[Adhikara 5]

[tad yatha bhagavan 1 buddhah sambuddho 'karasambhavah / akarah sarvavarnagryo maharthah paramaksarah

// (28)]

athedanim vajradhatumahamandaladvarena bodhicittavajrasya spharanadikam ityaha / tad yatha bhagavan ityadi2 / bhagavan buddhah 5

sambuddha iti gatarthah / akarasambhava iti akarajanitah3 / bhagavato jfianasattvasya pari^uddhatathagatajfianakayasvabhavatvat4 // tasya hrccandramandalopari yo 5 'sav akaras tasyedamm 6 viSesanam 7 aha

//

sarvavamagrya iti / agrebhavo8 'gryah sarvavarnanam sarvaksaranam / agryah ^resthas tanmukhatvat // mahartha iti maham^ casav arthaS ca 10

maharthah / maharthatvam9 sakalajagaddhitadhanahetutvat 10 // paramaksaraity aparaparyayenanaksaratlty11 * aksarah / bhutakotirupenavicalitatvat 12 * / paramaS casav aksarah ceti paramaksara iti // anena kramena jnanasattvadyaksarayoh13 * svarupabhidhanam 14 uktam //

Codd. n (B F II E C D J); B, (3 (F H E C D J); y (F H), 5 (E C D J); e (E C), C (D J). Lacking: A (fol. 19-22 lost); B, lines 11-13 (om.), [bhulakotijrupendvicalitvdt -» svaru­ pabhidhanam uktam. bhagavan NMAA[5, 4] : bhagavam NS.Dav.[51, v28a] ityadi (3 : ityaha B akarajanitah B : akarajjatah^ (akarajatahD) tathagata B : om. (3 yo By : yo 6 tasye P (tasya H) : tasyai B viSesanam Be : vis~esam yC agrebhavo BJy : agryabhavo eD maharthatvam B : < >£ : om. yC adhana By (adhara H) : adhyana 6 ksaratity B : raksati ity 6 : raksatity y avicalitatvat y : avicaritatvat 5 : mi g.yobaphir ro Tib.[191.3.4] sattvadyaksarayoh Dc : sattvadyaksara D : sattvadya eyJ : [ye shes] sems dpa' la sogs pa'i yi ge dag gi Tib.[191.3.5]) 14 svarupabhidhanam Dc Tib. (fJ91.3.5]rang bzhin brjod pa gsungs ba ste) : pari dhanam C : < >petidhanam y : < >panidhanam C : pranidhanam E

1 2 3 4 5 6 7 8 9 10 1J 12 13

258

Adhikara 5

Text

tad evaikaikam ca 1 * yoginam svacittodayasakalamandaleyadevata15

svabhavah2 samastavipaksavyavrttarupataya3 pratyaveksamyah4 / yatha 5 svasamadhinam6 * tadviparyasanupalambhasucakatvad7 upalambhapratipaksataya ca Sunyatavimoksamukhalaksanam / evam nimittamithyapranidhiviparyasabhisamskarapratipaksabhavena8 kramad animittapranihitanabhisamskaravimoksamukhatmakam9 * // tataS 10 canukramena sattvavajryadikulamatrsvabhavam 11 adhikrtyaha /

20

[mahaprano hy anutpado vagudaharavarjitah / sarvabhilapahetvagryah sarvavaksuprabhasvarah // (29)] mahaprano hy anutpada iti sattvavajrisvabhavena / tatra mahaprano hy akarah 12 sa canutpadasvabhavas 13 tasya dharmadhatusvabhavatvad 25

adarsajnanahetutvac ca // vagudaharavarjita iti ratnavajrimukhena / vacam udaharo vagudaharah / tena varjito 14 rahitah samatajnanena vagvyaparabhavat 15 // sarvabhilapahetvagrya iti dharmavajridvarena /

Codd. O (B F H E C D J); B, (3 (F H E C D J); y (F H), 5 (E C D J); e (E C), C (D J). Lacking: A (fol. 19-22 lost).

1 2 3 4

5 6 1 8 9 10 11 12 13 14 15

tad evaikaikam ca conj. (Tib. [191.3.5] de dag nyid re re yang) : tad evaikaikam tac ca (3 : om. B 4 tad evaikaikam ca yoginam om. B svacittodaya conj. (Tib. [191.3.5] rang gi sems las byung ba'i) : svacintodayah B : svacittadayah 3 4 mandaleya B : mandalaya P 4 devata 3 : devacata B taya Bey : maya C pratyaveksamyah BCy : tratyaveksaniyah C : om. E 4 pratyaveksanlyah -> upalambhapratipaksataya om. E yatha codd. : de Itar {tatha} Tib. [191.3.6J svasamadMnam F Tib.([191.3.6Jrang gi ting nge 'dzin rnams) : svam samadhinam cett. tadviparyasanupalambhasucakatvad B : catu< ~4 >vakatvat (3 (om. E) bhavena P (bhavana F) : bhaveta B samskaravimoksa BE : samskarapratipaksabhavena(-prati< ~4 > F) kramad animittapranihitanabhisarnskaravimoksace//. 4 atmakamBye : atmakamD : atmaka J tatasB : tac p vajryadi corr. : cajryadi B : vajradi P akarah B Tib.(yi ge a) : om. P sa canutpadasvabhavas em. (Tib.[191.4.1-2] de yang skyed ba med pa'i rang bzhin te) : sa canusvadasvabhavas B : evanutpadasva(-svam C)bhavas P varjito Bye : varjita C vyaparabhavat B : vyahara(-hara J)bhavat P

259

Adhikara 5

Text

abhilapanam abhilapah / sarvesam dharmanam skandhadmam abhilapah sarvabhilapah / tasya hetuh 1 karanam2 pratyaveksanajnanarn / ata 30

evagryah pradhanam ityarthah

//

sarvavaksuprabhasvara iti karma-

vajrisvarupena / sarva3 casau4 vak5 ca6 sarvavak / yatha devarutebhir nagarutebhir7 ityadi / ata eva suprabhasvara iti yathabhavyataya sarvasattvasamtanesu pratibhasodayad 8 anekamukhadharmadeSanayaS ca krtyanusthanajnanasvabhavatvat9 // ata evaitas 10 catasrah 11 sattvavajryadiparamita 12 * a a am ah iti

35

caturbhir aksaraih parinispanna 13 iti

//

tata£ ca yad eva cittam 14

dharmadhatusvabhavam prakrtya grahyagrahakadyakaravirahat 15 tad evagantukamalabhavaprabhasvaratvaj 16 jnanantarodayanimittatvac cadarsajnanatmakam // dharmalambanakrpadyatmakataya17 samtanantara40

bhedaparamarsac 18 ca 19 samatajnanam

//

sarvajneyaviparyasat20

Codd. n (B F H E C D J); B, (3 (F H E C D J); Y (F H), 5 (E C D J); e (E C), C (D J). Lacking: A (fol. 19-22 lost).

1 2 3 4 5 6 7 8 9 10 11 12

13 14 15 16 17 18

19 20

hetuh BH2 : hetu (3 karanam H : kamranam B : karana 6F sarva B : sarvams P casau P : case B vak BH2 : vaksa 6 : vakya y ca (3 : va B nagarutebhir B : klu'i skad dang gnod byin skad [ces pa la sogs pa Ita bu'o] Tib. fl91.4.5J om. P pratibhaso By : pratibhasa 6 jnana P : om. B evaitas" Bye : evainas C catasrah em. : catasro By : catasra 6 sattvavajryadi conj. (Tib.[191.4.7] rdo rje sems ma la sog pa) : vakvadi B : vajradi ye : cakradi t, parinispanna ByeD : parinispannatvad JH2 eva cittam y : evacintamB : evacattame : evavantamC grahya P (grahyah C) : om. B bhavaBF : bhave HeC atmakataya B : atmataya P samtanantara BJ : samtanantara De : samtanandara y 4 bhedaparamarsac corr. : bhedayaramarsac B : bheda aparam C : bheda aparamarsac D2 : bheda aparamrsad y : bheda aparamrsod e ca BD2 : va ye : C jneya B : jneya y : jfiaya 5

260

Adhikara 5

Text

pratyaveksanajnanam // samastalokahitahetutaya krtyanusthanajnanatmakam 1 // ataS ca2 yathakramarn s^sVatadyadhipatisvabhavarn3 // ata evaha

[mahamahamaharagah sarvasattvaratimkarah

/ (30ab)j

mahamahamaharaga ityadi // atra cadav amitabhasyopadanam4 ragasya

45

pradhanyakhyapanartham5 / ata evaha rago dvesas ca mohaS ceti / sarvatathagatasarvaka^avyapipujasamdar^anakale6 mahamaha ity ucyate / mahasabdena7 * pujabhidhiyate / mahan maho mahamahah8 / mahatmako ragaS ca maharagah9 * / mahamahas casau maharagaS ca mahamahamaharagah / 50

tatra 10 maharaga iti sattvapacanamocanalaksanah 11 sa ca12 bhagavan ^rivajrasattvah13 / maharago hi bhagavan vajrasattvas tathagata iti vacanat // sarvasattvaratimkara iti ragacaritanam sarvasattvanam 14 ratim pritim mahasukhatmikam karotiti sarvasattvaratimkarah / ayam amitabhasvabhavah // dvesacaritan adhikrtyaha /

Codd. n (B F H E C D J); B, p (F H E C D J); Y (F H), 5 (E C D J); e (E C), C (D J). Lacking: A (fol. 19-22 lost). 1 2 3 4 5 6 7 8 9 10 1] 12 13 ^4

atmakam Bye : atmaka C ataS ca By : ata eva 5 adhipati B : apipati (3 amitabhasyopadanam corr. : amitabhasyapadanam B : amita / tasyopadanam £ : ami< >tasyopadanam F : amita / tasyopadana J : amiti / tasyopadana D : abhitah / tasyopadana H pradhanya By : pradhanya 5 * artham BF : arthah 6H tathagata BJF : tathagatam eDH 4 puja B : pujana P 4 kale B : kala yC : kalaC : kalahE maha^abdena conj. (Tib.[19J.5.4Jmaha'i sgrani) : mahacchabdena (3 : < >SacchainaB mahan maho mahamahah B : om. (3 Tib. mahatmako ragaS ca maharagah conj. (Tib.[191.5.5] chen po'i bdag nyid kyi Mod chags dang 'dod chags chen po'o) : mahatmako maharagah (3 (maha< -13 illeg. >tmako maharagah C) : om. B tatra B : om. (3 pacana codd. Tib.([191.5.5-6] smin [cing grol ba'i mtshan nyid do]) caBTib.(19J.5.6Jydng) : om. (3 Srlvajrasattvah codd. : dpal rdo rje sems dpa' mtshan nyid yin no {srivajrasattvalaksanatvah} Tib.[191.5.6] sarva P : om. B

261

Adhikara 5

Text

55

[mahamahamahadvesah sarvakleSamaharipuh // (30cd)] mahamahamahadvesa iti* / mahamaheti gatarthah 1 / mahadveso 'ksobhyah sa ca sarvakleSamaharipuh / kle£opakle£adinam maharipur ity arthah // mohacaritan adhikrtyaha /

/ (3lab)]

[mahamahamahamoho mudhadhimohasudanah mahamahamahamoha iti

60

// mahamaheti purvavat / mahamoho

vairocanah2 / dhiyam3 moho dhlmohah* / mudhas casau dhlmohas ca4 mudhadhimohah* / tasya sudano vinaSakah // krodhacaritan adhikrtyaha /

[mahamahamahakrodho mahakrodharipur mahan //

(3led)]

mahamahamahakrodha iti / tathaiva mahamaheti / mahakrodha ity 65

amoghasiddhih / mahakrodharipur mahan iti / sattvasattvavisayatvad5 mahakrodhah / tasya6 sa eva maharipur iti // lobhacaritan adhikrtyaha /

[mahamahamahalobhah sarvalobhanisudanah

/

(32ab)]

mahamahamahalobha iti7 // mahamaheti purvavat8 * / mahalobho9 ratnasambhavah // sarvalobhanisudana iti / sarvalaukikalokottarasampatti70

pariSodhanarthatvat 10 //

Codd. n (B F H E C D J); B, (3 (F H E C D J); y (F H), 6 (E C D J); e (E C), C (D J). Lacking: A (fol. 19-22 lost). gatarthah (3 : gatartham B vairocanah J : vaicocanahB : vairocanam yeD dhiyam By : dhiyaS : [bio la gti mug pa ni] gti mug gi bio ste Tib.[192.1.2-3] ca B : ceti (3 visayatvad em. : visayatvata B : viSesatvat P : [sems can dang sems can ma yin pa'i] yul gyi phyir Tib.[192.1.4] 6 tasya B : om. 3 7 lobha iti ByC : moha iti C : moheti E 8 purvavat P : purvaca*ksu*tartham B : [mchod pa chen po'i don ni] gong ma zhin du bshad do Tib. [192.1.5] 9 lobhoBHC : lobha F : moho e 10 laukikaHS : lokikaBF 4 sampatti em. : saspatti B : phun sum tshogs pa Tib. [192.1.6] : om. |3 1 2 3 4 5

262

Adhikara 5

Text

Sunyatadicaturvimoksamukhanarp dharmadhatvayatanamukhena1 sarvasamadhihetutvat / tad eva2 cadhyatmaviparyasapratikulatayadhyatmaSunyatakaram / bahyopalambhapagamad3 bahirdhas"unyatasvabhavam / tada^raya^ariranupalambhasvabhavatayadhyatmabahirdha^unyatatmakam / 75

bhajanalokadhyavasayarahitatvenamahasunyatalaksanam // tata evadarsajnanaSrayavajrasattvadisamadhicatustayatmakam4 iti / ata aha5

mahakamo mahasaukhyo mahamodo6 maharatih // iti // (32cd)* tatra mahakama iti vajrasattvah / kamyata iti kamah / abhilasamya7 ity arthah / mahasaukhya iti vajrarajah / mahamoda8 iti vajraragah / maha80

ratir9 iti vajrasadhuh // ata eva caksurupadivasanaparavrttisvabhavadaiiajnananispatter10 adhyatmanupalambhadyadhisthanatvat / yathasvam 11 viparyasavipaksataya 12 bodhicittodayakarsanuraganasadhubhavahetutvac 13 caisarn yathakramam //

Codd. a (B A F H E C D J); B, a (A F H E C D J); A, (3 (F H E C D J); y (F H), 5 (E C D J); e (E C), C (D J). Lacking: A, lines 71-82 (fol. 19-22 lost); A, fol. 23r, starts, line 82 [viparyasavipajksataya bodhicittodaya... .

1 2 3 4 5 6 7 8 9 10 11 12 13

dhatvayatana yJ : dhatvapatanae : dhatvayamana D : dhatvalambana B eva (3 : evam H lambhapagamad B : lambhopagamat P catustayatmakam B : catuskayatmakam p iti / ata aha (3 : ity aha B mahamodo conj. (NMAA[5, 74] mahamoda; NS.Dav.[52, v32J mahamodo) : maha moho codd. (om. F) abhilasamya B : abhilaksaniya p : 'dun pa fzhes pa'i don to] Tib.[192.2.2] moda Y ' moha B5 ragah / maha om. B vasana B : vasata P (< >sata F) 4 paravrtti P : yanavrtti B yathasvam P (-tvarn F) : yathasva B viparyasaBy : viparyosa DC : vipayosa E : viparyase J akarsanuragana A : akarsa / nuragana B : akarsanuragena P

263

Adhikara 5

Text

evam tad eva Sunyatajnane 'py animittatvat 1 SunyataSunyata / 85

paramarthanimittaparamarsat2 paramarthaSunyata / samskrtapunyasambharanimittaparigrahat3 samskrtaSunyata / asamskrte 'pi bodhisambharanimittagrahad4* asamskrtaSunyata // ata evaha //

maharupo mahakayo mahavarno mahavapuh / mahanama5 mahodaro mahavipulamandalah // iti // (33) 90

tatra maharupo mahakaya iti vajraratnah6 / rupyata iti rupam vilokyata7 * ity arthah / mahad rupam yasyasau maharupah / mahattvam8 cintamanirupataya sarvatathagatanam traidhatukadharmarajyabhisekatmakarvat9 / ata eva mahakayo 'meyabuddhaksetravabhasanatvat 10 mahavapur iti vajrasuryah

95

/

//

mahavarno

mahamS casau varna^ ca mahavarnah

sadvarnopetatvat 11 / mahad vapur yasya sa 12 mahavapuh suvi^uddhadharma^rutacintabhavanamayajnanaprabhamandalatvat 13 // mahanama mahodara iti vajraketuh / mahan nama14 yasyasau mahanama mahakirtir

Codd. Q (B A F H E C D J); B, a (A F H E C D J); A, P (F H E C D J); y (F H), 6 (E C D J); e (E C), C (D J). 1 2 3 4 5 6 7

animittatvat ABH2 : anittatvat Cy : anityatvat E : anitatvat C paramarsat A : paramarsat BY : paramarsat e : paramarthat C punya ABFe . ^unya 1$, nimittagrahad conj. : nimitta*grahat B : nimittanungrahat A : nimittanugrahat 3 : mtshanmar(mar yang Tib.P.) mchog tu mi 'dzin pa'i phyir na Tib.[192.2.7] nama B(3 : namo A ratnah AB : ratna (3 vilokyata B : vinyasyata A : vinasyata (3 (na$yata F) : mi rtag [ces pa'i don to]

Tib. [19 2.2.8] 8 mahattvam corr. : mahatvam A6F : mahatva B : mahatvam H 9 rajya AB : raja {3 10 mahakayo 'meya corr. : mahakayah / ameya codd. (mahakagrah / ameya F) * ameya a : prameya B 4 avabhasanatvat AB : avabhasanatvat (3 11 sadvamopetatvat a : satvarnatvat B 12 sa ABE : om. yCC ^ 3 dharma ot : om. B 14 mahan nama AB : mahanama 5F : mahan namo H

264

Adhikara 5

Text

ity arthah / kuto danaparamitayah prakarsagamanat / tatas ca mahodara iti mahams casav udaras ca mahodarah sarvamdadatvat 1 100

// mahavipula-

mandala iti vajrahasah / madi2 bhusayam mandayatiti 3 mandah / mandam4 lati grhnatiti5 mandalam / vipulam vistirnam sakalalokadhatuvyapakatvat 6 / mahad vipulam7 hasamandalam8 * saram yasya sa mahavipulamandalah / trail okyabhasakaratvat9 // tataS ca samatajnanaparikaravajraratnadicatustayatmakam 10 klistamanah-

105

paravrttyatmani 11 samatajnane 12 grahakadimukhena yathoktanimittabhavat 13 / yathasvam 14 vipaksaprahanadvarena ca / abhisekatejo'bhivrddhidanaparamitaniyojanapr2modyanispattikaranatvac 15 ca 16 tesam 17 vajraratnadinam kramah // tad evatyantikasattvarthapranidhananabhinive^ad18 atyantaSunyata //

110

anavaragro maya samsaro na 19 parityaktavya20 iti pranidhananavagrahad

Codd. Q (B A F H E C D J); B, a (A F H E C D J); A, P (F H E C D J); Y (F H), 5 (E C D J); e (E C), C (D J). Lacking: B (ow.), lines 106-110, yathasvam -» iti prani­ dhananavagrahad.

1 2 3 4 5 6 7 8 9 10 11 12 13

14 15 16 17 18 19 20

sarvamdadatvat codd. (sarvadadatvat BH) madi AJF : mamdi £DH : madi B mandayatiti a : mandavatlti B mandam A : manda Bf3 grhnatiti E : grahnati iti A : grhnatiti ByCC sakala codd. (sakula Y) mahad vipulam ABF : mahavipula C : mahavipuram C : mahavipulam HE hasamandalam Y Tib.[192.3.7] : mahasamayamandalam A : mahamandalam BC : mahamanda E : mandalam CH2 lokyabhasakaratvat AB : lokyabhaskaratvat 5F : lokyabhasvaratvat H samata a : samaha B 4 vajra AB : om. (3 paravrtryatmani B : paravrttatmani A : paravrtyatmani P jnaneA2 : neA : jnanai B : jnanah 6H 2 : jnana Y yathokta ABJH : yathokta eDF 4 nimittabhavat a (nimiha- F) : nimittagraha bhavat B yathasvam Ac : yathasva Aa : yatha tvam P : om. B vrddhi AJ : vrddhir Y£D 4 pramodya p : pramoda A 4 karanatvac P : karanatvad A ca P : om. A tesam P : etesam A anabhinivesad A : abhiniveSat P samsaro na P : samsara A tyaktavya AY : tyakta 6

265

Adhikara 5

Tert

anavaragrasunyata // anupadhiSese 1 nirvanadhatau maya kusalam2 navakaraniyam 3 ity anabhinivesad4 anavakarasunyata

//

prakrtilaksanam

gotram5 maya sodhaniyam iti prarthanavirahat prakrtiSunyata // ata evaha //

115

mahaprajnayudhadharo mahakleSankus'o 'gramh6

/

mahayaSa7 mahaklrtir mahajyotir mahadyutih // iti // (34) tatra mahaprajnayudhadhara iti vajradharmah8 / tatra9 prajna pratyaveksanajnanam / tad evayudharn Sastram tad10 dharayatiti prajnayudhadharah / tajjnanasvabhavatvat 11 / mahamS casau prajnayudhadhara^ ca 12 120

mahaprajnayudhadharah // mahaklesankuso'granir iti vajratiksnah / klesa ragadayah / tesam anku^abhutatvat kleSahkusah / afiku^a ivanku^ah kle^avasitapraptatvat / tatas ca 13 mahamS casau kleSarikuSas" ca mahaklesankusah / ata evagramh14 ^restha ity arthah // mahayas~a 15 mahaklrtir iti vajrahetuh 16 / sattvaparipakopayabhutanam sarvamandaleyanam 17

Codd. Q (B A F H E C D J); B, a (A F H E C D J); A, p (F H E C D J); Y (F H), 6 (E C D J); e (E C), C (D J). Lacking: B (om.), lines 120-123, klesa ragadayah -» eva­ gramh srestha ity arthah.

5 6 7 8 9 10 11 12 13 14 15 16 17

anupadhi^ese A : nirupavi^ese pi B : nimpadhisesopi p kusalam A : ku£ala BY : ku^alo eD : kusale J navakaraniyam AP (-iya H, karanlyam J) : navantiraniyam B anabhinivesad A2B : anabhives"ad A . abhinive^ad p gotram AB : va gotram Y : va gatram 6 'granih corr. : granih AB : grani P yaSa ABFC : ya^o He dharmah AB : dharma P tatra AB : om. P tadAB : om. P taj corr. : tat p : tata A : ta B dharas"caAB : dharah P (dhara C) tata^ ca A : om. P evagranihH : evagram AF5 yaSa AB5 : ya^o Y vajrahetuh AF° Tib.([192.5.1J rdorjergyu) : vajraketuh Bp mandaleyanam B : mandalanam a

266

Adhikara 5

Text

125

tantranirdistanam 1 cakravartisvabhavatvat / mahad ya^o2 yasya sa mahaya£ah / ata eva£esanava£esalokadhatuvyapini 3 kirtir yasya sa4 mahaklrtir iti // mahajyotir mahadyutir iti vajrabhasah5 * / mahati rraidhatukaprasarpim jyotir diptir yasyasau mahajyotih6 / tataS ca vagdosanam drutamadhyavilambitanam7 pariharena tusnlmbhavasvabhavena8 japakale9 sarvasattva-

130

samtanesv^avabhasakatvan11 mahadyutir iti // tatas ca12 vikalpamanahparavrttilaksanapratyaveksanajnanasritavajradharrnadisvabhavam 13 * pratyaveksanajnane 14 yathoktaviparyasavivekena samastadharmatattvapratyaveksanat 15

/

tanmukhena 16 ca svabhava-

Suddhadharmatapraj naj nana vai vartikasamastabhisandhyarthaparij nanahetu135

tvac 17 ca / ata evaisam vajradharmadinam18 ayam kramah //

Codd. Q (B A F H E C D J); B, a (A F H E C D J); A, 0 (F H E C D J); y (F H), 6 (E C D J); e (E C), C (D J). 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17

18

tantra AB : tatra p ya$o ABH : ya^a F5 anava^esa ABJ : anavaSesa yzD sa a : sa B bhasah P Tib.([192.5.3J [rdo rje] smra ba'o) : bhasah AB jyotih a (jyoti CDF) : dyutih B madhya AB : mapya p ^ vilambitanam a (-anam A) : vilambitadmam B tusnim A : tusni BP japa a : japa B sarvasattva Bp J/^.(semscanthamscad) : sattvaA avabhasakatvan a : avabhasananB mabcacodd. : dernamskyang Tib. [192.5.5J laksana B Tib.([192.5.5] mtshan nyid) : laksane a 4 vajradharmadi A Tib. (/792.5.5yrdorjechosla sogspa) : dharmadi B : om. p 4 svabhavam B : sva bhaveA : om. P 4 pratyaveksanajnana^ritavajra dharmadisvabhavam om. P pratyaveksanajnane a : vikalpamanahparavrttilaksane pratyaveksanajnane B samasta ABey : samaste C * tattvacorr. : tatva codd. tanmukhena a : ta akhena B prajnajfiana a : jnana B 4 avaivartika a : avaivartikatva B 4 samastabhi sandhy a : meyestobhisandhy B vajradharmadinam AB : dharmadinam P (-adinam H)

267

Text

Adhikara 5

tad eva laksanavyanjanatmani 1 rupakaye 2 'nabhisamskaral 3 laksanaSunyata // balavaiSaradyadisvabhavadharmakayanabhisamskarat4 sarvadharmaSunyata // pudgaladharmasamaropanad 5 abhavas~unyata // yathokta^unyatalaksanapavadanabhisamskaranac 6 ca abhavasvabhavaSunyata // 140

atascaha 7 //

mahamayadharo vidvan mahamayarthasadhakah

/

mahamayaratirato mahamayendrajalikah // iti // (35) tatra mahamayadharo vidvan iti vajraviSvah / mayeti sarvatathagatavicitravipulodarapujanirmanam8 tad dharayatiti mayadharah / mahams 145

casau mayadharah ceti9 mahamayadharah10 / ata eva vidvan panditas tena pQjakarmana punyasambharoparjanat // mahamayarthasadhaka 11 iti vajraraksah / mayartheti 12 nirmanarthah / tena 13 kusidasattvani 14 sadhayati protsahayatlti 15 mayarthasadhakah / mahams' casau mayarthasadhakas ceti mahamayarthasadhakah // mahamayaratirata iti vajrayaksah / mayayam 16

Codd. a (B A F H E C D J); B, a (A F H E C D J); A, p (F H E C D J); y (F H), 6 (E C D J); e (E C), C (D J).

1 2 3 4 5 6 7 ^ 9 1° 1] 12 13 14 15 16

laksana AB : laksanam p rupakaye ABF : rupakayeh C : rupakayah JD : rupekaya H samskaral corr. : samskarat codd. (saskarat A) vaiSarady a : vaisa / rady B 4 anabhisamskarat AB : abhisamskarat p dharma AB : dharma P 4 aropanad AByC : aropetat e 4 pudgala corr. : pungala codd. 4 dharmasamaropanad codd. : [gang zag dang] chos la sgro 'dogs par 'du mi byed pa'i phyir na Tib.[192.5.8-193.1.1] apavadanabhi a (vadanabhi A, asyavadanabhi J) : apavadabhi B 4 samskaranac AB : samskarac P ata^ caha AB : ata aha P puja AB : om. P mayadharah ceti a : mayadharah B mahamayadharah om. B maha a : tamha B mayartheti A : mayartha iti P (mahamayartha iti Ec ) : mayarthohi B tena ABy : om. 5 kusidasattvani con/. (Tib.fl93.L5J semscan le lo can mams) : kusldasattvam codd. protsahayatlti ABye : protsahaya iti C mayayam AB : mayaya CH : maraya F : maroya C : mayaya E

268

Adhikara 5

Text

150

ratir mayaratih / tathagatapujanirmanaratir1 ity arthah / mayaratau ratah2 sakto mayaratiratah / mahamS casau mayaratirataS ca mahamayaratiratah // mahamayendrajalika iti vajramustih / mayendrajalam3 * nirmanasamuham kayavakcittavajranam eklbhavah4 /tad vettiti mayendrajalikah 5 * / maharnS casau6 mayendrajalikah7 * ca mahamayendrajalikah // vrddhyabhavo8 'tra9

155

drastavyah saddharmapathat 10 // tata£ ca krtyanusthanajnana^rayavajrakarmadicatustayatmakam 11 caksuradivijnanaparavrttyatrnani 12 krtyanusthanajnane 13 sati / uktaviparyasapahrasamukhena 14 samastasugatapujabhisamraksanavighnopa^amakayadi^unyatajnanakaranatvat15 * / ata16 eva caisam vajravi^vadinam krama

160

iti17 //

evam matsaryadauhsilyaksantikausidyapratipaksataya 18 tad eva 19 dana^ilaksantiviryaparamitatmakam iti / ata aha //

Codd. ft (B A F H E C D J); B, a (A F H E C D J); A, 0 (F H E C D J); y (F H), 5 (E C D J); e (E C), C (D J). Lacking: A, line 162 (fol. 26 lost); A, fol. 25v ends, ... -sila[ksdntiviryapa ramitat maka m]. nimiana A : nirmanai B : om. (3 4 tathagatapujanirmaiiaratir ity arthah om. (3 ratah a : raktahB mayendrajalam A : mayendralani B : mayendrajala (3 : mig 'phrulcan Tib.[193.1.8] ekibhavah H2 : ekibhavam AB : eklbhava (3 mayendrajalikah conj. : indrajalikahco^W. Tib.([J93.1.8J mig 'phrul can) casau cotr. : casav codd. mayendrajalikah conj. : indrajalikah co^. Tib.([193.2.1] mig 'phrul can) vrddhyabhavo corr. : vrrddhyabhavo B : vrddhyabha* A : vrddhabhavo ^ 'tra B : tra (3 : illeg. A saddharmapathat A : saddharmapathad va B(3 vajra A : om. B(3 4 atmakam a : atmakam B caksuradi ABy : caturadi 5 * vijnana conj. : jnana codd. 4 paravrtty atmani corr. : paravrtyatmani a : paravrttvatmani B 13 sthana oc : sthana B 14 viparyasapahrasa AB : viparyasa P 15 abhisarnraksana con). : abhisamskarana AH : abhisamskaraga B : abhisamskrena 6F : mngon par srung ba[r byed pa dang] Tib.[193.2.3J 4 kayadi A Tib.([193.2.3] sku la sogs pa) : karyadi Bp 4 karanatvat Bye : kara*atvat A : karanat C

1 2 3 4 5 6 7 8 9 10 11 12

16 ata a : atraB 17 krama iti em : karmeti codd. : las kyi go rim shes bya'o {karmakrama iti} Tib. [193.2.5]

18 matsaryaAB : tatparya 6 : gam(ga F)bhirya y * daus"Ilya p (Silya H) : dauhsrilyat B 19 tad eva AB : evam P

269

dauhsllya corr. : daus~ilya A :

Adhikara 5

Text

mahadanapatih Srestho 1 maha&iladharo 2 'gramh3

/

mahaksantidharo dhlro mahaviryaparakramah // iti // (36) tatra mahadanapatih srestha iti vajralasya / pujanugrahakamyaya diyata

165

iti danam / tasya danasya patir danapatih / mahamS casau danapatiS ca mahadanapatih / amisabhayadharmabhisekatmakacaturdanotsrjanataya sresthah sarvalokotkrstatvat // mahasiladharo4 'granir iti vajramala / kayavakcittanam samadhanac chilam / sila samadhav iti dhatvarthatvat5 / 170

dasakusalaviratilaksanam pratimoksasvabhavam6 ca / tad dharayatiti siladharah / maharnS casau siladharaS ca mahaslladharah / ata evagranih7 purahsara8 ity arthah // mahaksantidharo dhira iti vajraglta / ksamanam9 ksantis titiksa tarn 10 dharayatiti ksantidharah / maham^ casau ksantidharaS ca 11 mahaksantidharah / tataS ca dhirah 12 / mahatsv api duhkhesu

175

dhairyavalambanatvat 13 * // mahavlryaparakrama iti vajranrtya / vlryam 14 kusala15 * utsahah16 * / papadau tutsahah17 kusidatvat kausidyam* eva na

Codd. n (B F H E C D J); B, (3 (F H E C D J); Y (F H), 5 (E C D J); e (E C), C (D J). Lacking: A (fol. 26 lost). 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17

Srestho corr. : Sresthah yC : ^resta £ : 6esthau B maha^iladharo BCH : mahasllam dharo eF 'granihBH : 'grani 6F mahaslladharo BHy : maha^ilam dharo eF arthatvat B : arthat p pratimoksa BF : pritimoksa H5 evagranih By£ : evagrani C purahsara corr. : purassara codd. ksamanam BJ : ksamana yeD ksantis titiksa tarn B : ksantim astiti ksanta(-tam H) P ca B : ceti 0 tata^ ca dhirah om. B dhairyavalambanatvat conj. : dhairyavalambinatvat 6 : dhairyava(-ava F)lambina tvat y : vairyalambaratvat B : brtan par sran thub pa'i phyir ro Tib.[193.3.3-4] vlryam B : vlrya (3 kusala corr. : kuSale p Tib. ([193.3.4] dge ba la) : kus~alasu dharmesu B utsahah conj. (Tib.[193.3.4] spro ba ste) : ty utsahah codd. (ty utsaha EF) papadau tutsahah B : papady(-ody F)anutsahasya (3

270

Adhikara 5

Text

viryam / tasya parakramah sarvalaukikalokottarakusalamulaparipuranat 1 * // ata£ ca guhyapujacatustayatmakam2 yathakramam adarSadijnananispattau svavimoksamukhavi^uddhibalenanabhogabhirativisesotpadakatvat3 // tad eva dausprajfiaviksepasamyakpranidhivipaksabhibhutirahitatven-

180

aha4 //

mahadhyanasamadhistho mahaprajnaSariradhrk

/

mahabalo mahopayah pranidhir5 jnanasagarah // iti // (37) tatrasyam gathayam dhyanaprajnayor vyatyayo6* drastavyah / tatas 185

cayam arthah / mahaprajnaSariradhrg iti vajradhupa / mahati prajfia mahaprajna saiva7 Sariram tad8 dharayatiti mahaprajnaSariradhrk9 // mahadhyanasamadhistha iti vajrapuspa / samadhi^ cittasyaikagratalaksanas 10 tasmims tisthatiti samadhisthah / mahac ca tad 11 dhyanam ca 12 mahadhyanam mahattvam caturdhyanatmakatvat / tesu catursu 13 dhyanesu

Codd. fl (B F H E C D J); B, (3 (F H E C D J); y (F H), 5 (E C D J); e (E C), C (D J). Lacking: A, lines 177-189 (fol. 26 lost); A, fol. 27r starts, line 189, [caturdhyajnatmakatvdt.... 1 2 3 4 5 6 7 8 9 10 11 12 13

laukika BHe : lokika CF 4 mula B : mulanam P guhyapuja P : pujyapuja B balenaeDH : balanaF : balena BJH2 4 nabhogabhirati corr. (Tib.[193.3.6] Ihun gyisgrubpa {anabhoga}) : nabhogadibhirati (3 (nanabhogadi-H2 ) : bhogabhirati B dausprajna p : dausprarjfia B 4 viksepa corr. (Tib.[193.3.6] mam par g.yeng ba dang {viksepaSca}) : viksepat P : vaksepaB pranidhir Jy : pranidhi Be : pranidhi D vyatyayo con/. : vyatyaryo B : vyatyayasau CyC : vyatyasau E : bsnor bar Tib. [193.3.8] saiva BH : sa eva 6 : seva F tadB : om. P mahaprajna&riradhrk om. H laksanasBye : laksanam J : laksana D tadB : om. P caB : om. P catursu corr. : catusu A : caturBp

271

Adhikdra 5

Text

190

samadhisthah 1 // pranidhir2 iti vajradipa3 / pranidhiyata iti pranidhih4 / bodhisattvanam pranidhanavaSad utpatteh // mahopaya iti vajragandha / maham£ casav upayaS ca5 mahopayah / tata6 evopayabalenasesanavaSesasya7 sattvadhator yatha&ayabhiprayaparipuranat // ata eva prajnadhyanapranidhyupayaparamitalaksanataya bahyapuja-

195

catustayatmakam iti // sambhogikamargapritihetubhavena8 balajnanayor9 dharmadhatvavatarasvabhavatvat tayor eva 10 balajnanayoh pravibhagah 11 / tatra balam 12 Sraddhadipanca 13 / jnanam darsanamarge bhavanamarge ca / tatra daiianamarge 14 bodhyangesu15 / bhavanamarga aryastangesu 16 // tatas ca ^raddhaviryasmrtisamadhirupataya dvarapalacatustayatmakam

200

mahayanabhisampratyayapravrttitannairantaryasantustikaranatvat 1 7 // tad eva prajnapariSuddhimatrapeksaya 18 yathoktasunyataprabhedamukhena19 maitreyadisodaSasamadhyatmakam ity ata aha20 //

Codd. Q (B A F H E C D J); B, a (A F H E C D J); A, (3 (F H E C D J); y (F H), 6 (E C D J); e (E C), C (D J). 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20

sthah AB : stham (3 pranidhir a (pranidhi E) : pra&dhir B vajradipa ABy : om. 5 pranidhih a : praSidhih B upayaS ca a : upaya B tata codd. (ta A) : om. Tib. SesanavaSesasya B : Sesasya A : va^esasya (3 sambhogika AB : sambhogika (3 jnanayor a : jnanayad B eva AB : om. 0 pravibhagah ABy : pravibhogahS balam AB : balah 3 ^raddhadi ABy : ^resthadi 6 4 panca AB : pancabalam ca f) marge B : margoA : marga 5H : margam F bodhyangesu a : jnanarn bodhyangesu B : byang chub kyi yan lag bdun {sapta bodhyangesu} Tib.[193.4.7~8] aryastangesu a Tib. : jfianam aryastangesu B tannairantarya AB : tavairantarya(-arri eF) (3 parisuddhi a : parisuddhir B * matrapeksaya a : gatrapeksaya B yathoktaAB : yathoktam (3 4 prabheda a : prabhede B ity ata aha Bp (iti J) : ityaha A

272

Adhikdra 5

Text

mahamaitrimayo 'meyo mahakaruniko 'gradhih1

/

mahaprajfio2 mahadhiman mahopayo mahakrti* // iti 3 // (38)

205

tatra mahamaitrimayo 'meya iti maitreyah4 / mitram asyastiti5 maitram / paScan6 nibantam7 * / tatah8 sarvasattvesu maitrimayo9 maitrisvabhavah / maham£ casau maitrimayaS ca10 mahamaitrimayah / ata evameyah sakalasattvadhatav 11 ekaputraprematvat 12 210

//

mahakaruniko 'gradhir iti

manjusrih13 / mahakaruna analambanatmika 14 / sa yasyasti 15 sa16 mahakarunikah / agradhir iti suvi^uddhadharmadhatujnanam / tata£ caitad 17 uktam bhavati / audaryagambhiryatmako 18 'sau bhagavan iti // mahaprajno 19 mahadhiman iti gandhahasti20 / mahati prajna dharmapravicayalaksanatmika21 sa yasyasti22 sa23 prajnah24 / mahamS casau prajnas25 ca

Codd. Q (B A F H E C D); B, a (A F H E C D); A, p (F H E C D); y (F H), 5 (E C D); £ (E C), C (D). Lacking: J, lines 204-214 (fol. 29 missing); J, fol. 28v ends, line 203, ... -sodasasamadhyatmakam ity ata aha. 1 2 3

4 5 6 7 8 9 10 11 12 ^3 14 15 16 17 18 19 20 21 22 23 24 25

maharnaitrlmayo 'meyo mahakaruniko 'gradhlh om. F prajiio A : prada B : prajno P krtl // iti em. : krtiti AB : krtih(-ti H) p : krtih NS.Dav.f52, v.38d] (krtl NS.Dav. Vira) : mkhas pa [chen po thabs che ba] zhes pa'o Tib.[193.5.3J maitreyah AB : maitreya 5F : maitreyam H mitram asyastiti ABH : mitrasyastlti 5F (-syastiti E) pascan conj. : pa^caj P : pasyat A : pas*c*ata B nibantamA : jibantamBp

iatahcodd. : deni Tib. [193.5.4]

maitrimayo Bp (-nayo C) : om. A caAB : ceti P sakala AB : sarva P prematvatA : prematvotB : premakatvat eF : pramakatvat DH manjusrih AB : manju^ri P analambanatmika ABy : ana(-nam C)lambikatmaka 5 yasyasti AB : yasyastiti P saAB : om. P caitad ABy : cetadS audarya A : < >daya B : udaram DF : udanyam C : udaryam E : udara H gambhiryatmako AB : gambhiryatmako p prajno AB : prajno 6 : prajna y hastiBp : hasti A dharmaBp : om. A yasyasti AB : yasyastiti P sa AB : sa P prajnah ABy : prajnah 5 prajnah ABy : prajnaS 5H 2

273

+

Adhikara 5

Text

215

mahaprajnah 1 / ata eva mahadhiman2 // mahopayo mahakrtiti3 jnanaketuh / mahan upayo yasyasau mahopayah / tata£ ca mahakrti4 mahapanditah sarvasattvaSayanuSayajnataya saddharmaprakaSakatvat //

maharddhibalopeto 5 mahavego mahajavah / maharddhiko maheSakhyo mahabalaparakramah // iti // (39) tatra maharddhibalopeta iti bhadrapalah / rddher balam rddhibalam

220

tenopeta6 rddhibalopetah7 / mahams casav rddhibalopetas ca maharddhibalopetah // mahavego mahajava iti sagaramatih / mahan vego gatir yasyasau mahavegah / sarvatathagatapujopasthanahetutayanekalokadhatusamkramanat8 / ata eva mahajava ity ucyate // maharddhiko maheSakhya 225

iti aksayamatih9 / laukika^ravakapratyekabuddhanam 10 punyasambharopetatvan 11 * maharddhikah /

sarvatathagatajnanasambharaparipurana-

tayato 12 * mahe^akhya iti // mahabalaparakrama iti pratibhanakutah / mahanti 13 ca tani balani ceti da^a tathagatabalani sthanasthanajnanadini

Codd. n (B A F H E C D); B, a (A F H E C D); A, p(F H E C D); y (F H), 5 (E C D); e (E C), C (D). Lacking: J (fol. 29 missing); lacking B (om.), lines 225-7, aksayamatih —» mahesakhya iti.

mahaprajnah ABy : mahaprajnah 5H 2 mahadhiman AB : mahadhiman iti (3 (mahadhin iti F) mahakrtiti By : mahakrtir iti A : mahakrti iti 5 mahakrti By : mahakrtihA : mahakrti 5 maharddhi ABDE : mahariddhi Cy tenopeta AB : tenopetah (3 rddhibalopetah / maharnS casav rddhibalopetah ca maharddhibalopetah om. (3 pujopasthana Bp : pujyopasthana A + samkramanat B(3 : samkrtamanat A aksayamatih AE : aksayamati CDy : om. B laukikaAE : lokika CDHF : om. B punya conj. : punyajnana a Tib.([194.1.6] bsod nams dang ye shes) jriana E : punyajnana cett. Tib.([l94.1.6] bsod nams dang ye shes) 4 paripurana tayato corr. : paripuranataya 'ato A : paripuranataya hetor P (-taya hat or F, -taya heto E) : yongs su dzogs pa'i phyir na {paripuranat} Tib.[194.1.6] 13 mahanti a : mahati B

1 2 3 4 5 6 7 8 9 10 1] 12

274

Adhikara 5

Text

vaksyamanani 1 tesu parakramo viryarambho2 yasya 3 sa mahabala230

parakramah //

mahabhavadrisambhetta mahavajradharo ghanah

/

mahakruro maharaudro mahabhayabhayamkarah // iti // [40] tatra mahabhavadrisambhetteti mahasthamapraptah / bhavah pancopadanaskandhah / ta evadrir ajfianaSailatvat / tasya samyagaviparitajnana235

labhad4 bhanjanaSIlah sambhetta / mahamS casau bhavadrisambherta ca5 mahabhavadrisambhetta6 // mahavajradharo ghana iti sarvapayanjahah / mahavajram7 samantabhadrajnanam tad dharayatiti mahavajradharah / nibidatvan murtimattvac 8 ca ghanah // mahakruro maharaudra iti sarvasokatamonirghatamatih9 / mahamS casau krura^ ca10 mahakrurah 11 / dur-

240

dantanam 12 sattvanam mahakrodharupasamdarSanat 13 / maharaudra iti / Santo 'py 14 asau bhagavan manju^rih15 sattvavinayavaSad 16 * raudratam 17

Codd. ft (B A F H E C D J); B, a (A F H E C D J); A, p (F H E C D J); y (F H), 6 (E C D J); £ (E C), C (D J). Lacking: J, lines 229-230 (fol. 29 missing); J, fol. 30r starts, line 231, mahabhavadrisambhetta. .... 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17

vaksyamanani em. : vaksamanani codd. (vaksananani D) viryarambho Ay : vlryarasta B : vlryalambho 5 yasya ex : yasya B labhad AB5 : labhatvad y sambhetta ca AB : sambhettlti (3 mahabhavadrisambhetta AB5 : mahabhetta y vajram AB : vajra P murtimattvac corr. : murtimatvac P : murtitvac A : murticatvac B tamonirghata B : tamonirghatana AJ : tamondhakaranirgha(-gho H)tana Dye kruraScaBp : krurahA mahakrurah B Tib.fl94.2.5] : om.cn durdantanam AB : taryantanam P (taryakanam D) krodharupasamdarsanat AB : krodhasamdarsanat P Santo 7py corr. : Santo pyAB : satosyS : samtosyH : samtosa F manju^rih AB : manjuSri P vinayaBy : vineya A5 Tib. ([194.2.6} gdul ba) raudratam AB6 : raudratanC

275

Adhikara 5

Text

apt darsayatiti // mahabhayabhayamkara iti jalimprabhah / mahabhayo 1 mahabhairavarupadhari Samkara iti2 * / tasyapi bhayam karotiti mahabhayabhayamkarah //

245

mahavidyottamo natho mahamantrottamo guruh / mahayananayarudho mahayananayottamah3 // iti // (41) tatra mahavidyottamo natha iti candraprabhah / mahatyaS ca ta vidya§ ca mahavidyaS catasro dharanyo grantharthamantramudratmikah4 / tasam5 uttamah6* / ata7 eva nathah8 sarvapayebhyah sarvasattvanam a^esakaya-

250

vanmanahsamudacarasarn^odhanat9 // mahamantrottamo gurur iti amitaprabhah10 / maham^ casau mantra^ ca mahamantro mahamudratmakas 11 tasyottamah srestha ity arthah / ata eva guruh sarvasattvanam yathasayanusayajnatayabhisamayopadesat12 // mahayananayarudha iti gaganaganjah / mahayanam vaksyamanam 13 tasya nayo 14 margas tasminn arudhah15 sthita

255

ity arthah //

Codd. Q (B A F H E C D J); B, a (A F H E C D J); A, 0 (F H E C D J); Y (F H), 6 (E C D J); £ (E C), C (D J). 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

mahabhayo ABH : mahabhaya 6 Samkara iti conj. : sankara iti A : nankara B : gankarah F : gahvarah H5 om. Tib.[194.2.6] nayottamah a : mayottamo B mantramudratmikah AB Tib.(fl94.3.1J sngags dang phyag rgya'i bdag nyid rnams so) : mudratmikah P (mudratgikah J) tasam a : tsusam B uttamah B : uttama P : uttamah ^resthah A 4 tasam uttamah om. Tib. [194.3.1] ata ABY : agra 6 nathah a : narthahB samudacara AB : samudara P amitaprabhah A : amitabhadrah B : amrtaprabha P mahamudratmakas a : mahatmakasB yathasayanusaya BP : yathafrayanusraya A 4 jnataya ABFe : jnatatra 1C * samayopadesat ABF : samayapadesat H6 vaksyamanam A : vaksamanamBY : vaksamana8 nayoAB : tayoF : taya H5 arudhah AB : arudha P

276

Adhikara 5

Text

mahayananayottama iti sarvanivaranaviskambhi 1 / samsaranirvanayor anabhilapyadharmatadhigaman2 nirvikalpakam3 jnanam mahayanam4 iti / atha va trividhanuttaryena5 sambharopayagocaradharasvabhavo6* mahayanasya samgrhlta iti / yady evam katham asya mahattvam / sarvalokahita7 260

buddha bhagavantah / tena te8 mahantah / tesam idam yanam nispattaye9 niryanaya va 10 / tasmat tatsahitanam yanam11 mahayanam / mahatam va 12 yanam 13 mahayanam 14 // atha va saptavidhamahattvat 15 * / alambanapratipattijnanaviryopayakau^alyasamudagamakarmamahattvan 16 maha­ yanam / mahac ca tad17 yanam ceti 18 mahayanam / anyatra yane 19 tad-

265

abhavad20 iti / tad etat samasena trividham mahayanam / hetugocarasvabhavaparidlpanartham21 * veditavyam iti //

Codd. ft (B A F H E C D J); B, a (A F H E C D J); A, p (F H E C D J); Y (F H), 5 (E C D J); e (E C), C (D J). 1

2 3 4 5 6 7 8 9 1° 1J 12 13 14 15 16

17 18 19 20 21

nivarana BF : nivarana A : nivarana JC : nlvarna DE : nivarna H 4 viskambhl C : viskambhi AB : viskambhl E : bhiskambhl Cy anabhilapya A5 : anabhilapya B : anabhilasa y nirvikalpakam AB : nirvikalpaka P mahayanam AB : mahayananayottamam p trividha A : vrividha B : tridha P sambharopaya codd. : 'khor ba'i gnas thabs dang{samsaropaya) Tib.[194.3.5] hita BP (hito H) : mahitaA te AB : om. P nispattaye ABF : nispattaya H5 niryanaya va BF : niryanlya va H5 : niryanat A yanam AB : yana Y : yani eJ : yati D mahatam va em. : mahata va By : atha va mahatam A : ow.5 4 mahatam va yanam mahayanam om. 5 yanam AB : yana y : om. 6 mahayanam AB : mahayana Y • om. 6 vidha con/. (Tib.[194.3.7-8] mam pa [bdun]) : vidhenaa : vidhina B pratipatti ABY : pratipattir 6 * jnanaviryopaya con. : jnanavirya upaya A : jnanavirya utsaha B : jnanavirya p * kauSalya H5 : kausala ABF 4 mahattvanAB : mahatvatvat y : muhutvatvat(-tvan D) 6 tad ABY : tarn 6 ceti AB : citti p (ci*r*tta F) yane AB : yana 5F : yona H abhavad ABE : abhavad yCC hetugocarasvabhava codd. : rgyu dang yul dang rten gyi rang bzhin mams {hetugocaradharasvabhava) Tib. [194.4.1-2] 4 anartham ABH : anartha F6

277

Adhikara 5

Text

tad evam upayamatrapeksitaya caSesalaukikalokottarasamadhisvabhavatvad1 bhagavato jnanasattvasya2 tathagatajnanakayasya // tad yatha bhagavan buddhah sambuddho 'karasambhava ity evamadibhir 270

namamantraksarapadair grahanakavakyair viSesanaviSistaih 3 kriyapadarahitair viSesyo4 jnanasattvah sarvatathagatahrdayavihari manjuSrih5 Srimatam vara iti drastavyah // te 6 ca mandaleya 7 vajrasattvadayo8 mahamudrayogena9 svavarnapraharanopeta 10 vinihsrta 11 vinihsrtya 12 ca sakalalokadhatusu 13 sattvanam

275

buddhakrtyam krtva punar agatya namamantraksarani svakulavarnoparanjitani bhavanti 14* / yathanukramato 15 vajradhatumahamandalasya pragvinyastacandramandalesu vyavasthitani vicintya 16 taccakranabhisthitasya sarvatathagatajnanakayasya manjusrijnanasattvasya 17

Codd. £1 (B A F H E C D J); B, a (A F H E C D J); A, p (F H E C D J); Y (F H), 6 (E C D J); e (E C), C (D J). 1 2 3 4 5 6 7 8 9 10 1] 12 13 14 15 16 17

caSesa A : va^esa (3 : aSesa B sattvasya ABy : kayasya 6 visesana a Tib. [194.4.4] : anekavisesana B vi^esyo ABH : viSesya F5 mafiju^rih a : manjuSri B teB(3 Tib.([l94.4.5] de rnams) : tena A mandaleya AB : mandaleya (3 adayo a : adaye B yogena ABy : yoge 5 varna AB Tib. ([194.4.5] sku mdog) : kantha P vinihsrta B : vinisrta P : ni^rtah A vinihsrtya B : vinisitya A : om. P dhatusu AB : dhatu p bhavanti conj. : bhutva ABy : om. 6 : bsgyur\eTib.[194.4.6J yathanu AB : yamartha 5 : yathatha y vicintya a : cintya B manjusrijnanasattvasya om. B

278

Text

Adhikdra 5

prarinirdistarupam 1 atmano 'dhimucya2 tasyaiva bhagavato jnanamurter3 280

advayaparamartham4 namasamgitirn 5 nama cudamanim* yah kaS cit kulaputro va kuladuhita6 va mantramukhacaryacari bodhisattvas trihkrtva 7 kanthagatam avartayisyati / pustakagatam va pathamanah8 pravartayisyati // Sucir9 mauni 10 bhutva tusmmbhavena 11 yatha ka£ cin na 12 s"rnoti nama-

285

mantraksarapadani yathavad avartayan13 na cirad evanuSamsaparipathitan14 gunaganan anyamS ca prapsyati / paramarthanamasamgitisamdharakah 15 purusapumgavah16 * susambhrtapunyajnanasambharah ksiprataram sarvabuddhagunan 17 * samudamyanuttaram 18 samyaksambodhim abhisambhotsyate

/

279-80 cf. NS.Dav. 65, 3: jndnamilner advayaparamarthdm namasamgitim. 282 cf. NS.Dav. 66, 23: pustakagatam va pathamanah pravartayisyati 287-9 cf. NS.Dav. 68, 4-6: paramarthanamasamgitisamdharakapurusapumgavah susambhrtapunyajnanasambharah ksiprataram buddhagundn samudanlyanuttaram samyaksam­ bodhim abhisambhotsyate; cf. NMAA 12 [D.192r3; B.85r2] paramdrthandmasamgltisamdhdrakah purusapumgava(em. : -pumgala D : pungala B) /'// pradhdna ity arthah.

Codd. n (B A F H E C D J); B, a (A F H E C D J); A, (3 (F H E C D J); y (F H), 6 (E C D J); £ (E C), C (D J). 1 2 3 4 5 6 7 8 9 1° 1] 12 13 14 15

pran corr. : prag a : prig C 'dhimucya ABJ : 'simucya DC : 'mimucya H : 'pimucya FE murter a : murtir B paramartham co/r. (NS.Dav. [65, 3J) : paramartha codd. samgitim ABy : samgiti 5 kuladuhita a : kulahitaB trih B : tris A : tri P pathamanah(-mana H) a NS.Dav. [66, 23J : pathana B ^ucirABH : Suci F6 mauni By : mauni 6 : mom A tusnim ABH : tusni F5 na a : om. B yathavad avartayan A : yathavartayan P : tathavartayan B paripathitan A : pari< >tanB : paripathitani(-nani D) (3 paramartha B : paramartha a * samdharakah codd. (sanvarakah B) : sam dharaka NS.Dav.[68, 5] 16 purusa a : purusahB 4 pumgavah con/. (NMAA 72; NS.Dav. [68, 5]} : pungalah AByE : pumgalah JDC : [skyesbu] gang zag {pudgalah} Tib.fl94.5.3] 17 sarvabuddhagunan codd. Tib.([194.5.4Jsangs rgyas thams cad kyi yon tan rnams) : buddhagunan NS.Dav.[68, 6] 18 anuttaram ABJy : anuttarayam DeH2

279

Adhikara 5

Text

analpakalpan 1 aparinirvanadharma2 * sarvasattvanam anuttara-

290

dharmadesako3 * daSadiksaddharmadundubhir4 dharmaraja5 iti // asya ca vajradhatumahamandalasya namamantraksarapadanam ekatra pindlkrtanam sadasitir6 namani 7 bhavanti / tani ca mandaleyadevatanarn8 pravibhajya yojaniyani / tatra catvari jfianasattvacandramandalayoh / 295

akarasya trim / sattvavajryadinam panca / £a£vatadmam ekadaSa9 / vajrasattvad 10 arabhya 11 yavat sarvanivaranaviskambhlty 12 etatparyantam 13 s"esani* namamantraksarapadani viSesanaviSesyabhavena 14 yathakramato yojyani 15 // gatha£ 16 caturdasa / namamantraksarani 17 sadasitir 18 ankato 'pi / ity amnayah // aryanamasamgititikayam namamantrarthavalokinyam bodhicittavajrasya

300

vajradhatumahamandaladhikarah pancamah19 //

290-91 cf. NS.Dav. 68, 7-8: analpakalyanaparinirvdnadharmah sarvasattvanam anuttaradharmadesako 'dhisthito dasadiksaddharmadundubhir dharmaraja iti. Codd. n (B A F H E C D J); B, a (A F H E C D J); A, (3 (F H E C D J); Y (F H), 6 (E C D J); e (E C), C (D J).

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

19

analpa AB Tib.([194.5.4J mmgpo) : anatya P * kalpan conj. (Tib.[194.5.4] bskal pa) : kalyan codd. (kalyan E) : kalyan NS.Dav.[68, 7]) aparinirvana AB : aparimana (3 4 dharma codd. (dharma J) : dharmah NS.Dav. [68, 7] : [yongs su mya ngan las mi 'da' ba'i] chos su 'gyur Tib.[194.5.4] anuttara ABF : anuttaram H5 4 deSako codd. 7V^.([chos] ston par byed la) : de&ko 'dhisthito NS.Dav.[68, 8] NS.Tib.D.N. dundubhir AB : dundubhi P raja ABy : raja 6 sadasltir AB : sadaSIti (3 namani ABH2 : nani P mandaleya AB : mandaleya P 4 devatanam ABH2 : devatinam P ekadasa Ay : ekadasa 6 : ekadasah B vajrasattvad ABy : vajradS arabhya ABy : arebhya 5 (-bhyah J) nivarana ABy : nivarana 5 4 viskambhity corr. : viskambhity codd. (bhiskambhlty y) paryantamAB : paryanta y : paryenta CC : paryantah E vis"esanaAB : visesena 5 : om. y yojyani ABH2 : yojyati (3 (-jyani H, -jyeti J) gathaSB : gatha a aksarani codd. : yi ge'i tshig mams {aksarapadani } Tib.[195.1.1] sadasltir AH2 : sadasitir B : sadaSitir 3 pancamah By : pancama A : pancamah // gatha caturdasah 5

280

Insignificant Variants

Text

Insignificant Variants Adhikara 1 1.1 Sriyam codd. (Sriyam E) 1.2 namasamgltim A : namasamglti P (namasamgltirn H) 1.4 dharmao? Cy (anus"amsa mantra vinyasah suppl. H 2; 1.18 caturda^a codd. (catudaSah A)

anuSamsa mantra vinyasah suppl. C°) kayasthito ADy : kayagrasthito £

adhyesanah e

:

1.20 adhyesana ADy

1.19 4

Sakyamunir p (s^ikyamuni C) : om. A 1.21 sadhu 5 : sadhur y (sadhu H2) : om. A 1.22 AyE sakyamunir : ^akyamuni CDJ

1.24 arapacanaya Ay : arapamcanaya 6

tisr 5 (titrbhir D)

1.25 sambodhi P : sambodhih A 4

sucitah AF : sucitam H5

yavan codd. (yadvat F)

kramah A : krama P

4

ghoso P (ghosa D)

:

om. A

4

1.26 mahayananayottama P : 1.28 vajra codd.

mahamayanayottama^ A 4 iti P : om. A 4 gathabhir P : om. A (vrajaF) 4 paSa AH2 : pas"o p 4 iti AH2 : om. P

tisrbhir Ay :

4

(alapancanaya E, arapacanaya D) 4 nama A : namar p (nama H 2)

1.30 stutih AH2 : stuti p :

1.31 padonasardha Ay

4

padonasardham 6

(padanasardham C) 1.32 stutih AH2 : stuti p (stutir E) 4 jnanarcihAH2 : jnanarci P 1.33 dvacatvar P : catvar A 1.35 stutih AH2 : stuti p 1.39 stutih AH2 : stuti P

1.34 stutih AH 2 : stuti P 1.37 stutih AH2 : stuti p 1.41 stutih AH2 : stuti p

1.42 stutih Alf : stuti p 1.43 stutih A : stuti p

4

para codd. (om. A )

1.38 ta A5 : te yD 1.41 garbha p : om. A

4 uttara codd. (utara C, uttara C c)

1.44 adhyardha AH : addhyardha F6 4 s"atenanus"amsa A5H2 : Sata< >sa F : s"ate< >sa H

1.47 sarvesam codd. (om. A)

281

1.50 kathyeta A5H2

:

kathyete y

Insignificant Variants

Text

samdhavayantah CDJ (sam*avayantah C)

:

1.51 sambhavayantah AEy

1.52

1.53 pramanyat AH26 : pramat(mot F)yat y

vacaniyam A5H2 : vacamya Y

+

1.56 prakaSana codd. (prakaSanam Y)

pravartanta AH 2 : pravartata (3 (pravarta J)

1.57ganam P : gmam A 4 gitih AHE : giti CDJF 1.58 samgitirAH : samgiti 5F 4 tantra codd. (tantram J) 1.59 vinaya codd.(vinaye F) 4 laukika codd. (lokika F) sarvasthavarajangamani ca illeg. D

4

1.60 tesam namnam samgitir iti ACDJ : om. E :

tesam samgiti samgiti Y (corr. H2) 1.61 paricchedam (3 (pariksedam E) : paricchadam A 1.62 abhidhanabhidheya codd. (abhidhotadheya Y)

upayopeyalaksano codd.

*

(laksane J)

1.63 mukha codd. (mukhe H)

tattva codd. (sattva H)

1.64 bhyam

codd. (* E)

1.65 viddhaya Ae : ddha(dra H)ya Y : viccha(cha J)ya C

1.66 tata^ ca

bhavana om. F

4

yojanlyam om. H

4

1.67 prayojanaprayojanam P : prayojanam A

1.68 padarthas codd. (illeg. D) 1.74 namasamgltir codd. (namasamgltir iti J) 1.77 puto codd. (puto CB^ 4 agratahAH : agrata F6 1.78 sambandhah AEJH 2 : sambandha FDC

4

vajradharah AE.1H2 : vajradhara FDC

1.79 s~riyo codd. (^riya H)

4

vidyante codd. (-ate Y) 1.80 yasyasau codd. (yasyaso A) 4 snman codd. (^rlman iti J) snsamvare A5 : ^rlsamvara Y

1-83 Sasako codd.

(^asaka FC) 1.84 vineyakatvat codd. (vineyatvat J, vineyekatvat H)

1.8 8 bhavanlyatvat

codd. (bhagavanlyatvat D) 1.91 ragadibhih codd. (ragadibhi F)

1.9 3 sarvasattvams

1.81atmakoA5 : otmako Y

^

codd. (sarvasattvas JF) 1.94 kaya codd. (karya J) 1.95 paravittani codd. (paracittani e) 1.97 sattvarthat -* apnuyat illeg. D 1.101 kuli^varah codd. (kulis"e$o D)

1.99 avapnuyad codd. (adapnuyad F) 1.107 vibuddham

codd. (vibuddhas HE)

1.108 vibuddhapundarikam / vibuddhapundarlkam codd. (vibuddhapundarlkaceti / 1.110 prakarsenotphullam codd. (prakarsenautphullah Y)

vibuddhapundarlkam D)

1.117 mukhahco^. (mukhauh F) 1.120 durdantadamakair iti om. E 4 tair codd. (ttair H) blbhatsarupitah Y

*

4

kimrasa codd. (kimraso F)

ata eva 5 : atra eta Y

1.125ullalayadbhirco^. (ullalayedbhirY)

*

1.122 blbhatsarupinah 5 :

virablbhatsa 5 : dvlrablbhatsa Y

+ kaih codd. (om. F)

codd. (karunayo DC) 4 yesam codd. (yasam H) 1.132 karuna om. C

282

1.121 te om. E

1.128 kanmaya 1.133 sukham

Insignificant Variants

Text

4 jagac ca jagac ca jagac ca

5 : mukhamy 1.134 lambanatmika 6 : lambamatmikay

jaganti £ECC : jagac ca jaga Cy 1.136 aha codd. (aha C) 1.140 aSayo codd. (asaya E) 1.141 mudita codd. (mudito F, om. H) codd. (karar £)

1.147 rupibhir codd. (rupinir e)

1.149 vigraho codd. (vigrarhe E)

vaksya(ksa D)mana C : vaksyamanamana £ anuSasanat codd. (anus"asanan C) visinasti A

: viSinasti cett.

gunayogad Ac )

1.148 karair

1.151 vaksy(ksa A)amanam Ay : 1.155

1.154 natham codd. (natha* A)

1.156 sravakam codd. (Sravaka A) 1.157 vis"inasti H : 1.162 gunayogad cocW. (gunayogad D; gunayod A; 1.166 tathagatah codd.

1.164 prayatnasya codd. (prayannasya A)

4 tarn

(svathagatah E) 4 gata codd. (om. J) 1.167 gacchatiti codd. (gacchanti H) codd. (ta*t* A) 4 pranamyeti codd. (pranamy*ai*ti A) 1.175 jalabhisambodher codd. (lajabhi- E)

1.168 sthito codd. (sthita e)

1.176 yathaham codd. (ham D)

codd. (lobhi H) 1.181 duruttaratvac caco^rf.(duruttaratva< > F) (pratidya C)

1.192 chastety codd. (chastebhy

1.195 tattvajna codd. (sattvajna £)

1.196

1.202 namasamgltis AH : namasamgiti (3

1.203

Ae) 1.193 sattvahAEJH2 : sattva CDy as"ayavid codd. (a^ayadvid £)

1.1 8 4 pratigha codd.

1.188 tatpraptaya codd.

1.187 pratilambhah codd. (pratilambha D)

(tatpraptaye F) 1.191 s"a£vata codd. (s^asyata(tat E) e)

labhi

t

vaksyamanena AH2 : vaksamanena (3 (vahumanena J)

1.205 mahosnisas AJH

:

mahosnisa DF£ 1.206 abhisekatmakatvat codd. (abhisekayatmakatvat E) 1.208 murtir codd. (murtlh C, om. F) (svayambhu C)

4

murtis codd. (murtih C)

1.212 jnanasya codd. (< >sya F)

1.210 svayambhur codd.

1.214 manjus"rijnanasattvah codd. 1.216 hrdayaviharitvat

(manju^rir jnanasattvah A) 1.215 manjus'rih AH : manju^ri F5 codd. (dvayapariharitvat H)

1.218 advayajnanam codd. (yajnanam H2)

namasamgltis corr. : namasamgitih Ay : namasamgiti 6 gambhirartha codd. (gambhirartham J) 4 tarn codd. (ta H)

asamatvam codd. (asama< > F)

1.226 sa codd. (so J)

4

1.228 sa codd. (om. JE)

1.230 punar JHE

1.229 mahartha codd. (mahartho C)

1.222

1.243 atitair JE : atltai cett.

:

puna DCF

1.231

1.244 bhasita codd.

(bhasitahj) 1.248 hrstair HJ : hrstai cett. 1.255 pradarSitam y£ : dar^itam C

1.256

1.257 sambandhah codd. (sambandha F)

1.260

sarvasambuddha codd. (sambuddha C)

283

Insignificant Variants

* hanaya iti (?C :

yathavatCy : yathadat C : yathaca E 1.262$esaJy : aSeso eD

hanayctiE : hanaya < >y : hanayiti H2 1.263 grhyante codd. (grhyanta C) yJ : aSeso eD

1.264a$esajfianam yJ : aSesojfianam eD

*

aSesa

aSesajfianahanis (aSeso-

*

1.271 pranata codd. (prahva

F) codd. (ajnanahanis D) * tadartham CH : tadartha eF D) 4 'tra codd. (ca J) Adhikara 2

2.4 antaram corfrf. (amntaram C) 2.5 mauneya 6 : maunaya y * yogan C : yogat y E 5.8 bahalataram codd. (bahalatara H)

2.6 gatarthah corfrf. (gatartham C gatartha D) 2.9 ^afikha codd. (sakha C)

2.12 sattvanam om. D

2.13-14 pratyabhasateti codd.

(pratyabasat*ti C) 2.16 trini (trini E)5 : strini C * Srimayajala codd. (^rlmalajala C) 2.17 tantre codd. (tantra F) * pathyate codd. (pa< codd. (ca F)

*

trini codrf. (trini E)

> y; pathyale suppl. H2 )

(jihvandriya H)

2.23 bhur JEy : bhu CD

2.24 trayanam 6 : trayonam y

codd. (sahasryam H)

2.18 sahasryam

2.19 mukhena codd. (mukhana H) 2.20 tac codd. (ta H) »

2.21-22 jihvendriya codd.

svar iti yC : sver iti C : sveti 1^

* praka^ana codd. (prakasena H) 2.25 vajrapanim 2.27 anu&asanam

2.26 skandha AE : skanda cett. (skanda Ec)

codd. (vajrapatu H)

* ca

codd. (anu^asana H) 2.30 trilokam apurayantya codd. (trilokanam apurayantya H) vajrapanirn A(CF : vajrapani EH

*

4

2.3 2 opetatvat codd.

vaca Ay£ : vacam C

(opetatvatvat H, om. J) 4 brahmasvara codd. (brahmasvaram A) 4 ^abdo codd. (svaro E)

2.39tatrabhyantaracorfrf. (tathabhyantara J) 2.44 vajrapanim A£ : vajrapani ey

pratyabhasata AJ

:

pratyabhasatah eyD

*

2.46 bodha codd. (bodhah F)

2.54 vajradharetyadi codd. (vajradharatyadi H) 2.55 yat codd. (yatt H) 2.55-56 namasamgltis codd. (namasagitis C) 2.57 sadhu codd. (sadhum D)

*

kimartham AJEH :

kimartha CDF * prstah codd. (stah D) 2.63 ^uci codd. (rauci F) *

ukte codd. (ukta

C) 2.67 dvitiyah codd. (dvitlya C) Adhikara 3 3.4 Sakyamumr codd. (^akyamuni D) 3.6 gatha P : gatha*m* A * tarn codd. (m .IE) 3.7 vaksyamana AJ : vaksamana yeD 3.11 svabhavatvad JEH2 : svabhavatva ACDy

284

Insignificant Variants

Text

4 vairocana codd. (vailocana E) (aksobhya A)

3.15arthah codd. (artha C)

3.16 aksobhyah codd.

3.19 vairocana codd. (vailocana E)

Adhikara 4 4.2 tatradau codd. (tatradau C) 4 mantra codd. (matra CD) 4.3 imam codd. (ima F) 4 jnanasattvasya codd. (jnana F) 4 sarvatathagata codd. (sarvatathagatasya H) 4.5 kanio codd. (kama F)

4

bodhisattvo codd. (bodhisattva D) 4.6 atmakaS HE : atmakah /

ACCF 4 caturbhih A5 : catubhih y 4.7 sattvan codd. (sattvat D) 4.9 mam (3 : ma A

4.10 imam AEtH : ima CF

4.11 bodhicittam Aye

:

4

4 yavad (3 : yavad A

velam (3 : valam A

bodhicitta C

pratigrhnami codd. (pratigrahnami Ec)

4.17 mudram codd. (mundram C) 4

tattvatah

corr.

:

4

tatvatah codd.

4.19caturdanam codd. (caturdana C) 4 satkrtvaco^. (sankrtva C) 4.21 saddharmam ADEy :

saddharrna CJ 4 pratigrhnami codd. (pratigrhnati H) 4 bahyam guhyam

codd. (bahya guhya J) 4.23 tattvatah corr. : tatvatah codd. (bodhicittamm C) 4.30bodhim codd. (bodhi D) F)

4.31 bhavayet codd. (bhavayan F)

4.25 bodhicittam codd.

4.30-31 pratibimba codd. (pratimba

4 yatha (3 (tatha E) : yatha A

4.31-2

samnihitah codd. (samvihitah J) 4.34 adarSajnane codd. (adars"anajnane E) 4.36 magnas AH

:

magna 5F

4.38 asamvrtya codd. (samvrtya C)

4.40 adhimuncet codd.

(adhimucet D) 4.41 anatmanas codd. (a / natmanas C) 4.43 prajanati (3 : prajanati A 4 yatha (3 : yatha A 4.43-4 tatha prajanati sarvadharman (3 : tatha prajanati sarvadharman A 4.46 dharmanam codd. (dharmam D) 4.47 uccarya codd. (ucarya C, uccaya D) 4.48 mantri codd. (matri CD) 4.56prsthaA : prsta (3 4.57-58 bodhicittam Aye : bodhicitta C

4.58 vaiyava codd. (naiyava J)

4.60 akas"adesa codd. (akarades~a J)

4.62 sahasra codd. (sahasram E) 4 bhadram codd. (bhadra J) 4.65gatva codd (Satva H)

4.66 samsodhya codd. (suSodhya F)

4.67 ca coddd (ra H)

(cittayet F) 4.69 ksetra AJHE : ksatra CD : ksetre F

4.68 cintayet codd.

4.70 durdharsam AJF :

dudharsam £DH 4.71 Suklam codd. (sukla A) 4.73 caityam AyC : caitya e 4 dayam codd. (dayam E) 4.75 bhumim codd. (bhumim E) 4

adhitisthet codd. (adhisthet E,

-sthat F, illeg. D) 4.78 caturaSram AJH : caturasYa £DF 4 dvaramAvC : dvarae 4

285

Text

Insignificant Variants

catus codd. (catu CD)

4.79 patta codd. (patu C) 4

srak codd. (sras" C, srajas" E)

4.83 varnais" codd. (varnes" C, varnaS D) 4.89 kuryat codd. (ksaryat F) codd. (mamabhena F) 4.90 kuryad codd. (kurya F)

4

4.96 gajasanam codd. (gajasana D)

samasena

samdhisu (3 : sadhisu A

4.92 devatanam codd. (devatatam F) 4.93 madhye codd. (madhya J) 4 : asatam CD

4

asanam AyJE

4.100 mahavairocanam AJy :

mahavairocana eD 4.101 ciiiamcodd. (cintam D) 4.102mukhanam codd. (mukham E) 4.104 abharanam codd. (abharanaSm A, asaranam C, abharanas" H) codd. (vailocanam E)

4 catur codd. (catu F)

4.106vairocanam

4.107-8 pannam adibuddham codd.

(pannam < ~4 >m adibuddham C) 4.108 pancananam iti om. E codd. (varnopatam C) 4.111 haritam codd. (haritam F)

4

4.110 varnopetam

^vetam codd. (s"veta F)

4.112 vicitra codd. (vicaitra E) 4.113 caturdha codd. (catudha C) 4.114 karair AeH : karai CF

4.119 giram codd. (garam H) 4.120 sambandhah codd. (sabandhah A)

san codd. (sat A) 4.121 vaksyamana AJ : vaksamana eyD

4

petam codd. (peta D)

4 kimatmikam codd. (< >tmikamF) 4 dvayam codd. (hrstam F) codd. (grakams' H, gahakam F) 4.124sma codd. (smah A)

4

4.122 grahakam

4.124-125 kimvisistam

codd. (kimvisistam J) 4.132 amrta codd. (amr C) 4 dvitiya codd. (dviya F; dvitiya Fc ) 4.133 gagana codd. (gagana C) 4 gaganabhumi codd. (gaganabhumi CF)

4.134 sa

manorama codd. (so manarama F) 4.137 sasthi AJFE : sasti CDH 4.140 upamaco^W. (u>»»pama A) (anuttaro F)

4.141 sarvopama codd. (sarvoprama C)

4.144 yogi codd. (yami F)

4

4.142 anuttara codd.

prapuriketi codd. (prapurikati H)

4.145 parinispanna codd. (parinispanno D) 4.147 buddhanam codd. (buddhanams J) tryadhva codd. (adhva E) 4.148 buddhanam codd. (buddhana C) (om. F)

4.152 hrdaye codd. (hrdagre J, hrdaya F)

4

mula codd. (mulam E)

4

4.149 buddha codd.

4.155 pratityotpada codd.

(pratityatpada C, pratitpada D) 4.156 catur codd. (catu CD) (sabhrsya F)

4

4.158 sattvasya codd.

sitavarna codd. (sitavarne F)

4.159 mekhalayam codd. (mesalayam F) 4.162 varaa codd. (varta J) 4.164 murtaye codd. (murteye C) 4.165 om vagl^varaya A 5 : vaglSvaraya y

4.168 bhagavan codd.

(bhagavataE) 4.170 ah codd. (jnah E) 4.175 samapannam Aye : pannam C

286

4.176

Text

Insignificant Variants

Santarasopetam Ay (bantam- H) : Santasopetam 6 (Santarsepetam C)

4.177 vibhavya

codd. (vibhavyah CF)

4.182 gabhastimanti co^.(garbhasti F, gabhastirmalamta E)

4

vicintya codd. (vicintya F) 4.183ra$mibhih codd. (ras"mibhi D) 4.184ras'mayas' AJy : 4.186 ksetrani codd. (ksatrani CD)

raSmaya De

4.187 ksetresv codd. (ksatresv C)

tevu C bodhi C

:

4.192 bodhisattvo(-va F) Aye

4.193 alambayann codd. (alabayann F) 4

:

tesu Aye

4

4.194 tadrupam

kayena codd. (kayeca A)

codd. (tadrupam tadrupam J) 4.19 6 rupakayena codd. (rukayena A) 4.197 gaganatala rupa codd. (rupamD) 4.199 anena codd. (enena £)

codd. (gaganata*a C) 4

4.202

purvam codd. (purva CD) 4.204karodbhavam a : karaudbhavam B 4.207astasuBp : astasu A 4 tiksnaya a : tihnaye B 4.208 tiksnam a : tisna B (vastram H)

4 vama codd. (vamam H)

spharana codd. (sphalana CD)

4.209 vastra codd.

4.210 khadga a : khadgam B

4.212 saniharana a : sampharana B

4.211

4.213 varnam

ABye : varna C 4.215 purva van C : purva vat AC Y : purvat B 4.217 odbhutam a : audbhutam B

nilavarnam codd. (niravarna C)

4

4.218 pancaciramcinapatta-

paridhanam(-dhana D) a : pancavinapatapanidhanam B 4.220 kuryad codd. (kuyad B) vairocanam ABye : vairocana C

4.222 pari-

nispannam a : paripanispannam B 4.223 mantrair codd. (mantrai BCD)

4.225 clram

4.221 madhye codd. (madhya F)

4

codd. (cira CH) 4.226 vicitra codd. (citra E) 4.227candram codd. (candra F, ca*ndra* B)

4.229 mandala codd. (manda D, mandalamandala E)

purvavat BCJy

4

caresv codd. (carasv CD)

4.234 ^ekharam BJHFE : ^esaram CDF candram P : ca*ndra* B

4

4.231 purvavan DE

4.233 kayam By

:

:

kaya 6

4.235 taddhrdi 3 : taddlir taddhrdi B

4

dhyatva codd. (dhyatvat H) 4.239 mulamantradi codd.

(mu< 4 > F) 4 samyutam codd. (samyuktam F; samyukam H) 4.240 bijani codd. (bijaF) 4.241opetamp : aupetamB 4.243 ornkaracodd. (*kara C) 4.244$veta(3 : ^veja B

4.249arapacanam codd. (arapamcanam C)

4.252 candropary codd. (candrepari CD)

4.251 panim He : pani BFC

4.254 kramah codd. (krama J)

4.256 vedi-

tavyani p (viditavyani F) : om. B 4 mayajala codd. (mayala H) 4.257 avalokinyam p : avilokinyam B 4.258 tisrah codd. (tisra DF)

287

Insignificant Variants

Text

Adhikara 5 5.7 edanlrn codd. (edanl DF)

5.3 athedanim codd. (athedani F)

5.8 varnagrya codd.

(varnagraH) 4 'gryahBJ : agryah eDF : gragryahH 5.9 mahams" codd. (mahaS B) 5.14 yoginam codd. (yoginam E; om. B) 5.16-17 pratipaksataya codd. (pratiksataya H) 5.18 viparyasabhi codd. (viparyasobhi C) 4 animittapranihita codd. (animitt< >hita F) 5.23 vajri codd. (vamjrl B, vajra F) 5.26 tena codd. (sona D)

4

4

tatra codd. (titra B)

(abhilapanam abhilasah DH)

5.27 sarvabhilapa BCJ :

jnanena P : jnanai B

sarvabhilasa DHF 4 hetvagrya codd. (hetugrya y) 4

5.25varjitap : rjita B

5.28 abhilapanam abhilapah codd.

abhilapah Je : abhilayah B

: abhilasah Dy

5.30 pradhanam codd.

5.29 sarvabhilapah BEJ : sarvabhilapa C : sarvabhilasah Dy

(pradhana* B) 4 sarvavak P : sarvavaka B 4 suprabhasvara codd. (yaprabhasvara y) 4 karma (3 : kamma B 5.31 sarvavak codd. (sarvavaka B) 5.33 des~anaya$ p : de&tyaS B prakrbha B

4

yatha codd. (yaya J)

5.36aksaraih codd. (aksarah F) 5.37 prakrtya p :

5.38 odaya codd. (adaya F, edaya B)

dharmalabana(-taD)5 4 krpady codd. (krpa* B)

5.39 dharmalambana By

:

5.40 viparyasat codd. (viparyasata

B) 5.41 pratyaveksana P : < >tyaveksana B 5.43 kramam P : krama B 4 bhavam codd. (bhavaH) 5.45atraP : aB 4 cadav Bey : dav C

5.47 ucyate P :

acyateti B 5.51 maharago BJH : maharaga eDF 5.52 ratimkara codd. (ratikara H) raga ByC : raca £ (raga If) 4 ratim codd. (rail D)

4

5.52-53 mahasukhatmikam y^ :

mahasukhotmikam £ : mahalmikam B 4 karotiti P : kamotiti B 5.53 ratirnkarah Je : ratimkara DF : ratikarahH : ratimkah B 5.54 dvesa p : dvetha B 5.57 sa ca om. J 4 admamyJ : adinam eD : an B 5.57-58 ity arthahyeJ : ity artha BD 5.58 adhikrtyaha P : adhikrtyapha B 5.60 mahamoho codd. (mahamoha CF) 5.61 moho H : moha B : mohah 5F 4 H reads vairocanam dhiyam mohah (ditto.) after dhiydmmoho 5.62 mudha codd. (om. e) 5.64 mahamaheti cocW. (maheti E) 5.66caritanp : carritan B

4

adhikrtyaha codd. (adhikrtyaha B)

5.69 sarvalobhanisudana iti codd. (sarva-

mukhena sarvasamadhihetutvat sarvalobhanisudana iti ditto. J)

5.70 paris~odhanarthatvat

codd. (Sodhanarthatvat J) 5.71-72 H repeats: sunyatadicaturvimoksamukhanam dharma-

288

Insignificant Variants

Text

dhatvayatanamukhena sarvasamadhihetutvat (ditto.) 5.73bahirdha codd. (bahirva CD) 5.74 sariranupalambha Bye : Sariramupalambha C 5.75 lokadhyavasaya 6 : lokasya / dhyavasaya B : lokapyavasaya y 4 tata codd. (nata B)

4 mahamodo om. F 5.78 vajrasattvah BJ

H) 5.77 saukhyo (3 (sausyo C) : sokha B : vajrasattva DFe : vajrasattvam H 4 saukha B

4

5.76 samadhi codd. (samdhi

kamyata codd. (kamota F)

rajah codd. (rajrah B) 5.80 sadhuh (3 : sayuh B

5.79 saukhya (3 : 5.81 nispatter codd.

(nispanter D, nispattar F) 5.82 upalambhady (3 : uyalambhady B 4 adhisthanatvat 6F : adhisthanatvat H : edhisthanatvat B 4 viparyasa By : viparyosa DC : vipayosa E : viparyase J 5.83 odaya ASH : adayo F : audaya B 5.84 tad eva codd. (tad e C, tad evamH) 4 jnaneco^. (jnana F) 5.85 paramarthaSunyata codd. (paramarthah Sunyata E)

5.86parigrahat a : paratuhat B 5.88 rupo codd. (rupa F) 4 varno ex : varna B

5.89 mahodaro codd. (mahadaro £)

5.90 rupam codd. (rupa C)

yasya / sau B 5.93 ksetra codd. (ksatra C)

5.91 yasyasau a :

5.95mahad codd. (mad F) 5.96 maya

codd. (maye F) 4 jnana codd. (jana E) 5.97 mahodaracx : mahodana B 4 yasyasau S.lOOhasah a : phasah B

5.98 mahodaracx : mahadara B

a : masyasau B

5.101 lati codd. (lani E) 5.104 parikara a : parika B 5.105 mukhena AB : mu< >na y : muna 6 5.109 evatyantika ADy : evabhyantika j£

4 atmakam codd. (atmaka J)

5.107 nispatti codd. (nisprati J)

5.110 anavaragro (3 : anavaragro A

4

avagrahad codd. (acagrahad A, avaragrahat H) 5.113 maya B(3 : maya* A 4 virahat 4 jyotir

a: virahot B 5.115anku^oa : ankuSau B 5.116 klrtir Jy : kirti ABeD

codd. (jyoti BD) 5.117 tatra codd. (trata B) 4 vajraow. C 5.119 casau codd. (casav A)

4 prajnayudha codd. (ayudha A)

5.120 ankuSo codd. (anku^ro B)

codd. (yani J) 5.121tesam codd. (tesamm A)

4

4

'granir

anku^abhuta codd. (ankus'obhuta J)

4 ivanku^ah (3 : ivaankus'ahA 5.123 sYestha codd. (Srastha C) 5.125 nirdistanam a : niddistanam B

4

svabhavatvat codd. (svabhavatvot B)

5.126 lokadhatu codd.

(lokahdhatur H) 4 vyapini a : vyadhini B 4 klrtir codd. (kirti B) 5.127 mahajyotir codd. (mahajyotieD)

4 dyutir a : dyuti B 5.128 dosanam codd. (dosano H) 5.131

manah a : mana B 5.132 jnane codd. (jnate J, jnana F) 5.135 kramah codd. (krama J)

289

Insignificant Variants

Text

5.143 mayeti codd.

5.137 btiacodd. (< ~2 >bala C) 5.142 jalikah codd. (jalikah B) (om. J)

5.144 vipulodara codd. (-audara B, -adhara F)

4

mahamS codd. (mahas B)

5.145 vidvancofiW. (vidvana B) 5.146 oparjanat a : oyarjanat B 5.147 kusida ABF 5.148 mayarthasadhakah codd. (mahamayarthasadhakah Ec)

: ku&da 6 : ksuSIda H

5.152 mayendra

5.150 mayaratih ABJH : mayarati £DF 5.151saktoAB : s~akto (3

(samu* A)

5.155drastavyah codd. (illeg. A)

samuham codd.

4

(3 : mayendri A : mayaindra B 5.153 nirmana a : nirmas~a B

5.156 tataS ca a : tatrasra B

5.158

mukhena codd. (mukhera B) 4 samasta codd. (sammasta B, samasta F) 4 vighnopas'ama a : cighnayas'ama B 5.159 caisam(3 : caisaA : varnsamB 5.161 kausidya codd. (kausiya H)

5.164 dhiro codd. (dhira F)

codd. (pati CF) 5.167 dharmabhisekatmaka P : dharmabhirsekatmaka B krsta BJy : lokatkrsta eD

4

5.166 patir

5.165 vajra codd. (vajra C)

'granir codd. ('gram J)

5.168 lokot-

5.169 slla codd. (gala F)

4

5.171 siladharas

samadhav codd. (samadhar B) 5.170 das"aku£ala p : dasyaku^ala B

codd. (ladhara^ E) 5.175 vajranrtya B6 : vajranrtya(-tyah H2 ) y 5.178 nispattau codd. (nivpattau B)

5.180 abhibhuti codd. (abhibhutir H)

4

5.183 mahopayah P : mahopavah B 4 sagarah codd. (sara C)

rahita P

:

raphita B

5.184 prajnayor By :

prajfiayor(-yo E) 6 4 drastavyah codd. (drastevyah B) 5.185 cayam codd. (cam B) dhupaFS : supa B : dhupah H 5.187sthaBEJ : stho CDF : stha H 4

4

samadhiS

codd. (samadhi J) 5.188 tasmims codd. (tasmaims E) 4 tisthatiti codd. (tisthantlti J) 4 4

mahattvam J :

5.190 pranidhiyata codd. (-dhlyayata C, -dhiyayata E)

5.191 utpatteh

mahac 3 : maphac B 5.189 dhyanam codd. (dhyana D, dhyann B) mahatvam cett.

4

AeJH : utpanteh B : utpatte D : utpa< > F

mahopaya codd. (< >hopaya F)

5.192mahopayah codd. (mahopayah A) 4 evopayabalena a : evopayalena B 5.193 abhipraya codd. (ahipraya H) 5.196 avatara codd. (avatara A) 5.197 svabhavatvat a : svabha / vatvat B

4 tayor codd. (ayor C)

5.201 sampratyaya codd. (sapratyaya B) 5.204 maitrimayo codd. (maitrimeyo C, om. F) karunako E : om. F

5.199 bodhyangesu a : bodhyaga B karanatvat a

4 4

karanatvaka B

karuniko ABDH : karuniko C :

5.205 dhiman codd. (dhlmam A)

290

:

5.206maitrimayo codd.

Insignificant Variants

Text

(maitrimeyo C) 4 maitramABy : mamtram CD : metramE 5.208evameyah codd. (evamayah AF) 5.209 eka co^. (afika F)

4

5.210sa

'gradhir codd. (gra 'dhir B)

5.211

5.210-211 mahakarunikah codd. (maharunikah A; corr. A2)

codd. (so H)

suvisuddha codd. (savis~uddha B)

5.212 bhagavan iti ABy : bhagavann iti DE :

bhagavant iti C 5.213mahati ABEy : mahati C : mahanti D 5.215 mahopayo codd. (mahopaya F)

5.216 mahopayah codd. (mahapayah C)

codd. (mahopanditah B)

5.216-217 mahapanditah

5.217 aSayanusayajnataya a : aSaya anusayajnataya B 5.220 rddher codd. (rrddhe F)

5.218 vego codd. (veso H) 5.219 akhyo a : akho B

5.223 dhatu codd.

5.222 sagaramatih ABE : sagaramati CD : sara*mati H : om. F (patu F)

5.225 pratyeka codd. (pratyaka C, om. B

5.226 maharddhikah (3

:

maharddhiko A : om. B 5.228 smanasthana codd. (sthanasthana D) 5.229 parakramo codd. (parakrama F) 5.231 sambhetta a : sambheta B 4 ghanahcocW. (ghanah CDF) 5.233 sthama codd. (sthana H, stana B) 5.234 samyag a : myag B 5.236 sarva codd. (sarvae) 5.238 nibidatvan codd. (nibidatvat B) 5.239 soka ABCE : soga CF : sogata H

5.241 bhagavan codd. (bhagavat B)

nayarudho codd. (nayorudho H)

5.243 karotiti codd. (karobhiti A)

5.246

5.247 mahavidyaS

5.247vidya^a : vi*dy*o^ B

codd. (vidya^ A) 4 dharanyo corr. : dharanyahco^. (dharinyah J) 5.250 samSodhanat codd. (saSodhanat B) 5.252 tasyottamah ABH : tasyottama F5 ABJy : nayorudha De 4 anabhinayor H)

5.253 nayarudha

ganjah codd. (gajah H) 5.256 nirvanayor codd. (nirvanayor

5.257 adhigaman(-mon F) ABy : adhigama 6

5.25 8 anuttaryena

4

mahattvam corr. :

codd. (anurttaryena F) 5.259 samgrhlta B(3 : samgrahlta A mahatvam codd. (maphatvam B, mahatva F) bhagavantah codd. (bhagantah A)

4

5.260 buddha codd. (buddho y)

tesam ABJHE : taisam CDF

4 yanam codd.

(yana J) 5.261 tatsahitanam(na H) ABye : tatsattanam J : tatsatanam D yanam AB : mahayana (3

5.262 alambana codd. (alambata F)

4

4

maha-

5.263 samudagama

codd. (samuptagama C) 5.264 mahayanam codd. (mayanam B) 5.26 5 trividham codd. (trividha J)

5.266 paridipana codd. (yaridlpana B)

5.267 taya codd. (tayo F)

4

laukika codd. (lokika CF) 5.269 tad yatha codd. (tasratha A) 5.270 padair codd. (padai

291

Text

Insignificant Variants

CD) 4 grahanaka codd. (gahana B) 4 vakyair codd. (vavyair B) 4

viSesana codd.

(viSesena E) 5.271 jnana codd. (jana H) 5.272 srimatam codd. (Srisematam C) 5.275 punar codd. (puna* C)

4

agatya codd. (ogabhya A, agasya B)

ranjitani A(3 (varnepa- H, caraopa C) : varnoparinjitani B

5.275-276 varnopa-

5.279 bhagavato codd.

(bhagato A, bhagavrto J) 5.280 cudamanim codd. (cuSamani BH) codd. (avartayisyeti F) (evanamsamsa A)

5.285 mantraksara a : mantraksara B

4

5.282 avartayisyati evanu^amsa codd.

5.286 anyams codd. (anyaS B)

4

prapsyati codd. (prapryati J)

5.289 samudanlya codd. (samadanlya A, -niya H)

4

abhisam codd. (adhisam E)

5.291 dharmadesako codd. (dharmam deSako DH)

4

dasadik AyC ' da^adig e :

da^adika B

5.293 krtanam codd. (krtana D)

5.295 trini AB : trini 3 4 (na^vata B)

5.294 jnanasattva codd. (jnasattva C)

sattvavajryadlnam a : sattvavadlna B

5.298 caturdaSa codd. (caturndar^a H)

292

SasVata codd.

5.299 amnayah a : amnaya B

5.300 mantrarthavalokinyam(-kinya F) B(3 : mantravalokinyam A codd. (adhikara H)

4

5.301 adhikarah

Adhikdra 6

Text

Adhikara 6. Text Collated to Establish the Stemma Codicum *

lokottaraS 1 ca skandhadisv 2 anatmadyadhimoksapravrttas3 tattvadhigamapravibhavitaS4 ca // ata evaha madhyamakavatare5 candrakirtih // deyapratigrahakadatrSunyam 6 lokottara 7 paramiteti 8 danam 9 nirucyate 10 / tatra 11 yat 12 sarigajatam 13 tal 14 laukiki 15 paramiteti nirdistam 16 //

5

srutacintabhavanamayajnanodayakramena17 sarvakarajnatadhigamaprakarsaparyantam 18 etiti 19 vigrhya20 kvipi21 sarvapaharilope22 'nityam23

* The text collated for the purpose of establishing the stemma codicum covers 3 folios in MS A (33r6-36r6), commencing 2!/2 folios after the opening of the chapter (30v6). Codd. a (B A F H E C D J K G); B, a (A F H E C D J K G); A, (3 (F H E C D J K G); Y (F H), 5 (E C D J K G); e (E C K G), C (D J). 1 2 3 4 5 6 7 8 9 10 1] 12 13 14 15 16 17

18 19 20 21 22 23

lokottaraS codd. (lokottaro F) skandhadisv ABH : skandhadisu CDEGFK.T anatmady AH : an*o*tmady B : natmady CDEFGKJ 4 adhimoksapravrttas AB : avimoksapra\Ttta CDEFGKJ : avimoksapravrtya H tattvadhigama AJ : tattvadhigama B : tattvadhigamana CDEFGIIK 4 pravibhavitas"B : pravrtta^ celt. madhyamaka ABE : madhyamaka CCDFHJ : madhyakama CK : madhekama G deya codd. (daya F) 4 pratigrahaka AEFH : pratisrahaka B : pravigrahaka CJD : pravigahaka KG 4 datrsunyarn AB : da< >^unyam F : danaSunyam H : danasu CJ : daSuna Cc : danesu DE : data^u KG lokottara codd. : lokottaro H paramiteti ABFH : danaparamiteti EJDCCms : daramiteti CGK danam codd. (daram H) nirucyate AFHECDJKG : niru*dy*ate B tatra codd. (tatra KG) yat cow/. : ca A : ya B : yaj C^EFGHKJ sarigajatam conj. : sang*I*jatam B : jatasamgam ACDEFGFIKJ talB : saA : sa CDEFGHKJ laukiki JCDEH : laukiki*r* A : laikiki B : lokiki F : lakiki GK nirdistam HFEJ : distamAB : nidistam D : nidirstam C : nidirstha KG bhavanamaya codd. (vabhanamaya KG) 4 jnanodaya codd. (jMnadaya F) 4 kramena ABHFJE : kramena jnanodaya CGK : jnanodayakramena D jnatadhigama A : jnatadhimatparam B : jnatadhigamanat CDEFGHKJ 4 paryantam AF : paryantamm H : yaryam B : prayantam CDGKJ : prayatnam E etiti ABFH : atlti CDEGKJ vigrhyaABH : vigrhyam CDGKJ : vigraham E : vigrh*y*amF kvipi BH2 : ksipi A : kipi CDFGKJH : kimapi E sarvapaharilope AB : sarva(-rvam D)paharalapa CEGKJD : sarvapaharalopa FH 'nityam corr. : anityam ACDEFGHKJ : 'nityadB

293

Adhikara 6

Text

agamaSasanam 1 ity atuki2 /

tatpuruse3 krti bahulam ity aluki4 karma-

vibhakteh5 krte6 / paramir iti 7 bhavati tadbhavah8 paramita / danasya 10

paramita9 danaparamita 10 / evam Siladisv api yojyam // tata£ ca pravistabhumikanam 11 bodhisattvanam trikotipariSuddhilaksanam yad danam tad danaparamita / yat tu 12 prthagjananam 13 danam tad 14 danam eva na danaparamita 15 trikotipariSuddhajnanavirahat 16 / tatas ca danadinam parami tanam 17 prajnaparamitaiva18 caksur19 veditavya // ata evaha madhya-

15

makavatare20 candrakirtih / ekena21 pumsaksimata22 yathestam des~am23 samasto24 'ndhaganah25 sukhena26 /

Codd. O (B A F H E C D J K G); B, a (A F H E C D J K G); A, (3 (F H E C D J K G); y (F H), 5 (E C D J K G); e (E C K G), C (D J). 1 2 3 4 5 6 7 8 9 1° 1^ 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26

agamasasanam conj. : agamanuSasanam A : agamasasenam B : argamas asanan CDEGHKJ : argamasasana< > F ity atuki H : ity a*tu*ki B : ity atukim CDEGKJ : ity antaki A : < ~3 > F tatpuruse AJFH AA[\7I.3.14] : tatpurupe D : tanpuruse B : tadabhirupe E : ta*bh*purupe CGK aluki codd. (alaki GK) vibhakteh AB : vibhakta CDEFGFIJK krteAB : krta CDEFGHJK paramir iti BFH : param iti A : paramitar iti CDEGJK : paramiteti Hc tadbhavah AB : tadbhava CDEFGJKH2 : mahava H paramita AB : om. CDEFGHJK danaparamita codd. (om. B) pravista codd. (pravistah E) yat tu conj. : yastuA : yatuB : yatra CDEFGHJK prthagjananam ACDEJHKG : prthagjanani F : prthakunanam B danam tad AB : tu CDEFGHJK na danaparamita B : paramita ACDEFGHJK pari&iddhaAB : pariSuddhi HFDJECGK paramitanam codd. (paramitanam G) paramitaiva codd. (paramiteva B) caksur ABJ : caksura CDEFGHK madhyama codd. (madhya A; madhyama ACm«) ekena AB : etena CDEFGHJK pumsaksimata A : pumsa CDEGHJK : pumso F : ya*n*saksimata B desamA : desa BCDEHJKG : de*S*am F samasto codd. (samaste E; samasta J) 'ndhaganah AB : tvagana CDEFGJK : tvaguna H sukhena codd. (sukhenah A)

294

Adhikara 6

Text

akrsyate tadvad 1 ihaksihman2 adaya3 dhir yati4 gunafi5 jinatvam // iti // tad evambhuta 6 daSaparamitah prapto 'dhigato 7 yo8 bhagavan sa

20

tathoktah // tatraSraya9 adharah / tasam eva daSanam paramitanam adhararupatvat 10 tadadheyatvac u ca / tad12 etad uktam bhavati / anyonyadharadheyasambandhat 13 pratltyasamutpadarthah 14 sucito bhavati / iti 15 // daSaparamitaSuddhir iti trikotipariSuddhitvad 16 deyadayakapratigrahaka-

25

nupalabdhir17 asam 18 paramitanam19 iti // dasaparamitanaya iti da£aiva20 paramita nayo marga upayo yasyasau21 sa22 tathoktah23 //

Codd. n(BAFHECDJKG);B,a(AFHECDJKG);A, (3(FHECDJKG);v (F H), 6 (E C D J K G); £ (E C K G), C (D J).

1 2 3 4 5 6 7 8 9 10 1] 12 13 14 15 16 17 18 19 20 21 22 23

tadvad ABE : tadva*dd* CDHJ : tadva*dd* F : tadvah GK ihaksihinan ABHECGK : ihaksihin DJ : ahaksihinan F adaya AB : adaya CDEFGHK : adayah J dhir yati AFHCDEJKG : viryadiB gunafi conj. : gunamcodd. evambhuta ABJ : evambhutah CDEH : edhabhutoh F : evambhutamh GK prapto 'dhigato AB : praptadhigamo EJ : praptadhigato CDGFHK yo ABU : om. CDEFGJK tatra^raya ABJ : tatraSaya CDEFGHK adhararupatvat ABJ : ararupatvat CDEFGHK tadadheyatvac ACDEFHGKJ : tadadhipatvac B tad codd. (tata B) anyonyadharadheya AB : anyonyadharaya CDJKHG : nyodharaya E : anyanyadhareya F 4 sambandhat em. : sandhat A : sabandhat ACm^ : sambadhat B : sambandha CDJEFH : sambadha GK pratltya codd. (pratyamtya B) 4 arthah codd. (arthara F) bhavati / iti corr. : bhavati Iti A : bhavati CDEFGHJK : taratlti B pariSuddhitvad conj. : pariSuddhatvad AB : pariSuddhatva CDEGKFH : pari^uddhitva JH2 upalabdhir AB : upadeyadayaka codd. (dayadayeka F; diyadayaka H) 4 labdhit DJHK : upala*bdhit* F : upalamdhit E : upala*mdh*it C : upala*bdh*it G asam codd. (asa DHJ) paramitanam codd. (paramitam DHJ) da&ivaAB : da^eva CDEGJK : om. H 4 dasaiva — sa tathoktah om. H (***paramita nayo maga upaya yasya sa tathokta*m*h suppl.H 2mg) yasyasau codd. (yasya J; yasya H2) sa codd. (om. A) tathoktah AB : tathokta*m*h H2 : tathoktam iti CDEFGJK : om. H

295

Text

Adhikara 6

tah 1 punar bhumisu yojyah2 / tad yatha pramuditayam bhuvi danapara30

mita / atiriktataro3 bhavatiti4 / vimalayam 5 Silaparamita / prabhakaryam6 ksantiparamita / arcismatyam viryaparamita / sudurjayayam7 dhyanaparamita / abhimukhyam prajnaparamita / duramgamayam upayaparamita / acalayam pranidhiparamita8 / sadhumatyam balaparamita9 / dharmameghayarnjnanaparamiteti 10 //

35

tatraita 11 bhumayo dharmanairatmyamaitryadhigamaviSese 12 'pi danaparamitadivisuddhyadhisthanato 13 vi^esakhyam 14 pratilabhante // audaryam tu pranidhanabalad asam anantam 15 asadharanam ca sarva^ravakadibhir 16 iti // tatra bhumir iti jnanasyavastha 17 bhumir18 gocara iti yavat 19 / tatra bodhyadhigamaviSesad20 uttarottarajnanapratisthayogena21

Codd. Q (B A F H E C D J K G); B, a (A F H E C D J K G); A, (3 (F H E C D J K G); y (F H), 5 (E C D J K G); £ (E C K G), C (D J). 1 2 3 4 ^ 6 7 8 9 ^ 11 12 13 14 15 16 17 18 19 20 21

tahABH2 : ta CDEFGHJK + Before ta CDEGJK marks '2'. J inserts NS44. yojyah AB : yojya CDEFGHJK atiriktataro A : atiriktatara CDEFGHJK : atiriktatama B bhavatiti corr. : bhavati iti CDEFGHJK : bhavati AB vimalayam codd. (vimalam F) karyam codd. (*d*ayam C; kayam GK) sudurjayayam codd. (sudurjayam B) pranidhi codd. (prani FH; pranidhi H2) bala codd. (bara GK) paramiteti B : paramita iti cett. tatraita AB : tatreta CDEGHK : tatrata F : tatreta J dharma codd. (verma A) 4 maitryadhigama A : maitra< >gama B : maitradhigama CDEFGHJK viSuddhya AB : viSuddhya CDGHJKEF 4 sthanato codd. (sthatato F; sthanatatoE) visesakhyam AB : vitesakhyam CDEFGHJK anantam codd. (anantyam B; antam E) ^ravakadibhir ABFH : sravakabhir CDEGJK jnanasyavastha AB : jnanasya avistha H2 : jnanasyavistha CDEGK : jnanabhavio FH : jnanasyanistha J bhumir AB : bhumi CDEFGHKJ yavat ABFH : y a van CDEGJK bodhyadhigama A : bodhigama B : vakadhigamya CDEG.IK • < >dhigama F : vagadhigam*y*a H 4 visesadAB : visesancetf. uttarottara ABFHE : uttarotta CDGJK

296

Adhikara 6

Text

40

tattajjnanagunadharayogena 1 ca bhumaya ity ucyante2 // sarvatra sattvadhatoh samanyalaksanatvad3 yato dharmadhatuvinirmukto na kas cid dharmo4 'stiti tata5 atmasamatam6 pratilabhante7 nairatmyavis"esat8 / ekam9 api 10 taj n jfianam 12 da$ety ucyate / avasthavisesat // ataaha

da§abhumi§varo natho 13 daSabhumipratisthita 14 iti / (44ab) daSanam pramuditadmam 15 bhuminam is"varo 16 'dhipatir17 ity arthah / sa

45

yasyam 18 yasyam 19 bhumau20 danadiparamitanam21 vasitvalabhat22

Codd. O (B A F H E C D J K G); B, a (A F H E C D J K G); A, (3 (F H E C D J K G); y (F H), 6 (E C D J K G); £ (E C K G), C (D J). Lacking F (om.) H (lac. ), lines 43^48, [nairajtmyavisesat —> pratisthita iti // [dasa jiianani]. {For MS F naird concludes fol. 40v; dasajnanani starts fol. 41r}. 1 2 ^ 4 5 6 7 8 9 10 11 ^2 13 14 15 16 17 18 19 20 21 22

tattaj conj. : tatat A : tanta B : tatra DEFGH.TK : *tra C 4 gunadhara A : gunadhara B : samadhana CDEGHJK : illeg. F 4 yogena codd. (jogena A; illeg. F) ucyante AB : ucyate CDEFHJ : ucyamte KG samanyalaksanatvad ABDC : samanyatvad CDEGJK : samanyatvad FH dharmo codd. (*dh*armo CE) tataAB : atra CDEFGHJK atmasamatam AB : atmasamata CDHJKGF : atmasata E pratilabhante codd. (pratilabhate B; pratilabha*e FH) nairatmya AB : nairatmya CDGEK : naigamya JH2 : naira F : nai *< > H 4 visesat ABEKH : ni^esat CDGJ : om. F 4 [naira]tmyavisesat —» pratisthita iti // [dasajnanani] om. F; lac. H ekam AB : evam cett. (om. F; lac. H; suppl. H2) apiAB : avi Dc : a CDEGH2JK : om. F : lac. H 4 evam a< ~ 10 >t //ata aha CDEGH2JK (vi ta jfiana das"ety ucyate / avasthavisesat suppl. Dc) taj B : ta Dc : om. cett. jnanam AB : jnana Dc : om. cett. das'abhumls'varo natho AD Cw^ : das'abhumis'varau natho B : da^abhumipada CDEGH2JK : om. F : lac. H da&bhumipratisthita ABDC : pratisthita CDJEGH2K : om. F : lac. H adlnarn codd. (lac. H; suppl H2 ; om. F) bhuminam Isvaro A2 : bhumls'varo ABDC^ : bhuminamana^vara CDEGH2JK : om. F : lac. H 'dhipatir BDCm« : dhipatir A : jivapattir CDEH2J : jivapatirGK : om. F : lac. H sa yasyam A : sarvasya CDEGH2K : sarvasyam J : tasmat B : om. F : lac. H yasyam A : tasa B : om. CDEGH2JKF : lac. H bhumau codd. (om. BF; lac. H; suppl. H2) danadiA : danadiB : danadinam CGKEJ : danadinaniH2 : danadina D : om. F : lac. H 4 paramitanam CGKDEH2! : paramitadinam AB : om. F : lac. H vasitvalabhat A : bhumilabhat B : vaSatvam labhat CDEGJKH2 : om. F : lac. H

297

Adhikara 6

Text

natha 1 ity ucyate2 / tasv3 eva daSabhumisu pratisthanat4 pratisthita iti // daSa jnanani dharmajnanam anvayajfianarn 5 paracittajnanam6 samvrtijnanam duhkhajnanam7 samudayajnanam nirodhajnanam margajnanam8 50

ksayajnanam anutpadajnanam ceti // dharmajnanam katamat

/

kamapratisamyuktesu samskaresu9 yad

anasravam 10 * jnanam / kamapratisamyuktanam samskaranam hetau 11 yad anasravam jnanam / kamapratisamyuktanam samskaranam nirodhe yad anasravam 12 jnanam 13 / kamapratisamyuktanam 14* samskaranam praha55

naya 15 marge yad anasravam 16 jnanam // api ca dharmaj nane 17 dharma-

Codd. n (B A F H E C D J K G); B, cc (A F H E C D J K G); A, (3 (F H E C D J K G); Y (F H), 5 (E C D J K G); e (E C K G), C (D J). Lacking F (om.) H (lac. ), lines 43-48, [nairdjtmyavisesat -> pratisthita iti // [dasa jnanani]. {For MS F naird concludes fol. 40v; dasa jnanani starts fol. 41 r). 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17

natha AB : na CDEGH2JK : napa Dc : om. F : lac. H + na< ~7 >dasa bhumisu CDEGH2JK (pa ity ucyacyate tany eva suppl. Dc) ucyate B : ucyacyate D0 : ukte A : lac. cett. (om. F) tasv AB : tany Dc : lac. cett. (om. F) pratisthanat A : prasthanat B : pratisthitat DEH2J : pratitat CGK : om. F : lac. H anvaya ABJ : advaya cett. 4 jnanam codd. (jnana FH) paracitta codd. (parivartta F; p*** H) duhkhajnanam ABCDJEFKH2 : om. GH samudayajnanam nirodhajnanam margajnanam om. A (suppl A2ms) samskaresu ABDECJH2 : < 3 >su FH : svaskaresu GK anasravam ABHF : anasrava CDEJ : anaScava KG hetau AB : heto CDEGJK : hetoh H2 : < > HF yad anasravam codd. (yanasravam CGK) jnanam (jnana F) DH place kamapratisamyuktanam samskaranam prahanaya marge yad anasravam jnanam after kamapratisamyuktesu samskaresu yad anasravam jnanam (line51-2). prahanaya codd. (om. A) anasravam codd. (anarthavam G) dharmajnane codd. (dharmajnanai B)

298

Adhikdra 6

Text

jnanabhumau ca yad anasravam 1 jnanam 2 idam ucyate 3 dharmajnanam // anvayajnanam katamat4 / ruparupyapratisarnyuktesu samskaresu yad anasravam jnanam / ruparupyapratisamyuktanam5 samskaranam6 hetau7 60

yad anasravam jnanam / ruparupyapratisamyuktanam 8 samskaranam9 nirodhe yad anasravam jnanam / ruparupyapratisamyuktanam 10 samskara­ nam prahanaya marge yad anasravam 11 jnanam / api 12 ca 13 khalv 14 anvayajnane 15 anvayajnanabhumau 16 ca17 yad18 anasravam 19 jnanam idam ucyate 'nvayajnanam20 //

Codd. n(BAFHECDJKG);B,cc(AFHECDJKG);A, (3(FHECDJK G); y (F H), 5 (E C D J K G); £ (E C K G), C (D J). 1 2 3 4 5 6 7 8 9 10 11 12

13 14 15 16 17 18 19 20

anasravam ABFH : anasrava CDEGJK jnanam codd. (jnanas KG) ucyate < ~15 > pratisamyuktesu FH (dharmajnanam /ruparupya suppl H 2) > anvayajnanam katamat ABDC : advayajnanam katamat E : < - - -> JDa : < HF : < - - - > CGK : om. H2 pratisam codd. (pratisam GK) f ruparupyapratisamyuktanam samskaranam hetau yad anasravam jnanam om. B 4 ruparupyapratisamyuktanam samskaranam hetau -* nirodhe yad anasravam om. FH samskaranam codd. (skarana KG) hetau AFH : heto CDEGJK : om. B ruparupyapratisamyuktanam codd. (ruparupratisamyuktana GK) samskaranam AFHCDEJ : samskaradinam B : saskaranam GK samyuktanam codd. (samyuktanam om. B) anasravam codd. (anasrayam F) api AB : atha CDEFH2J : ya atha GK : om. H * api ca -»• ucyate 'nvayajnanam om. H (atha khalu dvayajnanesu anvayajnanesu bhumau yad asravam jnanam idam ucyate / anvayajnanam / suppl. H 2mg) caAB : om. CDEGHJKH2 khalv AB : khalu CEFGH2JKD : om. H anvayajnane A : dvayajnanesu CDEFGH2JK : om. BH anvayajnana AB : anvayajnanesu CDFH 2J : advayajnanesu E : anvryajnanisu GK : om. H ca ABF : om. CDEGHJKH2 yad codd. (om. F) anasravam ABE : asravam CDGJK : nasravamF : asravam H2 . om. H 'nvayajnanam coir. : anvayajnanam ACDFGH2JK : advayajnanam E : atvaya-

jnanah B : om. H

299

Text

65

Adhikara 6

paracittajnanam katamat / yaj 1 jfianam bhavanamayam2 bhavana phalam / bhavanamargasya3 pratilabdham4 avihmam 5 / yena6 jnanena kamavacaran7 rupavacaran8 samavahitan9 sammukhibhutan 10 paresam 11 cittacaitasikan12 dharman 13 janati 14* / ekatyamS 15 canasravan 16 dharman 17 idam ucyate paracittajnanam //

70

samvrtijnanam katamat / sasrava 18 prajna 19 * // duhkhajnanam katamat / pancopadanaskandhan20 anityato duhkhatah21

Codd. fl (B A F H E C D J K G); B, a (A F H E C D J K G); A, (3 (F H E C D J K G); y (F H), 5 (E C D J K G); £ (E C K G), C (D J). 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20

yaj AHFDEJCGK : sa B mayam codd. (maya GK) margasya AB : magamya CDEGHJK : < >gamye F pratilabdham codd. (pratila*py*am B) avihinam ABHF : avihina CDEGJKII2 yena AB : pana CDE : para JH2 : yana GK : yad F : yada H kamavacaran AB : kamavacarat CDEGJK : kamavacara F : kamavacara* H rupavacaran AB : rupavacarat CDEFGHJK samavahitan ABDC : sasatvahitan E : satvahitan CGJKH2 : satvahitan D : < >hitaya HF sam(sa-A)mukhibhutan ABDC : sanat E : sammukhK > HF : sammukhi< >nat H2 : samnat DJ : sam!nat C : sam // 1 // nat KG paresam A : parasam B : pare bhavanti paresam CDEGH 2JK : < >bhavanti paresam FH caitasikan codd. (-kana FGK) dharman codd. (dhamam KG) janati codd. (jinati GK) ekatyam^ codd. (ekatyam F) canasravan A : caSravanB : canyamscavan CDEHJ : canyam^cavan GK : nyam^cavam F dharman AB : dhavan CDEHJGK : dhaman F sasrava AB : sasra*dhv*a cett. prajna AB : prajna pari(pari- GK)jfieyam iti janati / samudaya prahino na punah prahatavya(-tavyah FH) nirodhasaksat(-saksa GK) tvata (*tr*to F vatta H) na punah saksatkartavyah (saksat om. HF; saksat H2) / samvrtijnanam ceti CDEFGHJK pancopadana ABDHJ : pancaupadan ECFGK 4 skandhan ABE : skandhat CFDHJGK

21 duhkhatah AB : duhkha cett.

300

Adhikara 6

Text

Sunyato 'natmatas 1 ca2 manasikurvato3 yad anasravam4 jnanam idam ucyate duhkhajnanam // samudayajnanam5 katamat / sasravam6 hetum7 hetutah8 samudayatah 75

prabhavatah9 pratyayata^ ca10 manasikurvato yad anasravam jnanam / tat 11 samudayajnanam 12 // nirodhajnanam katamat 13 / nirodham 14 nirodhatah Santatah pramtato 15 nihsaranatas 16 ca 17 manasikurvato yad anasravam jnanam idam ucyate nirodhajnanam // margajnanam katamat / margam 18 margato 19 nyayatah20 pratipattito21

80

nairyanikatas ca manasikurvato22 yad anasravam jnanam idam ucyate margajnanam //

Codd. Q. (B A F H E C D J K G); B, a (A F H E C D J K G); A, (3 (F H E C D J K G); y (F H), 5 (E C D J K G); e (E C K G), C (D J). 1 2 3 4 5 6 7 8 9 10 1] 12 13 14 15 16 17 18 19 20 21 22

'natmataS AB : natmana CDEJ : natmata GFK : natmatam II ca AB : om. cett. manasi codd. (manasa F) anasravam codd. (amasravarn B; anasrava J) samudayajnanam < ~48 > nirodhajnanam katamat HF (ie. katamat -»tat samud­ ayajnanam // om. FH) sasravam ABDC : CDEH2J : ni GK : om. FH hetum -» prabhavatah om. B hetutah codd. (heruta G) prabhavatah AJ : prabhavata CDEGK : prabhavata H2 : om. BFH pratyayataS ca ABDC : CDEGH2JK : om. FH tat codd. (tata B) jnanam / tat samudayajnanam AB : om. CDEGHJKFH2 katamat codd. (krtamat B) nirodham codd. (nirodha B) pramtato codd. (prasltato B; pranlmato D) niMaranata^ ABHF : nisaranata CDEGJK ca ABHF : om. CDEGJK margam A : marga CDEFGHJK : om. B margato ABHF : marga CDEJ : maga GK nyayatah CDEGHJK : n*m*ayatah A : nyacyatah B : nyayata F pratipattito ACDEFGH2JK : prativattito H : pratipattibhe B nairyanikata^ ca manasi ABDC : nai< ~8 > FH : nairyatimanasi CGK : nairyatamanasi E : naiyatitamanasi D : naiyatima nasi J : naiyatimanasi H2

301

Adhikdra 6

Text

ksayajnanam katamat / duhkham 1 me2 parijnatam 3 iti janati samudayo4 me 5 prahmah / nirodhah saksatkrto6 margo7 me 8 bhavita9 iti 10* janati tad 85

upadaya 11 jnanam 12 darsanam vidya 13 bodhih prajnaloko 14 * 'bhisamayah 15 / idam 1 6 ucyate ksayajnanam 17 // anutpadajnanam 18 katamat / duhkham me parijnatam 19 na punah parijneyam20 iti janati samudayah prahano21 na punah22 prahatavyah / niro­ dhah saksatkrto na punah saksatkartavyah / margo me bhavito23 na punar

90

bhavayitavya iti janati // tad upadaya24 jnanam25 dars"anam vidya bodhih prajnaloko 'bhisamaya26 idam27 ucyate 'nutpadajnanam ity ata aha //

das'ajnanavis'uddhatma da&ajnanaviguddhadhrg iti // [44cd]

Codd. £) (B A F H E C D J K G); B, a (A F H E C D J K G); A, 0 (F H E C D J K G); y (F H), 5 (E C D J K G); e (E C K G), C (D J). Lacking F, lines 85-92. Folio 42r ends ...bodhih pra< >. Folio 42v is blank; folio 43, missing. 1 2 4 5 6 7 8 9 10 1* 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 21

duhkham AB : duhkha cett. me AB : ma cett. parijnatam AB : parijnanam CDEGJK : patijnanam FH samudayo codd. (samudadayo A; samudaya F) meAB : ye CDEGHJK : ya F saksatkrto AB : saksattau CDEFGJK : saksattau H margo codd. (margau EH) me AB : ma CDFHJ : mama GK : om. E bhavita codd. (abhavita E; bhavitadu KG) iti codd. (iti na punar bhavayitavya iti A) upadaya ABFH : upayasa CDEGJK jnanam AB : jnana FH : tvana CDEGJK vidya ACDEGJK : vi*ghn*a B : < > FH : vidha H2 prajnaloko codd. (prajna'loko A; pra< > F) 'bhisamayah con. : abhisamayah AB : bhisamaya CEGJK : 'bhisamaya D : bhi samaye H : om. F idam codd. (dudam GK) ksayajnanam codd. (-jnana H) anutpadajnanam codd. (-jnana H) parijnatam AB : parijnanam CDEGJK : parijnana H : om. F parijneyam -* saksatkartavyah om. CDEGHJKF prahano A : praha* B : om. cett. punah A : om. cett. bhavito codd. (bhavita iti A; dhasito H) upadaya AB : upaya cett. jnanam AB : jnana cett. prajnaloko 'bhisamaya A : prajna'loko 'bhisamaya B : prajnalokabhisamaya CDEG­ HJK : om.F idam codd. (dam GK)

302

TEXTUAL NO TES

Textual Notes. Chapter 1

TEXTUAL NOTES Chapter 1 1.0

om namo manjughosasvamine 5

:

om namas te varadavajraya A

:

< -12 > Y : rje btsun sgrol ma la phyag 'tshal lo {namas bhattarikayai tarayai} Tib. [184.5.1] The variation between these opening salutations suggests that they are scribal additions rather than part of Vilasavajra's text. It is hard to view them as derived from an original that subsequently became corrupted. For example, A's reading cannot be seen as a corruption of S's or vice versa. A's om namas te varadavajraya probably alludes to NS 158a (namas te varadavajrdgjya), while S's om namo manjughosasvamine is a common salutation to ManjuSri (though hardly necessary here since a salutation to ManjuSri immediately follows) and Tib. contains a translator's homage to Tara. These salutations resemble the prasasti verses that precede the opening of some Buddhist sutras. 1.1

jfianendum H*2* Tib.([184.5.2] ye shes zla): jnanendu A|3 (jnane*du* C) aryamanjus'riyarn natva jnanendum tryadhvatayinam : dus gsum skyob pa'i ye shes zla / / 'phags pa 'jam dpal phyag 'tshal te / / Tib. Tib.'s zla confirms the reading of indu, weighing against an argument for emending jnanendu to a more familiar word such as jnanendra. Tib. also supports the uncompounded form jnanendum insofar as the genitive skyob pa'i indicates that it takes tryadhvatayinam to be in apposition to jnanendum. (Tib. could also be retranslated as tryadhvatayijnanendum, though not asjhanendutryadhvatayinam.) I take H's reading to be an result of contamination from outside a.

1.1

tryadhvatayinam em. (Tib.) : tryadhvatayinam A : pravatayimm y (tryadhvatayinim H2) : pradhvatayinam 6 (*dhvatayinam E, tpradhvatayinam C)

304

Textual Notes. Chapter 1

aryamanjusriyam natva jnanendum tryadhvatayinam : dus gsum skyob pa'i ye shes zla / / 'phags pa 'jam dpal phyag 'tshal te / / Tib.[184.5.2] I emend to the masculine tryadhvatayinam on the grounds that the word makes better sense as an epithet of ManjuSri rather than as an attribute of the Namasamgiti. Its position in the verse, before the tatha at the opening of the third pdda, which introduces the Namasamgiti as the second object of salutation, also suggests that the emendation is to be preferred. Tib. supports this in that it takes dus gsum skyob pa (tryadhvatayinam} to be in apposition to ye shes zla (jnanendum) and thus related to Manjusil. 1.4

anyanibandhanam A5H2 : anyatibandhanam F : anyapibandhanam H : rnam 'grel gzhan Tib.[184.53] sutrabhidharmapitakam vilokyanyanibandhanam A5H2 : mdo sde chos mngon sde snod dang / / rnam 'grel gzhan yang bltas pa dang / / Tib.[184.5.3] The reading -anyanibandhanam is problematic. It is witnessed by A6H2 and is thus almost certainly the reading of a. (F and H's /;' and/?/ are explicable as corruptions of ni, and MS B is unavailable here.) However, if nibandhanam means 'treatise' one would expect a plural termination, as well as a conjunction (api, ca or tatha) as found in the preceding and subsequent parts of this opening list of acknowledgements. In terms of sense -anyanibandhanam is also suspect. Glossing it as a karmadharaya ("other treatise[s]") gives an inappropriate meaning since no sdstras have been mentioned; only tantras and sutras. The alternative, taking it as a tatpurusa ("treatise[s] of others"), is also unsatisfactory if taken as implying that works by named authors have already been referred to. Sense could be made here if the 'others' are taken to refer to anyone apart from the writer (ie. Vilasavajra). Tib. has rnam 'grel for nibandhanam. In the MVy there is no entry for nibandhanam but rnam 'grel is given for vdrttikam ('critical gloss',

305

Textual Notes. Chapter I

'annotation') (MVy 1454). The term vdrttikam, however, is usually reserved for commentary on philosophical or grammatical works. Perhaps rnam 'grel was the best Tib. could do for nibandhanam. MVy 1450 gives rnam par 'grel ba for vivaranam ('explanation', 'exposition', 'interpretation', 'gloss', 'com­ ment'). Could nibandhanam be a corruption ofvivaranaml The sense may be better - "the explanation of others" - and there would no longer be the problem of number. Tib. takes the compound as a karmadharaya (rnam 'grel gzhan) "other explanation" - which seems unlikely in this context. Also a karma­ dharaya is, strictly speaking, ungrammatical and though pronouns like anya did come to be compounded in karmadharayas such grammatical laxity is untypical of Vilasavajra. Tib.'s yang {api} also deals with the problem of the lack of a conjunction. A further possibility is that both the Skt. and Tib. are corrupt, the latter inheriting the corruption from its Skt. exemplar. (For discussion of Tib. inherit­ ing error from its Skt. exemplar(s) see Introduction to the Text, section 4.4.) Reference to sastra at this point appears out of place. The texts previously listed, and the two items immediately following (jatakam cetivrttam ca) are all considered Buddhavacana, and their order of enumeration - from Yoga Tantra down to itivrtta - can be taken as one of decreasing importance. However, the list contains no reference to the vinaya, which, as the third pitaka, commonly follows any mention of the sutra- and abhidharmapitaka. The extant reading could represent a corruption of a word fulfulling such a function. On the other hand, the listing ofjataka and itivrtta might be taken as reference to the vinaya since in some divisions of Buddhavacana these two are placed under the head­ ing of the vinayapitaka (cf. Lessing & Wayman, 1968, 57). If this were the case, this might after all be an appropriate place for reference to bauddham sastram, since \aukikam sastram is mentioned later. (Against this, the

306

Textual Notes. Chapter 1

enumeration of Vijfianavada and Madhyamaka - vijnanavadam akhilam tathd madhyamakam ca yat - might be taken as fulfilling any need for mention of bauddham sdstram.) Prior to consultation of the Tibetan translation Prof. A. Sanderson suggested that viJokyanyanibandhanam could be a corruption of vilokydiyanibandhanam. This emendation, giving "treatises of the Aryas" (or "interpretation of the Aryas"), would make sense as denoting works by important commentators who were regarded as Aryas. Prof. Sanderson also pointed out that if -nibandhanam (or an emended reading) were to qualify sutrdbhidharmapitakam the need for a conjunction would disappear. The issue of whether and to what extent vilokydnyanibandhanam is corrupt does not seem capable of ready solution. In lieu of any satisfactory alternative that can be supported I have kept it, though with no great confidence in its being correct. 1.13

bodhicittam A Tib.D.([27v6] byang chub kyi sems) : < ~7 >bodhicittah FC : < ~7 >bodhicittem H (adarsajnanam bodhicittam H2) :

adarsanajnanam

bodhicittah ECC : om. D : byang chub kyi sems rdo rje {bodhicittavajram} Tib.P.[184.5.6] 4 bodhicittam —>• mahavairocanasvabhavena om. D bodhicittam s"rivajrasattvadvarena A : byang chub kyi sems dpal rdo rje sems dpa'i sgo nas dang Tib.D. : byang chub kyi sems rdo rje dpal rdo rje sems dpa'i sgo nas dang Tib.P. The hyparchetype a is likely to contain bodhicittam: it is the reading of A and H 2 . The readings adarsajnanam and adarsanajnanam are scribal corrections filling marked lacunae. They are surely erroneous, and probably borrow from the chapter title adarsajnanam aksobhyadvarena, which closely follows (NMAA 1, 14-15). Tib.D. supports bodhicittam. Tib.P. supports bodhicittavajrah, though bodhicittavajrah srivajrasattvadvdrena does not give any sense if bodhicittavajrah is the name of a deity.

307

Textual Notes. Chapter I

Though bodhicittam srlvajrasattvadvarena gives sense, there are some problems. Vilasavajra divides the NMAA into chapters following his chapterdivisions of the NS. With the exception of the present case, the NMAA chapter names correspond with his titles for the NS divisions. The fifth chapter of the NMAA, which corresponds to the present NS division, is called "The Vajradhatu-Mahamandala of Bodhicittavajra" (bodhicittavajrasya vajradhdtumahdmandala- NMAA 5, 300-301). The Nepalese NS MSS that have the chapter-divisions as enumerated by Vilasavajra also describe the fifth chapter as comprising verses on the Vajradhatu-Mahamandala (vajradhatumahamandalagdthds caturdasa NS.Dav. 52). One might thus expect more of a link between the name Vilasavajra gives to the NS chapter and the corresponding NMAA chapter, which suggests that the reading bodhicittam srlvajrasattvadvarena may be a corruption of a longer original. That this may be the case is further suggested by the lacuna marked before bodhicittam in all the MSS except for A. An emendation that straightforwardly incorporates the title of the fifth chapter of the NMAA is bodhicittavajrasya vajradhdtumahdmandalam srl­ vajrasattvadvarena. This would mean that a (with Tib.D. following) at some point lost -vajrasya vajradhatumahdmandala- (with Tib.P somehow pre­ serving bodhicittavajrafsya]). Although this emendation produces, on corruption, a lacuna after bodhicitta rather than before, it has the merit of mirroring the form of the following five titles of the NS chapters, those on the five jnanas, where eachymzna is explained via a particular Tathagata (mahavairocanasvabhavena, aksobhyadv arena, amitdbhamukhena etc.: lines 14, 15 & 16). However, in these cases, the 'explaining via' is an account of the panca jndndni, seen as non-dual Awareness (characterised by the absence of any

308

Textual Notes. Chapter I

subject-object dichotomy), being understood (and objectified) in more dualistic terms as the panca tathdgatdh. While preserving the title of NMAA 5, emendation to bodhicittavajrasya vajradhdtumahamandalam srivajrasattvadvdrena, gives a meaning ("the Vajradhatu mandala ... [is explained] via Vajrasattva") that moves in the opposite direction: greater differentiation (duality) is worked out in terms of less differentiation (non-duality). This does not make good sense. The form needed is the reverse of this, ie. that Vajrasattva is explained as the Vajradhatumandala. (The Skt. cannot be read as saying that Vajrasattva is the nature of the Vajradhatu mandala - this would require a bahuvrihi: bodhicittavajrasya vajradhdtumahamandalam srivajrasattvasvabhdvam.) In the passage of the NMAA that follows this one, where the opening and closing words of each NS chapter are given, Vilasavajra refers to the verses of the fifth chapter as being "The elucidation of Bodhicittavajra through the Great Mandala of the Vajradhatu" (vajradhdtumahdmandaladvdrena bodhicittavajrasydbhidhdnam NMAA 1, 26-7). This makes it clear that Vilasavajra sees the chapter as being about Bodhicittavajra, the hypostatisation of bodhicitta. If codd.'s srlvajrasattvadv arena is accepted, Vajrasattva is to be identified with Bodhicittavajra. If this is the case, then Vajrasattva can be seen as the embodiment of bodhicitta and the reading of A and Tib.D makes sense. It also follows the form of explaining non-duality in terms of duality found in the other chapter titles. An emendment to bodhicittavajram srivajrasattvadv arena, following Tib.P., could make sense but only if bodhicittavajra is not seen as the name of a Buddha by Vilasavajra. Unfortunately, there is no discussion of bodhicittavajra in NMAA 5, while in NMAA 3 the word is given (ambiguously) as the name of one of the six Buddha-families. Lacking a clear alternative and despite the

309

Textual Notes. Chapter 1

mismatch between the title given to the NS chapter and the corresponding NMAA chapter, I have accepted the reading of A, which can be seen to give good sense. 1.45

idam A : om. 0 Tib. tad ekatra dvasastyuttaraSatagathabhir adhyardhas'atenanus'amsagranthena stutih samagra // idam ca bhagavatah sarvatathagatajnanakayasya manjuSrijnanasattvasya namasamgiteh pindarthah A : tad ekatra dvasastyuttaraSatagathabhir adhyardhaSatenanuSamsagranthena ca bhagavatah sarvatathagatajnanakayasya manjuSrijnanasattvasya nama­ samgiteh pindarthah (3 : de Itar de dag gcig tu bsdus nas tshigs su (su om. Tib.P.) bead pa brgya drug cu rtsa gnyis rnams dang / yon tan brgya Inga bcu'i gzhung gis bcom Idan 'das de bzhin gshegs pa thams cad kyi ye shes kyi sku 'jam dpal ye shes sems dpa'i mtshan yang dag par brjod pa'i don bsdus pa'o / / Tib. [18 5. 2. 6] The choice of readings for this passage is between A and |3, where Tib. follows p (B is unavailable). The punctuation of both A and Tib.D. indicates that the end of the passage marks the end of a section: Tib.D. has two double shad (vertical strokes), which are usually found at the end of a chapter; A has two double dandas with a small circle in between (// o // ). A has these before as well as after the passage. A divides the passage into two sentences whereas p and Tib. have one. As a single sentence the passage is problematic: it is hard to see how either tad ekatra or the instrumentals, -gdthdbhir and -granthena, connect with what must be the subject, pindarthah. If there are two sentences the ca following -granthena of (3 now introduces the second sentence, rather than linking the two preceding instumentals. In terms of corruption, however, it is easier to understand the collapse of a sentence through scribal error than its expansion, unless there is an opportunity for insertion of additional material such as a gloss. As two sentences the sense is better: the first sentence makes good sense following the enumeration of the verses of each chapter of the NS; the second makes sense as

310

Textual Notes. Chapter I

the conclusion of the forgoing material that starts at NMAA 1,11 (idanim asya namasamglteh Sariravyavasthapanam abhidhiyate /). It might be suggested that the latter sentence should go with what follows, that is as the opening sentence to the discussion of sambandhah, abhidheyam, prayojanam and prayojanaprayojanam. (Such a proposal would require emendation of idam to idamm.} 776.'s 'dir {atra} at the beginning of this following passage could be read as indicating a link with the present sentence and thus supporting this proposal (yet Tib. may be itself in error here since the Skt. reads sarvatra rather than atra). However, as 'summary meaning' or 'summary' the term pindarthah better describes the enumeration of the chapters of the NS with their titles. If the two sentences belong together, A's idam can be retained, and understood to refer to the whole preceding passage on the divisions of the NS. Taking the two sentences in this way also follows the uncharacteristically decisive punctuation. While the punctuation of the Skt. MSS and of Tib. is in general unreliable, here it is striking that both A and Tib.D. indicate that the sentence concludes a section of text. Also, Tib.P. gives two single shad, a convention normally used for indicating verse line or chapter breaks. 1.47

sarvatra vakyaithasams'ayena conj. (Tib.[185.2.7] 'dir tshig gi don the tshom gyis) : sarvarra vakarthasamSayena Y : sarvatrai(tre A)va hy arthasamsayena AH26 (sarvatraiva hy arthasamsayena samsayena D) Tib.'s tshig gi supports vdk or vakya. Emendation to vakartha- leaves the syllable ka in Y's vakartha- and also hy in cett. unexplained. The alternative, of emending to vakyartha-, is preferable. The syllables kya and hya are close in Newari script and -kya- could corrupt to -ka- in Y- The reading sarvatraiva can be seen as a subsequent corruption (perhaps via A's sarvatrevd) to give sense after the creation of hy. Tib.'s 'dir supports atra rather than sarvatra, however.

311

Textual Notes. Chapter 1

1.47-8

abhidheya AEH 2CC : abhipraya yCa : abhidhepraya D

4

prayojanatat-

prayojana A : prayojanaprayojam y : prayojanaprayojanaprayojan! H2 : prayoja* Ca : prayojanaprayojana DECC sambandhabhidheyaprayojanatatprayojana A : 'brel pa dang / brjod bya dang dgos pa dang / dgos pa'i yang dgos pa'i Tib. [185.2.7-8] The issue here - whether A, or an emended version of H2 (prayojana­ prayojanaprayojan d] is to be preferred - is one of style and idiom rather than meaning, which is unaffected. Though interpretation of the variants is not simple it seems to me unlikely that a (B is unavailable here) read prayojanaprayojanaprayojand. One explanation of the reading of H2 is that -dbhipraya(yCa) is a corruption of -abhidheya-, the corrupt praya then attracting an additional jana to give the third prayojana in H2. (If this is the case D's -abhi­ dhepraya- may be the result of scribal misreading of a correction as an insertion.) The alternative explanation, that H2 is essentially correct, requires that -abhipraya- be a corruption of -dbhidheyaprayojana-. This account, which demands two two-syllable omissions seems less likely. (The first explanation also requires that (3 has omitted tat from tatprayojana-, though this is not impossible.) Unfortunately, since J is missing here it is not possible to establish whether or not H2 is following J, as it sometimes does. (See the discussion of contamination in the Introduction to the Text, section 4.3.) Nevertheless, H2 's reading could result from contamination from outside (3 and thus be a good reading. If prayojanaprayojanaprayojand were correct, tatprayojana could be an attempt to improve the style of the original since A's reading is certainly stylistically preferable to having prayojana three times in succession. But Vilasavajra's style is generally not clumsy. Tib.'s dgos pa dang dgos pa 'i yang dgos pa appears to support -prayojanaprayojanaprayojana- against -prayojanatatprayojand since if tatprayojana

312

Textual Notes. Chapter 1

were its exemplar one would expect de'i yang dgos pa or de'i dgos pa. However, Tib. may have translated as it did in order to avoid possible ambiguity in a literal translation oftatprayojana. Also, since this four-fold classification is found elsewhere in Buddhist Sastra, (for example, in the opening of Dharmottara's commentary on Dharmakirti's Nyayabindu: see Stcherbatsky, 1930, II, 1) Tib. may have used a familiar idiom rather than translate more literally. In any case, Tib. does not translate prayojanaprayojana literally - the presence of yang {api} gives pray oj ana sy dpi prayojanam as a retranslation. To conclude, the variants in the MSS of (3 suggest a path of corruption from A to H2 that is more convincing than one in the reverse direction, and A also seems more in keeping with Vilasavajra's style. Given that Tib. does not decisively favour either alternative, I have preferred A's -tatprayojana-. 1.56

dharmatavataranartharri 5 : dharmatadhvataranartham y : dharmatartha A : chos [gnyis su med pa] gzung ba'i don du Tib.[185.3.4] : dharmatartham NS.Dav.

:

[gnyis su med pa'i] chos nyid kyi don [dang ming]

NS.Tib.D.N.fMukherji 119, 3] yad utadvayadharmatavataranartharn namasamdharanaprakaSanatayeti 5 : yad utadvayadharmatartham namasamdharanapraka£anatayeti NS.Dav. 164, 17] I have followed 5 against A here; and y's -ddhvatdrand- is likely to be a corruption of -dvatdrand-, giving dharmatdvatdrandrtham as the reading of (3. Tib. 's gzung ba 'i don du indicates that it read something more than dharmata­ rtham, though gzung ba is puzzling. Tib. has it for -samdhdrana- immediately afterwards (mtshan gzung ba bstan to Tib.[185.3.4]\ so its presence may be a scribal error triggered by this use. Tib. also lacks an abstract suffix (nyid) to give -dharmatd- rather than -dharma-. This citation of the anusamsd differs, however, from the anusamsd as found in NS.Dav., which reads -dharmatartham. If this is taken as supporting

313

Textual Notes. Chapter 1

A's -dharmatdrtha (read as having lost an anusvdra), the element -vatdrandcould be seen as an interpretative addition. NS.Tib.D.N. follows NS.Dav. in having nothing for -dvatdrand- (chos nyid kyi don); it also reads a compounded -dharmatdrtha- rather than -dharmatartham (don dang ming), suggesting further that A may be correct as it stands. In Chapter 12, on the anusamsd, the NMAA paraphrases this passage as follows: yad utddvayadharmatd sunyatd tatrdvatdrandrtham (A.108v7). Since the anusamsd is not cited separately from this paraphrase it is not clear whether Vilasavajra read it as -dharmatdvatdrandrtham or -dharmatartham. In terms of sense he certainly took it as the former; and some such meaning has to be understood for sense to be made. Overall, I think the paraphrase makes it more likely that the NMAA also read this alternative in the present citation. If -dvatdrand- is a scribal addition, it is a sophisticated one, matching the later paraphrase, and has to have predated the Tibetan translation (unless Tib. here is itself contaminated). The NMAA cites or incorporates passages from the NS anusamsd at a number of points in Chapters 1-5. Tib. never follows NS.Tib.D.N., the Tibetan translation of the anusamsd that is found in the (Derge and Narthang) Kanjur editions, as printed in Mukherji. (See the Introduction to the Text, section 5.3, note 51, for a list of anusamsd citations.) 1.69

Citation of NS verses NS verses are given in bold, and those surrounded by square brackets have been inserted by the editor. MS J cites NS 1 as a whole, as it does the other NS verses of this chapter. These citations, which are usually accompanied with translation into Newari, are only found in J, and are clearly insertions. For a fuller description of MS J see Introduction to the Text, section 2.4. Of the NS

314

Textual Notes. Chapter 1

verses commented on in chapters 1-5 of the NMAA only verses 33^0 are cited in whole. Otherwise half-verses, padas, or single words are cited.

1.70

\n\Q\aNMAA [1,86] NS.Dav.[49, v.l] : trailokya J I follow NMAA[1, 85-6] (tan vijetum Silam asyeti trilokavijayi) as well as NS.Dav.[49, v.l] in preferring trilokavijayi to trailokyavijayl. The latter is only attested by J's inserted (see previous note) citation of NS 1.

1.72

mayajala codd. Tib. : mayajalat NS.Dav.[69, 8]

4

sahasrika A Tib. :

sahasrikat (3 : sahasrikan NS.Dav. aryamayajalaso^sasahasYikamahayogatantrantahpatisamadhijalapatalad A : aryamayajalat soda£asahasrikan mahayogatantrantahpatisamadhijalapatalad NS.Dav. [69, 8-10] : 'phags pa sgyu 'phrul dra ba khri drug stong pa mal 'byor chen po'i rgyud kyi nang nas ting nge 'dzin dra ba'i le'u las Tib.[185.4.2-3] : 'phags pa rnal 'byor chen po'i rgyud sgyu 'phrul dra ba stong phrag bcu drug pa ting nge 'dzin dra ba'i le'u las phyung ba NS.Tib.D.N.fMukherji 102] In this incorporation of the NS colophon into the NMAA, Tib. supports both codd.'s -mayajala- and A's -sahasrika-: there are no case particles after sgyu 'phyul dra ba or khri drug stong pa. The compound passage thus read yields good meaning. By contrast, the ablatives -mayajalat and -sahasrikat in NS.Dav. make no grammatical sense. It looks as if they have both resulted from attraction to the case of -patalad. However, that both |3 and NS.Dav. have -sahasrikat is puzzling. How likely is it that the same error could be made independently? Perhaps p's reading is a result of contamination from scribal familiarity with an already corrupt NS manuscript tradition. NS.Tib.D.N., which translates the passage differently to Tib., also supports the readings I have accepted, further suggesting that NS.Dav. is corrupt. (As with the translation of the other citations and incorporations from the NS anusamsd, NS.Tib.D.N. here differs from Tib.. See Introduction to the Text, section 5.3 (v) and section 5.5.)

315

Textual Notes. Chapter 1

1.73

bhagavacchakya Aye Tib.([185.4.3] bcom Idan 'das £a kya thub pas) : bhagavan Sakya £ : bhagavattathagataSakya NS.Dav.[69, 9] NS.Tib.D.N.[Mukherji 103] 1 have followed Aye (which by stemmatic analysis I take to be the reading of a) and Tib. against NS.Dav. and NS.Tib.D.N.. If the reading bhagavattathagatasakya- is correct then, since Tib. and a agree, Tib. must have inherited a corruption already present in its exemplar of the NMAA (independent loss of -tathagata- seems unlikely). However, it is also possible that the MS tradition of the NS itself became corrupted, adding the term tathagata, again before its translation into Tibetan. Lacking a strong reason for preferring one alternative over the other, I follow a on the grounds that it is the reading of Vilasavajra's text as far as can be ascertained.

1.73

bhasita conj. (Tib.[185.4.3] gsungs pa; NS.Dav. [69, 9f) : bhasitad codd. Sense requires emendation to -bhasita since it must qualify ndmasamgitih in the following line. The ablative, -bhasitad, may well have been attracted from the preceding -patalad.

1.88-9

s"ritattvasamgrahe panca guhyani pathyante guhyarat kuliSeSvarah NS Id : rdo rje dbang phyug gsang ba'i rgyal NS.Tib.fMukherji 2] Tib., in its translation of the NMAA on NS Id, follows the order of the epithets in NS.Tib., which reverses (for metrical reasons) that of the Skt.. Hence, it translates the short passage on kulisesvarah (rdo rje dbang phyug), namely, kuliso vajras tasya isvarah kulisesvarah (1.101; Tib. [185.5.4f), before the longer section on guhyarat (gsang ba'i rgyal), which starts with sritattvasamgrahe panca guhyani pathyante (1.88-101; Tib.[185.5.4-186.1.1]). For further discussion of 776.'s reordering of the NMAA to accord with the order of NS.Tib. see Introduction to the Text, section 5.3.

316

Textual Notes. Chapter I

1.100

guhyair va J : guhyaiva(va E) £ : guhyay : om. D : yang na gsang ba pa las Tib. [186.1.1]

4 guhyair va -+• guhyarat om. D

guhyair va rajata iti guhyarat conj. : yang na gsang ba pa las rgyal ba'i phyir gsang(gsar Tib.P.) ba'i rgyal po'o Tib.fl86.LJJ It is likely that J contains the reading of 6, which became progressively corrupted to guhyaiva and guhyaiva. The reading of Y, guhya, could perhaps give some sense (as corrupt locative, "rule over", though Jraj usually takes the genitive in this sense), but then vd, supported by Tib. would have to be supplied. Also y is easier to explain as a contraction than 6 as an expansion. Tib.'s gsang ba pa implies a reading of guhyaka: it translates as, "Alternatively, he is [called] guhyarat because he is victorious over the Secret Ones". Since las is the usual particle taken by the verb rgyal ba, it is hard to interpret it as supporting any particular reading in the Skt., though it could be taken as indicating an ablative. In terms of sense guhyair vd rdjata iti guhyarat is plausible: "Alternatively 'guhyarat' may be analysed as 'one who reigns because of the [five] secrets'". To take Jrdj in its sense of 'to shine', 'to be illustrious' is also possible: "Alternatively, * guhyarat' may be analysed as 'one who is radiant because of the secrets'". Neither the immediate nor broader context, however, suggests that this is how it should be taken. 1.106

vibuddhapundarikaksa iti conj. : vibuddhamti 5 (vibuddheti J) : om. y Tib. See the following note.

1.109

protphullakamalanana iti conj. (Tib.[186.1.3] pa dma rab tu rgyas pa'i zhal zhes pa ni {protphullakamalanana iti}) : om. codd. I have emended so as to include the lemma both here and for the first pdda of NS 2 (vibuddhapundarikaksa iti, line 106). Both glosses end with the word tathoktah (vibuddhapundarikam ivaksim locane yasya sa tathoktah - lines 1 OS-

317

Textual Notes. Chapter 1

9; prakarsotphullakamalavad ananam yasyasau tathoktah - line 111), implying a previous statement of what is being glossed. Since the NMAA does not cite the whole verse before its analysis, without such opening citations the two passages do not read with any sense. The evidence from the Skt. MSS is indecisive: for the first pada the lemma is partly supplied by 6, and omitted by y; for the second pada the lemma is not witnessed by any of the available MSS. Tib., on the other hand, omits the first lemma but has the second. Such a pattern could result from errors either of addition or of omission. The appearance of a lemma is on the whole easier to account for than its disappearance (see the note on 1.116 where the former is likely to have occurred) since it is a familiar unit. However, both MSS A and B are unavailable here, and (3 can have substantial levels of corruption in relation to AB. Given the ambiguity of the MSS evidence, the strength of the argument from sense suggests the proposed emendation. 1.112

vajrasya madhyam codd. Tib.([186.1.4] rdo rje'i dbus) Emendation to vajravarasya madhyam is suggested, as otherwise varam, a word from the NS, is neither mentioned nor glossed and this is uncharacteristic of Vilasavajra. Since Tib. supports vajrasya I have not emended, yet there remains a doubt over the passage since vajrasya may be the reading of |3 rather than ft, MSS A and B being unavailable. Tib., which misinterprets the latter part of this sentence, could have inherited a corruption (see the following note).

1.112

svakarena JH2 : karena y Tib.([186.1.4] lag pas) : sa svakarena De vajrasya madhyam svakarena grhitva muhur muhuh punah punah praksiptam vajram praticchan prollalayann idam aheti sambandhah : rdo rje'i dbus nas lag pas bzung (gzung Tib.P.) nas yang dang yang du rdo rje gyen du bsgyur nas gzung ba ni / rang gi lag gis (gi Tib.P.) rdo rje mchog / yang dang vang du gsor bved pa'o / 'di skad gsol to zhes pa dang 'brel to / Tib. [186.1.4]

318

Textual Notes. Chapter 1

The question that arises here is the same as that implicit in the discussion of the preceding note: is Vilasavajra incorporating the text of the NS into the sambandhah, ie. into his structural analysis of the half-verse and its relation to the rest of the sentence? It is usual for him so to do. The MSS support karena as well as svakarena (following the NS). Stemmatic analysis of the readings is not decisive: it is not possible to say what the reading of (3 may have been. The sa of De's sa svakarena could be a dittography, which J either corrected or failed to inherit. H2's svakarena, an instance of H's contamination, is not necessarily correct: svakarena may well be a scribal 'correction' to the familiar word of the NS. On the other hand svakarena may be correct while y's karena lost a syllable. Though Tib. supports karena, it could have inherited an error from its exemplar. (See the note on 1.45 for an example of Tib. inheriting an error, found in (3.) On balance I prefer svakarena on the grounds that it makes sense within the broader context of Vilasavajra's style. When he gives a sambandhah Vilasavajra usually incorporates words of the NS, sometimes with a gloss, (as here, with muhur muhuh punah punah praksiptam vajram praticchan prollalayann}. Tib. also changes the sense of the passage by inserting a citation of the halfverse after translating the sentence as far prollalayann. This is blended, though not very happily, with the rest of the sentence by the addition of the word shad ("speech") to the translation of idam aheti sambandah. See above for the Skt. as I have accepted it and Tib. - the citation of the NS by Tib. is underlined. 1.116

Srivajrapanyabhisekatantre conj. Tib.([186.1.5] dpal phyag na rdo rje dbang bskur ba'i rgyud las) : s~rivajrapanyabhiseke tantre Y • Srivajrapanibhiseke tantre £ : £rivajrapanya(-ni D)bhiseke tantre bhrkutl ityadi £CCH2 Srivajrapanyabhisekatantre £rivajradharanucara bhrkutitarangapramukhah panca s"atani krodhagananam pathyante conj. : dpal phyag na rdo rje dbang bskur ba'i rgyud las / dpal rdo rje 'chang gi rjes su

319

Textual Notes. Chapter I

spyod pa khro gnyer rim par(pa Tib.P.) Idan pa la sogs pa khro bo'i tshogs Inga brgya brjod Tib.[186.1.5-6] There are two issues here: firstly, the emendation from snvajrapdnyabhiseke tantre to srivajrapanyabhisekatantre; and secondly, the possibility that the reading bhrkuti ityadi may be (part of) a lemma that has become displaced from its original position. The emendation to srivajrapdnyabhisekatantre creates good Sanskrit. It is the standard way that Vilasavajra refers to other texts, and it is supported by Tib, (the connective particle 'i is used to translate such a compound). As for the reading bhrkuti ityadi, it is not supported by Tib. and is not required by the sense of the passage, which can stand alone without a lemma (bhrkutitarangapramukhdh can be taken as an allusion to the NS that is incorporated into the text). I therefore take bhrkuti ityadi to be a scribal insertion: the manuscript readings suggest that it was not present in |3 (being absent in both y and e). The corrected readings of CCH2 could represent contamination from corrupt exemplars, for example from J in the case of H (see Introduction to the Text, section 4.4). 1.120-1 ata eva viras te / tair anabhibhavaniyatvat codd. : de nyid kyi phyir dpa' bo ste / de de rnams kyis zil gyis mnan par mi nus pa'i phyir ro Tib.[186.1.7-8] Taken strictly, tair must be understood to refer back to durddntadamakaih, yet what is required is reference to durdantaih, the 'hard-to-tame'. It is they who cannot overcome the attendant Vajrapanis. In other words, taih would be glossed as durddntair iti taih. Could taih be a remnant of durdantair! Tib. supports taih (taking the first de as a misplaced translation of te} or tair eva (treating de de as emphatic). In the absence of any concrete evidence for emendation it is probably safer to retain taih; its proper referent is clear from the context.

320

Textual Notes. Chapter 1

1.125

ullalayadbhih svakaraih prasphuradvajrakotibhih NS 4ab : rdo rje rtse mo rab 'phro ba / rang gi lag gis gsor byed pa / NS.Tib.[Mukherji 3, 7-8] Tib. follows the order of NS. Tib. and rearranges the NMAA, placing the passage on prasphuradvajrakotibhih (1.126-7; Tib. [186.2.1-2J) before that on ullalayadbhih svakaraih (1.125; Tib.[186.2.2]}.

1.125

svakaraih y Tib.[186.2.2] : muhur muhuh svakaraih 6H2 ullalayadbhir iti urdhvam utksipadbhih / kaih / svakaraih // : gsor byed pa zhes bya ba ni gyen du sgyur ba'o / gang gis zhe na / rang gi lag gis so/ Tib.[186.2.2] I follow Y and Tib. here, rather than 6H2 . The reading muhur muhuh could be a scribal insertion, added under the influence of the phrase svakarena muhur muhuh in NS 2. That H2 has muhur muhuh svakaraih is not necessarily evidence of (3 having it, since H2 is sometimes the result of contamination from J (see Introduction to the Text, section 4.3).

1.126

prasphuranti conj. (Sanderson; Tib.[186.2.1] rab tu 'phro ba) : prasphullanam 6H2 (prapraphullanam J) : prasphurad eva y prasphuradvajrakotibhir iti prasphuranti vajranam kotayo 'grabhaga yesam te praspuradvajrakotayah / taih conj. : rdo rje rtse mo rab 'phro ba zhes ba ni rab tu 'phro ba'i rdo rje'i rtse mo ste / thog ma'i cha dang gang Idan pa de ni / rdo rje rtse mo rab 'phro ba zhes so Tib. [186.2.1] Within the context of the bahvrihi analysis of praspuradvajrakotayah the extant readings cannot be correct since a word is needed to agree with kotayo 'grbhagah. A present participle is suggested in order to accord with the stem form, prasphurat, in the NS pdda. Tib. supports prasphuranti: the same verbal adjective, rab tu 'phro ba, is used in the translation of the citation of \hz pada and its analysis (though abbreviated to rab 'phro ba to conform to the metre in the pdda citation).

321

Textual Notes. Chapter 1

1.129

tatra sarvadharmanam codd. : de la snying rje dang Idan pas snying rje'o / bde ba 'gog par byed pa snying rje ste / Tib.[186.2.3-4] prajnopayamahakarunajagadarthakaraih paraih NS 4cd[NS.Dav. 49, v.4] : snying rje che dang shes rab dang / thabs kyis 'gro don byed pa'i mchog NS.Tib.[Mukherji 3, 9-10] Tib. does not correspond with the Skt. since it rearranges the NMAA passages that comment on the terms prajnd, upayah and mahdkarund in conformity with their order in NS.Tib.. Thus, the passage following de la {tatra} above is Tib. for the passage on mahdkarund, which is translated first (1.131-4; Tib.[186.2.3-5]). This is followed by the passage on prajnd (1.12930; Tib. [186.2.5]}; with that on upayah being translated last (1.130-31; Tib.[186.2.5]). Tib.'s bde ba 'gog par byed pa snying rje ste translates kam runaddhiti karund (emend 'gog par, 'to tear out' to 'gogs par, 'to suppress'?), but the preceding passage has nothing that corresponds with it in the Skt.. Although MSS A and B are missing for this passage, the Skt. does not give the impression that some may have been dropped, and it is not clear how the Tibetan (meaning something such as "[the word] compassion [is used] since [he is] compassionate") could fit in. Also Tib. does not attempt to translate kam iti sukhasydkhyd, the second part of the nirvacana.

1.129

pravicayah conj. (Tib.[186.2.5] rab tu 'byed pa) : pravicarya 5 : prasiddha-

yey Emendation to pravicayah is supported by Tib. as well as on grounds of sense. Tib.'s rab tu 'byedpa means 'analysis', 'investigation'. The longer form, rab tu rnampar 'byed pa, is given for pravicayah by the Mahavyutpatti (MVy 846, 990). The second of the seven bodhyangdni, dharmapravicayah is defined as one of the components of prajnd by Vasubandhu: smrtyupasthdndni

322

Textual Notes. Chapter 1

dharmapravicayasambodhyangam samyagdrstis ca prajnaiva (AKBh 1017, 10). 1.130

prajna / upayo conj. (Tib.[186.2.5] shes rab bo / / thabs ni) : prajnopayayo(-yau C) 5 : prajnopayayoh y The context here must be that of a discussion, in turn, of prajna, upayah and mahakaruna, in which case |3's dual form makes no sense (A and B are lacking). Emendation to prajna / upayo is supported by Tib..

1.131

sakalasattvarthakaranam conj. (NMAA[1, 142] \ Tib. [186.2.6] sems can rnams kyi don byed pa) : sakalasattvakaranam codd. The context suggests that a word such as artha has been lost from codd., and the parallel phrase at 1.142, sattvarthakaranad (em. : -karanani codd.), suggests such an emendation. Tib.'s don confirms this. The emendation from -karanani to -karanani is supported by byed pa, which is also used at 1.142 (see the note on 1.142 below).

1.135

arthakarair iti Cay : arthakaraih parair iti 5 Tib. I have preferred the reading arthakarair iti since this passage concludes the material commenting on prajnopayamahakarunajagadarthakarair by citing it in the (ablative) form used in the NS. Since paraih, the last word of NS 4, is not part of what has been commented on - it is glossed immediately afterwards - there is no point in its inclusion. Scribal familiarity with the half-verse could well result in its addition to the text, and it is the type of error that could occur independently in 5 and Tib..

1.142

sattvarthakaranat conj. Tib.([186.3.1] sems can gyi don byed pa'i phyir ro) : sattvarthakaranani codd. (om.H)

323

Textual Notes. Chapter 1

The plural termination -ant makes no sense in this context. An ablative of reason is supported by 7/6. 's phyir. Since MSS A and B are unavailable here codd. represents the reading of |3 only. 1.144-6 mahabhairavattahasa conj. : mahabhairavo ttahasa y : mahabhairava attahasa 5 (mahabhairava athasa C) 4 alamkrti conj. : alamkrta kirn tad ruparn / mahabhairavattahasanekasirsakaracaranavikrtanayanordhvakes^pihgalabhriibhangakapalamalalanikrtisarpabharanavyaghracarmanivasanadikarn / tad vidyate yesam te krodhavigraharupinah conj. : khro bo'i lus kyi gzugs ci 'dra zhe na / 'jigs byed chen po dang / drag tu bzhad pa dang / dbu dang phyag dang zhabs du ma dang / spyan mi sdug cing skra kham pa(sa Tib.D.) gyen du 'greng ba dang / rdzi ma ser po(em. : ba Tib.) dang / khro gnyer dang thod pa'i phreng bas brgyan par byed pa dang / sbrul gyi rgyan dang / stag Ipags kyi(kyis Tib. P.) smad g.yogs la sogs pa de mams gang la yod pa de ni khro ba'i lus kyi gzugs Idan pa ste / Tib. [186.3.2]. Neither the Skt. nor Tib. is completely satisfactory here. The bahuvrihi termination -adikam requires the elements of the compound giving the attributes of the Krodhas to be a list of substantives. However, if mahabhairava is part of the compound and if it is read as an adjective ("very terrifying") it cannot stand alone but has to qualify attahasa. Yet given the existence of mahabhairava and attahasa as separate epithets of Siva one might expect them to be separate attributes. Furthermore, the available MSS (AB are lacking) separate mahdbhairava from attahasa, as does Tib. If the separation is taken to indicate that mahdbhairava should be outside the compound, emendation to mahdbhairavam attahasa- is required, to give independent qualification of rupam (thus, "[that form] is very terrifying, with loud laughter, many heads ..."). Though this gives good sense, the process of corruption to mahdbhairavo (y) and mahdbhairava (5) is hard to see. Also, Tib., although it separates mahdbhairava from attahasa, takes them on the

324

Textual Notes. Chapter 1

same syntactic level (ie. jigs byed chen po is followed by dang, as are the other attribues). Alternatively, if mahabhairava is part of the compound the Skt. readings could be the result of an earlier separation of mahabhairava and attahdsa to show the word break within the compound (y then taking them as separate words and giving a 'correct' sandhi and 6 reading mahabhairava for mahdbhairavo). But mahabhairava could be taken as both a separate attribute and part of the compound only if it can be read as a noun. For bhairava, MW records the meaning "n. terror or the property of exciting terror" (s.v. bhairava, where it is given as a meaning recorded by H. H. Wilson). This could give some sense ("great terrifyingness"), though it seems less likely than the karmadharaya "greatly terrifying", which is based on the more common adjectival meaning of bhairava. To summarise, the evidence from the MSS points more towards mahabhairava being compounded, and this is supported by Tib.. The evidence for emendation to mahdbhairavam attahdsa- is inadequate, and as a result I have emended to mahdbhairavdttahdsa-. As for meaning, I have translated it in line with the requirements of the syntax (pace Tib., which could possibly have misunderstood the original) as an adjective qualifying attahdsa. Yet reservations must stand. The description of the Krodhas might have been incorporated from elsewhere. If such an original or parallels were traced, further light might be cast on the passage. I have also conjecturally emended -kapdlamdJdlamkrta- to -kapdlamdldlamkrti- on the grounds that a substantive is required here (though alamkj-ti is a little unusual in prose). Tib.'s brgyan par byed pa, a nominal form, supports such an emendation. Two other features of Tib. can be noted: for urdhvakesa it has skra kham pa gyen du 'greng ba adding kham pa ("brown"), which is not

325

Textual Notes. Chapter 1

found in the Skt; and it takes pingalabhrubhahga as two attributes (rdzi ma ser po dang / khro gnyer dang), "tawny eyebrows and a frown".

1.155

tamconj. : tarn natham A : natham (3 Tib.([186.3.6] mgonpo) I take tarn to conclude the gloss on natham by signifying its (accusative) case in the NS. This is necessary since the gloss itself is carried out in the nominative. This method of indicating the case of a glossed word at the end of a gloss is seen elsewhere, eg. at 1.127 (prasphuradvajrakotibhir iti prasphuranti vajranam kotayo 'grabhaga yesam te praspuradvajrakotayah / taih // ). The word natham following tarn in A is redundant. Tib. would appear to support (3's reading of natham. However, the evidence from Tib. at this point is not clear: the punctuation makes mgon po the first word in a new sentence that initiates a rearrangement of the text of the NMAA (... gtso bo ni mgon po'o / / mgon po yang ci 'dra ba'i mgon po yin zhe na / gsungs pa / de bzhin shegs pa zhes so Tib.[186.3.6-7J). There should probably be a shad after mgon po with yang translating punar api. This would leave Tib. supporting |3. Nonetheless, natham on its own seems stylistically unlikely: a citation of the root text in such circumstances is usually introduced by atah or ata eva. Also, |3 omits natham after the immediately following punar api kimvisistam. (That A's punar api kimvisistam natham is correct here is shown by the parallel forms in this passage on NS 6ab: 'bhuyo 'pi natham - line 157; punar api kimbhutam natham - line 165.) The first natham could possibly be a dittography present in a, with the second (and correct) instance subsequently omitted in (3. Despite the evidence, supported by Tib., that tarn natham (or natham} is present in a, I have diagnostically emended to tarn on grounds of style and the existence of other instances of the same pattern elsewhere in the NMAA.

326

Textual Notes. Chapter 1

1.156

punar api kimviSistam natham ity aha sambuddham iti : yang ci 'dra ba'i mgon po yin zhe na / gsungs pa / de bzhin shegs pa zhes so Tib.[186.3.6-7] pranamya natham sambuddham bhagavantam tathagatam NS 6ab [NS.Dav. 50, v.6] : mgon po bcom Idan de bzhin gshegs / / rdzogs sangs rgyas la phag 'tshal te NS.Tib. [Mukherji 4, 9-10] Tib. arranges the text of the NMAA that comments on NS 6ab in a different order to the Skt. at this point. The Skt. takes each word of the half verse in turn. Tib., on the other hand, follows the Skt. order until the conclusion of the comment on natham (ie. ...andthdndm lokdndm anusdsandt ndthah svdmi / tarn, line 154-5 ), after which it takes the glosses on the three remaining words in reverse order (ie. tathagatam, Tib.[186.3.7]; followed by bhagavantam, Tib.[186.3.8]; followed by sambuddham, Tib.fl86.4.3J). Tib. returns to the order of the Skt. at pranamyeti purvena... (line 1.167). The reason for this order is not clear. Changes by Tib. to the order of the Skt. usually reflect the order of the Tibetan translation of the NS used by Tib. in the citations of the NS in the NMAA (ie. the order of NS.Tib.). For the present verse, NS.Tib. does translate the three terms in a different order, but that of bhagavantam tathagatam sambuddham rather than tathagatam bhagavantam sambuddham. (Surely in NS.Tib., bcom Idan was placed where it was so that the following de could be also heard as 'das}. One solution is that this is an instance where the Tibetan translation of the NS used by Smrtijfianakirti differed from that usually ascribed to Rin chen bzang po (taking NS.Tib. to reflect Rin chen bzang po's translation). Unfortunately, since neither the Skt. of the NMAA nor Tib. cite NS 6ab as a whole we cannot see what that translation may have been. (See Introduction to the Text, section 5.)

1.157

va conj. : om. codd. na Sravakam pratyekabuddham va conj. : nyan thos dang rang sangs rgyas ma yin pa'i phyir ro Tib.[186.4.3-4]

327

Textual Notes. Chapter 1

The emendation is diagnostic, the minimum necessary to give some sense. Tib. makes this passage a reason for what precedes: "How does it, [that is, the Namasamgiti,] also describe the Lord? As the perfectly enlightened one (sambuddham > samyaksambudddham), since he is not a £ravaka or Pratyekabuddha." To render this literally in Sanskrit one would have to emend to something such as net sravakapratyekabuddhatvat. There is no justification for this in the Skt., though it may be argued that this is its implicit sense as emended. 1.159

avrti conj. (AAA 1.2-3) : avrtis codd. : [nyon mongs shes bya'i] sgrib Tib. [186. 4. 1J The evidence from MS B is unavailable at this point. In emending to the dual form I follow the (unattributed) citation of this verse by Haribhadra, taking klesajiiey avrti to stand for klesavrtijney avrti, ie. short-hand for the two hindrances more commonly known in the form klesavaranam and jneyavaranam. Tib. supports the word avrti but is ambiguous as to number (sgrib pa is used to translate both avrti and avaranam into Tibetan).

1.164

bhaga iti A AAA; AbhayaP; CSViv; SpPr : iti bhaga(gah J) 0 4 smrtih AH2

CSViv : smrtah P SpPr : frutih AAA; AbhayaP aisVaryasya samagrasya rupasya ya&asah Sriyah / jnanasyatha prayatnasya sannam bhaga iti smrtih // conj. : dbang phyug dang ni gzugs bzang (bzangs Tib.D.) dang / / grags dang dpal dang rab mchog dang / / de nas ye shes brtson 'grus dang / / drug po rnams la Idan par bshad Tib. f186. 4. 2-3J. The readings of the MSS representing (3 are clearly corrupt in reversing the order of bhaga and ///. The final word is more problematic since both smrtih and smrtah are witnessed by testimonia, as is a third option, srutih, which, understood as a synonym for sabdah, gives good sense: "The word 'bhaga' ...". The different readings of the testimonia suggest that the verse circulated in

328

Textual Notes. Chapter I

more than one form; the present task, however, is to try to establish Vilasavajra's reading. AH2 's smrtih is likely to be the reading of (at least) a since when H's corrections do not follow J they come from outside P. (See Inroduction to the Text, section 4.3 for discussion of H's contamination.) The reading smrtih is attested in the CSViv, and although it could be a corruption of srutih, smrtih gives sense if understood in the sense of 'tradition': "There is the tradition that [the word] 'bhaga' ...". The reading smrtah, attested by |3, and also witnessed in a £aiva testimonium (SpPr 89, 2-4), is more problematic. Being a participle, smrtah needs a subject. If it were to be understood as, "What is called bhaga [denotes] six things ..." one would expect ...bhaga iti smrtam. Also, smrtih could, through an error of omission, easily corrupt to smrtah, whereas corruption from smrtah to smrtih is less straightfoward. I have, therefore, preferred the reading smrtih, as attested by AH2 . (Corruption to smrtah could also have been influenced by its occurence in line 160, at the end of the preceding citation exemplifying the first nirvacana of bhagavan - though |3 has smrtah here.) Tib. is puzzling in a number of respects. The phrase rob mchog dang seems to make a list of seven items rather than six (though rob mchog is an adjective); dang after ye shes brtson 'grus would usually imply another item to follow; and gzugs bzang indicates that samagrasya has been taken with rupasya, rather than with aisvaryasya as the Skt. pada break would suggest. (That aisvaiyasya samagrasya belong together is further suggested by the £aiva testimonium, which has aisvaryasya samagrasya jnanasya yasasah sriyah, ie. jnanasya substitutes for rupasya suggesting the two are separate items; but aisvaryasya samagrasya stay together.) Tib.'s treatment of the final part of the citation is not literal and so cannot throw any light on the Sanskrit: "As for the six,

329

Textual Notes. Chapter I

(sovereignty etc.) [they are] explained as possessions" (drug po mams la Idan par bshad) (Das 709, s.v. Idan pa Inga "the five possessions"). (I am indebted to Prof. A. Sanderson for the testimonia for this and the preceding citation.) 1.183

klesah sat A Tib.([186.4.8] nyon mongs pa drug go) : klesa ragadayah sat |3 The reading of A is to be preferred: it is supported by Tib.; also the list of klesas that follows starts with mdnah, not rdgah. Tib. also supports the order of klesas as given in the Skt.. Since most of the standard lists of the six klesas do, however, start with rdgah the presence of ragadayah can be understood as a likely scribal gloss. (See note to the translation.)

1.189

bhagavan conj. (NMAA 1, 191 prakaSayatu bhagavan ityaha) : bhagavam NS.Dav. 50, v.9 In inserting the text of NS 9, I follow the reading bhagavan given by codd. in its citation of this part of the verse against NS.Dav. on the grounds that my intention is to give the NS as read by Vilasavajra. (In fact for one MS, Mukherji [6, v.9] has bhagavan. Davidson omits to record this.) It might be argued that bhagavan is a scribal regularisation of an originally irregular form, bhagavam. I do not think such a position sustainable. There is no evidence that bhagavam is any more than an orthographic habit of some Newar scribes, where an anusvdra substitutes for a final nasal. (This is probably an extension of the more common habit of substituting an anusvdra for medial nasals.) NS.Dav. also has bhagavam against codd.'s bhagavan in NS 17 and NS 23 (see notes on 2.1 and 3.1 respectively).

1.201

bhagavan J : bhagavan cett. : bcom Idan'das kyi [ye shes sku] {bhagavaj-} Tib.fl87.1.1J

330

Textual Notes. Chapter 1

atha bhagavan jnanakayasyetyadi // hebhagavanJ : de nas bcom Idan 'das kyi ye shes sku zhes bya ba la sogs pas ni / kye bcom Idan 'das Tib.[187.1.1] The context makes it clear that Vilasavajra reads bhagavan here and takes it as a vocative: the gloss he bhagavan is attested not only by codd. but is supported by Tib. (kye bcom Idan 'das ). That this is the case is reinforced two lines later when Vilasavajra treats jndnakdya as a separate epithet of Manjusrijnanasattva: kimvisistasya jndnakdyasyeti (line 203). Tib. follows the gloss of bhagavan as a vocative despite using an inherited translation of the NS in its opening citation which takes it as compounded in stem form w\\hjnanakayasya to give bhagavajjndnakdyasya (bcom Idan 'das kyi ye shes sku = NS.Tib.[Mukherji 1, v.10]). 1.212

evavabhotsyate / iti conj. : evavabhotsyatheti A : evavabhautsatheti |3 : [kun tu bzang po ye shes kyi / / bdag] nyid mngon par byang chub bo (em. : po Tib.) Tib.[187.1.6] I have emended diagnostically to the third person singular future since -avabhotsyatha is not possible, ova Jbudh being atmanepada. Corruption to -eti could be the result of double sandhi (ie. evavabhotsyata iti -*• evavabhotsyateti), but it is not straightfowardly clear how -yateti could become -yatheti. The context provides no clue as to the original person or number, but emendation to the singular maintains the metre and gives sense. Tib.'s mngon par byang chub can be read as supporting the emendation. (Tibetan prose would use a more periphrastic form.)

1.225

gambhirartham ityadi codd. : mtshan ni yang dag brjod pa'i mchog ces bya ba la sogs pa ste {namasamgltim uttamam ityadi} Tib.[187.2.3-4] gambhirartham udarartham mahartham asamam s~ivam / adimadhyantakalyanim namasamgltim uttamam NS llfNS.Dav. 50] : mtshan ni yang dag brjod pa'i mchog / / don zab don ni rgya che zhing / / don chen

331

Textual Notes. Chapter I

mtshungs med rab zhi ba / / thog ma bar dang mthar dge ba // NS.Tib. [Mukherji 7-8] For this verse NS. Tib. has as its first quarter verse a translation of the fourth pada of the Sanskrit (namasamgitim uttamam); and Tib., following NS.Tib., takes the opening of NS.Tib. as the pratika rather than translating the Sanskrit. As a result Tib. rearranges somewhat the order of the commentary to accord with the order of epithets in NS.Tib. by next translating the gloss on namasamgitim uttamam from line 235 (ie. Tib. translates iti gatdrtham). Tib. then returns to the Skt. order, translating the gloss on gambhirdrtham (don zab ces pa ni... Tib.[187.2.4]). From this point there is no conflict with the order suggested by NS.Tib. since after the initial change the order of NS.Tib. and the Skt. coincide for the remainder of the verse. The sDe dge edition of the NS revises the order of the translation to accord with that of the Skt., placing the first pada of the earlier translation last. It also revises the translation of the fourth pada to mtshan yang dag par brjod pa mchog (see Mukherji 7, note 5). Although this accords more closely with the Sanskrit, the correct metrical weighting of NS.Tib. is lost. 1.227

udaraS casav arthas codarartho conj.

: udara< ~14 > F : udarartham

< ~8 >rtha H : udararthamrtha C : udarartham udarartha D : udara­ rtham udarartha JH2 : om. E Tib. udarartham iti / udaraS casav arthaS codarartho vaipulyartha iti yavat / sa yasyam vidyate sa udarartha / tarn ca // conj. : don ni rgya che zhing zhes pa ni rab tu rgyas pa'i don to / de gang la yod pa de ni don ni rgya che zhing zhes so / Tib. 187.2.5 With the exception of E all the MSS available (those representing |3; A and B being unavailable) have some corrupted material that I take to be the remnant of a karmadharaya analysis of udarartham. I have emended diagnostically following the immediately preceding gloss on gambirdrthdm since its structure and that of the gloss on uddrdrthdm are the same: an opening citation followed

332

Textual Notes. Chapter 1

by a karmadharaya analyis (for both if the emendation is correct) and semantic gloss (sunyatarthah and vaipulyarthah); a bahuvrihi analysis of the whole (here the form is identical in both cases: sa yasydm asti sd gambhirdrthd; sa yasydm vidyate sa uddrdnhd) and finally a return to the accusative case of the NS. The grounds for emendation are persuasive given that the two glosses are so clearly meant to run in parallel. However, the process of corruption to the present state of the MSS is not clear, and with the exception of F the lacunae are not large enough to accomodate the proposed restoration. Stemmatic analysis of the readings suggests that |3 contained a lacuna, witnessed in varying sizes by FHC, which later scribes attempted to correct. Tib. (following an inherited NS translation with don ni rgya che zhing) has no passage that corresponds to the karmadharaya analysis: it appears to have read uddrdrthdm iti vaipulydrthah / sa yasydm vidyate sa uddrdrthd. Either it omits the karmadharaya gloss, though it does not omit the corresponding analysis of gambhirdrthdm, or it is not present in its exemplar. It is possible, though it would be uncharacteristic, that Vilasavajra dispensed with a karmadharaya analysis in this instance. Nonetheless, the corrupt passages in the Skt. MSS remain to be accounted for and they are more easily explicable as errors of omission than of addition. 1.230

tarn ca conj. (NMAA[1, 228]} : om. codd. This is a purely diagnostic emendation based on the fact that Vilasavajra usually returns to the case of the citation when giving a grammatical analysis of a compound. I have adopted tdm ca from two lines earlier. If MSS A and B were available here and supported the reading of (3,1 would be more cautious in suggesting such an emendation, but (3 often omits words and phrases. Tib. is of no help since it does not translate this part of these grammatical analyses.

1.231

sasama conj. : sa ity asamam codd. (om. F; ity asamam H2 )

333

Textual Notes. Chapter 1

asamam iti / na vidyate sama yasya sasama conj. : mtshungs med ces pa ni mnyam pa med pa gang yin pa de ni mtshungs med do Tib. [187.2.6] I take iti to be a scribal addition, perhaps an attempt to create some sense after an earlier corruption to asamam. It is attested by the MSS of (3 only, A and B being unavailable. The sequence of corruption may have proceeded as follows: sasama -»• sd asama (resolving the sandhi), sd asama -* sd asamam (the accusative termination perhaps resulting from attraction to the case of the original citation), sd asamam -»• sd ity asamam (asamam marked off as a restatement of the original word). Tib. is somewhat interpretive - for mtshungs, an inherited translation of asamam, it gives a gloss (mnyam pa) - but there is nothing to suggest it had a second iti. Nevertheless, a problem remains. The usual restatement of the case of the epithet following a bahuvrihi analysis is lacking. (See the following note, on 1.232.) 1.232

tarn caconj. (NMAAfl, 228J) : om. codd. My reasons for adopting this diagnostic emendation are the same as those given in the note on 1.230, above.

1.245

punah ye : punah punah £ • om. Tib. Stemmatic analysis supports the reading punah (ie. being in YE it must be in P), unless Y and £ independently omitted one punah. This seems unlikely given that punah punah would be familiar as the reading of the NS. An error of commission is more likely in this case than one of omission. (See the following note for discussion of the context and of Tib..)

1.246

punar codd. : yang dang yang du {punah punar} Tib.[187.3.2] punar api kimviSistam ity aha / yatltair ityadi / atitair buddhair ya bhasita anagataS ca ye samyaksambuddha hi yasmad yarn bhasisyante punah / pratyutpannas" ca buddha bhagavanto bhasante yam namasamgitim punar apiti // : khyad par gzhan yang ci 'dra ba fdra Tib.P.) zhe na / gsungs pa / 'das pa'i sangs rgyas rnams kyis gsungs

334

Textual Notes. Chapter I

zhes pa la sogs pa ste / 'das pa'i sangs rgyas rnams kyis gang gsungs pa dang / ma 'ongs pa'i yang dag par rdzogs pa'i sangs rgyas rnams kyis gsung bar 'gyur ba gang yin pa dang / yang da Itar byung ba'i sangs rgyas bcom Idan 'das rnams yang dang yang du gsung ba (gsungs pa Tib.D.) gang yin pa'i zhes bya'o Tib.[187.2.8-187.3.2J Rather than emend to punah punar I follow codd. on the grounds that it is unlikely that one punah would be omitted in error given the familiarity of the NS context. It seems that Vilasavajra reads NS 12d's punah punah in such a way that one punah qualifies andgatah samyaksambuddhdh (see the note on 1.245, above) and the other pratyutpannd buddhdh. Tib., however, does not have a punah qualifying andgatah samyaksam­ buddhah, but has both (yang dang yang du) qualifying pratyutpannd buddhdh (da Itar byung ba'i sangs rgyas). Possibly Tib. has been influenced by its inherited translation of the NS: for the opening pratlka (yatitair ityadi) Tib. cites the whole of the opening pdda of NS 12 ('das pa'i sangs rgyas rnams kyis gsungs) and where the Skt. paraphrases the fourth pdda Tib. cites NS.Tib. (yang dang yang du gsung ba gang) and omits any translation of ndmasamgitim. There is a case for emending to pratyutpannds ca buddhd since elsewhere the Skt. incorporates the text of the NS into its paraphrase. However, Tib.'s da Itar byung ba'i sangs rgyas supports codd. (pratyutpannaS ca buddha); it does not have sambuddhd (da Itar byung ba'i rdzogs sangs rgyas = NS.Tib.[Mukherji 8, v.12]). 1.247

mayajaletyadi ... codd. : des 'di skad gsung bar (par Tib.D.} 'gyur te ... {anenaitad uktam bhavati ...} Tib.[187.3.2] Following its translation of .. .ndmasamgitim punah apiti, Tib. translates the passage anenaitad uktam bhavati (line 251) -* pradarsitam bhavatiti (line 255)

335

Textual Notes. Chapter 1

before returning to mdydjdletyddi, ie. where the Skt. has mdydjdletyddi, Tib. des 'di skad gsung bar (par Tib.D.) 'gyur te. In many ways this is a better order than the Skt. since the passage in question relates to NS 12 rather than to NS 12-14, which is how it has to be understood in the Skt.. MS A and B are unavailable here, so the Skt. is that of (3, and not necessarily that of O. On all other occasions (except for at 1.156: see note) when the Skt. and Tib. order the NMAA differently, it is clear that it is Tib. that contains the rearrangement under the influence of an inherited translation of the NS used for the NS citations. Here there is no such influence. However, if the order as preserved in Tib. is correct, the process by which |3 became corrupt is not clear from any of the MSS. One possibility is that an exemplar of f3 initially omitted the passage and then incorrectly reinserted it as a marginal correction, which was subsequently incorporated into (3. Although I have decided to follow the extant Skt., the evidence from Tib. and from the context makes a persuasive case for emendation of the Skt. order. 1.252

sa conj. : om. codd. tryadhvavartibhis tathagatair ya bhasita sa namasamgitih conj. : dus gsum gyi de bzhin gshegs pa rnams kyis mtshan yang dag par brjod pa gsungs pa Tib. [187.3.2-3] Since MSS AB are unavailable codd. represents (3 rather than O. I have emended diagnostically by inserting sd as a necessary correlative of the relative pronoun yd. The evidence from Tib. is neutral since it does not translate the passage with a relative construction (which does not mean that the Skt. did not contain one). Loss of a syllable, sd, through scribal omission is not improbable given that there would be three consecutive syllables ending in -a.

1.252

tulyatam conj. (Sanderson) : tulya CJ : tulyam DEy : [nam mkha'] dang 'dra bar Tib.[187.3.3]

336

Textual Notes. Chapter 1

anenaitad uktam bhavati / tryadhvavartibhis tathagatair ya bhasita namasamgitih / ato 'sya dharmaparyayasyakaSatulyatam samdars~ayati conj. : des 'di skad gsung bar(par Tib.D.) 'gyur te / dus gsum gyi de bzhin gshegs pa rnams kyis mtshan yang dag par brjod pa gsungs pa de'i phyir chos kyi rnam grangs 'di ni nam mkha' dang 'dra bar bstan pa yin te Tib. [187.3.2-3] MSS AB are unavailable here. The emendation to -tulyatam is diagnostic: an abstract noun is required for any sense, given the genitive asya dharmaparydyasya. Tib. (lit. "therefore, as for this Dharma teaching, it has been explained as [being] like space") supports the meaning of the Skt. as emended although it does not translate literally (and so has no genitive or abstract particles.)

1.254

tathagatanam conj. (SekUdTi; AKBh) : om. codd. Tib.[187.3.3] tatha coktam / utpadad va tathagatanam anutpadad va sthitaivaisa dharmanam dharmateti pradarSitam bhavatiti conj. : de skad du yang gsungs pa / de bzhin gshegs pa mams byung yang rung ma byung yang rung chos nyid ni kun tu (du Tib.D.) gnas pa'o zhes bstan par 'gyur ro zhes so // Tib.[187.3.3-4] This passage is cited in a number of other sources. YaSomitra, requoting it from the Abhidharmakos~a, states that the passage comes from a sutra (AKBhVya 452, 20), though I have not been able to identify which one. Against all the testimonia, both codd. and Tib. have just one tathdgatdndm. However, the MS evidence is that of the descendents of (3; A and B, generally the most reliable MSS, are unavailable at this point. If a second tathagatdndm has been omitted, its omission by Tib. could be the result of Tib. working with a corrupt exemplar. Alternatively, since the testimonia differ in the latter part of the passage (sthitaivaisa dharmanam dharmatd as against sthitaiveyam dharmatd} suggesting it circulated in more than one form, Tib. could be following a version of the passage with which it is already familiar rather than translating afresh (as with citation of the NS verses). That it also fails to witness

337

Textual Notes. Chapter I

dharmdndm, as does the version cited in the AKBh and its Vyakhya, perhaps supports this view. If Tib. is following another version, rather than being corrupt, that version has tathdgatdndm only once, yet all the Sanskrit testimonia witness it occurring twice. A further argument for not emending is that the Vairocanabhisambodhisutra, with which Vilasavajra was familiar - he quotes from its first chapter in NMAA 6 - contains what looks like a version of this quotation in chapter two. Wayman (1985, 51, note 3) cites this passage in English translation (presumably from the Tibetan): "Master of the Secret Ones, besides, the mantra character was not made, nor arranged to be made, nor rejoiced in, by any of the Buddhas. Why so? It is like this: There is a continuum (dharmatd) of natures (dharma) whether a Tathagata arises or does not arise. The continuum of dharmas remains immemorially. And this is the mantra-character of mantras...". Yamamoto translates (less intelligibly) the same passage from the Chinese: "Again oh master of mysteries, all the Buddhas did not make the form of mantras, nor did they let others make, nor did they rejoice by making of others. What is the reason? Because many dharmas (things) are like this as the dharma (sic). All the dharmas remain naturally, whether many Tathagatas appear or not, for many mantras remain innately." (Yamamoto 1990, 28). Here we have just one reference to the Tathagatas. Could this, perhaps, have been Vilasavajra's source? Yet Vilasavajra was also familiar with the Abhidharmakos'a - he cites it a number of times in chapter 6 - where tathdgatd­ ndm occurs twice. In the present instance, however, the search for a source may be misplaced. Since the quotation appears to have been popular and well-known it is likely to have been part of a general body of quotations from sutra familiar to the scholar of his time. That Vilasavajra introduces the citation with tathd coktam supports such a view.

338

Textual Notes. Chapter 1

In conclusion, I have emended so as to include a second tathagatdndm. Although it is not certain that Vilasavajra did not know the quotation as it has come down to us in the extant MSS of the NMAA, the fact that all the Skt. testimonia have tathagatanam twice is strong evidence that it generally had such a form. 1.254

bhavatiti conj. (Tib.[187.3.4] ) : bhavati codd.

pradarSitam bhavatiti conj. (Tib.[187.3.4] bstan par 'gyur ro zhes so) I emend to bhavatiti, following Tib., since, given that the quotation is introduced by tathd coktam, pradarsitam bhavati makes sense only as part of the citation. 1.269

adhyesya coir. : adhyesye codd. Newar scribes often confuse ya and ye following a consonant.

339

Textual Notes. Chapter 2

Textual Notes: Chapter 2

2.1

bhagavan co/i/. (NMAA[2, 5-6]) : bhagavam NS.Dav.[50, v.17] bhagavan sambuddha iti gatharthah NMAA[2, 5-6] In inserting the text of NS 17, 1 follow the reading bhagavan given by codd. in its citation of this part of the verse against NS.Dav. (though Davidson neglects to record that Mukherji [1 1, v.17] has bhagavan). See also the notes on 1.189 and 3.1 for the reading bhagavan vs. bhagavam in NS 9 and NS 23 respectively.

2.12

kim krtva / smitam ityadi codd. : ci zhig mdzad ce na / 'jig rten gsum po snang byed cing zhes pa la sogs pa ste {kim krtva / trail okyabhasakaranam ityadi} Tib. [187. 5.4] Tib. does not translate smitam, the first word of NS 1 8, but cites the first line of an inherited translation of the verse, which corresponds to the third pada (trailokyabhasakaranam). Tib. follows the citation of NS 1 8c with that part of the NMAA that comments on it, thereby changing the order of the original. The translation of the remaining material on NS 18 follows the same pattern, ie. it is rearranged so as to follow the pa da order of the translation of the NS, namely, c-d-b-a.

2.34

prasuyate A : prasuyate iti D : itieJ : < ~3 >y : mahar*dya*m ca iti H2 : rab skye ba Tib. [188. 1.5] 4

'nekasahasra conj. : anekasahasram (3 :

anekasahasrya A 4 parsadi / iti em. : parsaditi A : parsadam |3 brahmam punyam prasuyate 'nekasahasraparsadi / iti conj. : tshangs Idan bsod nams rab skye ba'o (pa'o Tib.D.) zhes bshad pa Ita bu'o / stong phrag du ma rnams las te Tib. [188. 1.5] This passage is problematic in number of ways. The placing of iti before anekasahasram parsadam by (3 is puzzling. It is the lectio difficilior; it is supported by Tib.; and brahmam punyam prasuyate scans as a second or fourth anustubh pada (ie. iti may mark the citation of a final quarter verse). Yet

340

Textual Notes. Chapter 2

sahasra, part of the next word, is subsequently glossed implying that 'nekasdhasraparsadi (in this or some other form) is part of the quotation. Tib. also differs from the Skt. as a whole. It does not have anything for the word parsad, and it has a nominal form (rab skye ba) for prasuyate, giving, "There is a production (or 'producer') of great merit" as a translation. The ablative and semi-final particles las te following stong phrag du ma mams {anekasahasra} are also problematic: they could represent a corruption of the locative la ste, in which case there would be some support for A's locative. Since the context clearly implies that sahasra is part of a citation I have preferred A's positioning of iti and I have also diagnostically emended to sahasra- giving a pdda that scans. There remains the question of the case termination for parsad. Although f3's parsadam is possible as an accusative of extension ("throughout the assembly"), I have followed A's parsad7. This decision is linked with one to accept A's reading against [3's for the sentence after the gloss on sahasra (ie. anekasdhasramayydm parsadlty arthah against anekasdhasramayi parsad ity arthah). Here, the locative is more likely to be correct, understood as following the case of the citation although it could be taken as glossing an accusative of extension (parsadam) and so supporting (3. Though this is possible it involves accepting (3 at this one point when other considerations suggest that A has better readings for the passage overall (see the following two notes). Prof. A. Sanderson has pointed out that an emendation to prasuyetdnekawould give a regular sound half-verse. In this case the optative could be taken as being used as a preterite, as it sometimes is in Buddhist Sanskrit (see BHSG § 32.85). 2.36

mayyam conj. : maryyam A : mayl (3 4 parsadlty AJ : parsad ity ytD sahasraSabdena trisahasramahasahasro lokadhatur avagantavyah / anekasahasramayyam parsadlty arthah conj. : stong phrag gi sgra ni

341

Textual Notes. Chapter 2

stong gsum gyi stong chen po'i 'jig rten gyi khams la shes par bya ste / stong phrag du ma'i rang bzhin gyi dkyil 'khor zhes don (conj. : ston Tib.) no (to Tib.D) I Tib.[188.1.5-6] I take anekasdhasramayydm parsadity arthah to be a concluding paraphrase of the latter part of the citation following the equation of sdhasra with trisdhasramahdsdhasra in the preceding gloss. It would be natural to use the same case as the original. Corruption from mayydm to mayi is feasible, involviong loss of a y and anusvdra, followed by a change in vowel from d to /, not a great change in Newari script. A second corruption, from parsadity to parsadity, to give agreement of case termination, might follow (J's -mayi parsadity may represent the intermediate stage). 776."s ... dkyil 'khor zhes don no could indicate a nominative; but since Tib. is not always literal a case particle may have omitted. Alternatively, Tib. could be following (3 - it seems to be following some version of (3 in the original citation (see the preceding note, on 2.34). 2.40-1

tena mahattvad brahma / tasya svaro brahmasvarah conj. : tena mahattvad brahma / tasya suro brahmasvaram A : tena mahattvad brahmasvarah (3 Tib.([188.1.7] de bas na che ba'i phyir tshangs pa'i gsung ro) I follow A (emending to svaro and brahmasvarah) against |3 and Tib.. MS A gives better sense: tena mahattvad brahma reasserts the earlier statement that the term brahma in brahmasvara should be understood to refer to what is great, or what is of great quality (brahmasvara iti brahmaSabdo mahato mahagunasyadhivacanam: lines 32-3); tasya svaro brahmasvarah then gives a tatpurusa analysis of brahmasvarah, which I take as showing that Vilasavajra was aware that the compound could not be taken as a karmadharaya although his explanation of the term suggests the translation 'great sound' to be appropriate. Tib. follows (3 here, as it generally does in this passage on the NS epithet brdhmyd. The reading of (3 is straightforwardly explained as a

342

Textual Notes. Chapter 2

corruption of A, an omission resulting from a saut du meme au meme. It is harder to explain A's reading as involving an addition. In terms of sense, a case might be made for f3 being correct if, in line 32, brahmasabdo is taken as glossing brahmasvara rather than specifying the word brahma. This would give, "[His voice is also described as] sublime (brahmya) because it is endowed with brahma-sound. [The phrase] brahmasound (brahmasvarah > brahmasabdah) is the name for that which is great, that is to say, for that which is of great quality". The difference in interpretation amounts to the difference between saying that Sakyamuni's voice is sublime because it has great qualities and saying that it is sublime because its sound has great qualities. Adopting the former, the interpretation that could support |3's later reading, tena mahattvad brahmasvarah, involves accepting a less likely use of the word sabda (as 'sound') in this context. It also makes less sense of the following citations, which are concerned with the sound-quality of Sakya­ muni's voice. 2.42

aihavaconj. (Tib.f188.1.8] yang na) : tatra codd. The emendation is diagnostic, though supported by Tib. (yang na usually renders atha vd). It is required here since a second explanation of brahma­ svarah is being introduced.

2.49

vajradhara conj. (NMAA[2.54-5J) : vajradharah NS.Dav.[51, v.20] sadhu vajradharetyadi / te tava vajrapaneh sadhu / sriman he vajra­ dhara NMAA[2.54-5] The NMAA clearly reads the vocative, vajradhara. Davidson, though adopting vajradharah, notes the reading vajradhara in a number of the MSS and editions of the NS (NS.Dav. Vira- Ne> **• Mmaev^ The nominative form could result from a desire to make vajradharah agree with sriman (if read as nominative). See note following.

343

Textual Notes. Chapter 2

2.49

sriman conj. (NMAA [2.54-5J) : §rimam NS.Dav.f51, v.20J sadhu vajradharetyadi / te tava vajrapaneh sadhu / Sriman he vajradhara NMAA [2.5 4-5] I follow the NMAA reading, taking the final anusvara of NS.Dav. to be, as with bhagavam (see notes on 1.189, 2.1 and 3.1), an example of an extension of the use of the anusvara in place of medial nasals to stand for a final nasal. It is not clear whether Vilasavajra takes sriman as a vocative, though it is hard not to read it as such. However, it is not explicitly glossed as such, as is the name Vajradhara (srlman he vajradhara ...). Edgerton gives some examples of the nominative singular being used for the vocative (BHSG p.50, § 8.28) but not for '-nf stems.

2.50

yat conj. (NMAA[2, 55] yat yasmat; NS.Tib.) : yas NS.Dav.[51, v.20] sadhu vajradhara Sriman sadhu te vajrapanaye / yat tvam jagaddhitarthaya mahakarunayanvitah // mahartham namasamgitim pavitram aghanaSanim / manjuSrijnanakayasya mattah s"rotum samudyatah // tat sadhu des"ayamy esa aham te guhyakadhipa / (NS 20-22ab) conj. : snying rje che dang Idan gyur pas / / 'gro la phan pa'i don du khyod / / ye shes lus can 'jam dpal gyi / / ming brjod pa ni don che ba / / dag par byed cing sdig sel ba / / nga las mnyan par brtson pa ni / / legs so dpal Idan rdo rje 'chan / / lag na rdo rje khyod legs so / / gsang ba'i bdag po de phyir ngas / / khyod la legs par bstan par bya / / NS.Tib. [Mukherji 13-15] In reading yat tvam in NS 20c against NS.Dav.'s yah tvam, I follow the NMAA, which has yat (glossed as yasmat) at 1.55. Tib.[188.2.3] reads gang dang gang gi phyir, translating yat yasmat. This is grammatically more satisfactory since, whereas yat can be picked up by tat in NS 22a, yah has no correlative. NS.Tib. follows Vilasavajra's analysis of the structure of NS 2022ab, with gyur pas corresponding to yat, and de phyir ('therefore') - in the final line of NS. Tib. above - corresponding to tat. However, the revised NS

344

Textual Notes. Chapter 2

translation of the Derge and Narthang editions of the Kanjur substitutes the connective particles cing and te respectively for the pas ofNS.Tib.'s gyurpas, while keeping de phyir for tat. (For these changes and a change in order of the stanzas to accord more with the Skt., see Mukherji 14, note 1; 15, notes 1& 2.) This evidence suggests that yat tvam... and yas tvam... may have existed as alternatives in the Sanskrit transmission, withes tvam... possibly being later and hence incorporated only into the revised translation of the NS. If it had come to be regarded as the correct reading by this period, it is not surprising that the extant Skt. MSS of the NS attest it. Tib. reads snying rje che dang Idan gyur pas zhes pa la sogs pa ni (Tib.[188.2.2]} for sddhu vajradharetyadi (1.54), ie. citing the opening of NS.Tib. for NS 20-22, rather than translating the Skt. as it stands. However, Tib. keeps to the order of the NMAA in what follows. 2.53

guhyakadhipa conj. (NMAA[2, 56] he guhyakadhipa)

:

guhyakadhipah

NS.Dav.[51, v.22J I follow NMAA against NS.Dav. (Davidson does not note that Mukherji [14, v.22] has guhyakadhipa). The context indicates that the sense here must be vocative. The reading guhyakadhipah could arguably be a survival of an original nominative functioning as a vocative, in the same way as srlmdn in NS 20a (see note to 2.49, above) or it could be a straightforward scribal addition. In any case, whereas the unexpected form srlmdn is preserved, there is no evidence that Vilasavajra had guhyakadhipah before him.

345

Textual Notes. Chapter 3

Textual Notes: Chapter 3

3.1

bhagavan conj. (NMAA[3, 4] atha Sakyamunir bhagavan)

:

bhagavam

NS.Dav.f51, v.23J NS.Dav. has the reading bhagavam in NS 23 (though Davidson fails to note that Mukherji [15, v.23] has bhagavan}, as it does for NS 9 and NS 17. However, all the available MSS read bhagavan in the NMAA's citation of NS 23a. For further discussion of the grounds for accepting bhagavan against bhagavam see the note on 1.189 (NS 9). For the reading bhagavan in NS 17 see the note on 2.1. 3.1

sakalam conj. (NMAA[3, 4J, NS.Dav.[51, v.23])

:

sakala NMAA[3, 4],

NS.Dav.Ne" A sakalam mantrakulam mahad iti NMAA[3, 4] (sakalam A : sakala (3) I have accepted A's sakalam against f3's sakala in the citation of NS 23b in the NMAA. NS.Dav.[51, v.23] in fact reads savakalam, but I take this to be a misprint of sakalam, in which case it supports A. However, Davidson does record two readings of sakala-. I have preferred sakalam mantrakulam mahat against sakalamantrakulam mahat as the reading of the (irregular) quarter verse that Vilasavajra had before him. Though Vilasavajra does not gloss sakalam, his statement that the phrase is a summary statement (grahanakavakyam) is in keeping with sakala qualifying kula rather than mantra (ie. the 'entire great family of mantras' rather than 'the great family of all mantras'). The Tibetan translations of the NS give both possibilities. NS.Tib. [Muhkerji 15, v.23] has gsang sngags rigs chen thams cad dang (sakalam mantrakulam mahat) but the revised Derge and Narthang xylograph editions have gsang sngags mtha' dag rigs chen po (sakalamantrakulam mahat). The existence of the different inter­ pretations in the Tibetan translations of the NS suggests that the reading sakalaprobably appeared quite early. It may have been accepted as the correct reading

346

Textual Notes. Chapters

at the time when the earlier Tibetan translation of the NS was revised, or it may be that the reviser had a corrupt exemplar before him. Thus where it is found in NS MSS it might be a genuine variant reading and not the result of a lost anusvdra, though that may explain its origin. f3's sakala- in the NMAA citation could be due to the influence of scribal familiarity with that NS reading. 3.4

ityadi / tatra conj. (Tib. [188.3.2] ces pa la sogs pa ste / de la) : om. codd. atria Sakyamunir bhagavan ityadi / tatra sakalam mantrakulam mahad iti conj. (Tib. [188.3.2] de nas bcom Idan s"a kya thub ces pa la sogs pa ste / de la gsang sngags rigs chen mams cad dang zhes pa ni) : atha Sakyamunir bhagavan sakalam mantrakulam mahad iti codd. I have emended by inserting ityadi tatra, which is supported by 776.'s ces pa la sogs pa ste / de la. The reading of codd. is not necessarily that of H, since the testimony of MS B is not available. A number of additional considerations support the emendation. Firstly, it makes better sense in relation to the NMAA passage that follows. The phrase grahanakavdkyam etat sdmdnyenopdttam applies now to just sakalam mantrakulam mahad, and not the whole half verse. As an explanation of the half verse grahanakavdkyam etat makes no sense as it cannot stand on its own (ie. it is not a vdkyam). An alternative solution to emending as Tib. indicates would be to see grahanakavdkyam etat as descriptive of both verses that deal with the mantra families (NS 23-4). In this case one would have to emend codd.'s mahad iti to mahad ityadi. However, though one could see these two verses as an introductory or summary statement (grahanakavdkyam), in this context Vilasavajra wants to mark sakalam mantrakulam mahad as introducing the six families that are named immediately afterwards. This makes the reading of Tib. more plausible.

347

Textual Notes. Chapter 3

Secondly, the emendation suggested by Tib. gives a Skt. text that is in stylistic harmony with what precedes and follows. If codd. is correct then the NMAA is citing here a complete half verse of the NS. This is atypical thus far in the NMAA. In fact, there is no citation of more than a quarter verse until chapter five, where NS 32cd is cited. Subsequently, whole NS verses are given. Furthermore, when emended, the opening of this chapter becomes close in style to that of the previous one. Chapter two opens, adhyesananantaram prativacanam aha / atha sakyamunir ityadi (see NMAA 2, 3). Thirdly, Tib. is unlikely to interpolate a passage such as ces pa la sogs pa ste / de la into a citation of the NS. In general Tib. tends to give more of any NS citation in the NMAA than is found in the Skt. MSS (for example, it gives NS 30cd where the O has NS 30c: see the note on 5.56 for this and similar instances). Thus it would be uncharacteristic of Tib. to shorten an opening citation by adding an extra passage. 3.4

sakalamAA/S.Dffv./'j;, v.23] : sakala 0 NS.Dav.Ne-* See note 3.1 above.

3.11

vairocanakulam codd.

: rnam par snang mdzad kyi rigs zhes pa'i don to

(vairocanakulam ity arthah} Tib.[188.3.5] The question of whether to emend to vairocanakulam ity arthah, following Tib., is not one of meaning, which is unaffected, but rather one of style. The discussion of kulatrayam in the NMAA differs from that of the other mantra families in its brevity: kulatrayam iti kayavakcittasvabhdvatvdd vairocanakulam (codd.). This makes good sense as it stands; an additional ity arthah is redundant. I have, therefore, accepted the reading of codd., and take Tib. to represent an addition, influenced perhaps by the instances of ity arthah concluding the passages on the other families.

348

Textual Notes. Chapters

The discussion of kulatrayam also differs from that of the other families in that the others are all analysed as genitive tatpurusas. A concluding ity arthah is more appropriate in such a context. See note on 3.18.

3.17

mahad cony. (Tib.[188.3.8J cten) : om. codd. tasya kulam lokalokakulam mahad vajrakulam ity arthah conj. (Tib.[188.3.7-8] de'i rigs ni 'jig rten snang byed rigs (rin Tib.P.) chen dang zhes pa ste / rdo rje'i rigs so //) : Following Tib., I emend from lokalokakulam to lokalokakulam mahad, thereby giving the full name of the kula. It could be argued that since the sentence analyses lokalokakulam as a tatpurusa, to have mahat is inappropriate. However, the sentence also has the role of identifying the family as the vajrakula, for which the full name lokalokakulam mahad is more appropriate. The other family that contains an uncompounded mahat as part of its name is given its identity in this way: tasya kulam mahosmsakulam mahad ratnakulam ity arthah (line 22).

3.18

ity arthah H2 : iti a : om. Tib. Though little in terms of meaning turns on it, the assesssment of the presence or absence of ity arthah at the conclusions of the passages on the six families, and the agreement or disagreement of the Skt. and Tib. witnesses, is not simple. For the passage on Vairocana's family, which differs in structure from the others, see the note above, on 3.11. Of the remaining five families four are analysed in the same way, as genitive tatpurusa's followed by an identification of the family symbol, using the form, tasya kulam X Y-kulam ity arthah (see lines 10, 14-15, 17-18, and 22), where T is filled by karma, padma, vajra and ratna respectively, and X is the family name given in the NS. The final ity arthah is attested for each of these by all the available Skt. MSS and Tib., with the exception of vajrakulam ity arthah, which is witnessed by just one corrected MS (H2). The fifth family, the mahdmudrakula, is also

349

Textual Notes. Chapter 3

analysed as a genitive tatpurusa, but no family symbol is given afterwards. (Its identification as the family of Bodhicittavajra is given before the concluding tatpurusa analysis.) Here a final ity arthah is also witnessed, again, only by the one same corrected MS, giving the reading tasya kulam mahamudrdkulam ity arthah (ity arthah H2 : iti a : om. Tib.}. How are H2 's readings to be assessed? As representing scribal additions made for the sake of an apparent consistency? I am inclined to think not. H2 can have good readings from outside a, and the readings of MS B, which come from a separate branch of the stemma, are unavailable here. An assessment of H2's readings depends on an understanding of the function of the phrase ity arthah. This could be to make the family symbol identification, in which case vajrakulam ity arthah can be accepted. Such a reading is consistent with the text of the three other family identifications. Alternatively, the scope of ity arthah could be wider, having as its reference the tatpurusa analysis, with the family symbol identification, when present, seen as part of that analysis. The two options can be illustrated by means of punctuation: 1. asya kulam lokalokottarakulam / padmakulam ity arthah; 2. asya kulam lokalokottarakulam padmakulam ity arthah. Each can give sense. For the purposes of editing, I have taken ity arthah to have the wider scope and have accepted asya kulam mahamudrdkulam ity arthah. In both cases I have accepted H2's reading where ity arthah is supported only by H2 (ity arthah H2 : iti a : om. Tib.). To have assumed a narrower scope for ity arthah would have meant accepting Tib. in the second case (line 20) against not only H2, but also a (insofar as a attests iti). Assuming it is correct that ity arthah functions to point to the tatpurusa meaning of the NS names, it accords with the decision to accept codd.'s reading of vairocanakulam rather than Tib.'s {vairocanakulam ity arthah}.

350

Textual Notes. Chapter 3

3.19-20 mahamudrakulam codd. : phyag rgya chen po'i rigs mchog {mahamudrakulam agryam} Tib.[188.4.1] I have followed the reading of codd. here. It could be argued that the same grounds for accepting Tib. apply in this case as in the previous one (see preceding note), namely that the full NS description of the kula should be given as it is for mahosnisakulam mahad (line 22). However, the mahamudrakula is analysed differently from the mahosnisakulam: its identification as Bodhicittavajra's family is made immediately after the opening citation of the NS pada, rather than at the end of the discussion, after the concluding tatpurusa analysis. As a result, there is no need to state the full NS description of the family in the tatpurusa analysis of mahamudrakulam. Also, since agryam is explained earlier (bodhicittavajrakulam ata evagryam: lines 18-19), further reference to it is unnecessary. 3.20

ityarthahH2 : iti a : om. Tib. See note above, on 3.18.

351

Textual Notes. Chapter 4

Textual Notes: Chapter 4 4.4

namasamgitim nama cudamanim codd. Tib.([188.4.5-6] mtshan yang dag par brjod pa zhes pa gtsug gi nor bu) mantramukhacaryacari imam bhagavato manju^rijnanasattvasya sarvatathagatajnanakayasya jnanamurter advayaparamartham namasamgitim namacudamanim NS.Dav. [65, 3-5] : namasamgiri namna cuda­ manim NMAA 12 [D.185r2] The issue here is whether to take nama cudamanim as two words or one. They form part of a passage, cited above, that is incorporated into the NMAA from the NS anusamsa. I have followed Tib. in taking them as two words, making what precedes nama into a title, "this crest-jewel called the Namasamglti". The alternative, of reading them as one word, namacudamanim, is followed by Tib. when part of this same anusamsa passage is incorporated into chapter five (NMAA 5, 280: see note 5.280): "this crest-jewel of name[recitations], the Ndmasamgiti" (mtshan yang dag par brjod pa mtshan gyi gtsug gi nor bu 'di Tib.[194.4.8-194.5.1J). Of the two interpretations the first is preferable since the second involves interpreting namacudamanim as ndmasamgiticuddmanim to give any sense. The decisive evidence for Vilasavajra's understanding is his gloss on the same passage in NMAA 12 (on the anusamsa), which confirms that he took nama to be uncompounded. It should be observed that Davidson (65, note 4) cites Tib. for the chapter 12 passage to confirm Minaev's reading of namacudamanim, but Tib. here mistranslates (mtshan gyi gtsug gi norbu Tib. f2 24.1.6J). In contrast, NS.Tib.D.N. translates imam ... namasamgitim nama cuda­ manim (the passage cited by NMAA) differently from either of the two trans­ lations given by Tib., reading mtshan yang dag par brjod pa gtsug gi nor bu zhes bya ba 'di (imam namasamgitim cudamanim nama} (NS.Tib.D.N. [Mukherji 119, 9-10]), which could be rendered in English as "this Ndmasamgiti,

352

Textual Notes. Chapter 4

called a crest jewel". Though it does take nama as 'called' rather than as 'name', NS.Tib.D.N. does not appear to translate the Skt. as attested by NMAA and NS.Dav.. On the one other occasion where the phrase ndmasamgltim nama cuddmanim occurs in the NS (also in the anusamsa: see NS.Dav. 66, 21), NS.Tib.D.N. does not translate any differently (see Mukherji 123, 4-5) and Tib. again reads namacuddmanim (mtshan gyi gtsug gi nor bu Tib. [224.5.3]). 4.6

kit A Tib.([188.4.6-7] byang chub chen por smon pa'i sems ) : krta (3 Tib. ("the aspiration mind [directed] to great enlightenment") favours A's mahdbodhikrt pranidhdnacittah. This makes sense in the context of the practitioner who is about to take the vows of the Buddha families as a way of expressing a further stage in his resolve to gain full Buddhahood.

4.12

traiyadhvika codd. : traiyadhvika SDPS[146, 9] I follow codd. here since it gives a half verse that scans.

4.13

kuSala conj. (SDPS[146, 10]\ Tib.[188.5.2] dge ba'i [chos ni sdud pa dang]) : kus"alam codd. I emend to kusala, following Tib. and SDPS, giving kusaladharmasamgraham ("the collecting of good dharnas") for the second pdda of the verse, which makes better sense than kusalam dharmasamgraham ("the good collection of dharmas"). SDPS.Tib., however, opts for taking it as two words (dge ba dang ni chos sdud dang SDPS 147, 11).

4.17

vajram Y : vajra A5 SDPS[146, 14] 4 ghantam AJ : ghantamn y : ghanta

eD vajram ghantam ca mudram ca : vajraghantam ca mudram ca : rdo rje dril bu (em. : dril bu'i Tib.) phyag rgya dag Tib.[188.5.3] : rdo rje dril bu phyag rgya yang SDPS.Tib. [147, 15-16] According to an analysis of the readings in relation to the stemma, a should have vajraghantdm but since MS B is unavailable it cannot be assumed that this

353

Textual Notes. Chapter 4

is the reading of O. Also, A and 6 could independently have dropped an anusvara from vajram. Tib. and SDPS.Tib. are ambiguous as to number. The former's dag can indicate both dual and plural, and in the absence of other indications SDPS.Tib.'s yang suggests a plural. In context, vajram ghantdm ca mudram ca makes better sense. It is unlikely that only a vajra-bell and mudra would be mentioned, given the ritual importance of the vajra and ghanta as a pair. If one of the three were to be left out it would most likely be mudra. I have therefore preferred vajram, the reading of v, against vajra-. Loss of the anusvara may have been influenced by scribal familiarity with vajraghantd as an alternative term for the bell used in Buddhist Tantric ritual. SDPS also reads vajraghantdm ca mudram ca, though Skorupski trans­ lates, "In truth I will grasp the vajra, the bell, and the Mudra.," (SDPS 18, 35). 4.54

sarvasattvarthakriyavirahito |3 Tib.([l89.2.1] sems can thams cad kyi don byed pa dang bral bas [na rnal 'byor pas]) : sarvasattvarthakriya'virahto A A's avagraha indicates a reading of sarvasattvdrthakriyd-avirahito. How­ ever, I take this use of the avagraha to be a scribal addition rather than part of the textual transmission. Tib. supports kriya-virahato, as does the context, which suggests a transition to kriyd in the fifth abhisambodhi, away from the absorption injndna accompanied by karund, characterisic of the first four.

4.60

yogi conj. (Tib.[189.2.4] rnal 'byor pas) : yoginaS codd. The subject of the sentence cannot be cittam with yoginah as a genitive singular form since cittam could hardly function as the subject of cintayet. and also vicintya would lack an object. The subject of the sentence is more likely to be 'the Yogin', and the Tibetan translation supports this view, the final instrumentive s of rnal 'byor pas indicating the subject. Yoginas could be a thematised nominative a-stem derived from yogin. This hybrid form, though found, is not at all typical of the pure Sanskrit of the NMAA. It is probable that

354

Textual Notes. Chapter 4

Vilasavajra would have used the classical form yogi since he uses it just a few lines earlier (at line 54: sarvasattvarthakiiyavirahito yogi). If this is correct, yoginas has to be understood as a corruption, perhaps overlaid by scribal correction to a familiar form. 4.61

akaram vicintya cow/. (Tib.[189.2.4] absams te) : vicintya codd. tato yogi cittam mayopamakaram aka£ade£astham prakrtiprabhasvaram tarakakaram vicintya : de nas rnal 'byor pas sems nyid sgyu ma'i rnam pa nam mkha'i phyog su gnas pa rang bzhin gyis 'od gsal ba skar ma'i rnam pa 'gra ba'i a bsams te Tib. This conjectural emendation is suggested by Tib., which indicates that the word akaram may have been lost from the Skt, though Tib. has just a bsams te rather than the more normal yi ge a bsams te (ie. it has a word for a but not for karam}. The loss of an akaram following tarakakaram would be explicable as a simple scribal omission of the saut du meme au meme type. It makes sense that the yogin should start by visualising his consciousness in the form of the letter A, given the fundamental role that Vilasavajra gives this letter (See NMAA 5, 10-11). However, the text still has sense without the emendation: the yogin imagines his consciousness to be naturally radiant, like a star etc.. If the emendation is correct, tad in the following tadudbhutam can refer to akaram ("arising from that [letter A]"); otherwise it would refer back to cittam. A further possibility is that Tib. has inserted some text. If the following text in brackets is removed, Tib. reads as the Skt.: ...rnam pa 'dra baf'i a] bsams te. It is not obvious why Tib. should make an insertion here if indeed this is the case. An omission on the part of a (MS B being unavailable here) is more understandable. Provisionally, I have followed the emendation suggested by Tib..

4.63

atmanam brhad dhumkaram codd. Tib.([189.2.5] bdag nyid yi ge hum chen po bsgom par bya'o)

355

Textual Notes. Chapter 4

Although the use of brhat is unusual, the word not being found ordinarily in prose, it is supported by Tib.. 4.68

pravis"anrlti conj. (Sanderson) : pravis"atah A : pravis"anta(-ta£ J) 5 : praviSantarn y Though the meaning is fairly clear, the passage from tasmdd dhumkdrdd to cintayet is syntactically problematic (lines 63-8). If anekdkdrarasmayah is the subject of the sentence it needs a verb. The diagnostic emendation pravisantiti provides one, the /'// putting the whole of the sentence into indirect speech. Otherwise cintayet would have no object. The alternative is to emend anekdkdrarasmayo, vinihsrtds and sarvatathdgatds from nominative to accusative, -rasmin becoming the object of cintayet, and keep A's pravisatah as an accusative plural masculine present participle qualifying -rasmin. This latter possibility recommends itself as pravisatah is the lectio difficilior. Even though it would require a greater number of individual emendations they could arise from a single error that 'knocks on' through the whole section. A further difficulty is the role of tais ca sarvatathdgataih. It could be interpreted as taking over the role of the subject of the sentence in the instrumental, but though such an interpretation could yield a good meaning there is no corresponding verb in the passive. Assuming -rasmayah stays nomin­ ative, then a diagnostic emendation to te ca sarvatathdgatdh would be appropriate. (If we are dealing with some form of reverberating error, requiring the emendation to -rasmin, then tan ca sarvatathdgatdn is needed). An alternative is to take taih ca sarvatathdgataih as taih ca sarvatathdgataih saha so that -rasmayah stays as the subject ('the rays ... and together with those Tathagatas, having gone to all the Buddha-fields' etc.) and this is the alternative I have adopted. In such visualisations it is usually the rays that are finally reabsorbed into the seed syllable. Later in this chapter there is a parallel

356

Textual Notes. Chapter 4

visualisation that contains a similar syntactic problem where it is clearer that it is the light-rays that are emanated and reabsorbed into the seed syllable (see note on 4.191). 4.81

samyutaih conj. (Sanderson) : samyutam codd.

khacitarn vajraratnais tu ardhacandrena samyutaih conj. : rdo rje rin chen rnams dang ni / / zla gam (kham Tib.P.) rnams kyis yang dag brgyan (mnan Tib.P.) Tib.[189.3.5] If codd. is accepted then samyutam, like khacitam, has to be taken as qualifying kutdgdram and the interpretation of the verse is problematic since it is not clear what "joined with a sickle-moon" might mean in this context. Mandala representations show what is being described is a single decorative motif: a crecsent moon with a half-vajra emerging from between its horns. The motif is found exactly as described, in the corners and at the door joints of the mandala. As illustrative examples, the mural of the Dharmadhatumandala of Manjus"ri (C12th-C14th) in Sum-tsek at Alchi, Ladakh (Pal & Fournier, 1988, S 76) and the line-drawing of the Vajradhatumahamandala (Candra, 1987, fac­ ing p.30) may be mentioned (see also Olschak, 1973, pp. 35-41). For the half verse to give this meaning samyutam needs to be emended to samyutaih, so that vajraratnaih is qualified rather than kutdgdram (I am grateful to Prof. A. Sanderson for suggesting this diagnostic emendation). Since this passage is a citation, there remains the possibility that it was corrupt (or irregular) from the start. The parallel passages cited in the note to the translation suggest that there were a number of standard pddas descriptive of mandalas in circulation. The STTS (63, 15-16) reads, konabhdgesu sarvesu dvdraniryuhasandhisu / khacitam vajraratnais tu sutrayed bdhyamandalam //. In this example the quarter verse ardhacandrena samyutaih could have been replaced by sutrayed bdhyamandalam, or it could be an independent phrase that in some contexts is combined with khacitam vajraratnais tu. If the latter is true,

357

Textual Notes. Chapter 4

the original form may have contained samyutam, which remained uncorrected in the new context. Although I have found no other testimonia, the half-verse in the NMAA exactly describes a particular motif, as has been seen. This suggests that the two quarter-verses may have originally belonged together. I have therefore adopted the emendation to samyutaih. Tib. does not translate this half verse literally but has one word, yang dag brgyan, for both khacitam and samyutaih, giving "... it is adorned with vajragems and half-moons". (See Introduction to the Text, section 5.3, for further discussion of Tib.'s reading.) 4.96-7

tatrasanavikalpah A : tatrasanam vikalpetah |3 : de la gdan rnams brtag pa Tib. f189.4.2] 4 simhasanam A Tib. : simhasanam varam |3 4 gajasanam A Tib. : gajasanam jneyam (3 tatrasanavikalpah / madhye simhasanam purve gajasanam daksine turagasanam paScime mayurasanam uttare garudasanam iti A : tatrasanam vikalpetah madhye simhasanam varam / purve gajasanam jneyam daksine turagasanam / paScime mayurasanam uttare garudasanam / iti |3 : de la gdan rnams brtag pa ni dbus su seng ge'i gdan dang / shar du glang po'i gdan dang / Iho ru rta'i gdan dang / nub tu rma bya'i gdan dang / byang du nam mkha' Iding gi gdan zhes so Tib.[189.4.2-3] I have preferred the readings of A as against those of (3. Tib. supports A in giving a simple list of the thrones of the deities whereas (3 continues in metrical form. The additional words, varam and jneyam in (3 are metrical space fillers adding nothing to the meaning. In itself, this is not sufficient reason for concluding that the original was unmetric, though they could easily have been added to transform an original prose passage into verse. In general, MS A is more often correct against |3 than vice versa, though A could have become corrupt and Tib. be based on a similarly corrupt text. Given that MS B is unavailable here, there is no way of inferring the reading of O. Consideration of A's tatrasanavikalpah tips the balance in favour of A and Tib.. As a genitive

358

Textual Notes. Chapter 4

tatpurusa, asanavikalpah makes sense, with vikalpah understood as meaning 'arrangement' or 'order'. In contrast, |3's tatrdsanam vikalpetah is problematic. It has the extra syllable to make it metric, but needs to be emended to the optative vikalpeta to give sense; though this leaves the singular dsanam, where a plural (asanani) is needed. Also, 7/6.'s gdan rnams brtag pa suggests a compound and gives the plural particle rnams. One might have expected rnam nog for vikalpah - brtag pa is the usual equivalent for vicayah (see MVy 8522) - though perhaps it is too strongly associated with its meaning of (false) 'conception' or'discrimination'. There is also the question of whether there are one or two quotations here. It may be argued that the iti following garuddsanam acts to close the list of different thrones rather than signal the end of a citation. If this is the case, the iti at the end of line 91 (...torandni na kalpayed iti) can be seen as closing the quote but this leaves the question of why the following four lines are in verse. It is unlikely to be Vilasavajra's composition: the only other instance of verse in the NMAA, apart from that in citations, is at the opening of the work. I therefore take the iti after the list of thrones to indicate the closing of a citation. It could be that if this passage on the Buddha-thrones is a citation, it is more likely that the whole is in verse than that it changes into prose after four lines. This argument has some force, but in addition to the problems involved in accepting (3 as against A and Tib. already discussed, taking the passage as verse also means that the citation uncharacteristically finishes on a half verse. Though Tib. also indicates an /// after the list of thrones, it does not after torandni na kalpayet. In this case I have accepted the Skt. against Tib.. Since there is a shift in the subject matter from a discussion of the construction of the mandala to a description of the thrones, I take the iti to indicate the end of the first of two quotations immediately suceeding each other.

359

Textual Notes. Chapter 4

4.102-5 tad eva caturmukham ^unyatadicaturvimoksamukhanam dharmadhator alambanatvena sarvasamadhiprasutihetutvat / Suklavarnam dharmadhatusvabhavatvat / jatamakutopetam nirabharanam ca santacittarvat / bodhyagrimudrayuktam prajnopayatmakatvat / The Tibetan translation of this passage suggests a different punctuation (and hence interpretation) from that above, where I follow the punctuation of MS A (except that A has a double danda after -hetutvdt and nothing following -atmakatvat at the end). None of the MSS support the punctuation implied by Tib.: those remaining ((3) differ from A in punctuating after bodhy­ agrimudrayuktam - generally a single danda - but not before it. However, any support of the MSS in regard of puncuation cannot be conclusive evidence in its favour. Tib. and the punctuation of the Skt. that it implies follow: de nyid zhal bzhi pa ste / stong pa nyid la sogs pa'i rnam par thar pa bzhi'o / / chos kyi dbyings la dmigs pa nyid kyis ting nge 'dzin thams cad rab tu skye ba'i rgyu yin pa'i phyir sku mdog dkar po'o / / chos kyi dbyings kyi rang bzhin nyid kyi phyir ral pa'i dbu rgyan dang Idan pa'o / / zhi ba'i sems nyid kyi phyir sku la rgyan mi mnga' ba'o / / thabs dang shes rab kyi bdag nyid kyi phyir byang chub mchog gi phyag rgya dang Idan pa'o / / Tib.[189.4.4-6] {tad eva caturmukham ^unyatadicaturvimoksamukhanam / dharmadhator alambanatvena sarvasamadhiprasutihetutvat Suklavarnam / dharmadhatusvabhavatvat jatamakutopetam / nirabharanam ca £antacittatvat / bodhy­ agrimudrayuktam prajnopayatmakatvat /} There are substantial problems with Tib. in this passage. In the opening sentence (de nyid -+ thar pa bzhi 'o), which corresponds to tad eva catur­ mukham sunyatddicaturvimoksamukhdndm it has bzhi pa {caturthah}, no translation of -mukhdndm, and concludes with a final particle ( 'o). It seems therefore to read something like tad eva caturtham mukham sunyatddicaturvimoksdh, which yields little sense.

360

Textual Notes. Chapter 4

For the remainder Tib. translates the (remaining) material explaining the significance (visuddhi) of the various attributes of Mahavairocana. With the exception of bodhyagrimudrayuktam the links differ from those preferred in the present edition. Tib.'s reading of the Skt. produces a text that is stylistically odd; a consistent form of word order, with Mahavairocana's attribute being followed in each case by its visuddhi in the ablative, is more probable. This would confirm caturmukham -* -hetutvdt as the first (and most complex) visuddhi. Not only is the word order of the Skt. as read by Tib. unlikely, but the resultant visuddhis make little sense. It makes sense to say that Mahavairocana is white since he is of the nature of the Dharmadhatu (whiteness symbolising the purity of the Dharmadhatu), but it is not clear what the connection might be between whiteness and the generation of samddhis (Tib. has, "he is white in colour since he is the cause of the production of all samadhis..."'}. Again, following Tib., it is not clear how having a braided hair-crest (ral pa'i dbu rgyan dang Idan pa) is to be understood as the result of having the Dharmadhatu as one's nature. On the other hand, braided hair (jatamakutopetam) and lack of adornment (nirabharanam ca) are signs of the simplicity of the renunciate and his attainment of tranquility (santacittatvdf). 4.105

bodhyagri A Tib.([189.4.6] byang chub mchog); NMAA[4, 221-2]

:

bodhyarigi (3 NMAA 4, 221-2: vairocanam sitavarnam bodyagrimudrdyuktam dhkdraparinispannam (bodhyagri AB Tib.([190.5.5] byang chub mchog) : bodhyangi(-ga F) |3) I have accepted the reading of A, supported by Tib., against |3 (MS B is unavailable). This mudra, here said to be that of Mahavairocana, is attributed to Vairocana later in the chapter. For the latter reading, B is available and bodhyagri is attested by both AB suggesting that it is the reading of Q. (see note on 4.222). Yet f3's bodhyahgi is the name of a recognised mudra. In the

361

Textual Notes. Chapter 4

Vajradhatumandala it is the mudra of Vairocana (cf. NispY 44, 7). (3 could be correct if H were corrupt and its reading were emended by a scribe familiar with the names of mudrds. If this hypothesis is correct then Tib. has to be seen as inheriting the error of H. However, bodhyagri is also found as the name of a mudra. J.P. Losty, describing a miniature of Vairocana in a twelfth century Nepalese manuscript of the NS, gives bodhyagri as the name of his mudra (see Zwalf W., ed., 1985, 130: item 174). According to Losty, in the bodhyagri mudra the tips of the thumbs and forefingers are together (but see Introduction note 117). Lacking further evidence, I have preferred the reading suggested by the stemmatic analysis of the readings here and at 4.222 and supported by Tib.. 4.109

tasyaiva conj. : tad eva codd. Emendation to tasyaiva is made on the grammatical grounds that bandhanat requires a genitive.

4.112

yasya tarn conj. (Sanderson) : om. codd. This diagnostic emendation presents a solution to the syntactic problem of mukham needing to be picked up by a participle (eg. 'endowed with') or, as here, within a relative construction.

4.114

samdharayamanam em. : samdharyamanam codd. (sadharyamanam D) This emendation to the causative present participle overcomes the grammatical problem of samdharyamanam being passive. Since both atmanepada and parasmaipada forms are found, a more substantial emendation to samdharayantam is unnecessary.

4.117

sanmantrarajanasamyuktam conj. (NMAA[4, 120-1]} : sadmantrarajanam samyuktam NS.Dav.[51, v.25] (-rajanasamyuktam NS.Dav.Minaev) Cf. NMAA 4, 120-1: kimbhutam sanmantrarajanasamyuktam iti vaksyamanasanmantropetam A : ji Ita bur gyur pa zhe na / gsang sngags

362

Textual Notes. Chapter 4

rgyal po drug Idan zhing zhes pa ni phyis 'byung ba'i sngags drug dang Idan pa la bya'o / Tib.[189.5.3-4] The citation and analysis of this passage in the NMAA suggest that Vilasavajra read the NS here as sanmantrarajanasamyuktam, one word straddling the first two padas of NS 25. Though only MS A attests sanmantra­ rajanasamyuktam in the citation in line 120 (-rajana- A : -rajanam 6H : -rajanah H2 : -*g*ajatam F), leaving the possibility that its reading is the result of a dropped anusvdra, the comment immediately following, vaksyamanasanmantropetdm, which glosses samyuktam with upetdm, shows that Vilasavajra read one word. Davidson prefers to read two words, construing sanmantardjdnam as a bahuvrihi agreeing with gdthdm, and therefore displaying a confusion of gender (NS.Dav. 22, note 64). This analysis leaves the problem of how to understand samyuktam, which Davidson appears to take with giram patch: "[Sakyamuni] pronounced this mystic verse ... joined with the vocal lord" (NS.Dav. 22), but the word order makes this interpretation unlikely and in any event, my concern is to establish Vilasavajra's reading. Given that sanmantrardjdnasamyuktdm makes good sense and does not involve any presumption of gender confusion, it must be the preferable alternative despite the resultant lack of hiatus at the end of the quarter verse. For the translation of kimbhutdm sanmantrardjdnasamyuktdm iti Tib. follows NS.Tib., which also treats sanmantrardjdnasamyuktdm as a single unit (gsang sngags rgyal po drug Idan zhing NS.Tib.fMukherji 17]}. The Derge and Narthang Kanjurs do not change this line, although its position is altered (see Mukherji \ 7, note 4). 4.153

kimvis"istam ity aha codd. Tib.([190.1.6] ci 'dra ba zhig gsungs she na) Despite the support of Tib. this passage is suspect: it is generally used by Vilasavajra to introduce the next part of the text to be glossed.

363

Textual Notes. Chapter 4

4.154

araih codd. I tentatively retain the non-standard form, drah. Turner (1966, 26: s.v. ara-) cites it for 'spoke' in Oria and Bihari instead of the usual arah. It may be a vernacularism surviving within the quotation. Certainly, there is no obvious reason why an original araih should become corrupted to araih. Alternatively, though not mentioned by Edgerton in BHSD, the spelling may be a standard one in such (Buddhist Tantric) contexts. For example, Advayavajra's Saptdksarasddhana reads tatah satsv dresu vdmdvarttena sad devatih pasyet, "Then, on the six spokes, proceeding anticlockwise, he should visualise the six goddesses" (Sadhana 251; SaMa 492, 1-2).

4.156

sadvarnopetam sarasatkam conj. : sadvarnopetam arasatkam codd. : kha dog drug dang Idan pa rtsibs drug pa Tib.[190.1.7-8] Emending to sarasatkam makes the word an adjective qualifying pajndcakram as the unstated subject of the sentence, so giving grammatical sense to the passage. It also removes any need to justify the apparent use of arah (rather than arah) by Vilasavajra when he is commenting rather than citing from agama or sdstra (though see the preceding note). Corruption to sad­ varnopetam arasatkam would not be difficult, involving loss of an anusvdra and a subsequent misreading of sa as ma. (The syllables sa and ma are easily confused in Newari script.)

4.163

namah A : te namah 0 Tib.[190.2.3] Should the form of this and the following mantras be 'om x namah' or l om x te namah '? When they appear later in the sddhana, placed on the spokes of their respective prajndcakras, they are given in the form, i om x namarC (see NMAA 4, 20If). In this latter context, the number of spokes of each prajndcakra should correspond with the number of syllables of the mantra associated with it, since each mantra-syllable occupies a single spoke. The

364

Textual Notes. Chapter 4

correspondences all work out when the mantras are of this form (ie. 'om x namah') and therefore it seems likely that this should be the form of the mantras here. Where necessary I have emended accordingly. The MSS do not give consistent readings. For the third mantra codd. reads om prajndjndnamurtaye namah, and for the first two mantras this same form is also supported by MSS close to the archetype. For each of the last three mantras codd. attests the form 'om x te namah\ Tib. transcribes all six mantras as ...te namah. Corruption to this form is understandable both in Tib. and in the Skt., given that it is a common formula for mantras found in many other contexts. 4.174

mandalacchaya em. : mandalacchaya codd. balarkamandalacchayaprabhamandalam em. : nyi ma 'char ka'i dkyil Itar 'od kyi dkyil 'khor gyis bskor ba Tib.[190.2.6-7] I emend to mandalacchaya since a short 'a' is required for bdldrkamandalacchdya to function as a bahuvrihi within the larger compound. The Tibetan translation ("encircled by a halo like the disc of the newly risen sun") is not literal. It fails to translate -chdya- treating the compound as if it were bdldrkamandalaprabhdmandalam. It also adds the interpretive bskor ba rather than using the more standard possessive particle can.

4.191

sthita iti conj. : sthitam A : sthitam (3 (sthita D) The meaning is clear but this sentence (ending ...cintayet) is syntactically problematic. The apparent subject, uddyolayanto rasmayah, has no verb, exact­ ly mirroring the problem in a passage describing a similar visualisation, using a string of purvakriyds, earlier in this chapter (see note on 4.68, above). As before my emendation is diagnostic and puts the passage into indirect speech so that sahaikibhuydvasthitdh agrees with rasmayah. (Possibly, A's -sthitam is a survival of an original -sthita.)

365

Textual Notes. Chapter 4

An alternative solution would be to emend uddyotayanto rasmayah to uddyotayato rasmm, rasmm becoming the object of cintayet. The problem appears to have been inherited by Tib.. It takes rasmi as the subject ('od kyis Tib. [190. 3. 3J), yet has cintayet as the main verb (gcig tu gyur nas gnas par sgom par bya'o Tib. [190. 3.7]). 4.201

idanim conj. (Tib. [190.4, 3] fani) : i dam idanim ABy : itam idanim C£ : itah idanim E Emendation to idanim is supported by Tib.. The reading of ABv, which by stemmatic analysis must be that of O, yields no sense; it could be explained as a form of dittography with itam and itah stemming from idam as secondary corruptions.

4.207

astasv astau conj. : astasu castau codd. tatas caresv astasv astau bijani om vajratiksnaya nama iti vinyaset conj. : de nas yang rtsibs brgyad la sa bon rnams ni om badzra tiksna ya te na mah zhes bskod de Tib.[190.4.6] Some emendation is required here as the second ca in the unemended sentence is redundant. The emendation to astasv astau is diagnostic, suggested by a parallel difference in readings in the passage on Bodhicittavajra's prajndcakra later in the chapter, at 4.249 (bijaksarany aresu B : bijaksarani caresu |3). Alternatively, castau could be derived from a dittography whose original form was perhaps astau. Tib. could be read as supporting this hypothesis since it gives eight as the number of the spokes (rtsibs brgyad) but gives no number for the bljas (sa bon). However, there is reason to suspect that Tib. has dropped a word. If the NMAA spells out clearly the arrangement of one seed syllable per spoke, it would be convenient for Tib. not to specify that there are eight bijas in the mantra since the addition of te before namah in 776. 's transcription of the mantra results in nine syllables to be placed on the eight spokes.

366

Textual Notes. Chapter 4

4.207

namaAB : namah (3 : te namah Tib.[190.4.6] In this and the following passages giving the mantras to be placed, syllable by syllable, on the spokes of the sixprajndcakras, codd. gives them in the form 'om x namah' rather than l om x te namah' (the exception being om arapacandya namah where four MSS have te namah: see NMAA 4, 248). That this is the correct form is clear since in each case the number of syllables in the mantra corresponds to the number of spokes of the prajndcakra. By contrast Tib. transcribes te namah in every case except Vairocana's prajnacakra, where the mantra is given without the te: om pra dznya dznya na mu rta ye na mah {om prajnajnanamurtaye namah} Tib. [190. 5. 5]. Scribal familiarity with mantras ending with te namah could easily result in corruption of the original and Tib. could have inherited the corruption from its exemplar or could have incorporated it itself. A possible consequence of the addition of the syllable te in 776. is discussed in the preceding note. (See the note on 4.163 for discussion of the variants in the earlier citations of these mantras.)

4.222

bodhyagri AB Tib.([190.5.5] byang chub mchog ) : bodhyahgi(-ga F) This mudrd, said here to be that of Vairocana, is attributed to Mahavairocana earlier in the chapter (see the note on 4.105). As before, I have accepted the reading bodhyagri against bodhyangl. According to stemmic analysis of the present readings, since bodhyagri is in AB, it must be the reading of £1, unless it is an error that has arisen independently, which is unlikely. AB's reading is also supported by Tib. .

4.227

ahkarap : ahkara B : yi ge a {akara} Tib. [1 90.5. 6] : om. Sak. [23,19] vamena cakradharam daksinena prajnakhadgam ahkaraparinatam vibhavya taddhrdi candrarn tadupary ahkaram dhyarva (3 : vamena cakra­ dharam daksinena prajnakhadgadharam ahkaram dhyarva Sak. : g.yon gyis 'khor lo / g.yas kyis shes rab kyi ral gris 'dzin pa bsgoms nas Tib. [190. 5. 7]

367

Textual Notes. Chapter 4

I follow p here since ah is the bija attested elswhere in this prajnacakra (see lines 222 & 227) as that of Vairocana and Prajfiajnana. Tib. appears to be an unreliable witness for the bljdksaras (see the following note). Sakurai omits dhkdraparinatam vibhdvya taddhrdi candram tadupary, perhaps as a result of a saut du meme an meme from dhkdraparinatam to dhkdram. Also for prajndkhadgam he has prajndkhadgadharam, which is unattested by any of the MSS and is not acknowledged by him as an emendation. Although 776.'s 'dzin pa qualifies both 'khor Jo (cakra) and shes rob kyi ral gris (prajnakhadga), shes rab kyi ral gris 'dzin pa may have influenced Sakurai in giving the reading prajndkhadgadharam. Tib. also reorganises the order of the sentence so that dhkdrapannatam comes earlier, making more idiomatic sense in Tibetan (yi ge a las byung ba'i [bcom Idan 'das shes rab ye shes] Tib.[190.5.6J}. That it is not in the position that might be expected in a literal translation may also have influenced Sakurai in making his omission. 4.230

ahkara B : akara (3 Tib.([190.5.8] yi ge sa bon a) MS A is unavailable here. The choice between B and |3 has therefore to be made on grounds of sense since either could equally hold the reading of O. Given that the same bija is used throughout each prajnacakra, I have accepted the reading of B since the stemma indicates that this is the reading of Cl when it occurs in lines 234 & 235. Also, the syllable A is used as the bijdksara for the prajnacakra of Bodhicittavajra and it is unlikely that the same syllable would be used twice. Further indication that Tib. is unreliable is demonstrated by it also giving A throughout as the bijdksara for the prajnacakra of Vairocana instead of AH as attested by the Skt.

4.245

garvataya nisannam conj. : gar*vt*ayamsarnnam B : gardbhayadivarnam C : garbha adivamam E : garbhapavarnam J : garbhapadavamam D : garda-

368

Textual Notes. Chapter 4

bhavarnam H*2* :

garbha< 2 >varna F : rdo rje snyems pas bzhugs pa

Tib.[191.1.8] B's nisarnnam is clearly a corruption of nisannam, which is supported by Tib.'s bzhugs pa and which I take as the correct reading. The short / of nisannam is preserved in CE (e), though in the corrupted syllable di. B's reading may also show how corruption from [ni]sannam to -varnam ((3) progressed: -sannam -*• -sarnnam -*• -varnnam -> -varnam. (The first step exemplifies the generation of a parasitic supercript r before a double consonant, characteristic of many Nepalese MSS: see Introduction to the Text, section 2.5.) Tib.'s snyems pa ('pride') suggests that it read some form of the word garva preceding nisannam. Emendation to -garvanisannam, however, gives one syllable less than found in B and all the MSS of |3 except for H. Assuming DJ's [-garbhjapa- is a corruption of f-garbhjayd- - the result of the similarity of the syllables pa and ya in Newari script - B and 5 (from CJD) have the element -fgarvjayd. This suggests that Q. may have read vajragarvayd nisannam, although this is still unsatisfactory since garvayd, being masculine in gender, cannot be instrumental singular. B's gar*vt*ayd may provide a clue to the correct reading. In the uncertain ligature vta, the letter / is clear, suggesting that the original text may have read -taya, the abstract suffix -ta in the instrumental. If this were the case the original could be [vajrajgarvataya, which would give good sense as an adverb qualifying nisannam. On this account, a syllable has been lost in all the extant MSS, a loss probably also in Q. The alternative - emending to vajragarvanisannam - presupposes corruption via the addition of a syllable rather than the omission of one. 7/6.'s instrumental [snyems pa]s could translate either, though since it has no abstract particle nyid it could be taken as supporting vajragarvanisannam. The balance of prob­ abilities as to which reading is correct seems fairly even. Since a decision is

369

Textual Notes. Chapter 4

required I have emended to -garvatayd: it explains the presence of -aya in the MSS and it gives the required sense more accurately (vajragarvanisannam has to be understood as vajragarvatvanisannam). 4.251

pustakankam B : pustaka(-ke D)|3 : phyag na po ti {pustakapanim} Tib.

[191.2.3] pustakankam khadgapanim vicintya B : pustakakhadgapanim vicintya (3 : phyag na po ti dang ral gri 'dzin pa bsgoms te Tib.[191.2.3] B''s pustakankam is the lectio difficilior and corruption to pustaka- involves loss of just one syllable (nkam) and the stroke for the preceding a. Although Tib. does not support pustakankam as such, its use of phyag na po ti and ral gri 'dzin pa {khadgadharam} suggests two words here rather than |3's one. Also, all the other prajndcakra visualisations have separate words for the emblems held in the two hands of the central deity. This is the only case where there is no ascription of the emblems, one to each hand. 4.253

samharana conj. (Tib.[191.2.3] bsdu ba dang) : spharana B

4

samharanam (3 :

sam-

nilaya conj. (Tib.[191.2.3-4] gnas pa [brtan par gyur pa]) :

nilayam |3 : lilaya B purvavat spharanasainharananilayadrdhibhavadikarn krtva bhavayed iti sarvatra yojaniyam conj. : gong ma bzhin du spro ba dang bsdu ba dang zhugs pa dang gnas pa brtan par gyur pa la sogs pa byas nas bsgom par bya'o zhes kun la sbyar bya'o Tib.[191.2.3-4] The word order here is odd. It is probable that sarvatra yojaniyam ("in each case ... should be added") refers to -nilayadrdhlbhdva- ("making firm the dwelling place") since this is the element that has not been mentioned before in the visualisation instructions for the previous prajndcakras. If this is correct, it would make more sense to separate -nilayadrdhlbhdva- from the instruction to emanate and withdraw light rays (spharanasamharanam krtvd bhdvayet). A

370

Textual Notes. Chapter 4

possible emendation, but one with no support, would be to purvavat spharanasamharanam kftvd bhdvayed iti nilayadrdhibhdvam krtvd sarvatra yojaniyam. Tib. follows the Skt. except for the phrase zhugspa dang ("[ ...and rejentering and [making firm...]"), which has no Skt. counterpart and may be an explanatory addition. It is possible that 7/6.'s exemplar may have already been corrupt.

371

Textual Notes. Chapter 5 Textual Notes: Chapter 5 5.11

ksaratity B : raksati ity 6 : raksatlty y f3's raksati is a metathesis of ksa and ra in ksarati.

5.11-12 avicalitatvat y : avicaritatvat 5 : mi g.yo ba phir ro Tib.[1913.4] The verbal roots Jcal and Jew being near synonyms, avicalitatvat and avicaritatvat are equally possible. Given the confusion of ra and la syllables in Newari MSS, the reading of either 6 or y could represent a corruption of the other. On balance I have preferred y since the Tibetan mi g.yo ba is commonly used for the Skt. acala (Jcal) and MW records avicalita but not avicarita. (In terms of probability, y is more likely to be correct since overall it has the correct reading more often than 5.) 5.13

sattvadyaksarayoh Dc : sattvadyaksara D : sattvadya eyj : om. B : [ye shes] sems dpa' la sogs pa'i yi ge dag gi Tib.[191.3.5]) Though jnanasattvadyaksarayoh is only fully attested by Dc, it is supported by Tib.. However, Tib. misinterprets the dvandva, "of [Manjusri]jnanasattva and the letter A" (jndnasattva;ddyaksarayoh} , taking it to be a karmadharaya, "of the two syllables jndnasattva, etc." {jnanasattvadi:aksarayoh}. This yields no sense.

5.14-15 tad evaikaikam ca conj. (Tib.[191.3.5] de dag nyid re re yang) : tad evaikaikam tac ca (3 : om. B

4

svacittodaya conj. (Tib. rang gi sems las byung ba'i) :

svacintodayah B : svacittadayah |3

4 pratyaveksaniyah BCy : tratyaveks-

amyah £ : om. E tad evaikaikam ca yoginam svacittodayasakalamandaleyadevatasvabhavah samastavipaksavyavrttarupataya pratyaveksaniyah conj. : tad evaikaikam tac ca yoginam svacittadayah sakalamandaleyadevatasvabhavah samastavipaksavyavrttarupataya pratyaveksaniyah {3 : de dag nyid re re yang rnal 'byor pa rang gi sems las byung ba'i dkyil 'khor gyi lha thams cad kyi rang bzhin mi mthun pa'i phyogs ma lus pa

372

Textual Notes. Chapter 5

las bzlog pa'i ngo bo nyid du so sor brtag bar bya ba ni Tib.[191.3.56] The MSS evidence for the text here, which has a number of points of difficulty, is restricted: A is lacking and B omits the opening words, ending a two and a half line omission that started at 5.11. It is not immediately clear whether the passage contains one or two sentences. The latter part, yogindm svacittodayah sakalamdndaleyadevatdsvabhdvah samastavipaksavydvrttarupatayd pratyaveksaniyah, could stand on its own giving, "The arising of consciousness of yogins, since it is by nature free from every obstacle, should be understood to be of the nature of all the mandala deities". However, the meaning is strained and what remains - tad evaikaikam tac ca - is unintelligible. I think it more likely there is just one sentence. Tib. treats the passage as a single unit though not as a complete sentence - it finishes with the emphatic particle «/'; and reads de Itar {tatha} instead of/7 liar for the following vat ha. If there is one sentence, however, the opening words as attested by (3, tad evaikaikam tac ca, yield no sense as they stand; tac ca could start a new sentence but this would leave tad evaikaikam on its own, again with no clear meaning. I have thus diagnosticaly emended to tad evaikaikam ca taking tac as a scribal addition. This gives a sentence that opens with tad eva where tad can be taken to refer to svampdbhiddnam from the preceding sentence. Although Tib. opens with de dag, a dual or plural, which probably refers back to the previous (mis)translation of sattvddyaksarayoh (see the preceding note, on 5.13) it supports the emendation to tad evaikaikam ca insofar as it lacks any term that corresponding to tac. However, it is not clear what Tib. takes as the grammatical subject of the sentence: it could be either de dag nyid re reyang ("And they themselves individually ...") or dkyil 'khor gyi lha thams cad kyi rang bzhin ("The nature of all the mandala deities ...").

373

Textual Notes. Chapter 5

If tad is the subject of the sentence and refers to svarupabhidanam, the passage can be taken, in essence, as making the statement that the nature of the mandala deities is the same of that of ManjiiSrijnanasattva, in which case praty­ aveksaniyah should perhaps be emended to pratyaveksamyam so that it agrees with tad. However, if a participle is used in the place of a finite verb it can be attracted in gender to a nearby substantive predicate, rather than agreeing with the subject. Given this possibility that pratyaveksaniyah may be correct I have not emended. Also, both B and MSS from p have pratyaveksaniyah making it likely to be the reading of Q,. The relationship between the reading svacittodayah and sakalamandaJeyadevatdsvabhdvah is problematic. Since B and (3 read two words it is likely that O does. (B's svacintodayah is clearly a corruption of svacittodayah, the ligatures nt and tt being often confused in Newari script.) However, given that in visualisation sddhanas mandala deities arise in the yogins' consciousness, one would expect svacittodaya to be compounded with sakalamdndaleyadevatdsvabhdvah. Tib. supports svacittodaya (rang gi sems las byung ba) but could be construed either as taking it in apposition to sakalamdndaleyadevatdsvabhdvah or as compounded with it (the genitive particle 'i following byung ba is ambiguous). I have emended to the componded form since otherwise the sense is obscure. The result is a sdpeksasamdsa with the genitive plural yoginam qualifying svacittodaya-, but this can be tolerated. It also means that ft is corrupt, unless B and (3 independently added a visarga. If corruption were present in O resulting in a separation of svacittodaya from sakalamdndaleyadevatdsvabhdvah, the reading pratyaveksaniyah may be a 'knock on' error, an attempt to make the rest of the sentence agree with svacittodayah as subject, in which case pratyaveksamyam would be correct.

374

Textual Notes. Chapter 5

A further difficulty with the passage is the position of ekaikam. If it is adverbial it would be more usually placed before pratyaveksamyah (or pratyaveksaniyam), though if it is taken in the sense of 'one by one', it more properly qualifies sakalamdndaleyadevatd. 5.16

svasamadhmam F Tib. ([191. 3. 6] rang gi ting nge 'dzin rnams) : svam samadhmam cett. F's reading may the result of dropping the anusvara found in the remaining MSS of |3. However, since Tib. supports sva- (gang gi) rather than svam and svasamadhmam gives sense, I have accepted F against cett.

5.19

samskaravimoksa BE : samskarapratipaksabhavena(-prati< ~4 > F) kramad animittapranihitanabhisamskaravimoksacert. -sarnskarapratipaksabhavena kramad animittapranihitanabhisamskara/7ra//paksabhdvena kramad animittdpranihitdnabhisamskdravimoksa. The passage in italics above can be seen to be a dittography of the immediately preceding passage, resulting from a saut du meme au meme from the second -samskdra- back to the first. The dittography is found in all the MSS of (3 with the exception of E.

5.35

sartvavajryadi conj. (Tib. [191.4.7] rdo rje sems ma la sog pa) : vakvadi B : vajradi ye : cakradi C The context and Tib. make it likely that this diagnostic emendation is almost certainly correct, though sems rdo rje ma might be expected for Sattvavajri. Such a rendering, which is characteristic of the Tibetan translation of Skt. compounds that keeps the order of the original, is found in Tib. where the name Sattvavajri appears twelve lines earlier (line 23): sems ma rdo rje ma

5.48

mahaSabdena conj. (Tib. [191. 5.4] ma ha'i sgra ni) : mahacchabdena (3 : < >Sacchaina B

375

Textual Notes. Chapter 5

maha£abdena pujabhidhiyate conj. : ma ha'i sgra ni mchod pa la brjod yin no Tib. [191.5.4] Though the word maha is usually neuter (mahas), it is also found as a masculine a-stem. That it is so treated by the NS is shown by the form maha in mahamahamahdragah and by the masculine mahan in the NMAA analysis that follows the present gloss of maha (mahan maho mahamahah). As a result, I have emended to mahasabdena. f3's mahacchabdena can be understood as a corruption to the more familiar form mahat with subsequent sandhi. 7/6.'s "As for the word maha, it is a name for worship" is a reversion to another familiar form. 5.48-9

mahan maho mahamahah / mahatmako ragas ca maharagah conj. : mahan maho mahamahah B : mahatmako maharagah (3 (maha< ~13 illeg. >tmako maharagah C) : chen po'i bdag nyid kyi 'dod chags dang 'dod chags chen po'o Tib. [191.5.5] MS B contains mahan maho mahamahah but omits mahatmako ragas ca maharagah, whereas the remainder of the MSS (|3) just have mahatmako maharagah, which gives no sense as it stands. Could B be correct and p a corruption of B? Tib. has nothing corresponding to mahan maho mahamahah but is retranslatable as mahatmako ragas ca maharagah, which suggests that (3 lost ragas ca and is thus a corruption of mahatmako ragas ca maharagah. Further support for the presence of this reading in the original comes from the consideration that if it were not present there would be no grammatical gloss of maharagah and such neglect of a NS epithet would be uncharacteristic of Vilasavajra. This leaves the question of B's mahan maho mahamahah. It could be a corruption of mahatmako maharagah, which may have been the reading of ft, already corrupt; it could be a scribal gloss by, or inherited by, B. This, in its turn, would leave the compound mahamaha unanalysed in the original and

376

Textual Notes. Chapter 5

again I think this is unlikely given that it is a refrain carried through the next two and a half verses. Also, Vilasavajra states, on each occasion it re-occurs, that the term mahamahah has been already explained. My conclusion is that both passages were in the original text, their respective omissions coming from an error of the saut du meme au meme type. MS C might be interpreted as providing further support for this thesis. It has approximately thirteen illegible syllables, which appear to have been intentionally scrubbed out. Could C's reading have initially been mahdn maho mahamahah

/

mahdtmako mahdrdgah, later emended to mahdtmako

mahdrdgah to conform with (3? This would require C to be contaminated from outside (3. In the present part of the text there is no evidence for this (see Introduction to the Text, section 4.3). Also, the length of the text deleted in C is some five syllables more than is required. There could well be another (and simpler) explanation for the deletion, such that it was a dittography, discovered during or after the copying of the MS. 5.56

mahamahamahadvesa iti codd. : mchod pa chen po zhe sdang che / / nyon mongs kun gyi dgra che ba zhes pa ste {mahamahamahadvesah sarvakleSamaharipuriti} Tib.[191.5.8] Tib. cites the whole half verse here rather than the first pada only, as attested by codd. The same is true at lines 60, 64 and 68: Tib. cites NS 3lab, 3led and 32ab where codd. cites NS 3la, 31 c and 32a respectively (see Tib.[192.1.1-2], Tib.[192.1.3J, and Tib.[192.L5J). In each case, although MS A is lacking, B is present. The readings must therefore be present in O, and there is no reason to suspect Q. of being corrupt. Also, the citation of the first pada on its own is stylistically appropriate in the context of the commentary; it contains the Name that has to be identified with a deity of the Vajradhatu mandala. The beneficial activity of the Name-as-deity is stated in the second part

377

Textual Notes. Chapter 5

of the half verse and any discussion of this follows after the deity identification. NS 31 d and 32b are thus cited independently (lines 65 & 69) before such discussion, and NS 30d is incorporated into the text (sa ca sarvaklesamaharipuh: line 57). When a complete half verse or verse of the NS is cited (pddas c and d of NS 32 are cited together and thereafter each verse is cited as a whole), then, following the citation, the part of the verse that is then to be commented on is requoted before the ensuing comment. Thus after citing NS 32cd we have tatra mahakama iti ... (line 5.78). In the present instances, if the whole half verses were cited as attested by Tib., one would also expect to a following find tatra... or some equivalent passage. I present the text with the NS half-verses inserted in bold (a design to render it readable by those unfamiliar with the NS), which gives a version close to how the text would look if the half-verse were cited. To incorporate the citation fully one would have to add .. .iti II tatra ... after it. 5.61

dhiyam moho dhimohah By : bio la gti mug pa ni gti mug gi bio ste

Tib. [19 2.1.2] 5.60: mahamahamahamoha iti codd. : me hod pa chen po gti mug che / / gti mug bio ste gti mug sel zhes pa ste Tib.[192.1.1-2] 4 mchod pa chen po gti mug che / gti mug bio ste gti mug sel / NS.Tib.fMukherji 21] The NMAA's analysis of the first three words of NS 31b, mudhadhimohasudanah, is somewhat unexpected. It takes -dhi- with -moha-, rather than with mudha- to give the karmadharaya {mudha:dhtmoha-}. Tib., for the citation of NS 31a in line 60, gives a translation of the whole half verse, following, as elsewhere, an inherited translation (see NS.Tib.); and its management of the rest of the NMAA on NS 3lab is thereby confused, by both the inheritance of gti mug for both mudha and moha and by the translation's implied construal of mudhadhimoha- as {mudhadhr.moha-}.

378

Textual Notes. Chapter 5

Thus, Tib. either mistranslates dhiyam moho dhimohah or is corrupt. For dhimohah it has gti mug gi bio, which retranslates as mudhadhih, rather than dhimohah. It looks as if gti mug gi bio is the result of influence from the inherited translation, gti mug bio ste gti mug sel. (See also note following.) 5.62

mudhas" casau dhimohaS ca mudhadhimohah : gti mug kyang de yin la bio yang de nyid yin pas gti mug blo'o {mudhaS casau dhis ca mudhadhih} Tib. [1 9 2. 1.2-3] Continuing the confusion (see the previous note), Tib., here, gives a karmadharaya analysis of mudhadhi .

5.68

purvavat P : purvaca*ksu*tartham B : [mchodpa chen po'i don ni] gong ma zhin du bshad do Tib. [192. 1.5] Tib. indicates that B may be correct in having something more than |3's purvavat. A few lines earlier Tib. translates mahdmaheti purvavat (line 60) with the shorter, chod pa chen po ni gong ma zhin no. I am unable, however, to suggest an emendation. It is possible that Tib. is augmenting the original to clarify the meaning, which in any case is not materially affected here.

5.77

mahakamo mahasaukhyo mahamodo maharatih // iti // (32cd) Hereafter in this chapter the NMAA cites each NS verse in full.

5.87

nimittagrahad conj. : nimitta*grahat B : nimittanuhgrahat A : nimittanugrahat (3

:

mtshan mar(mar yang Tib. P.) mchog tu mi 'dzin pa'i phyir na

Tib. [192.2. 7] The sense of the passage and Tib. suggest a synonym of aparigrahah as an emendation, to parallel the previous line's samskrtapunyasambharanimittaparigrahat. Tib. does not support emendation with aparigrahah itself: the adverbial particle chog tu stands most often for the Skt. verbal prefixes para- and pra-, and where pan- is found above (in the preceding -nimittaparigrahat), the usual

379

Textual Notes. Chapter 5

adverbial particle yongs su is used. As well as the lack of support from Tib., there is no obvious path by which -dparigrahdt could become corrupted to -dnugrahdt. Taking Tib. into account, one might consider emendation to -pragrahdt but neither MW or BHSD record pragrahah. An alternative that provides a suitable meaning is -nimittdgrahdd. B's -nimittd*grahdt provides slight evidence that this may be correct since the reading covers a line break, nimittd* falling at the end of a line. The illegible syllable could be an end of line mark in which case B would read -nimittdgrahdt. However, Tib.'s chog tu remains to be explained (an interpretive addition on the part of the translator?) as does the extra syllable present in A and (3. In the absence of further evidence I have adopted -dgrahdd as a diagnostic emendation. 5.90

vilokyata B : vinyasyata A : vinas"yata (3 (na$yata F) : mi rtag [ces pa'i don to] Tib. [192. 2. 8] I take Tib. to be an abbreviation of mi rtag pa ces pa 7 don to, "[in other words], it refers to what is impermanent, [that is, the transitory world]". This could be a rendering of |3's vinasyate, "that which perishes". However, one would expect the usual parasmaipada form vinasyati rather than vinasyate. Though vinasyate is supported by Tib. as a further gloss on rupyata iti rupam ("rupam means that which is represented"), B's vilokyate makes better sense. In the present context, that of mahdrupah as an attribute of ManjuSrijiianasattva, rupam must denote 'what is seen', the object of the sense of sight rather than matter or the external world.

5.102

mahad vipulam ABF : mahavipulam HE : mahavipuram C : mahavipula 4

hasamandalam y Tib. [192.3. 7] : mahasamayamandalam A : maha-

mandalam BC : mahamanda E : mandalam

380

Textual Notes. Chapter 5

mahad vipulam hasamandalam saram yasya sa mahavipulamandalah : gang gi bzhad (em. : bshad Tib.) pa'i yangs pa chen po'i dkyil 'khor gyi snying po de ni (di Tib.D.) dkyil 'khor chen po yangs pa yin te. Tib.[192.3.6-7] The Tibetan provides a possible clue to the understanding of the several difficulties of this passage. The term bshad pa, 'to explain', makes no sense but it could well be a corruption of bzhad pa, 'laughter' - I am indebted to Dr P. Harrison for pointing this out - suggesting that y's hasamandalam may be correct. Certainly, the opening identification of the epithet mahavipulamandalah with Vajrahasa makes it likely that Vilasavajra will try to establish, as elsewhere, a connection between the epithet and the name of the corresponding deity. If hasamandalam is right, the reading mahamandalam could be a corruption of hasamandalam (through metathesis of ha and sa followed by confusion of ma and sa) rather than vice versa. An alternative route of corruption may be seen in A's mahasamayamandalam. Starting as hasamandalam, a scribal gloss, maya, could have been incorporated to give hdsamayamandalam ("a mandala consisting of laughter"). The move to reading this as mahasamayamandalam (ie. mahd-samaya-mandalam) through addition of an extra syllable ma- could result from the previous word, vipulam, initially being written with a virdma after the final -m. If this were subsequently lost and an anusvara placed over -la, the text would then have vipulam mahasa-. It may be objected that B, from another branch of the stemma, also witnesses mahd- as against hdsa- and that the present account - where hdsa- is preferred - implies that corruption to mahd- occurred independently more than once (ie. in B, A and £), but perhaps this is possible. The reading hasamandalam also makes sense in the context of trailokydbhdsakarafrdt, the reason given after the gloss and bahuvrihi analysis, with its allusion to the smile (smitam, near in meaning to hdsah) of £akyamuni

381

Textual Notes. Chapter 5

(see NS 18) which illuminates the three worlds: smitam samdarsya lokanam apayatrayasodhanam / trailokyabhdsakaranam caturmardrisasanam //. Yet the passage on the epithet mahavipulamandalah as read is, to my mind, not wholly satisfactory. Continuity of thought lapses between the first part, the initial gloss of mandala as that which receives adornment, and the concluding combination of bahuvrihi analysis and further gloss. The presence of saram alludes perhaps to a gloss of mandalam where saram glosses manda- (see note 170 to the translation for an example of this gloss of mandalam). Why Vilasavajra analyses mandalam as he does here rather than with a more metaphysical alternative is unclear. (Could it be because NS 33 deals with appearance.) Tib.'s gang gi bzhad pa 'i yangs pa chen po 'i dkyil 'khor gyi snying po can be taken as a translation of mahad vipulam hdsamandalam saram yasya (though the positioning of bzhad pa might suggest hdsamahdvipulamandalam as a retranslation). Also, in the phrase yangs pa chen po 'i dkyil 'khor, chen po should probably be taken as qualifying yangs pa suggesting an original mahavipulam rather than mahad vipulam. Yet the latter, attested by ABF, must be the reading of ft. Vilasavajra gives no other indication about how maha- in the NS epithet should be taken; both mahad vipulam ("great, extensive mandala") and mahavipulam ("greatly extensive") are possible analyses, the former being the lectio difficilior given the familiarity of mahavipula- from the NS. Thus in the light of the Skt. MS evidence and Vilasavajra's tendency to take maha- in a non-adverbial sense, I have preferred mahad vipulam. Tib.'s dkyil 'khor chen po yangs pa yin, the translation of the epithet maha­ vipulamandalah, is, as usual, inherited rather than original (see Mukherji 22). It is in itself a little unexpected; surely dkyil 'khor yangs pa chen po would be better? As we find it, vipulamahamandalah ("with an extensive great mandala") seems a more likely exemplar than mahavipulamandalah.

382

Textual Notes. Chapter 5

5.127

bhasah p Tib.([192.5.3] smra ba) : bhasah AB I have accepted P's vajrabhasa (Diamond Speech) against A and B's vajrabhasa (Diamond Light). It is supported by Tib. (rdo rje smra ba). According to stemmatic analysis, Q should have -bhdsa, but sibilant confusion between sa and sa could have resulted in A and B independently writing -bhdsa for -bhdsa. This account is persuasive given that other texts describing the Vajradhatu mandala only have the name vajrabhasa (STTS 11,21; NispYA 45, 14) yet vajrabhasa makes particularly good sense as the name of the mandala deity who has the Names mahdjyotir mahddyutih and Vilasavajra may well have intended it. However, at the end of the commentary on NS 34 Vilasavajra links the deity with 'understanding the intended meanings of all [speech?]' (...samastdbhisandhyarthaparijndnahettttvdt 5.134-5). If this is the correct interpretation of the passage, it makes sense as the function of Vajrabhasa, but is hard to link with Vajrabhasa.

5.132

laksana B Tib.([192.5.5] mtshan nyid) : laksane a

4 vajradharmadi A Tib.

([192.5.5] rdo rje chos la sogs pa) : dharmadi B : om. P 4 svabhavam B : svabhave A : om. P : pratyaveksanajnana^ritavajradharmadisvabhavam om. P tata£ ca vikalpamanahparavrttilaksanapratyaveksanajnana^ritavajradharmadi svabhavam : de rnams kyang rnam par rtog pa'i yid gnas gyur pa'i mtshan nyid so sor kun tu (du Tib.D.) rtog pa'i ye shes la brten pa'i rdo rje chos la sogs pa'i rang bzhin te Tib. [192. 5. 5] I have acccepted -laksana- against -laksane. If the latter were correct, ie. -laksane taken as qualifying pratyaveksandjnane, one would expect a different word order, with vikalpamanahpardvrtti laksane immediately preceding pratyaveksandjndne (see the parallels in lines 104-5 and 156-7 of this chapter). Line 81's -svabhdvddarsa- provides a parallel to the present case (caksurupadivasanaparavrttisvabhavadar^ajnananispatter). Also, Tib. supports -laksana-: it

383

Textual Notes. Chapter 5

has no locative particle after mtshan nyid, and finishes the passage with a semi­ final particle. In any case, the meaning of the passage is unaffected. In accepting -svabhavam I have followed B and Tib. against A. Such a reading follows the structure of parallel sections of the commentary in this chapter: those on NS 33 (lines 104-5), and NS 35 (lines 156-7). It also gives better sense than having a locative (-svabhdve) agreeing with the following pratyaveksandjndne. As with the parallel passages, one has to assume here an unstated subject: "that same non-dual mind". Tib. takes the subject to be plural, "those same" (de rnams kyang), but this will not do if -svabhavam is the singular adjectival predicate. What 776.'s de mams might refer to is unclear. It cannot be 'deities' since the subject is qualified as having the four deities, Vajradharma etc., as its nature. 5.152

mayendrajalam A : mayendrajala (3 : mayendralam B : mig 'phrul can {indrajalikam} Tib.[193.1.8] I have accepted mayendrajalam against Tib.. A reading of indrajdlam nirmanasamuham, suggested by Tib., has nirmanasamuham glossing indra­ jalam. However, given that may a is glossed by nirmana in each of the three preceding quarter verses of NS 35 (see lines 143-4, 147, 150) it is more likely that nirmanasamuham glosses mayendrajalam (i.e. samuham glosses indra­ jalam).

5.153

mayendrajalikah conj. : indrajalikah codd. Tib.([193.1.8] rmg 'phrul can) I have emended to mayendrajalikah against codd. and Tib. on the grounds that if mayendrajalam nirmanasamuham kdyavdkcittavajrdndm ekibhdvah is correct then the next step in the analysis of mahdmayendrajdlikah must be to explain mayendrajalikah, not indrajalikah. Also, mayendrajalikah must be correct if the grammatical analysis that follows takes the whole compound to be a karmadharaya with mahd- as its first element (see note following).

384

Textual Notes. Chapter 5

5.154

mayendrajalikas' conj. : indrajalikah codd. Tib.(fl93.2.1] mig 'phrul can) mahams" casau mayendrajalikaS ca mahamayendrajalikah conj. : sgyu 'phrul chen po yang de nyid yin la mig 'phrul can yang de nyid yin pas sgyu 'phrul chen po mig 'phrul can no {mahamayas' casav indrajalikas ca mahamayendrajalikah} Tib./193.1.8-193.2.1] I have emended to mayendrajalikas since the context indicates that maha­ mayendrajalikah is treated as a karmadharaya {mahd:mdyendrajdlikah}, mahd being analysed in the same way as in the other pddas of this stanza. The alternative of taking the compound as mahdmdyd'Jndrajdlikah would require substantial emendation of the present passage and would be hard to square with the preceding analysis of the compound where mdyendrajdlam is taken as a unit. The emendation also makes sense in the context of the idea of mdydjdlaabhisambodhikramah, key to both the NMAA and NS. Such an interpretation would also be consistent with the treatment of the term may a in the rest of the verse: mahdmdyd is never taken as a unit. The corruption could be due to an error that has reverberated through the passage. Q, could have taken indrajalikah from the preceding (erroneous) tad vettiti / indrajalikah, or vice versa. On the face of it Tib. appears to take the compound as a karmadharaya composed of two bahuvrihis, sgyu 'phrul chen po yang de nyid yin la mig 'phrul can yang de nyid yin, lit. "[He is] a great magician and a conjurer" {mahdmdya.indrajdlikah}. This analysis also does not give any sense in the context of the previous part of the gloss. Tib. could have inherited the initial corruption from the Skt, and translated as well as it could on the assumption that indrajalikah was correct, compounding the error by giving mig 'phrul can for mdyendrajdlam in line 152 (see note above, 5.152). Such a move would be encouraged by Tib.'s use of an inherited translation of NS 35d (sgyu 'phrul chen po mig 'phrul can (NS.Tib.[Mukherji 23]), "Possessing the magic of great illusion") for its citation in the NMAA. This translation, reading the compound

385

Textual Notes. Chapters

as a bahuvrlhi (mahamayd-jndrajalikah), could have influenced Tib. to accord with it, especially if the Skt. were doubtful at any point. Although 776.'s karmadharaya analysis {mahdmdya:indrajdlikah} does not yield any sense, it can be seen as an attempt to render something that does not contradict the received Tibetan translation of the lemma mahdmdyendrajdlikah. (See Introduction to the Text, section 5.3, for further discussion of the effects of Tib. using an existing translation of the NS for the NS citations, rather than translating afresh in accordance with the NMAA.) 5.159

abhisamraksana conj. : abhisamskarana AH : abhisamskaraga B : abhisamskarena 6F : mngon par srung ba[r byed pa dang] Tib.[193.2.3] The emendation is tentative: neither MW or BHSD record abhisamraksana. However, 776.'s srung ba ('protection') makes good sense as an effect produced by the deity Vajraraksa, which makes emendation to some form of the root Jkr unpromising (-abhisamskara- makes no sense). The verbal prefix abhi- is preserved by the Tibetan mngon par, though sam- is not. Since both prefixes are attested by all the MSS and the combination abhisam- is commonly found in Buddhist material, I have retained it. An original raksana could be corrupted to skarana through metathesis of ra and ksa, followed by scribal alteration of ksa to ska to create a recognizable word.

5.175

dhairyavalambanatvat conj. : dhairyavalambinatvat 5 : dhairyava(-ava F)lambmatvat y : vairyalambaratvat B : brtan par sran thub pa'i phyir ro Tib.[193.3.3-4] The two main options here are -avalambanatvat and -alambanatvat. The MSS of |3 attest -avalamb- whereas B has -dlamb-. Errors of omission are on the whole easier to make than those of commission, so -avalamb- is perhaps more likely as the reading of O. For the latter part of the word, B's -lambaratvat is almost certainly a corruption of -lambanatvdt, na and ra being

386

Textual Notes. Chapter 5

commonly confused in Newari script. The final element -natvat is also present in the MSS of (3; and the variants -lambinatvdt and -lamblnatvat could well be corruptions of -lambanatvdt. A. proposal to emend to dhairyavalambitatvat (which is sometimes found but is, strictly speaking, ungrammatical - the correct form would be avalambitadhairyatvdt] is hard to justify given the variants and would be uncharacteristic of Vilasavajra's chaste Sanskrit. In terms of meaning, -avaJambanatvdt is preferrable to -dlambanatvdt. Though almost synonymous, Vilasavajra is familiar with the latter's technical meaning in Buddhist contexts of 'object' (of sense), which makes it less likely that he would use it in the present context (eg. see 5.210, where great compassion is defined as andlambandtmika). Also, for ova -Jlamb MW records the usage 'to enter a state or condition (as mdydm, mdnusyatvam, dhairyam. etc.)'. Tib. does not translate the passage literally ("...since he can endure with firmness...") though the sense does not contradict a reading of dhairydvalambanatvdt. 5.176

kuSala corr. : kuSale (3 Tib.([1933.4] dge ba la) : kuSalasu dharmesu B I have followed (3 since kusale, as well as being supported by Tib., mirrors the following pdpddau (or |3's pdpddy-) giving a contrast between good and evil. This seems more likely than one between good dharmas and evil. The reading of dharmesu in B could be an incorporated scribal gloss, perhaps with the final 'a' of an original kusala retained in kusalasu.

5.176

utsahah conj. (Tib.[193.3.4] spro ba ste) : ty utsahah codd. viryam ku£ala utsahah conj. : brtson 'grus che ba ni dge ba la spro ba ste {maha viryam kuSala utsahah} Tib.[193.3.4] Tib. has no particle corresponding to ty of codd.. Perhaps it should be emended to tv to give viryam kusale tv utsahah. However, this would still not make any sense in the context of a definition of the term virya. The tv, if that is

387

Textual Notes. Chapter 5

what it was, could have been picked up from the definition of laziness that follows. 7/6.'s brtson 'gms che ba {mahaviryam} is puzzling. It could be an influ­ ence of the inherited translation of the NS used for its citations in the NMAA. Hence, for mahdviryapardkrama Hi (NMAA 5, 175), Tib.[193.3.4] has brtson 'grus chen po brtul ba yin ni (= NS.Tib.[Mukherji 24]) where maha is taken as qualifying virya (brtson 'grus chen po). On the other hand the Skt. does not, as is usual, give a karmadharaya analyis (ie. mahams casau viryaparakramaS ca mahaviryaparakramah) for the relation between maha and the rest of the compound. Perhaps then, Tib.'s reading and interpretation of the compound is correct and the Skt. should be emended to mahdvuyam kusala utsdhah. Though I do not think it is possible to reach a firm conclusion here, I am inclined to see che ba as a corruption on the grounds that to take mahaviryaparakramah as "strength of great energy" is unlikely since it would require emendation of the following passage to kausidyam eva na mahaviryam. Such an emendation is supported by neither the Skt. nor Tib. and would also be inconsistent with Vilasavajra's interpretation of the other padas of this and previous verses in this chapter. Against this is the fact that in his commentary on the following verse Vilasavajra does analyse maha as qualifying dhydna and prajnd rather than the final elements of the two compounds of which they are a part. But in these cases the grammatical relationship is clearly spelt out as is, in the instance of prajnd, the reason for it. 5.176

papadau tutsahah B : papady(-ody F)anutsahasya (3 papadau tutsahah kusidatvat kausidyam eva na viryam conj. : sdig pa la sogs pa la spro ba ni dam pa'i phyir na le lo nyid de brtson 'grus ma yin no Tib. [193.3.4]

388

Textual Notes. Chapter 5

I have preferred pdpddau tittsdhah as against emending (3. B's locative pdpddau is supported by Tib. (sdig pa la sogs pa la), though there is no Tibetan for the Skt. tu. It makes good sense without emendation, and contrasts well with the preceding definition of virya, whereas (3's pdpddyanutsdhasya does not yield sense in context. If it is emended to pdpddyutsdhasya, giving viryam kusala utsdhah / pdpddyutsdhasya kusidatvat kausidyam eva na viryam, meaning is obtainable: "vigour is effort for the good; since effort for evil etc. is [being] lazy, [that] is just laziness, not vigour". However, (3's genitive ending -utsdhasya is unnecessary and if the emendation is accepted corruption to pdpddyanutsdhasya has to be accounted for (involving the insertion of a syllable nit and changing yu to yd). On the other hand, given that u / u and t / n are easily confused in Newari script, the extra syllable of (3 is explainable as a corruption of B if dan is corrupted to dya: (pdpddaututsdha- —> pdpddyanutsdha-). The genitive ending -utsdhasya could represent a later scribal emendation attempting to make sense of the passage. Codd.'s kusidatvat is problematic since it appears redundant. However, if it is regarded as corrupt - perhaps a remnant of an incorporated scribal gloss (kusidatvam) of kausidyam - and removed, the eva that follows kausidyam becomes stylistically odd. Also, 7/6.'s dam pa'iphyir na shows it read a word in the ablative preceding kausidyam, though dam pa ('excellent', 'the good') must be corrupt. A reason in the ablative case makes good sense in context, though it is possible that kusidatvat is the remnant of a different reason. Nonetheless, I am not convinced that kusidatvat is without force here (see the translation). Both B and (3 attest kusidatvat so that to emend is to conclude that O is corrupt. Since there is no further evidence for an alternative I have followed the Skt..

389

Textual Notes. Chapter 5

5.177

mulaB : mulanam (3 tasya parakramah sarvalaukikalokottaraku^alamulaparipuranat B : de'i brtul ba ni 'jig rten dang 'jig rten las 'das pa thams cad kyi dge ba'i rtsa ba las yongs su rdzogs par byed pa'o Tib.[193.3.4-5] Both readings are possible. I have accepted that of B on the grounds that Vilasavajra usually gives reasons in the form of one or more compounds rather than a number of separate words. No ambiguity results in preferring -miilaparipuranat that would be better resolved by the alternative, -mulanam paripuranat. Arguably, Tib. also favours B since there is no plural or genitive particle for mula (rtsa ba). Tib., however, is odd in that the verb yongs su rdzogs par byed pa is active and causative, giving a different sense to the Skt.: "As for the strength of that [vigour], it is that which brings (others?) to completion as a result of (las) (his own?) roots of merit...". The past form, byas pa, would be more satisfactory, giving "As for the strength of that [vigour], it is that which is brought to completion from...". (I am indebted to Dr P. Harrison for this suggestion.)

5.184

vyatyayo conj. : vyatyaryo B : vyatyayasau CyC : vyatyasau E : bsnorbar Tib.[193.3.8] This emendation is tentative. Though the meaning is clear, the synonym vyatyaso could serve as an alternative emendation. Cv£'s -dsau could be seen as giving some support to vyatyaso, though the extra syllable ya requires explanation (dittography of [t]yal]. However, B's vyatyaryo strongly supports vyatyayo, requiring only the gemination of a superscript r as corruption. Tib. (bsnor. past tense ofsnorpa, 'to confound, mix, disturb') does not help here.

5.205

krti // iti em. : krtiti AB : krtih(-ti H) 0 : krtih NS.Dav.[52, v.38d] (krti NS.Dav. Vira) : mkhas pa [chen po thabs che ba] zhes pa'o Tib.[193.5.3] mahopayo mahakrti em. : mkhas pa chen po thabs che ba Tib.[193.5.3J

390

Textual Notes. Chapter 5

A number of factors suggest that the reading of the NS citation here should be krti ('learned', 'skilful') rather than krtih ('action', 'activity'). Given that both A & B read krti, stemmatic analysis indicates that it must also be the reading of O. Added support for krti is found in the readings for the two citations of the word in lines 215 and 216 in the commentarial passage that follows (mahakrtiti By : mahakrtir iti A : mahakrti iti 5

4 mahakrti By :

mahakrtih A : mahakrti 5). (There is a difficulty in accounting for y having -krti. Since both A and 6 have a short /, so should a. Nepalese scribes commonly confuse / and /, so it is possible that both A and 6 could miscopy independently.) More importantly, the gloss of mahdpanditah (line 216) indicates that Vilasavajra read mahakrti. Also, Tib.'s mkhas pa chen po ("greatly learned") translates mahakrti. It is noteworthy that Tib. reverses the order of the epithets in its citation of the verse, as if translating mahakrti mahopdyah. Despite the reversed order, Tib. then follows the Skt., commenting on thabs che ba first. This retention of the correct order is uncharacteristic (see Introduction to the Text., section 5). In its citation of the NS, Tib. is again surely using a previously existing NS translation - here it follows NS Tib.'s mkhas pa chen po thabs che ba for NS 38d (see Mukherji 25). The Derge and Narthang Kanjur editions read thabs chen byed pa chen po ste (Mukherji 25, note 1), which must be based on mahopdyo mahakrtih. Perhaps there was some confusion as to the Skt. reading for this pdda, with mahakrtih being accepted as the preferred reading by bLo gros brtan pa whose revised edition of the NS was used for the Kanjur editions. This confusion of readings is reflected in NS MSS where both mahakrti and mahakrtih are recorded (see list of variants above). 5.207

paScan em. : pas"caj |3 : pas"yat A : pa$*c*ata B antam Bf3

391

4

mbantam A : jlb-

Textual Notes. Chapter 5

paScan nibantam em. : mjug ni nge'i (de'i Tib.D.) rkyen dang Idan pa'o Tib. [193.5.4] According to the stemma codicum H should read jibantam since both B & (3 have it. This (or the palaeographically identical jivantam) makes no sense within the present context, which is concerned with the derivation of maitri from maitram. A's nibantam (nip-antam) containing the grammatical term mp is more likely to be correct. Nip can stand on its own, meaning either 'the suffix nip' or 'a word ending in «/>'. A possible translation would be, "Subsequently, it [that is, maitram,] has the suffix mp [added] at the end". I take mbantam as a bahuvrihi qualifying maitram. Tib. is also problematic since mjug ni, though supporting some form of the Skt. word pascal, is best translated by, "As for the end part [of the word]...". Also, de'i in Tib.D. is meaningless. Perhaps it and Tib.P.'s nge'i are corruptions of an originally transcribed nip. However, rkyen dang Idan pa does suggest there is a bahuvrihi here. 5.226

punya conj. : punyajnana a Tib.([l94.1.6] bsod nams dang ye shes) See the note following.

5.227

jnana E : punyajnana cett. Tib.([l94.1.6] bsod nams dang ye shes) Here, I have preferred the reading of E, and in the preceding instance (see the variants above, on 5.226) I have emended. In both cases the choice is determind

on grounds of sense. Separate reasons are supplied for the

attribution to ManjuSri of the epithets maharddhikah and mahesdkhyah so it is more likely that they would be justified respectively in terms of the attainment of punyasambhara and jnanasambhara rather than in terms of the attainment of both sambhdras in either case. The fact that the all the available MSS (with the exception of B, which omits this whole passage) as well as Tib., read punyajndnasambhdra on both occasions may be due to an early scribal error,

392

Textual Notes. Chapter 5

punyajnanasambhara being a stock phrase of Mahayana Buddhism. MS E's (correct) reading jhanasambhara cannot be taken as the reading of £1. It could be an independent scribal correction, though as E has punyajnanasambha.ro- in the explanation of the first epithet this is unlikely. It could be due to extraneous contamination or, more likely, to erroneous copying si punyajnanasambhara. The attribution of the sambharas in the first instance to Laukikas, £ravakas and Pratyekabuddhas and in the second to Tathagatas also suggests that Vilasavajra had in mind a division between the two sambharas. Jnanasambhara, being the wisdom characteristic of the Mahayana (ie. dharmanairdtmya), is not attainable by the first group who are, however, able to attain the lower sambhara - that of punya - as a result of their ethical behaviour. Being associated with means (upayd) rather than wisdom (prajna) punyasambhara is also productive of the magical power of the first epithet (maharddhikah). 5.241

vinaya By : vineya A5 Tib.({194.2.6] gdul ba) sattvavinayavaSat By : sems can gdul ba'i phyir Tib.[194.2.6] Applying stemmatic analysis to the readings, a must have vineya since it is attested by A5. If a has vineya, y's vinaya must be the result of scribal error (even if it is the correct reading) rather than faithful transmission from H. Such error is quite possible: ne is often hard to distinguish from na in Newari script and is copied as na\ misreading na as ne is less likely. The choice of readings is thus between vinaya (B) and vineya (a). (If Y's vinaya is accepted as the result of accurate textual transmission the readings call into question the structure of the stemma codicum.) Tib. supports vineya (vinaya would be 'dul ba). Being an adjective, vineya makes no sense here (ie. in sattvavineyavasat). Tib. gives the meaning "on account of beings that are to be disciplined", which would require vineyasattvavasat (taking -vasat as 'on account of). The sense therefore requires a noun

393

Textual Notes. Chapter 5

and dictates that the correct reading, following B, must be vinaya, giving "[Manjusri displays ferocity] through the desire for the dicipline of living beings". Corruption to vineya may have occurred in part of the transmission before the translation into Tibetan, which managed as best it could with a corrupt exemplar. If B's reading is itself an erroneous copy of vineya, occuring in the same way but independently of y's corruption then Q is in error and the point of corruption has to be placed further back in the transmission. 5.243

samkara iti conj. : sankara iti A : nankara B : gankarah F : gahvarah H5 :

om. Tib.[194.2.6] mahabhayo mahabhairavarupadhari s"amkara iti conj. : 'jigs chen ni 'jigs byed chen po'i gzugs 'dzin pas na 'jigs chen no Tib.[194.2.6] I have retained A's iti though since it is absent in B{3 stemmatic analysis suggests that Cl did not have it. It makes sense as punctuating a name and it is not out of the question that it could have been lost independently in B|3. Tib., with its redundant repetition of mahdbhayah (jigs chen ni... jigs chen no) in place ofsamkarah is surely corrupt. The ease of corruption from sankara to H6's gahvarah through confusion of syllables in an unfamiliar word is particularly apparent in the Newari script of MS F where the syllables sa and ga are very similar, as are the ligatures nka and hva. 5.249

uttamah B : uttama (3 : uttamah Sresthah A

4

tasam uttamah Sresthah om.

Tib.[194.3.1] According to the stemma, H should have plain uttamah unless both B and (3 have dropped sresthah independently, which is unlikely. The reading of sresthah could be a scribal gloss taken from the passage closely following that comments on NS 41b, the next pada (mahamantrottamah): tasyottamah srestha

394

Textual Notes. Chapter 5

ity arthah (see line 252). Some doubt over the reading remains as one would usually expect the NMAA to comment on uttamah. The corruption of the following ata eva by 6 to agra eva may be an attempt to provide a gloss. 5.258

sambharopaya codd.

:

'khor ba'i gnas thabs dang {samsaropaya} Tib.

[194.3.5] atha va trividhanuttaryena sambharopayagocaradharasvabhavo mahayanasya sarngrhita iti : yang na 'khor ba'i gnas thabs dang spyod yul dang rten gyi rang bzhin te / bla na med pa rnam pa gsum gyis theg pa chen po bsdus pa yin zhes so / Tib. [194.3.5] Orthographically, codd.'s sambharopaya and the reading indicated by Tib., samsaropaya, differ in one syllable only. Tib. could have preserved the correct word, the Skt. MSS being corrupt. Alternatively, codd. could be correct and Tib. could have either misread the Skt. or worked from an already corrupt MS. As each reading gives some sense, the choice of which to accept depends on context for support. The preceding sentence in the NMAA defined mahayana in ultimate terms (paramdrthasatyam) as non-objectifying Awareness (nirvikalpam jndnam). The present passage is concerned with the conventional meaning of the word (samvrtisatyam). Assuming that sambhdra is the correct reading, one can take the subject of the sentence to be sambhdropdyagocarddhdrasvabhdvah, a karmadharaya compound {sambhdropdyagocara:ddhdrasvabhdvah} whose first member is a dvandva {sambhdra;updya;gocara}. Here the threefold supremacy of the Mahayana is of sambhdra, updya and gocara: "Alternatively, the nature of the Great Way as a support can be summarized as a threefold supremacy: of provisions, method, and of scope". Here, the distinction is between Mahayana as non-objectifying Awareness and as a support. If, on the other hand, the correct reading is samsdra, the subject of the sentence is the same but it has to be interpreted as a locative tatpurusa

395

Textual Notes. Chapter 5

{samsarakipayagocaradharasvabhavah} whose second element is a karmadharaya {updyagocarddhdra:svabhdvah} that in turn contains a dvandva as the first element {updya;gocara;ddhdra}- There could initially have been two words, samsara updyagocarddhdrasvabhdvah, in which case the threefold supremacy is of updya, gocara and ddhdra: "Alternatively, the nature of the Great Way with respect to [the context of] Samsara can be summarized as a threefold supremacy: of method, of scope, and as a support". Yet further analyses are possible. For example, for sambhdropdyagocarddhdrasvabhdvah the trividhdnuttarya could be sambhdropdya, gocara, and ddhdra. I have chosen to follow the reading of the Skt. MSS, although without the testimony of another Sastric reference to the Mahayana's 'threefold supremacy' it is not possible to settle which of these alternatives is correct. Further complication ensues with the summary statement at the end of the discussion, tad etat samdsena trividham mahdydnam / hetugocarasvabhdvaparidipandnham veditavyam (NMAA 5, 265-6), which I take as saying that the word mahdydna, "should be understood as having meanings which explain [its] hetu, gocara, and svabhdva". However, here Tib. reads, rgyu dang yul dang rten gyi rang bzhin rnams {hetugocaradharasvabhava-} (Tib. f194.4.12f). The position of the plural particle rnams indicates that Tib. took hetu­ gocaradharasvabhava as a dvandva {hetu\gocara;ddhdrasvabhdva} meaning "...should be understood as having meanings which explain [its] hetu, gocara, and ddhdrasvabhdva" . If this is correct and not a dittography from trividhdnuttaryena sambhdropdyagocarddhdrasvabhdvah then it suggests that ddhdrasvabhdvah could be a unit in this formula of the 'threefold supremacy' of the Mahayana. Interpretation of this passage depends on whether it is understood to refer to the trividhdnuttarya or to the total number of explanations given to the term mahdydna, in other words whether or not its definition as

396

Textual Notes. Chapter 5

non-objectifying Awareness is to be included as the first of the trividham mahaydnam. Tib. involves another difficulty in that it may have had the correct Skt. (namely, hetugocarddhdrasvabhdvaparidipandrtham veditavyam) but misinter­ preted it. An alternative analysis would be to take svabhdva as the second part of a karmadharaya {hetugocarddhdra:svabhdva}. Lacking a clear way out of this tangle of problems I have again stayed with the Skt. and assumed that the reference of trividham mahaydnam is to the three explanations given, which includes non-objectifying Awareness as the first.

5.262

vidha conj. (Tib.[194.3.7-8] mam pa \bdun]) : vidhena a : vidhina B atha va saptavidhamahartvat conj. : yang na chen po rnam pa bdun gyi phyirte Tib.[194.3.7-8] saptavidhena gives no sense as it has no noun or pronoun, implied or stated, with which to agree. The emendation is diagnostic, though supported by Tib. 's chen po rnam pa bdun.

5.266

hetugocarasvabhava codd. : rgyu dang yul dang rten gyi rang bzhin rnams {hetugocaradharasvabhava} Tib. [194.4.1-2] See the latter part of the note above, on 5.258.

5.276

bhavanti conj. : bhurva ABy : om. 5 : bsgyur te Tib.[194.4.6J This is a diagnostic emendation to create grammatical sense. Without it there is one sentence starting with tena ca mdndaleyd and ending avartayisyati but with two subjects, mdndaleydh and kulaputro vd kuladuhitd. 776.'s bsgyur is the perfect and future stem of the verb sgyur ba, ' to transform'.

5.280

namasamgitim nama cudamanim : mtshan yang dag par brjod pa mtshan gyi gtsug gi norbu 'di {namasamgitim T\amacudamamm}Tib. [194.4.8-194.5.1] This is part of a passage incorporated into the NMAA from the anusamsd of the NS. This same passage is also incorporated into the fourth chapter of the

397

Textual Notes. Chapter 5

NMAA (see the note above, on 4.4). As then, I have taken nama cuddmanim as two words rather than one. Tib. translates them as one word (mtshan gyi gtsug gi nor bu), although in the previous instance it took them as two. I do not think that there is any contextual difference to warrant taking them as compounded here. 5.287

purusa a : purusah B

4

pumgavah conj. (NMAA 12; NS.Dav.[68, 5]) :

purigalah ABvE : pumgalah £C : [yang dag par 'dzin pa'i skyes bu] gang zag

{pudgalah} Tib.[194.5.3] paramarthanamasamgitisamdharakah purusapumgava(em. : -pumgala D : pungala B) iti pradhana ity arthah NMAA 12 [D.192r3; B.85r2] : don dam pa'i mtshan yang dag par brjod pa 'di yang dag par 'dzin pa'i skyes bu gang zag de ni gtso bo zhes pa'i don to Tib.[225.4.1] I have emended to -pumgavah, following NS.Dav. and Vilasavajra's later gloss on this anusamsa passage in chapter 12, against codd., which can be taken as reading pudgala - pungala and pumgala being common spellings for pudgala in Nepalese MSS (see BDSD s.v. pudgala; Regamey, 1954, note 4). The gloss in chapter 12 is clear in showing that Vilasavajra read -pumgavah and read it in its sense of 'best', 'most eminent'. Emendation to purusapumgavah produces good sense ("best of men", lit. "bull among men") whereas it is hard to see what sense purusapudgalah could make, purusa and pudgala being more or less synonymous. Corruption from pumgavah to pungalah or pumgalah is easy. Since pungavah is an alternative spelling for pumgavah (the anusvara being written with an homorganic nasal), the change involves the misreading of just one syllable (va as Id). What is puzzling, however, is that Tib. for the both the NMAA citation and its gloss in NMAA 12, as well as NS.Tib.D.N. (see Mukherji 126, 8) read gang zag (pudgalah). It is especially so given that purusapumgavah was sufficiently common to have its own entry in the Mahavyutpatti (MVy 7360: skyes bu skyes

398

Textual Notes. Chapter 5

mchog or skyes mchog). Comparison of Tib. and NS.Tib.D.N. for the wider passage reveals a number of differences showing that Tib. has, as elsewhere with anusamsa passages, not followed the NS translation of the

Kanjur

editions but translated afresh or followed another (earlier) translation as is the case for the NS verse citations. If this is so, both Tib. and NS.Tib.D.N. independently read or thought they read pudgala. How likely is it that such an error could occur independently in both MS traditions? Could pudgala have somehow been seen as a gloss of purusal As I have indicated in the Introduction to the Text (section 5) the textual transmission of both Tib. and the Tibetan translation of the NS is neither straightforward nor clear. Error could occur at a number of points increasing the chance of independent error. A more satisfactory solution is that Tib. is following a pre-existing translation of the anusamsa here that contained gang zag, and when it was revised to give NS.Tib.D.N. the reading gang zag was retained. A few lines before the present passage Vilasavajra, again citing the anusamsa, describes the practitioner - the Bodhisattva - who recites the NS as being either male or female: "a son of [good] family or a daughter of [good] family" (kulaputro va kuladuhita va NMAA 5, 281). On the face of it this conflicts with the present description of the sadhaka in masculine terms as the "best of men" (purusapumgavah). If there is a problem here it is one that Vilasavajra has inherited from the anusamsa, and one that he may well have been unaware of. 5.288

sarvabuddhagunan codd. Tib. (sangs rgyas thams cad kyi yon tan rnams) : buddhagunan NS.Dav.[68, 6]\NS.Tib.D.N.([Mukherji 128, 9-10] sangs rgyas kyi yon tan rnams) I retain sarva-, following codd. and Tib. against NS.Dav. and NS.Tib.D.N.. The difference in readings suggests that Tib.'s translation of this anusamsa

399

Textual Notes. Chapter 5

passage was independent of NS.Tib.D.N. See also the notes on 5.287 and 5.291. 5.290

analpa AB Tib.([194.5.4] mang po) (Tib. [194. 5.4] bskal pa)

:

:

anatya p

+

kalpan conj.

kalyan co«W.(kalyan E NS.Dav.f68, 7]}

4

aparinirvana AB : aparimana |3 analpakalpan aparinirvanadharma conj. : analpakalyanaparinirvanadharmah NS.Dav. : bskal pa mang por yongs su mya ngan las mi 'da' ba'i chos su 'gyur la Tib. [194. 5. 4] Except for the ending -dharma (see the following note, on 5.290), codd. might appear to support NS.Dav. for this passage. But the sense of such a reading is problematic, especially analpa as a qualification of parinirvana. Tib.'s bskal pa mang por {analpakalpan} clarifies the Skt.; kalyan (codd.) should be emended to kalpan (the ligatures lya and Ipa are easily confused in Newari script). There is thus a word break after -kalpan and the following compound starts aparinirvdnaAll the NMAA MSS except one have a dental n, supporting -kalpan, rather than a retroflex, as in -kalydna- (though the common confusion between dental and retroflex nasals in Newari MSS reduces the weight of this testimony). Also, NS.Tib.D.N. agrees with Tib. in reading bskal pa mang por, providing further evidence that NS.Dav. is corrupt (see Mukherji 126, 1 1-12). 5.290

dharma codd. (dharma J) : dharmah NS.Dav. [68, 1] : [yongs su mya ngan las mi 'da' ba'i] chos su 'gyur Tib.[194.5.4] I have retained -dharma, taking it as the nominative singular masculine of the older -an stem form dharman. It is the lectio difficilior; the reading of NS.Dav. could be a result of scribal regularisation. Tib., though it supports the Skt. in terms of sense ("becomes ones who has the mark of not entering final Nirvana"), is of no help in the choice of termination.

400

Textual Notes. Chapter 5

5.291

desako codd. Tib.([194.5.5] [chos] ston par byed la) :

desako 'dhisthito

NS.Dav.f68, 8] NS.Tib.fMukherji 126, 12-14] sarvasattvanam anuttaradharmadesako daSadiksaddharmadundubhir dharmaraja iti conj. : sems can thams cad la bla na med pa'i chos ston par byed la / phyogs bcur dam pa'i chos kyi mga sgra sgrogs pa'i chos kyi rgyal por 'gyur ro Tib.[194.5.5] : sarvasattvanam anuttaradharmade£ako 'dhisthito daSadiksaddharmadundubhir dharmaraja NS.Dav. [68, 7-8] : sems can thams cad kyi bla na med pa'i chos ston pa po dang / phyogs bcur chos kyi rnga bo che lhag par gnas pa'i chos kyi rgyal por 'gyur ro NS.Tib.D.N.[Mukherji 126, 12-14] In retaining -desako rather than emending to -desako 'dhisthito, I follow codd. and Tib. against NS.Dav. and NS.Tib.D.N.. As well as the omission of 'dhisthito (lhag par gnas pa) by both codd. and Tib., comparison of Tib. and NS.Tib.D.N. reveals other differences (eg. bla na med pa'i chos ston par byed la Tib. : bla na med pa'i chos ston pa po dang NS.Tib.D.N.) that lend support to the hypothesis that they are either independent translations or that NS.Tib.D.N. is a revision of one used by Tib.. See the notes above, on 5.287 and 5.288. 5.297

sesani namamantraksarapadani codd. I take sesani as being used adjectivally here.

401

AP PEND ICES

Appendix I

Appendix I: Works and Authors Cited in the Namamantrarthdvahkini Work / Author Advayasamatavijaya Abhidharma Abhi dha rma sutr a AmoghapaSatantra _ Astasahasri [Prajnaparamita] Asadharanaguhyalamkaravyuhatantra Aryadeva Aryamanjus'ritantra Kamalacarya Krodhavijaya Krodhendutilaka Guhyakosa Guhyamanitilaka Guhyasamaja Guhyalamkara Guhyendutilaka Cakrasamvara

Yogatantra (Toh 492?)

Folio no. (MS B) 38v3, 39v6 ...... 5v2,27v8,30r7,44r I,68r7, 83 v6 ______...._____ 52r5 ._„„....._„„ 38r8 46r3 (Aryasta-), 67v5_______ 39r5-6; see Guhyalamkara.

Kriyatantra (T5h 543)

33 v3, 50v6 40 v5, 40 v6 _

Tantra Class j Yogottaratantra (Toh 452-453)

• Kriyatantra (Toh 686)

",

Kriyatantra (Toh 604)

J Yogatantra (Toh 493) Yogatantra (Toh 492?) : Yogottaratantra (T6h477) (= SBSYDJS? Yoganiruttara tantra Toh 366-7)

Candrakirti Jajmbhalastotra of Candragomin_ | Jnanapada_ _ j Tattvasamgraha \ Yogatantra (Toh 479) TriSatika [Prajnaparamita] Trailokyavijaya Dignaga

Yogatantra (Toh 482)

Dharrnakirti [Dhatupatha]

Nagarjuna

Paramadya [Paiiini] Bhutadamaratantra Madhyamakavatara of Candra­ kirti Madhyantavibhaga

Yogatantra (Toh 487-8) Knyatantra (Toh 747)

403

37v8,39r3 _ 39r4___ 38v8, 39r8 39yl See Samaja. 39r9; see Asadharanaguhyalamkaravyuhatantra 22vl, 38v2, 38v4 38r7, 39r2; see Sarpvara, Satpraj nanayasamvaratantra. 61v9; see Madhyamakavatara. 58r9 3v3-7, 34v4, 39vl, 40v2 (Satsahasrika Tattva-), 69v4 32r6, 70r9 (Arya-) 39v6, 40v4

6r8 [PPA 1] 25v8 JPPA_1_9J 49v8 9rlt [DhP 1.76] 17v2t [DhP 10.49] 19r3 [DhP 25.16] 26r7, 27r3, 31vl, 31v6 [MMK XXV: 20]; see Mulamadhyamaka, Ratnavali. 6r4, 39r8 23 v7 [AA 6.3.14] ""'""""" 3 8vZ, 23 v5 [I 16] 24rl |V1 2] 23r6 [V 5] 25 v4 [III 15-16ab] 25v9 [ffl 1-2]

Appendix I

[Madhyantavibhagabhasya]

25v5 26r2 28v3 f

Mahabhasya [of PatanjaliJ Mahavidyottama tantra Mahavairocanabhisambodhi_ tantra Mahasamayatantra __ Mayajalatantra Mulamadhyamaka[-karika]

Kriyatantra (Toh 746) Caryatantra (Toh 494) Yoganiruttaratantra (Toh 390) 1^ Yogottaratantra (Toh 466)

Ratnavall

Vajraghanoccaya Vaj rapanyabhi sekat ant ra Vajrapatalatantra Vaj rabha ira vatantra VajraSekharatantra Vasubandhu

._

jJCaryatantraJToh 494) \ Caryatantra (Toh 499) ' Yogottaratantra (Toh 468, 470) Yogatantra (Toh 490) Yogatantra. (Tohj48Q)

Vairocanabhisambodhitantra

Caryatantra (Toh 494)

Satprajnanayasamvaratantra

(= SBSYDJS? Yoganiruttara­ tantra, Toh 366-7)

SodaSasahasrika [Prajnaparamitaj Samvara (= SBSYDJS)

Samaja [Guhya-] Saryatantrasamuccaya Subahukaratantra Suvisiddhikaratantra Unattributed Citations ity aha tatha caha

[Ill 15-16ab] [ffl 1-2] [115] [III 7cd-8a] [III lOcd-llabJ

74r5____.,.,...,....,.....„.....,...._ 38r5__.____..„.„.„ __ 22v3-7]l], 39v4-5 38vl, 39v2-3 37v9, 39rl, 39v3, 39v4_ __ [XXTV 8-9]; 28v7 see Nagarjuna.__ 31vT [128] 32r9^ [122] [I 25J; see Nagarjuna. 32v2 38rl,_3_9v9__.__ 3 8 v6 37v5, 39v7 ____ __ __..... ,.. 12r5-12v2, 38r4-_5, 38v7, 40r3 [AK: III 95-6] 27v8 32v5-6 [Trims 20] [Trims 2 lab] 32v9 [Trim^ 2 led]; 33r2 see Madhyantavibhagabhasya. See Mahavairocanabhisambodhitantra. 40v3; see Samvara, Cakrasamvara. 38v4

Yoganiruttaratantra (Toh 366-7) 3r9, 28rl, 37v6 (-tantra), 39r8, 72v5 (-tantra), 75r9 (-tantra), 78r3; see Cakrasamvara, prajnanayasamvaratantra. Yogottaratantra (Toh 442) 38r9, 39r8, 55v5 (Maha^risamaja), 70rl ____ 3 9r5__ 40r9 [Kriyatantra (Toh 805) 40vl Kriyatantra (Toh 807) llr4-9 5r2, 5r4 [BuBhuS?], 9vl-9, 10r6, 12v5, 23rl,33r3 6v9, 8r7, 27v2, 29r5, 34rl, 36r5, 50r3 24r3 8r4 8r3 10r8, 31r7

tathd coktam

tad eiad ukiam bhavati yathoklam

Incorporations

404

Appendix I

Notes to Preceding Table 1. The work or author is given as in the NMAA. Where necessary, corruptions in the reading of MS B have been emended with reference to the other MSS. 2. Entries enclosed in square brackets, eg. '[Panini]', indicate unattributed citations. Other material in square brackets is additional and explanatory. 3. Versions of titles that differ slightly from that given in the first column are given in parentheses after individual folio references. 4. For citations of Tantras, the Tantra class and Tohoku catalogue number have been given where possible. (The Tantra class is that of the Kanjur classification, 'Yogottara' and 'Yoganiruttara' standing respectively for the Father and Mother divisions of the Anuttara Tantras. (See also Introduction note 11.) 5. An obelus following a text reference indicates an incorporated citation. 6. The folio reference 39r8 for the Guhyakosa, Paramadya, Samvara and Samaja is notional since MS B has a lacuna at the beginning of this line. In MS A these citations are found on fol. 57vl.

Works and Authors cited in NMAA chapters 1-5 Text Reference

Work / Author Abhidharma Tattvasamgraha Dignaga [Dhatupatha] Paramadya[tantra] Vajramandalamkaratantra Samvara[tantra] ity aha tatha caha

tatha coktam yathd coktam yathoktam Incorporated

1.183 1.88-9 [PPA 1] 1.118-9 [DhP 1.76] 3.14t [DhP 10.49] 5.100* [DhP 25. 16] 5.169 1.210-11 4.130 1.81 4.77 1.158, 1.162,4.8,4.42,4.154 1.253, 2.44 2.36 2.33 4.48-9

Tantras cited in the NMAA listed according to their Tantra Class. Yoga­ Unidentified niruttara Mahasamaya Krodhendutilaka

Kriyatantra

Caryatantra

Yogatantra

Amoghapas'a

Vajrapanyabhiseka Vajrapatala Vairocanabhisambodhi

Asadharanaguhy- Advayasamataalarnkaravyuha vijaya Guhyamanitilaka Guhyendutilaka Samvara • Mayajala Tattvasamgraha

Aryamanjus'ri Krodhavijaya

Yogottara

Vajrabhairava

Bhutadamara

Trailokyavijaya

Mahavidyottama Subahukara

[Guhya-]samaja Paramadya _ Vajramandalamkara Vajratekhara

Susiddhikara

405

GuhyakoSa _ _ Vajraghanoccaya Sarvatantrasamuccaya Cakrasamvara Satprajnanayasamvara

Appendix I

Citations of the Samvaraftantra]. Four citations attributed to the SamvaraftantraJ have been identified by Mr K. Tanaka as coming from the Sarvabuddhasamdyogaddkimjdlasamvaratantra (see Introduction note 20) as follows. 1 1.B.3r9; Text 1.81: ata evoktam srisamvare / sarvdkdsdvakdse srivajrasattvas tathdgata iti // This is found in the Tibetan translation of SBSYDJS, chapter 1: nam kha' kun gyi skabs na dpal / rdo rje sems dpa' de bzhin gshegs / 2. B.37v6-7, A.55r4: tathd coktam snsamvaratantre / atyanladustaraudresu saumyatd nopayujyate / prajnopdyamayam manyum cakruh sarvatathdgatd iti // (B reads prajnopdyamayam guhyam cakuh ...) This is found in SBSYDJS chapter 5: shin tu gditg par gtum po la I zhi has phan par mi 'gyiir te / shes rob thabs kyi ngo bo yi I khro bar de bzhin gshegs kun mdzad I 3. A.57vl-2 (lacuna at B.39r8): mahdsukha iti srisamvare I talra mahdsukha iti yat tathdgatam andsravam sukham tat mahdsukha ity ucyate / tatraivoktam sukham sam iti vikhydtam sarvabauddham mahdsukham / iti // This is found in SBSYDJS chapter 1: sham zhes bya ba bde bar bshad / sangs rgyas kun gyi bde chen vin / 4. B.78r3: tathd coktam srisamvare / kvacic ca paratirthydndm saha dharmena nigraha iti // This is found in SBSYDJS chapter 4: la lar pha rol mu stegs mams / chos dang 'thun par tshar good do /

J I am grateful to Prof. A. Sanderson for passing on these findings of Mr Tanaka, who had examined the citations at his request. Unfortunately Mr Tanaka gave no folio references to the Tibetan translation of the SBSYDJS.

406

{SamuddeSa} Pindikramatippani 18

15 16

14

{Kramasatka} { Samayac ittaprakaSa } { Mayaj al aghudrst anta svaSrayakrama} { Mayaj alasahaj ajnanaloka }

11 12 13

9 10

Rim pa drug pa 12 Dam tshig gsal bkra 13 sGyu 'phrul dpe chung rang gnas kyi rim pa 14 sGyu 'phrul dra ba lhan cig skyes pa'i ye shes snang ba 15 rGyun bshags 17 bsDus pa'i rim pa'i dka' "grel

Aryanamasamgititika Namamantra- Thags pa mtshan yang dag par brjod pa'i rgya cher 'grel pa rthavalokini3 mtshan gsang sngags kyi don du rnam par Ita ba { Samanyasamayasamgraha} sPyi'i dam tshig mdor bsdus pa4 'Chol ba'i sgrub thabs {Ucchusmasadhana|£_ rGyud kyi rgyal po chen po dpal { Sriguhyagarbhamahatantrarajagsang ba'i snying po'i 'grel pa tika} 7 Thugs kyi thigs pa Cittabindu9 Thugs kyi thigs paf'i man ngag] Cittabindiipades'a 1 1

5

6 7 86

&iy amantakamui amantrallhavaj raprabheda2

4

Thig le chen po'i rim pa gSang ba 'dus pa'i rgyud kyi gleng gzhi bla ma'i man ngag gi bshad pa dPal gsang ba 'dus pa than cig skyes pa'i sgrub thabs dPal gshin rje gshed po'i rtsa ba'i sngags don rdo rje rab tu 'byed

§rl satiaj aguhy asamaj asadhana

Mahatilakakrama 1 { Guhyasamaj atantranidanagurupades"abhasya}

Tibetan Title

3

1 "2"""""""

Sanskrit Title

Appendix II: Works Attributed to Vildsavajra (sGegpa'i rdo rje}

Not given Dana^ila rMa Rin chen mchog8 Not given rMa Rin chen mchog

UpayaSrimitra & Chos kyi bzang po Smrtijnanaklrti

Buddhas"rijfiana

Not given Not given

Translator

Jftanakumara Not given dGe bsnyen Saudita & bLo Idan shes rab

rGeg pa rdo rje 16 sGeg rdor sGeg pa rdo rje

sGeg pa'i rdo rje 10 Sangs rgyas gsang ba (Buddhaguhya), Vimalamitra, dGeg pa rdo rje sGeg pa rdo rje Not given sGeg pa rdo rje Not given Not given sGeg pa rdo rje

sGeg pa'i rdo rje sGeg pa rdo rje sGeg pa rdo rje

sGeg pa'i rdo rje

sGeg pa'i rdo rje

rDo rje sgeg pa

sGeg pa'i rdo rje sGeg pa'i rdo rje

Author

LXXV 26 LXXV 29 LXXV 33 12 6 5

1 24

2'/2

LXXVTT4 LXXXI 4

LXXVI 9

LXXV 8 LXXV 23 1 11

79

LXXII 18 LXXII 38 LXXV 3

LVIII 2

XLIII 90

XLI 10

XXI 66 XLI7

Cordier No. (Peking)

l'/2

5'/2

104

83

1

No. of Folios (Peking) 4 9

Toh 1836

—— - --•-•

Toh 3723 Toh 3742

Toh 2533

Toh 2014

Toh 1913

Tohoku Cat. No. (Derge) Toh 1290 Toh 1910 "

a

1

o

'^This is Cordier's restoration of Pindikramatribadu, though dka' 'grel usually translates panjikd rather tha tippam. Toh 1836 gives Samuccayakramapanjikatripada as the Skt. title. This work is placed in the Guhvasamdjatantra section of the Tanjur in the Derge edition.

17The title is from the sub-title. The dkar chag has the title bShags pa gtso bor ston pa rgyun bshags, retranslated by Cordier as Mukhyadesaiiasasanasamuddesa.

"•'sGeg pa'i rdo rje is only given as the author in the dkar chag volume. 1 'The Sanskrit is restored by Cordier from Cittabindupatresa. ' 2The title is taken from the second colophon, the opening title being missing. The colophon and chapter colophons have dPal gsang ba 'i snying po de kho na nyid nges pa las bsdus pa {Sriguhyagarbhatattvanirnayasamgraha}. title is from the second colophon, the opening title being missing. title is from the colophon, the opening title being missing. title is from the sub-title, the colophon and the dkar chag, the opening title being missing. 16The dkar chag gives sGeg gtor (= sGeg rdor?) as the name of the author.

^Cittabindu is Cordier's restoration ofdtabidu. The colophon has sGyu 'phrul thugs kyi thigs pa {Maydcittabindu} as the title and the dkar chag has sGyu 'phrul dra ba thugs toi thigs pa {Mdydjdlacittahindu}.

3The Index volume (dkar chag) describes the NMAA as the 'grel pa bar ma, the 'Mediumf-sized] Commentary' on the NS. The 'Great Commentary' ( 'grel chen) is the Tikd of Manjus"rlklrti (Toh 2534) and the 'Small Commentary' (grel chung) is Manjusrimitra's Vrtti (Toh 2532). ''This title is taken from the colphon and dkar chag. It is missing from the beginning of the text. 5The Peking Tanjur gives Ucitsmd for the Skt. title; T5h 3742 has Ucusmasadhana. This sadhana is of the wrathful form of Jambhala. The dkar chag gives the full name rMugs 'dzin 'chol ba'i [Ucchusmajambhala]. ^This and the following seven works do not appear to be contained in the Derge Tanjur. With the exception of the last two (nos. 15 and 16) they are concerned with the two Mahayogatantras, the Mayajala and the Guhyagarbha. 7Cordier notes that a 'mutilated' Skt. title, Mahdrdjaiantrasriguyagarbhandmatika, is given in the colophon of the Peking edition It seems likely that this is a reconstruction from the Tibetan rather than the transcription of an actual Skt. title. On the title Guhyagarbhatantra see Introduction note 42. °Thc translator is not mentioned in the Peking Tanjur but is stated to be rMa Rin chen mchog in a rNying ma edition (Sanje Dorje (ed.) Commentaries on the Gnhvagarbha Tantra and Other Rare Nyingmapa Texts from the Library of Dudjom Rinpoche, New Delhi, 1974, 222.5). See also Introduction note 43.

2Toh 2014 gives the Skt. title as Sri Yamantakavajraprabhedandmarmllamantrdrtha.

!This work is placed in the Hevajratantra section of the Tanjur. According to Dudjom Rinpoche (1991, 464) it is concerned with the perfection stage of Hevajra. If this work is indeed by Vilasavajra it raises a doubt over giving him a date of the latter part of the eighth century.

Notes

Appendix II (cont.)

Appendix II

Appendix II (cont.) The preceding table lists works attributed to Vilasavajra (as sGeg pa'i rdo rje) in the Tibetan Tanjur. They are given in their order of appearance in the Peking edition and, unless otherwise noted, the information is taken from the same edition, as recorded by Cordier (Catalogue du Ponds Tibetain de la Bibliotheque Nationale, Parts II & III. 1909, 1915). Where a transcribed Sanskrit title is lacking, Cordier's reconstructions are given in brackets. Of these works only the NMAA is extant in Sanskrit. There is one further work surviving in Sanskrit and attributed to (a) Vilasavajra that I am aware of one, the short Samksiptavajravdrdhtsddhana, which is found in the Guhyasamayasamgraha, a collection of 47 works devoted to Vajrayogim and Vajravarahi, where it is the 26th text. Whether the author of this and the NMAA are one and the same remains a matter for further investigation (see also Introduction note 56). (For a palm-leaf MS of the Guhyasamayasamgraha see Wintemitz& Keith No. 1455, p.264-5 - the title is given incorrectly as Sddhanamdld Tantra. In this MS the Samksiptavajravarahlsadhana is one folio in length (fol. 101rl-101v6). See also Dhlh I, pp. 7-41, for a summary of the contents of the Guhya­ samayasamgraha.) Cordier (XLVI 5) lists one further work attributed to sGeg pa'i rdo rje, the dNgos po gsal ba grub pa {vyaktabhdvasiddhi} , but though it is itemised in the dkar chag volume it does not exist in the Tanjur itself. The dkar chag also states that sGeg pa'i rdo rje was the disciple of the authoress of the previous work (XLVI 4), Bhagavati Laksmi (= Laksmimkara). This places sGeg pa'i rdo rje in a guruparampard given in the dkar chag linking the authors of Cordier XLVI 2-8 in such a way that (except for XLVI 8) the author of each work is said to be the disciple of the author of the previous one. The following teacher-pupil lineage results: Anangavajra —»• Indrabhuti —» Laksmimkara —* sGeg pa'i rdo rje -»• Darika pa -» Sahajayogini Cinta. Domblheruka, the author of XLVI 8, is said to be the disciple of sGeg pa'i rdo rje. B. Bhattacharyya uses this lineage as a basis for a chronology of Tantric authors (1928, p. xl ff.; 1932, 62ff). Attributions of authorship and statements such as these concerning teacher-pupil relationships should be treated with considerable caution if attested in the dkar chag alone. They are likely in many cases to be the addition of the compilers of the individual Tanjur edition, perhaps made in order to fill a lacuna in the text itself. If this lineage were accurate it would suggest a date of the ninth century for Vilasavajra, since Anangavajra, Indrabhuti and Laksmimkara are more likely to be ninth century than eighth (see Introduction note 10).

409

Appendix III

APPENDIX III: NMAA Colophons Chapter Colophons Chapter Colophon

3_ 4 5

A

7 8 9 10 11 12 13 14

aryanamasamgit itikayam namamantrarthavalokinyam adhyesanaya prathamo 'dhikarah aryanamasamgititikayam namamantrarthavalokinyam prativacanadhikaro dvitlyah Osatkulavalokanadhikaras trtiyah °mayajalabhisarribodhikramasyadhikara^ carurthah °bodhicittavajrasya vajradhaturnahamaridaladhikarah pancamah Osuvisuddhadharrnadhatujnanastutyadhikarah sasthah Oadar^ajnanasvabhavastuter adhflcarah saptamarj Opratyaveksanajnanamukhena stutyadhikaro 'starnah °samatajrlanasvabhavena^utyadhi.]karp navamah__ _ _____ °ikrtyanusthanajnanastutyadhikaro daSamah _ __ _ °upasamharastutyadhikara ekadaSamah parisamaptah °anusamsadhikaro dvadaSamati Oprajnacakramantravinyasadhikarastrayoda^amari "upasamharadhikaras' caturdas'amah samaptah_____

Final Colophon The final colophon is preserved only by MS A, 1 which is in poor condition, with damage to the left margin and elsewhere loss of the surface layer of palm leaf. Tib. has the final colophon, though in an expanded and slightly reordered form. It is clear that at one point Tib. mistranslates the Skt., but it nonetheless enables emendation of the Sanskrit. A transcription of MS A, the text of MS A as emended, and the colophon as found in Tib. follow. The translators* colophon in Tib. is also included. (Note. An asterisk indicates illegible aksaras resulting from damage to the manuscript. Semi-legible aksaras are enclosed in parentheses. Square brackets indicate text missing as a result of loss of the MS margin.) granthapramanam asyah [ ](dha)sahasradvitayam / krtir acaryavilasavajrasya ratnadvipanivasi(nah) / s"rimadagra(b )*(bha)gineyasya pra****sVampabMdhanasya A(114v5-6) granthapramanam asyah sardhasahasradvitayam / krtir acaryavilasavajrasya ratnadvipanivasinah Srimadagrabodhibhagineyasya prasiddhasYIvisVarupabhidhanasya conj. 'phags pa mtshan yang dag par brjod pa'i rgya cher 'grel pa mtshan gsang sngags kyi don du rnam par Ita ba zhes bya ba rgya gar gyi mkhan po sgeg pa'i rdo rje nyi ma can gyi rin po che gling na bzhugs pa dpal Idan byang chub mchog gi skal ba dang Idan pa gang gi mtshan dpal sna tshogs gzugs can du grags pa des mdzad pa rdzogs so // // rgya gar gyi mkhan po smr ti dznya na kl rtis bsgyur ste bshad nas gtan la phab pa'o / / phyis rgya gar gyi mkhan po phag na rdo rje dang / lo tsa ba (lo tsa ba om. Tib.P. ) klog skya (kya Tib.P.) shes rab brtsegs kyis bcos nas gtan la phab pa'o / / 'di la gzhung tshad ni shlo ka nyis stong Inga brgya'o // Tib. [226.1.6-226.2.1]

* Of the remaining manuscripts, BH2 have the name of the author and the rest have nothing.

410

Appendix III

Notes 1 1. The Skt. opens with a statement of the length of the commentary. Tib. places this after the translators' colophon. 776.'s nyis stong Inga brgya ("2,500") supports emendation of the Skt. to sdrdhasdhasradvitayarn.

2. Emendation of srimadagra(b >- to srimadagrabodhi- is supported by Tib.'s dpal Idan byang chub mchog. Although skal ba dang Idan pa (= bhdgin) confirms the semi-legible syllable bhd, to give the genitive singular srimadagrabodhibhdgineyasya, the following yang gi [mtshan] suggests that Tib. read -yasya as a separate word (leaving dpal Idan byang chub mchog gi skal ba dang Idan pa as a name). 3. Tib. 's dpal sna tshogs gzugs can suggests emendation of pra****svarupdbhi- to pra**srivisvarupdbhi- or pra*srimadvisvarupdbhi-. However, grags pa suggests that prasiddha is likely to have been the first word of the compound (MVy 2622 has grags pa, without a prefix of rab in, for prasiddha) and therefore I emend to prasiddhasrivisvarupdbhidhdnasya. 4. Tib.'s rin po che gling na bzhugs pa confirms ratnadvlpanivdsinah, but the Skt. has nothing corresponding to the appositional nyi ma can {suryavat]. 5. Tib. describes Vilasavajra as apandita (mkhan po); the Skt. has dcdrya. The emended Skt. and the parallel section of the Tibetan may be translated as follows: [Here ends] the work of Acarya Vilasavajra, inhabitant of Ratnadvipa, son of the sister of £ri Agrabodhi [and] whose name is [also] known as Sri Visvarupa. [The work] called "An Explanation of the Meaning of the Name-Mantras", a commentary on the Aryandmasamgiti - composed by the Indian pandita Vilasavajra, dweller on the jewel island, 'Endowed with sun', *Srimadagrabodhibhagin, whose name is [also] celebrated as Sri Visvarupa - is complete.

] For discussion of the colophon relative to the life and date of Vilasavajra, see Introduction, p.lSff. On Smrtijnanakirti and kLog skya shes rab brtsegs see Introduction to the Text, section 5.1.

411

Appendix IV Appendix IV: NS Commentaries Surviving in Sanskrit and in Tibetan Translation. Commentaries in Tibetan Translation The following table lists NS commentaries in the Derge Tanjur. As well as commentaries on the NS as a whole it contains Kalacakrapada's commentary on the anusamsd section and Advayavajra's on the upasamhara. The information is taken from the TShoku catalogue. (Names in square brackets are my additions.) Author

Title

[Advayagupta] gNyis su med pa'i sbas pa [Advayavajra] gNyis su med pa' rdo rje

AryamanjusYinarnasamgltivrtti

1. Aryamanjus"rinainasarngititika sarabhisamaya nama ^Namasa/rigityupasarnharavitarka nama Anupamaraksita Manjus"rinamasamgiti amrtabindupradipalokavrttinama _ Ayadhutipa Aiyamafiju^rinamasarngitivrtti^ _ __ __ [Kalacakrapada] Dus zhabs pa Aryamafiju^riiianiasarngityanusamsavrtti _ [KumarakirtiJ gZhonjm ,grags pa Aryamanju^rinamasarngrtyupade^a \Ttti dGa' rab rdo rje ___ __ AiyanT^rljusrinamasamgityarthai^ rGyal po Padma dkar po Aryamafijusnnamasarngititika. vimalaprabha [CandrakirtiJ Zlaba gragspa Aryamanjusrinamasamgltivrtti Candragomin Aryamafijusrinamasamgiti nama mahatlka Dombiheruka Thags pa 'jam dpal gyi mtshan yang dag par brjodpa'i 'grel pa [Narendrakirti] Mi'i dbang po Aryamanju^rinamasamgitivyakhyana gragspa dBu ma la dga' ba Aryamanj u^rinamasamgitiparlj ika [Manjusriklrti] 'Jam dpal grags pa Aryamafij u^nnamasarngi ti tlka [Mafijusrimitra] 'Jam dpal bshes Namasamgltivrtti gnyen Ramakaragupta Aryamarljusrinamasarngitipanjika samgraha nama __ __ _ ___ _ Arnrtakanika nama aryanamasamgititipparii [Ravi^rijnana] Nyi ma'i dpal ye shes Namasamgltivrtti namarthaprakas"akaranadipa Vimalamitra nama NamasamgititikanamamantrarthayalokinI [Vilasavajra] sGeg pa'i rdo rje 1. Aryamanj u^rinamasamgitiguhyavadvidhiSmrtijnanakirti vrtti jnanadipa nama 2. 'Jam dpal mtshan brjod kyi bzhad 'bum 1. GaganamalasuparisuddhadhannadhatuAuthor Unknown jnanagarbha nama 2. mTshan yang dag par brjod pa'i 'grel pa tshul gsum gsal ba 3. £rimafijuvajrodayamandalavidhisarvasattvahitavaha nama

412

No of Ton Folios no. 2537 20 34

2098

28

2094 1396

27

26 18 48 27_ 27 40

253 6_ 1399 2539 2093 1398 _2535_ 2090 2542

60

1397

76 187 26

2540 2534 2532

14

2541

60

1395

37

2092

43

2533 2584

22 53"

51 69

2538 2589

19

2091

49

2590

Appendix IV

Altogether there are 129 works related to the NS in the Tanjur. They are found in three places, according to the way in which they were understood to interpret the NS. 1 The following table summarises the classification in the Derge edition. Tanjur Division Kalacakra Tantra Anuttara Tantra Yoga Tantra Total

No of works Toh nos. 1395-1400 6 32 2090-2121 91 2532-2622 129

Commentaries Surviving in Sanskrit Apart from Vilasavajra's NMAA only two other NS commentaries are known to survive in Sanskrit: Ravisrljnana's Amrtakanika (Toh 1395) and the Gudhapada, by an unknown author. The Amrtakanikd comments on the NS from the perspective of the Kalacakratantra. It is unedited and survives in a number of MSS, some palm-leaf (see Tsukamoto, 1989). The Gudhapada survives in one palm-leaf MS in the possession of the Royal Asiatic Society, London, and does not appear to have been translated into Tibetan (see Appendix VII). A subcommentary to the Amrtakanika, the Amrtakanikoddyotana of Vibhuticandra, also survives. 2

^avidson (1981, 15) points out that the classifications were sometimes rather arbitrary. Candragomin's commentary, in the Anuttara Tantra section of the Derge edition, is placed in the Yoga Tantra section of the Peking. Vimalamitra's commentary is found in the Anuttara division though it has a closer affiliation to the NMAA than, say, works by Advayavajra. 2The title is taken from the title sheet of the NGMPP manuscript, A 117/10. There is another MS in the Asha Saphu Kuthi (Asha archives) in Kathmandu. Here the text is referred to as the Amrtakarnikodyota nama nibandha (see Yoshizaki 4259, where the MS is also identified as Takaoka DH 366).

413

35v4

_

6^v4

32r4

\ 33v5_ 84r3 . - •

62r9

18r8

4_7v4

14.

13.

Total no. of folios

Upasamharah

1 Prajnacakramantravinyasah

i

5 (NS 163 167)

_

7()vll

115

86

(113) j 85v4

i

134r8 135

72 194 80

|

77

77r7

133v8JJ77r2 72r2

71v8

L 64r6_ JJ^lL'^iL1

\

193r5

; 192v3

175vl

80rll

i (112) j 85r7 J 80r4

. . . ..... ....... 103rl j^ 76r5

68v6

121

(121r)

88

114v4

104

225.5.2

• •-225.4.5

H4r7 120v4

221.4^.4

221.3.8

218 -4.6_

__2H.5.2 __

204.5.5

M2!2.™ 105v2

; Illr3

105r6

Anusanisa

!

12.

j. . . 64r2

119r6

JJ •

76rl _ , 7()v5 _ 175r2.

; .1()lvl

5 (NS 158-162)_ 102v5

}

Pancatathagatajnanastutih

, — .— .

! 99r5,

63 v8 -

J 1()vLlA3r2

-"'-*

58r6 159r4 ~. -»««J— . • .

s 69vl - -

95v5

15 (NSJ43; 157)

Krtyanusthanajnanaslutih

10.

~»«*v^

[ 87r2__iJJOrl 93v9 _ £53v8

53rll................135v5 59v4 ,. . . 50r9 . ..........

82v7

i 70rl

24 (NS 1197 142)

i 37r8

^ 60vl

203.1.1 65v6 j 33v2 j 54vl i

195.1.2

r

i 48v3 32r5

;. Samatajnanastutih

37r4 _ 93 vl

19V3 j,...„..„.

9. ™4» •

60vl

! 42 (NS 77-118)

^•... ........... Pratyaveksanajnanastutih 41r7

3J7v3

54v6

. . . 10 (NS 67 76)

1 AdarSajnanastutih

7.

_

30v6 .. .... ., ; 22r4

36vl j 20r8

191.2.7 ' 41r2

31v4

[ 22r6

14r3

13v8

;

] 24r5

15r7

! 12r5

;

(20v)

] 14 (NS 28-41)

25 (NS 42-66)

Vajradhaturnahamandalam

! SuviSuddhadharmadhatujnanastutih

r

188.4.4 35v2 f

i 15r7

i 8v3

14r7

;7rll 9r6 _)..8v9 _ . 18v5 . ....... . .__r

Ilv3

3 (NS 25-27) :

, . ——— 188.3.1 .... ,..,...._. ' 35r2 .....-,.I \4r6

8r3

13r9

7r2

17v4

8r9

8v6

_j_ 1 Irl

2 (NS 23 24)

..........

6v8

Ijr4 _

187.4.8

Tib.P.

1 2r 1 __ ! 33 v4

Tib.D.

184.4.8

j

I27vl ( _. ...

Ivl j . . ....„jjvl

J

6r2_

Ivl

i

14v4

Ivl

H

\ 7r4

Ivl

F

7r7

Ivl

E

(9r)

Ivl

D

6 (NS 17-22)

Ivl

ABC

Manuscripts and Tibetan Xylograph Sigla

6.

5.

4. _ ^Mayajalabhisambodhikramah

———?™ • 1

• ••

Prativacanam

2.

\ Satkulavalokanam

16 (NS 1-16)

Adhyesana

1.

3.

NS Verses Commented on

Ch­ Corresponding NS Section apter

Appendix V: NMAA Chapters: Folio Numbers for the Skt. MSS and Tibetan Xylograph Editions

g, >

Suggest Documents