Early Tantric Vaisnavism: Three Newly Discovered Works of The Pancaratra (The Svayambhuvapancaratra, Devamrtapancaratra and Astadasavidhana)

Citation preview

Early Tantric Vais.n.avism: Three Newly Discovered Works ˜ cara ¯ tra of the Pan The Sv¯ayambhuvapa˜ ncar¯atra, Dev¯amr.tapa˜ ncar¯atra and As.t.a¯da´savidh¯ana

L’Institut Fran¸cais de Pondich´ery (IFP), UMIFRE 21 CNRS-MAEE, est un ´etablissement ` a autonomie financi`ere sous la double tutelle du Minist`ere fran¸cais des A↵aires Etrang`eres et Europ´eennes (MAEE) et du Centre National de la Recherche Scientifique (CNRS). Il est partie int´egrante du r´eseau des 27 centres de recherch´e de ce Minist`ere. Avec le Centre de Sciences Humaines (CSH) ` a New Delhi, il forme l’USR 3330 du CNRS “Savoirs et Mondes Indiens”. Il remplit des missions de recherche, d’expertise et de formation en Sciences Humaines et Sociales et en Ecologie dans le Sud et le Sud-est asiatique. Il s’int´eresse particuli`erement aux savoirs et patrimoines culturels indiens (langue et litt´erature sanskrite, histoire des religions, ´etudes tamoules. . . ), aux dynamiques sociales contemporaines, et aux ecosyst`emes naturels de l’Inde du Sud. The French Institute of Pondicherry (IFP), UMIFRE 21 CNRS-MAEE, is a financially autonomous institution under the joint supervision of the French Ministry of Foreign and European A↵airs (MAEE) and the French National Centre of Scientific Research (CNRS). It is a part of the network of 27 research centres under this Ministry. It also forms part of the research unit 3330 “Savoirs et Mondes Indiens” of the CNRS, along with the Centre de Sciences Humaines (CSH) in New Delhi. It fulfils its missions of research, expertise and training in Human and Social Sciences and Ecology in South and South-East Asia. It works particularly in the fields of Indian cultural knowledge and heritage (Sanskrit language and literature, history of religions, Tamil studies), contemporary social dynamics and the natural ecosystems of South India. IFP, 11, St. Louis Street, P.B. 33, Pondicherry – 605001, India. Tel: (413) 2231600 Website: http://www.ifpindia.org Email: [email protected] ´ L’Ecole fran¸caise d’Extrˆeme-Orient (EFEO), fond´ee en 1900 ` a Hano¨ı, est un ´etablissement relevant du minist`ere fran¸cais de l’Enseignement sup´erieur et de la Recherche dont la mission scientifique est l’´etude des civilisations classiques de l’Asie. Son champ de recherches s’´etend de l’Inde ` a la Chine et au Japon et, englobant l’ensemble du Sud-Est asiatique, comprend la plupart des soci´et´es qui furent indianis´ees ou sinis´ees au cours de l’histoire. Autour de ses dix-sept centres et antennes, install´es dans douze pays d’Asie, se sont constitu´es des r´eseaux ´ de chercheurs locaux et internationaux sur lesquels l’Ecole a pu s’appuyer pour construire son essor. L’EFEO aborde l’Asie par des recherches pluridisciplinaires et comparatistes, associant l’arch´eologie, l’histoire, l’anthropologie, la philologie, et les sciences religieuses. Pondich´ery, les projets de l’EFEO portent essentiellement sur l’“indologie” classique : sanskrit, tamoul ancien, histoire, histoire de l’art et des religions. The mission of The French School of Asian Studies (EFEO), founded in 1900 in Hanoi and today under the aegis of the French Ministry of Higher Education and Research, is to study the classical civilizations of Asia. Stretching from India, in the West, across the whole of South-East Asia to China and Japan, the EFEO’s research areas cover most of the societies which have been ‘Indianised’ or ‘Sinicised’ over the course of history. A network of international scholars working at the EFEO’s seventeen centres and branch offices, which are spread across twelve Asian countries, has been essential in the development of the School’s research programme. Interdisciplinary projects bring together leading scholars in the fields of anthropology, archaeology, history, philology, and religious studies. In Pondicherry, the projects of the EFEO focus mainly on classical Indology: Sanskrit, Old Tamil, History, and History of art and of religions. EFEO, 22, avenue du Pr´esident-Wilson, EFEO, 16 & 19, Dumas Street, 75116 Paris, France. Pondicherry – 605 001, India. Tel: (33) 1 53 70 18 60 Tel: (91) (413) 2334539/2332504 Website: http://www.efeo.fr/ Email: [email protected]

the early tantra series Tantric scriptures form the basis of almost all the various theistic schools of theology and ritual in post-Vedic India, as well as of a major strand of Buddhism (Vajray¯ ana). Among these schools, those centred on the Hindu ´ deities Siva and Vis.n.u spread well beyond the Indian subcontinent to Kambuja (Cambodia/Laos/Thailand), Champa (Vietnam) and Indonesia, while Buddhist tantrism quickly became pan-Asian. With this mini-series, launched within the ‘Collection Indologie’, we are releasing some of the fruits of a Franco-German project funded from 2008 to 2011 by the Agence Nationale pour la Recherche and the Deutsche Forschungsgemeinschaft. The series takes its name from the project’s title: ‘Early Tantra: Discovering the interrelationships and common ritual syntax ´ of the Saiva, Buddhist, Vais.n.ava and Saura traditions’. Our aim was to study the interrelationships between the tantric traditions on the basis of fundamental source-material, which we would edit and publish for the first time. For this we made use of some of the exceptionally rich manuscript resources gathered in the twentieth century and studied by German and French research institutes at opposite poles of the sub-continent. In recent centuries, the Tamil-speaking South is the only area where a vast corpus of Sanskrit texts of what was long the dominant school of tantric ´ Saivism continued being copied and so transmitted to the present day. So when, in 1956, Jean Filliozat secured a foothold in Pondicherry for French indological research, he created an ideal institutional base for the study of a forgotten chapter in the religious history of Asia. Gradually, the largest ´ specialised manuscript collection of texts relating to the Saiva Siddh¯anta was amassed, recognised in 2005 by UNESCO as a “Memory of the World” ´ collection: The Saiva Manuscripts of Pondicherry. At the other end of the subcontinent, the cool climate of Nepal has preserved ancient manuscripts of texts of virtually every branch of Indian learning. Much of the early history of tantrism is thus preserved in the vast archive of Nepalese manuscripts microfilmed over more than three decades by the Nepal-German Manuscript Preservation Project (NGMPP), then partly catalogued by the Nepalese German Manuscript Cataloguing Project (NGMCP), run from Hamburg and from the NGMCP’s base in Kathmandu, the Nepal Research Centre (NRC). One Nepalese treasure, also included, in 2013, in UNESCO’s register, formed the corner-stone of our project: the ninth-century manuscript of the Ni´sv¯ asatattvasam a. . hit¯ The Early Tantra Series is not a closed collection: as well as the editions and translations promised as part of the ‘Early Tantra’ project, studies of numerous related works were inspired or further advanced during the project’s workshops.

Collection Indologie 129 Early Tantra Series 2

Early Tantric Vais.n.avism: Three Newly Discovered Works ˜ cara ¯ tra of the Pan The Sv¯ayambhuvapa˜ ncar¯atra, Dev¯amr.tapa˜ ncar¯atra and As.t.a¯da´savidh¯ana Critically edited from their 11th- and 12th-century Nepalese palm-leaf manuscripts with an Introduction and Notes

by

Diwakar Acharya

´ry institut franc ¸ ais de pondiche ´ ˆ ecole franc ¸ aise d’extreme-orient ¨ t hamburg asien-afrika-institut, universita

CONTENTS

Preface

ix

Introduction The Manuscripts . . . . . . . . . . . . . . . . . . . . . . . . . Editorial Policy . . . . . . . . . . . . . . . . . . . . . . . . . . The Language of these Texts . . . . . . . . . . . . . . . . . . The Significance of these Texts . . . . . . . . . . . . . . . . . The Ritual System these Texts Teach . . . . . . . . . . . . . Contents of the Sv¯ ayambhuvapa˜ ncar¯atra . . . . . . . . . . . . Chapter 1: Introduction and Synopsis . . . . . . . . . . Chapter 2: On Vis.n.u’s Nature . . . . . . . . . . . . . . Lost Chapters 3 and 4 . . . . . . . . . . . . . . . . . . . Interpolation in the Manuscript . . . . . . . . . . . . . . Chapter 5: Consecration of Pavilion . . . . . . . . . . . Chapter 6: Characterisation of the Consecrator . . . . . Chapter 7: The Deity inhabits the pavilion . . . . . . . Chapter 8: Image Installation . . . . . . . . . . . . . . . Chapter 9: Internalisation of Vis.n.u? . . . . . . . . . . . The As.t.¯ ada´savidh¯ ana . . . . . . . . . . . . . . . . . . . . . . Contents of the As.t.¯ ada´savidh¯ana . . . . . . . . . . . . . . . . Contents of the Dev¯ amr.tapa˜ ncar¯atra . . . . . . . . . . . . . . Chapter 3: Sculpting Vis.n.u’s Image: Material and Size . Chapter 4: What makes a Good Image . . . . . . . . . . Chapter 7: The Pavilion is Divine, not Demonic . . . . Chapter 8: Consecration of the Pavilion . . . . . . . . . Chapter 9: Ritual Ablution of the Deity . . . . . . . . . Chapter 10: The Consecrator and his Assistants . . . . Chapter 11: Deposition of Mantras . . . . . . . . . . . . vii

. . . . . . . . . . . . . . . . . . . . . . . . .

xiii xvi xviii xxii xxv xxix xxxv xxxv xxxvi xxxviii xxxix xl xlii xlii xlvi l l lviii lxxi lxxi lxxii lxxiv lxxvi lxxviii lxxx lxxxi

Chapter 12: The R . gveda, Vis.n.u’s Image, and his Incarnations . . . . . . . . . . . . . . . . . . . . . . . . . . lxxxiii Chapter 13: Assembling the Image-Limbs Ritually . . . . lxxxv Sanskrit Text Sv¯ ayambhuvapa˜ ncar¯ atra . . . . . . . . . . . . . . . . . . . . . . As..t¯ ada´savidh¯ ana . . . . . . . . . . . . . . . . . . . . . . . . . . Dev¯ amr.tapa˜ ncar¯ atra . . . . . . . . . . . . . . . . . . . . . . . .

1 1 35 57

Critical Notes

107

Works Consulted

127

Facsimiles

135

Indices Index of P¯ adas in all three texts . . . . . . . . . . . . . . . . . Index of Tantric mantras and other prose lines . . . . . . . . . Index of Vedic mantras cited . . . . . . . . . . . . . . . . . . .

185 187 215 219

General Index

225

PREFACE Vais.n.avism was well-established in the Kathmandu Valley already in the early Licchavi period and it enjoyed a strong following in subsequent periods too.1 I had therefore always hoped that the Nepalese ¯ archives might preserve some early Vais.n.ava Agamas alongside its early ´ ¯ manuscripts of Saiva Agamas. The chances seemed high because more than one manuscript of the Jay¯ akhyasam a and several manuscripts of . hit¯ the Vis.n.udharma, an early Vais.n.ava text belonging to a less esoteric current of Vais.n.avism, were already found there. So, I was not unduly surprised when I discovered the texts presented in this volume while cataloguing manuscripts for the NGMCP in Kathmandu and Hamburg. I quickly prepared electronic transcriptions of these texts, and when I moved to Kyoto University in 2006, I started to edit them one after the other and to read them in my seminars with my colleagues and students. The result is in your hands. The texts presented in this volume have special value as they provide ´ revealing evidence for the influence of Saivism over Pa˜ ncar¯atra Vais.n.avism 1

I postulate this on the basis of tidbits of evidence coming mainly from Licchavi inscriptions. The most prestigious Vais.n.ava temple of those days appears to be the Changu Narayan temple, referred to as Dol¯ a´sikharasv¯ amin in inscriptions. This is one of the four Vis.n.u temples on hilltops in four corners of Kathmandu Valley. According ¯ ca ¯ rya & Malla 1985: 28, 76, 123), the earto the Gop¯ alar¯ ajavam ¯val¯ı (cf. Vajra . ´sa liest of Nepalese chronicles, these temples were built by King Haridattavarman a few generations before M¯ anadeva I. In front of the Changu Narayan temple, M¯ anadeva I ´ erected a stone pillar with a Garud.a on top, in Saka 386 (464 AD) after subjugating ¯ ca ¯ rya 1973: inscription no. 2). Three years later he his revolting vassals (cf. Vajra installed two Vis.n.uvikr¯ anta images for the sake of accumulation of his mother’s merits (cf. Id: inscription nos. 4 and 5). King Am . ´suvarman in Licchavi Year 31 (606 AD) repaired the golden jacket of the deity in the same temple, though he was inclined to ´ Saivism (cf. Id: inscription no. 76). One inscription from the reign of this king records that this temple received the same amount of revenue as the Pashupati temple (cf. Id: inscription no. 77). King Narendradeva’s inscription from the 80s of Licchavi era (655–664 AD) makes certain arrangements for the celebration of an annual ceremony in the temple of Lokap¯ asv¯ amin near presentday Bhaktapur; it also mentions a festival of Var¯ aha (cf. Id: inscription no. 129).

x

Early Tantric Vais.n.avism

in the early mediaeval period, at a time when Vais.n.avism was beginning a process of self-transformation that involved partly recasting itself in the ´ mould of tantric Saivism. Besides, these also speak of the influence of late Vedic and Sm¯ arta ritual systems over the Pa˜ ncar¯atra ritual system. Therefore, with the discovery and publication of these texts, I hope that some light will be shed on the early history of Pa˜ ncar¯atra Vais.n.avism, ´ which is otherwise obscure, and that its affinity with both Saivism and Vedism in its formation process will be more widely recognised. I am obliged to the Early Tantra Project (initially funded for three years by the Deutsche Forschungsgemeinschaft and the Agence National pour la Recherche), first, for organising workshops on early Tantras and giving me a chance to talk about these texts, and now, for taking the responsibility of publishing this volume. I am also grateful to the Oxford Centre for Hindu Studies for inviting me to Oxford in spring 2010, where I delivered a lecture about the early Vais.n.ava texts I discovered in Nepal. A part of that lecture has now been incorporated in the introdution. I also thank the Japanese Society for the Promotion of Science (JSPS) for providing me with a research grant (Start-up 2011–2012) to finalise this edition of Vais.n.ava Pa˜ ncar¯atra texts. Above all, I am eternally grateful to my two kaly¯an.amitras, Profs. Dominic Goodall (EFEO) and Harunaga Isaacson (Hamburg University), who have always helped me in their various incarnations of intimate friends, teachers, academic colleagues, and sources of digital resources. They have read an earlier draft of this book and made valuable suggestions. I have also benefitted immensely from the suggestions of Prof. Yuko Yokochi, my colleague at Kyoto University, for we have studied these texts at various reading sessions, and from those of Prof. Judit T¨orzs¨ok, who has reviewed my manuscript very carefully, made useful suggestions, and even painstakingly corrected numerous typos. Helpful also were the suggestions of my two students Mr. Koreto Ikehata (Kyoto) and Mr. Nirajan Kafle (Kathmandu/Pondicherry). Besides, Dr. Marion Rastelli (Vienna) as well as Dr. Taisei Shida (Kyoto) have occasionally corrected my mistakes and they deserve words of gratitude. I am also thankful to the sta↵ at the Nepal Research Centre and the National Archives (Kathmandu), my family, and other friends for their co-operation in various stages. At the end, with this book, I should like to pay tribute to my father Loknath Acharya (1931–2010) who shaped my childhood with a tradi-

Preface

xi

tional Sanskrit education and also introduced me to certain Tantric texts. ¯ I should also like to remember my Tantra/Agama teachers in various times and places and in various capacities: the late Sw¯am¯ı Vidy¯aran.ya, better known as M¯ urkh¯ aran.ya, and the late Shyam Chetan Baba, both from Vasuki Ashram in Pashupati, who were my informal mentors in my childhood days in Kathmandu; the late Pt. Hemendra Chakravarti, the late Prof. Vraj Vallabh Dvivedi, and Prof. Ramji Malaviya, and all my ¯ Tantra/Agama teachers at Benares. Diwakar Acharya Kyoto, September 14, 2014

INTRODUCTION We know from some literary sources, including the Mah¯ abh¯ arata and a number of inscriptions, that a form of Vais.n.avism existed even some centuries before the commencement of the common era.2 However, the Pa˜ ncar¯ atra system that has come down to us even in the earliest available Pa˜ ncar¯ atra Sam as does not reflect this early stage of the religion. The . hit¯ Jay¯ akhyasam hit¯ a has long been regarded as the earliest of the Pa˜ ncar¯atra . 3 Sam as but, as Sanderson has recently stated, even the ritual system . hit¯ of this Sam a is the product of a competitive reformation of their system . hit¯ ´ along Saiva lines. Although Embar Krishnamacharya, the editor of the Baroda edition of the Jay¯ akhyasam a , had dated this text to the . hit¯ age before 700 a.d. (cf. Sanskrit introduction, pp. 52–53),4 it is now a long while since K. V. Soundara Rajan published his conclusion, after an analytical study of the architectural chapter, that the age of the text, excluding later interpolated portions, was ‘likely to range between 600– 850 a.d.’ (cf. Rajan 1967–1969: 80). But I should now like to add that, even in the main body of this text such as we find it transmitted in Nepalese manuscripts,5 comparatively modern elements are included, such as the fourfold classification of the initiate, which do not feature in the ¯ ´ early Agamas of the Saiva fold such as the Ni´sv¯asa and the early recensions of the K¯ alottara texts. So it is safe to say that the Jay¯ akhyasam a does . hit¯ not represent the earliest stage of the ‘competitive reformation of the 2

The Besnagar Garud.a pillar inscription of Heliodorus and Ghos¯ un.d.¯ı inscription of Sarvat¯ ata, both belonging to the second century BC, provide evidence for an ancient form of Vais.n.avism which centred on worship of V¯ asudeva along with Sam . kars.an.a and other subordinates. For both inscriptions, see, Sircar 1965: 88–91. 3 Sanderson 2009: 61. 4 Upstaging the editor of the text, B. Bhattacharyya, the general editor of the series, dates (pp. 26–34) the composition of the Jay¯ akhyasam a to about 450 a.d. . hit¯ 5 I would like to recall the fact that a large portion towards the end of the Baroda edition of the Jay¯ akhyasam a is not found in Nepalese palm-leaf manuscripts of the . hit¯ text. I will compare the contents of these manuscripts with those of the printed edition in the introduction to my forthcoming edition of the Jayottaratantra, which I hold to be the Urtext of the Jay¯ akhyasam a. . hit¯

xiv

Early Tantric Vais.n.avism

´ Pa˜ ncar¯ atra system along Saiva lines’; rather it reflects a late stage of that reformation, perhaps at a time when it was already settled in the South and beginning to evolve there. Moreover, I have found what I believe to be the Urtext of the Jay¯ akhyasam a preserved in a palm-leaf manuscript . hit¯ in Nepal and, as mentioned in fn. 5 above, am preparing a critical edition of the work. Its text is much shorter and contains only what I regard to be the core of the printed Jay¯ akhyasam a. . hit¯ However, as we might readily imagine, there must have been earlier Pa˜ ncar¯ atra texts, and we have some evidence for their existence. For example, the srotonirn.aya chapter of the Brahmay¯amala6 mentions some ritual manual of the Pa˜ ncar¯atra school (pa˜ ncar¯atravidh¯ana) as well as the 7 ritual manuals for the worship of Narasim . ha, Var¯aha and Vaikun.t.ha. Similarly, we know the name, but no more, of another Vais.n.ava text, ´ the M¯ ay¯ av¯ amanik¯ a, as it is mentioned by Kashmirian Saiva exegete 8 Ks.emar¯ aja in his commentary on the Netratantra. In the same way, the opening passage of an extra Vais.n.ava text copied at the end of one of the Nepalese manuscripts of the Jay¯ akhyasam a enumerates a few . hit¯ names of Vais.n.ava scriptures which were allegedly taught in the past. These previously taught texts are Jay¯a, Vaih¯ayas¯ı, M¯ay¯a, Paus.kar¯ı, Jyotis, P¯ at¯ al¯ akhy¯ a, Laks.m¯ı, K¯alavai´sv¯anara, P¯an ˜car¯atrarahasya, and Kulake´sarika.9 But except the Jay¯a, Paus.kar¯ı, and Laks.m¯ı, none of these works has survived. Even among these three, the available versions of the Paus.kar¯ı and Laks.m¯ı contain quite a lot of relatively new matter; com6 Cf. Brahmay¯ amala 39.80–81: n¯ arasim anan tu vidh¯ anam a | . havidh¯ . ks.etrakalpan¯ vivar¯ antargat¯ ah. kalp¯ a [var¯ a]hasya vidhis tath¯ a k pa˜ ncar¯ atravidh¯ anan tu vaikun.t.hasya vidhis tath¯ a | kul¯ ac¯ arasamopet¯ a evam . vai bhairavo ’brav¯ıt k 7 Pa˜ ncar¯ atra is treated here as something di↵erent from the cults of N¯ arasim . ha, Var¯ aha and Vaikun.t.ha. Does this mean that N¯ arasim aha and Vaikun.t.ha were . ha, Var¯ not worshipped in the original Pa˜ ncar¯ atra? Such a possibility may be entertained, and in that case, the quartet of V¯ asudeva, Sam . kars.an.a, Pradyumna, and Aniruddha would be at the centre of Pa˜ ncar¯ atra worship, if not merely N¯ ar¯ ayan.a or V¯ asudeva with or ´ ı. without Sr¯ 8 Ks.emar¯ aja introduces a text-portion of the Netratantra (13.10↵) that teaches a cult of an eight-armed form of Vis.n.u, informing us about the source of that cult, thus: evam asam adr.s.t.yoktv¯ a m¯ ay¯ av¯ amanik¯ asthity¯ apy a ¯ha. . ´sr¯ıjay¯ . hit¯ 9 NAK MS No. 1–49, fol. 113r , ll. 2–3: jay¯ am ayas¯ım ay¯ am . vaih¯ . m¯ . paus.kar¯ım . jyotir eva ca k p¯ at¯ al¯ akhy¯ am a laks.m¯ım alavai´sv¯ anaram an ˜car¯ atram . tath¯ . k¯ . ca yat | p¯ . rahasyam . ca kulake´sarikam ak . tath¯ I assume that the works from Jay¯ a to Laks.m¯ı, all of which have feminine endings, are thought of as Sam as and that the rest are Tantras. . hit¯

Introduction

xv

pared to these two, the Jay¯ a contains relatively old materials, but this too, in its Sam a form, has been considerably diluted with additions . hit¯ over time. Nevertheless, it is fortunate that the treasure trove of manuscripts preserved in Nepal also contains some rare Vais.n.ava texts. These little known archaic texts preserved in Nepalese palm-leaves provide some evidence for earlier stages of the competitive reformation of the Pa˜ ncar¯atra system that Sanderson has talked about (see above, p. xiii, footnote 3). I am working on five of them and would tentatively put them in the following chronological order: the Sv¯ ayambhuvapa˜ ncar¯ atra, As..t¯ ada´savidh¯ ana, Dev¯ amr.tapa˜ ncar¯ atra, Jayottaratantra, and the V¯ asudevakalpa ascribed to the now lost Mah¯ alaks.m¯ısam a. All these texts are, I be. hit¯ lieve, earlier than the surviving recension of the Jay¯ akhyasam a . The . hit¯ Sv¯ayambhuvapa˜ ncar¯ atra has affinity with the bracketing layer of the Ni´sv¯ asatattvasam a , that is to say with the two latest of its five books, . hit¯ namely the introductory Ni´sv¯ asamukha and the concluding Guhyas¯ utra.10 The second text, as its title tells, is a precept of eighteen rituals and is related to the tradition of the first text. It is actually embedded in the first text, but I have separated it and presented it alongside the Sv¯ ayambhuvapa˜ ncar¯ atra, for, as will become clear from the following pages, I have come to regard it as a separate work. The third text is obviously based on the Sv¯ ayambhuvapa˜ ncar¯ atra, and the fourth text, the Jayottaratantra is the core base of the Jay¯ akhyasam a . The fifth text, the V¯ asudeva. hit¯ kalpa, contains some sort of Kaula influence, and can be compared, as far as its nature and time is concerned, to the early Yogin¯ı Tantras and Krama Texts. The cults of the last two texts are already known to the Netratantra. In this volume, I am going to present the first three of the abovementioned texts, the Sv¯ ayambhuvapa˜ ncar¯ atra, As..t¯ ada´savidh¯ ana and Dev¯ amr.tapa˜ ncar¯ atra, and the other two will appear soon in separate volumes.

10

For the order and mutual relationship of the five books of the Ni´sv¯ asatattvasam a, . hit¯ see Goodall, Sanderson, Isaacson, et al., 2015: 17–18.

xvi

Early Tantric Vais.n.avism

The Manuscripts The Sv¯ ayambhuvapa˜ ncar¯ atra is transmitted in a palm-leaf manuscript preserved in the National Archives, Kathmandu. Its manuscript number is 1–1648 and the NGMPP reel number is A 54/9. Only nine folios of this manuscript are available and all of them are damaged on both edges. The leaves are relatively long and the text is written in three columns separated by two string holes. These folios contain 5–6 lines on both sides, a line containing about 120 aks.aras. The second-last folio is severely damaged and very little of the text has survived in the first column. The best preserved folio of this manuscript measures 55 x 4.5 cm. The manuscript is dated, the colophon telling us that it was copied on the 11th of the ¯ .¯ bright half of As ad.ha, NS 147 (1027 CE). Because the manuscript is damaged in the margins, all folio numbers are lost except on one folio, numbered 6 with a letter symbol. Because the text is continuous at this point, we can identify two folios preceding this as the fourth and fifth folios and the one following it as the seventh. We can also identify the last and second-last folios on the basis of textual continuity and colophons. The remaining two folios which come one after the other must be placed after the seventh and before the last two folios. But it is not sure how many folios are missing after the seventh folio and before the second-last folio, and so we cannot number these intervening folios. Therefore, I have numbered these two folios, which are microfilmed in reverse order, as *X and *Y. A transcript was made from this palm-leaf manuscript in the first quarter of the 20th century, originally for the private library of Rajaguru Hemaraj Pundit. It is now preserved in the same archives, where it bears Manuscript number 5–1966, and it has been microfilmed twice by the NGMPP under reel numbers B 114/20 and B 237/16. Because it is an inaccurate copy that provides no help, I have not used it for my edition. It appears that this text had nine chapters. As the colophons of the first and sixth to ninth chapters are found in the manuscript, we can say that the first and seventh to ninth chapters are complete if we ignore the small portions lost in damaged margins. The whole of the first chapter and the first fourteen verses of the second are contained in the first folio. We can recover the missing portion of the second chapter from the Dev¯ amr.tapa˜ ncar¯ atra, which recasts a large part of the Sv¯ ayambhuvapa˜ ncar¯ atra. Nothing of the third and fourth chapters survives; they are lost with the

Introduction

xvii

missing second and third folios. In the beginning of the 4th folio we are at the end of the first unit of a precept of eighteen rituals called the As..t¯ ada´savidh¯ ana, a text interpolated into the Sv¯ ayambhuvapa˜ ncar¯ atra in our manuscript. A few verses, possibly 14, from the beginning of this precept must have been covered in the second of the lost folios. However, 119 verses of this precept have survived, and they extend to the third line of the recto of the 7th folio. I have separated this interpolated text and placed it after the conclusion of the Sv¯ ayambhuvapa˜ ncar¯ atra. After the As..t¯ ada´savidh¯ ana concludes, there follows the fifth chapter, as I have assumed it to be. This remains incomplete after 29 verses, as the text breaks o↵ in the middle of the 30th verse. The rest of the fifth and almost the whole of the sixth chapter are lost again in missing folios. This time, we do not know how many of them are missing, but at least one. Then two folios are available in consecutive order which contain the last four verses of the sixth chapter, the whole of the seventh chapter, and the first twelve verses of the eighth chapter. One folio is again missing, and lost with it is a large portion of the eighth chapter. Then follow the last two folios of the manuscript that contain the latter part of the eighth chapter and the whole of the ninth chapter, which is extremely short. The Dev¯ amr.tapa˜ ncar¯ atra, too, is preserved in the same archives in a single incomplete manuscript. Its manuscript number is 1–1078 and the NGMPP reel number is B 29/2. It has 26 folios of 28 x 5 cm in size. Each folio contains 5–6 lines on both sides, and each line contains an average of 40 aks.aras. The manuscript is not dated but after comparing its paleographical features with other dated Nepalese manuscripts I have come to the conclusion that this manuscript can be assigned to about one hundred to one hundred and fifty years after the date of the Sv¯ ayambhuvapa˜ ncar¯ atra manuscript; so it belongs to the twelfth century CE. Two transcripts of this manuscript are available, one preserved in the same archives and the other in a private collection. The first is dated in Vikrama Year 1985 (1928 CE) but the second is undated. The first bears Manuscript number 5–1965 and is filmed by the NGMPP twice under reel numbers B 114/15 and B 237/20. The other one can be found in the NGMPP microfilms under reel number E 529/2. I checked these transcripts in the hope of being able to decipher a few more aks.aras at the beginning of the eighth chapter, but found no help there.

xviii

Early Tantric Vais.n.avism

Editorial Policy Editing a text surviving in a single manuscript is a risky task because of the possibilities of misjudgement and the obvious limitations involved. In spite of this, I could not help but want to edit these texts after realising their importance, thinking it necessary for the sake of presenting possibly the last evidence of these rare and archaic texts for future readers. I have aimed in this edition of these texts to establish a readable text that is in all possible ways closest to the text in their singular manuscript sources. I have corrected minor mistakes, emended transmissional errors, and also conjectured possible readings that I think may have been originally intended. Although the manuscripts I used are fairly old, the text in them is quite corrupt, and I have often been forced to emend the text by conjecture. In all cases of emendation I have recorded the manuscript reading thoroughly and accurately. This edition is thus not a diplomatic one. In spite of the fact that each of the three texts presented here is edited from a single manuscript, I have earnestly strived for a critical edition. The Dev¯ amr.tapa˜ ncar¯ atra, the third text, for the most part recasts the first text, the Sv¯ ayambhuvapa˜ ncar¯ atra, and the As..t¯ ada´savidh¯ ana is a paddhati text related to the tradition of the former and so, although it has some extra materials, it shares much in common with the other two texts. Thus we have practically two sources for the larger part of all these texts. Many of the prat¯ıkas of Vedic mantras cited in these texts for ritual use are corrupt, but they can be emended after identifying them in similar ritual texts belonging to the ´ Vedic or Saiva traditions. The language of these works is comparable to the language of many other Tantric scriptures. All these conditions have enabled me to strive for a critical edition. I have presented text-critical information in three layers of the critical apparatus. The first register, at the bottom, is the main critical apparatus recording manuscript readings. It is wholly positive, and each time the entries are presented carefully. Given first is a lemma preceded by the concerned verse number and followed by a lemma sign ( ] ). This draws the attention of the reader to the portion of the constituted text that is read in the manuscript di↵erently. The lemma sign is followed either by the abbreviation ‘em.’ to indicate a proposed emendation or by the abbreviation ‘conj.’ to indicate a conjecture that is somewhat less secure. A semi-colon is placed after this and then the corrupt manuscript reading.

Introduction

xix

The manuscript reading is indicated by the common abbreviation MS; the superscripts ac (ante correctionem) and pc (post correctionem) are used together with the abbreviation MS to indicate readings respectively before and after correction. The abbreviation ‘ow.’ is used to report overwritten syllable(s). In cases where there is more than one entry reported for the same verse, a thick dot (•) is given to separate two entries and the verse number is not repeated. The second register, the second from the bottom, is meant to record parallels. Whenever a parallel is recorded in this register, it is always accompanied by the verse or unit number of the constituted text and followed by an exact reference to the source. As mentioned earlier, the Dev¯ amr.tapa˜ ncar¯ atra adopts with minor changes a considerable number of verses from the Sv¯ ayambhuvapa˜ ncar¯ atra and thus we have parallel passages in these two texts. Some passages of the As..t¯ ada´savidh¯ ana are either parallel or at least comparable to parts of the Dev¯ amr.tapa˜ ncar¯ atra and they can be seen in this layer of the apparatus. In very rare cases, some ´ passages from Puranic or Saiva texts feature in this apparatus. Some pages contain a third register at the top of the apparatus. A good number of prat¯ıkas of Vedic mantras are embedded in these texts, and I have tried my best to identify them in this third register of the apparatus. Many pages, however, have only two registers, because one or other of the two registers of parallels and quotations can be missing. I have employed two types of punctuation in the constituted text: parentheses ((. . . )) for partially visible aks.ara(s) on a broken edge or elsewhere and brackets ([. . . ]) for the aks.ara(s) restored in the lacunae. Where I am not able to fill in the lacunae but have calculated the approximate number of aks.ara(s) missing, I have added the number between a pair of t: thus ‘t12 t’ indicates that twelve aks.aras are missing. I have used a plus sign (+) to indicate an aks.ara lost in a broken edge, and a dash (–) to indicate an aks.ara unreadable due to damage to the surface of the palm-leaf when the leaf itself is not broken. The crux signs (†. . . †) have been used to indicate that the marked portion of text is so corrupt that I am unable to provide a convincing conjecture. At least once I have used double angular braces (⌧ ) to supply a portion of the text that is missing in the manuscript because of an eyeskip or lack of attention on the part of the scribe but that is logically needed and actually supported by a parallel in the testimonia. At some places words or parts of words are marked with a wavy line

xx

Early Tantric Vais.n.avism

to alert the reader that the marked portion of the constituted text is quite insecure, being based on a conjecture about which the editor himself is not certain. This does not of course mean that all other conjectures not marked with such a wavy line are secure. The verse-numbering and -division are my own. Normally a separate number is not assigned to a half-verse; it is instead attached to the preceding lines, creating a six-p¯ada verse. In the As..t¯ ada´savidh¯ ana a single free-standing p¯ ada sometimes serves as a heading (e.g. AVi 9, 29, 58), but this is not always so (e.g. AVi 40, 42, 54, 80). We cannot therefore avoid accepting some verses with odd number of p¯adas, such as three or five. Since the manuscript preserving the Sv¯ ayambhuvapa˜ ncar¯ atra and the As..t¯ ada´savidh¯ ana is more difficult to handle, the margins being damaged and folio numbers lost, as can be seen on the images presented in pp. 133–152, I have indicated in the edition the change of every line. But in case of the Dev¯ amr.tapa˜ ncar¯ atra, I have indicated only when a folio is changed or turned. Wherever the text is interrupted because of the loss of a folio, I have supplied a set of five flowery signs in the centre of a separate line (⌦⌦⌦⌦⌦). Because each of these texts comes from only one independent manuscript source, often while editing the text I have faced a dilemma in deciding whether a particular linguistic phenomenon in the manuscript is scribal or authorial. I had to make delicate considerations before I decided, and in this process, I have often relied on the metre. I have adopted regular form where they do not violate the metre, and an Ai´sa form if a regular one would infringe the metre. Thus, the metre has proved to be an important ground for many decisions. I would also like to inform the reader that I have corrected inconsistencies in the gender of qualified nouns and qualifying adjectives, at least those in the same p¯ ada, wherever the grammatically correct form fits in the metre. I am however aware that some of these inconsistencies may be authentic. For, it is not problematic to have alternation of the masculine and neuter gender in Ai´sa Sanskrit. I agree that I could sometimes have maintained the apparent incoherence of the original. But I have decided to make the text somewhat more coherent, especially because the manuscript itself is corrupt and it is anyway impossible to be sure that these consistencies are authorial. In many places, because of a gen-

Introduction

xxi

eral scribal sloppiness, we cannot fully rely on the manuscript readings, particularly when only a visarga or an anusv¯ara determines the gender. When conjectures are made, I have taken care that they are metrical, and in this process I have sometimes been forced to cut a word in the prat¯ıka of the Vedic mantra being cited. But this does not seem to be a problem and there is no need to strive for a perfect prat¯ıka, either. For at some places in the text (e.g. AVi 34cd) it is clear that the text stops in the middle of a word while citing a prat¯ıka and sometimes even changes the order of words for the sake of metrical regularity. A good number of prat¯ıkas of Vedic mantras are embedded in these texts, and I have tried my best to identify them. In this e↵ort I have derived much help from the S¯ amavidh¯ana Br¯ahman.a and occasionally from some other primary and secondary sources. All these Vedic mantras are identified and recorded in the topmost register of the apparatus. They are also listed separately in alphabetical order in one of the indices appended. There are also portions of prose in the Sv¯ ayambhuvapa˜ ncar¯ atra and the As..t¯ ada´savidh¯ ana. These include enumerations of tattvas (AVi 12), the actual reading of mantras (e.g. AVi 29), and ritual instructions that would be difficult to fit metrecally. They are all collected in the last of the three indices given at the end of this volume. The As..t¯ ada´savidh¯ ana contains a good number of unmetrical p¯adas, and occasionally, as we have just seen, lines of prose. Unmetrical p¯adas and prose lines in the Sv¯ ayambhuvapa˜ ncar¯ atra are found particularly towards the end of that work, but these are not many. The Dev¯ amr.tapa˜ ncar¯ atra does not contain any prose, but a few unmetrical p¯adas can be found in the eighth and thirteenth chapters. I have noted them in the critical apparatus and occasionally discussed them in critical notes. An avagraha is inserted in the edited text, although it is hardly used in the source manuscript, to mark the elision of an initial a preceded by e or o, but not to mark a homophonic sandhi. The anusv¯ara is usually given preference over class nasals as far as it is grammatically permitted, but when the nasal falls at the broken edge of a folio the manuscript reading is followed. I have not recorded the doubling of consonants in ligatures with semivowels in the manuscripts, and have normalised the text in this regard. The same is true with the confusion of the dental s and palatal ´s.

xxii

Early Tantric Vais.n.avism

The Language of these Texts I would like to draw the attention of the reader to the grammatical irregularities encountered in these texts. These irregularities, as could be ´ ´akta expected, are often the same as those found in early Saiva and S¯ Tantras, and the language of these texts, too, can be called Ai´sa Sanskrit. In the following I collect and categorise these irregularities. My account is by no means exhaustive, and for more detailed account of Ai´sa Sanskrit, the language of early Tantric texts, the reader should resort to Goudriaan and Schoterman 1988: 44–109, Goodall 1998: lxv–lxx and 2004: ¨ rzso ¨ k 1999: xxvi–lxix, and Goodall, Isaacson & lxxvii–lxxxvii, To Sanderson 2014: 107–130; and for the sake of a comparative outlook, to Oberlies 2003 also. Hiatus and hiatus-bridger. Hiatus at the seam of two p¯adas11 seems to be an acceptable feature even in the Pur¯an.as but in the texts presented in this volume it is encountered frequently even within a p¯ada ´ ´akta traditions.12 Nevjust like in the Tantric texts of the Saiva and S¯ ertheless, much more frequently we can see such a hiatus within a p¯ada has been avoided by inserting sandhi consonants m, d or r as a hiatusbridger.13 Sometimes, even an allowed hiatus resulting from a regular sandhi is bridged.14 Such a hiatus-bridger can even be found bridging two members of a compound.15 And more, once we find ra inserted as a hiatus-bridger instead of r inside a compound (SP 8.12). Double sandhi and irregular sandhi. In these texts one can now and again notice that a due sandhi has been ignored but at times double sandhi 16 or irregular sandhi 17 has been practised. However, a sandhi 11

See e.g. SP 7.45ab: annapr¯ a´sanamantras tu es.a, AVi 1cd: tasm¯ at sarvaprayatnena anga ˙ . 12 See e.g. SP 7.23a = DP 11.32a: tu are nyasya, DP 7.21c: aviditv¯ a im¯ am .. 13 See e.g. SP 1.16c = DP 1.13d: nakha m-anguly ˙ m, ¯ SP 2.12b: s¯ amavedo r-atharvan.ah.), AVi 44b: p¯ adau m-upari, DP 12.12b: savyapaks.am alabhet. . tu d-¯ 14 See e.g. SP 5.21d = DP 6.29b: p¯ urva-m indr¯ aya. 15 See e.g. SP 5.26d: vedi-m-agratah., AVi 69d: v¯ ayu-r-ambarayor yath¯ a. 16 See e.g. DP 7.15b: naivedyausadhayah. smr.t¯ ah. for naivedya ausadhayah. smr.t¯ ah. which would be hypermetrical. DP 12.6d: tatordhvag¯ a for tata urdhvag¯ a. We can also list SP 5.12a: nyagrodhodumbara´s caiva for nyagrodha udumbara´s caiva, but the occurrence of ´sam¯ıpal¯ a´sadumbaryah. in SP 7.39c tells that this text uses dumbara as a stem alternative to udumbara. 17 See e.g. DP 7.2a: n¯ asuredam asuram idam abhimantritam . for n¯ . , DP 9.15d: dady¯ for dady¯ ad abhimantritam.

Introduction

xxiii

between two p¯ adas is hardly ever employed. Loss of an anusv¯ ara. A final anusv¯ara or a homorganic nasal resulting from a sandhi is dropped wherever needed in order to fit the word in the metre.18 I have even found a case of such elision in the middle of a word: DP 9.12b: satyaparya’tam. Loss of the last component of a word. The last syllable of a word, mostly a case-ending, may be dropped for the sake of the metre.19 Loss of an initial vowel. An initial vowel, mainly the initial i of iti, is elided in many conditions: when preceded by a final e,20 by a final ¯a remaining after the disappearance of a visarga,21 and by a final o.22 Some very rare cases involve the loss of the final consonant of the preceding word and the initial vowel of the following word as well.23 I have restored the final t/d at the end of an optative singular verb and also an a-stem ablative singular noun if it is allowed by the metre. For such a loss occurs only when the final t/d is followed by a consonant either at the end of the odd p¯ adas or inside a p¯ada and I diagnose this as scribal error, not as a linguistic peculiarity. There is not a single case of the loss of the final consonant at the end of even p¯adas. The loss of the final t/d at the end of the odd p¯adas cannot be original, because, unless constrained by the metre, our texts do not observe even a normal sandhi on the basis of any component present in the beginning of the following p¯ada. I therefore set the criterion that if a final consonant needs to be dropped to keep the metre intact, it should be regarded as authorial, and if otherwise, it should be regarded as a result of scribal error or confusion. I am aware that others would have decided otherwise at least in some of these cases. At this point, I would like to remind the reader that our manuscripts are full of scribal errors and that I have therefore taken the position that unless there is compelling evidence it is better not to identify an abnormal feature as a linguistic peculiarity. To indicate this kind of metri causa shortening and also a loss of a syllable after double sandhi, I have appended an apostrophe to the 18 See e.g. SP 7.31b: ´sa ¯rnga’ ˙ vinyaset, SP 8.49: gandhac¯ urn.a’ sam¯ alikhya, AVi 48d: b¯ ahu’ vinyaset. 19 See e.g. DP 3.4d nirdis.t.a ¯’ for nirdis.t.a ¯ni, DP 3.11a siddhi’ for siddhid¯ a, DP 7.15 digas.t.a’ vasav¯ as.t.akam for digas.t.ake vasav¯ as.t.akam, AVi 49d: jam . gha’ for jam . ghayoh.. 20 See e.g. SP 5.14: agnim¯ıle ’ti, SP 5.25: yoge yoge ’ti, AVi 17c: jat.hare ’ti. 21 See e.g. AVi 92: bhadr¯ a ’ti, AVi 94 ´sr.ng¯ ˙ a ’ti. 22 See e.g. DP 9.20b: dadhikr¯ abn.o ’ti. 23 See AVi 93c: k¯ an.d.a ¯t k¯ an.d.a ¯’ ’ti.

xxiv

Early Tantric Vais.n.avism

shortened word. For instance, SP 7.31b: ´s¯arnga’, ˙ AVi 25b: bak¯ara’, AVi 71d: j¯ıva’, DP 1.1b: brahm¯a’, and DP 13.4b: n¯asik¯am¯ urti’. Change of gender. There are two types of such deviations: the first is a simple change of the formal gender of a noun24 and the other is the usage of attributes of one gender with the nouns of another gender.25 Irregular inflection. A range of irregularly inflected verbal and nominal forms are found in these texts.26 There are also some denominative verbs which have been irregularly formed.27 Irregular compounds. We encounter a few such irregularly formed compounds which are not allowed in the standard high register of Sanskrit.28 Incorrect usage of cases. There are some cases where the nominative case is used in the sense of the locative,29 the genitive in the sense of the ablative,30 and the locative in the sense of the dative.31 There are a few rare cases of interchange of the locative and instrumental (e.g. SP 5.5), some of which could perhaps be justified by a grammarian. Likewise attested in these texts is the interchange of the nominative and accusative cases (e.g. SP 6.3-4, DP 10.3-5) which basically results from the blending of active and passive voices. Irregularity of number. A few cases of incongruity of numbers are also encountered. At least once the nominative plural is treated as singular.32 The N¯ asatyas are twins and should be in the dual but our text uses the plural for them.33 In another case a predicate in the singular is applied to a subject in the dual.34 24

See e.g. SP 1.2a: devataih. for devat¯ abhih., SP 5.21a: pat¯ ak¯ ani vicitr¯ an.i for pat¯ ak¯ a vicitr¯ ah., DP5.3a naks.atr¯ a´s ca for naks.atr¯ an.i, DP 8.11a: m¯ ale for m¯ al¯ ay¯ am. 25 See e.g. SP 5.27a: d¯ıpam¯ al¯ a vicitr¯ an.i for d¯ıpam¯ al¯ a vicitr¯ ah.. 26 See e.g. SP 5.28b: dh¯ upitv¯ a, AVi 89d: gr.hn.et, AVi 111b: n¯ am¯ an¯ am, DP 8.22b yaj˜ napum . sam . , DP 10.11a: hanati, 10.11c: hanate. 27 See e.g. DP 12.11b pratis.t.hayet, 13.1↵: several occurrences of sandhayet. 28 See e.g. SP 8.12d: daityagandharva-r-¯ apsar¯ ah., SP 8.51b: yath¯ avitt¯ anus¯ aratah., AVi 48b: devadevajagatpateh., DP 3.5d: ekadvitricatuh., DP 10.1d: v¯ ajimadhyandinam . , DP 11.3: n¯ asikopari. 29 See e.g. SP 5.2: pratipac ca dvit¯ıy¯ ay¯ am ncamy¯ am . pa˜ . tu trayoda´s¯ı, AVi 43d: daks.ine caran.au nyaset, AVi 44b: p¯ adau-m upari kalpayet. 30 See e.g. DP 2.5ab: yasya sarv¯ an.i bh¯ ut¯ ani. 31 DP 9.7d: dhenum at tu sth¯ apake. . dady¯ 32 SP 2.8c=DP 2.6c: naran¯ ar¯ıprakart¯ aras tena n¯ ar¯ ayan.ah. smr.tah.. 33 DP 8.11b n¯ asaty¯ ah. for n¯ asatyau. 34 See e.g. SP 1.16a: katham os.t.hau tu kartavyam.

Introduction

xxv

Consonant-stems treated as vowel-stems. Nominal stems ending in s are most of the time treated as vowel-ending stems after dropping the final s.35 However, there are exceptional cases in compounds where the actual s-stem words are used.36 In one case, a consonant-stem is turned into a vowel-stem by adding an a.37 Elliptical syntax. Sometimes the syntactical structure of a verse is so elliptical that without supplying extra words it is not possible to construe the sentence naturally.38 Incorrect usage of lyap with a simple uncompounded root, too, is attested. This phenomenon appears to be a widespread feature in Tantric and even Puranic texts, but in our texts it occurs only once.39 There are also cases of interchange or confusion of cardinal and ordinal numbers.40 These texts also contain, but very rarely, irregular word-formations such as yath¯avitt¯anus¯aratah. (SP 8.51b) and s¯annidhyakaran.¯ arth¯ aya (DP 9.27c)—the regular s¯annidhyakaran.¯aya or -karan.¯ artham would have been unmetrical—and some clever ones like kuyastam ˙ (DP 11.40), formed to avoid metrical . (DP 7.23c) and ´sr.ngi problems. These kinds of formation are rather frequently found, especially in Tantric but sometimes also in Pur¯an.ic texts.

The Significance of these Texts The first text, the Sv¯ ayambhuvapa˜ ncar¯ atra, is mentioned and actually paraphrased in the Paddhati of Soma´sambhu (the second half of the 11th century CE), in the chapter on the consecration of an image of Vis.n.u.41 Just the name of this text appears in the Agnipur¯an.a (39.4) 35

See e.g. DP 3.25a vaks.a ¯d for vaks.asah., DP 7.14c: ´siropadh¯ ane for ´sira¨ upadh¯ ane, DP 8.15b d¯ıpajyotim . for d¯ıpajyotis.am . , AVi 55c: ure for urasi, AVi 58c: ´siram for ´sirasam .. 36 See e.g. DP 9.25a: ´sirah.sn¯ anam ayine. . , DP 9.28d: brahmatejonap¯ 37 See DP 7.6c: r.tvij¯ ah., DP 10.1b: r.tvij¯ an¯ am . . 38 See e.g. SP 5.9: nad¯ısam aluk¯ am apayet, DP 4.4b: h¯ın¯ ang¯ ˙ a . gamat¯ırthes.u b¯ . tatra d¯ maran.am . bhavet. Even more elliptical and problematic syntax is found in SP 1.16-19 and 2.12-13. 39 See DP 2.25 ku´sa ¯m . s t¯ırya. 40 See e.g. DP 3.5d-6d, AVi 18a. 41 On the identity of this text with the Svayambh¯ upa˜ ncar¯ atra of Soma´sambhu, see Sanderson 2009: 62, fn. 68. As he remarks, the exact “title Sv¯ ayambhuvapa˜ ncar¯ atra appears nowhere in the surviving folios but is reconstructed here from the analytic

xxvi

Early Tantric Vais.n.avism

when it presents a list of 25 Pa˜ ncar¯atras before launching into a section that covers several chapters and deals with the consecration of Vais.n.ava deities. The core components of the ritual system of the Sv¯ ayambhuvapa˜ ncar¯ atra can be traced back in the Ni´sv¯asa Guhyas¯ utra (15.227–233),42 ´ agama, the the latest and bracketing layer of the earliest known Saiv¯ Ni´sv¯ asatattvasam a . In the choice and ritual application of Vedic hymns . hit¯ (cf. 7.34–39), too, our text stands close to the Ni´sv¯asaguhya (cf. 2.77– 80). A verse in praise of the twelve-syllable mantra found in both of our Pa˜ ncar¯ atras is traced in the Brahmapur¯an.a, and another found only in the Dev¯ amr.tapa˜ ncar¯ atra is found in the Vis.n.upur¯an.a (see fnn. 77 and 143). A few lines of the Sv¯ ayambhuvapa˜ ncar¯ atra (8.22cd–23) from the context of deposition of precious stones, minerals and so on match with those of the Sarvaj˜ n¯ anottara (9.110cd–111). A large part of the Sv¯ ayambhuvapa˜ ncar¯ atra is reintroduced in the Dev¯ amr.tapa˜ ncar¯ atra in a slightly edited and rearranged form. Most of the time, the latter text follows the former, borrows from it, and acknowledges this fact clearly. These texts in their extant forms overlap with each other in large portions, but each also provides material that is not shared among or has been lost from the others. Thus certain elements missing in one text may be recovered from another. For example, the third and fourth chapters are entirely missing in the Sv¯ ayambhuvapa˜ ncar¯ atra, but there is a chance that, as in the case of the first and second chapters, they are the same as in the Dev¯ amr.tapa˜ ncar¯ atra, because it is highly probable that the latter has borrowed them from the former, especially since the fifth and sixth chapters of the Dev¯ amr.tapa˜ ncar¯ atra are to be found in the As..t¯ ada´savidh¯ ana and Sv¯ ayambhuvapa˜ ncar¯ atra respectively.43 These two Pa˜ ncar¯atras are unique as they insist on the identity of the three deities, Brahm¯a, Vis.n.u and Mahe´svara. This phenomenon is seen in the second half of the N¯ar¯ayan.¯ıya section of Mah¯ abh¯ arata,44 but hardly in any Vais.n.ava Sam . hit¯as. Both of these texts use a fair number of Vedic mantras and s¯ uktas, particularly in Pratis.t.h¯a rituals, as if they were perhaps consciously outdoing the Ni´sv¯asaguhya in this respect. In addition, the Dev¯ amr.tapa˜ ncar¯ atra (12.2–8) introduces a man.d.ala ten lines equivalent seen in the colophon of the eighth Adhy¯ aya.” 42 See the next section for the analysis of the ritual system. 43 On the possible relationship of the Dev¯ amr.tapa˜ ncar¯ atra and As..ta ¯da´savidh¯ ana, see below pp. lii–liii. 44 Cf. e.g. Mah¯ abh¯ arata 12.328.19–26.

Introduction

xxvii

of which representing the ten incarnations of Vis.n.u are equated with ten Vedic mantras from ten man.d.alas of the R . gveda. The same set of mantras ´ankh¯ is found in the S¯ ˙ ayanagr.hyas¯ utra in the context of the up¯akaran.a ritual.45 There is however some di↵erence in the use of this set of mantras. In the Gr.hyas¯ utra the first and last prat¯ıkas stand for one stanza and the rest for two stanzas, but our text does not tell us such a thing. Next, our text uses the last mantra of the R utih., as a unifying . gveda, sam¯an¯ı vah. ¯ak¯ entity, but the Gr.hyas¯ utra does not use it; and of course in the account of the Gr.hyas¯ utra, none of these mantras is said to be related to Vis.n.u in any way. When it comes to the issue of the deity’s images, the Dev¯ amr.tapa˜ ncar¯ atra prescribes the veneration of the set of five images (pratim¯apa˜ ncaka) of Narasim aha, V¯ amana, Trivikrama, and V¯asudeva in Vi´svar¯ upa . ha, Var¯ form (cf. 4.6). As mentioned above, it is possible that the whole of the fourth chapter of the Dev¯ amr.tapa˜ ncar¯ atra, like many others, comes from the Sv¯ ayambhuvapa˜ ncar¯ atra 46 and that this component originally belongs to the latter. In any case, this unique component is found neither in the extant Sv¯ ayambhuvapa˜ ncar¯ atra nor in the As..t¯ ada´savidh¯ ana, and apparently, it has not managed to survive in later texts. One very important piece of information, however, I have found in Srinivasan 1979 (pp. 45– 46). She has described “an interesting Vishnu-Caturmukha” image in the Gwalior Museum and published photographs of all five sides of it. According to Srinivasan, this image “comes from Badoh and dates approximately to the eighth century a.d.” “One side shows a four-armed, wellornamented Vishnu seated on Garuda. . . . On the three other sides, continuing counterclockwise, are representations of Trivikrama, Narasimha, and Varaha; a lotus configuration is carved on top.” As I can see in one of the photographs she has published, the panel of Trivikrama includes the dwarf V¯ amana. If we count him separately, we have here all five divinities listed by our text. It is also interesting that the lotus on top is depicted with eight petals as prescribed in our texts. But in this com45 ´ankh¯ Cf. S¯ ˙ ayanagr.hyas¯ utra 4.5.3–9: . . . pratyr.cam ad iti haika a ¯huh.. . vedena juhuy¯ s¯ ukt¯ anuv¯ ak¯ ady¯ abhir iti v¯ a. adhy¯ ay¯ ars.ey¯ ady¯ abhir iti m¯ an.d.ukeyah.. atha ha sm¯ aha kaus.¯ıtakih.. agnim ¯ıle purohitam ity ek¯ a. kus.umbhakas tad abrav¯ıt, a ¯vadams ˙ tvam . ´sakune bhadram ¯ a vada, gr.n.a ¯n¯ a jamadagnin¯ a, dh¯ aman te vi´svam . bhuvanam adhi´sritam ., gant¯ a no yaj˜ nam niy¯ ah. su´sami, yo nah. svo aran.ah., prati caks.va vi caks.va, a ¯gne y¯ ahi . yaj˜ marutsakh¯ a, yat te r¯ aja˜ n chr.tam ah.. tac cham ¯ vr.n.¯ımaha ity ek¯ a. . havir iti dvyr.c¯ . yor a 46 About the lost third and fourth chapters of the Sv¯ ayambhuvapa˜ ncar¯ atra, see below pp. xxxviii–xxxix.

xxviii

Early Tantric Vais.n.avism

position and the pentad of the Dev¯ amr.tapa˜ ncar¯ atra, the order of Var¯aha and Narasim . ha is di↵erent. This composite image otherwise closely corresponds to the pentad of the Dev¯ amr.tapa˜ ncar¯ atra. But our text in all probability is not talking about one composite image but five separate images. Furthermore, the Dev¯ amr.tapa˜ ncar¯ atra describes V¯asudeva as Vi´svar¯ upa and the teacher of the world and, in the immediately following text, gives instructions to make the image with two, four or eight arms. At this place, the text does not speak of Garud.a, but from the second last chapter we know that the Vis.n.u of this text mounts on Garud.a. These Pa˜ ncar¯ atras, including the As..t¯ ada´savidh¯ ana, are special in yet another respect: they focus on the all-pervasive nature of Vis.n.u and the application of this concept in the ritual in the form of a ny¯asa.47 In this process, all kinds of living beings, cosmic elements, and Vedic as well as non-Vedic divinities are incorporated in Vis.n.u’s body, and he is designated as ‘One consisting of all deities’ (sarvadevamaya). I find that this phenomenon is typically Vais.n.ava, something coming from the earlier layer ´ of Vais.n.avism before it was reformed along Saiva lines, because this concept is found highlighted also in the Bhagavadg¯ıt¯a. Although any deity might be characterised as all-pervasive, all other deities are not mapped onto its body. It is interesting that this component disappears when we come to the Jayottaratantra, the Jay¯ akhyasam a and other subsequent . hit¯ Sam hit¯ a s. . There are further noteworthy features in both of these Pa˜ ncar¯atras that indicate that at an earlier stage Vais.n.avas had close connections with ´ contemporary tantric as well as non-tantric cults, especially the Saivas. I will analyse them occasionally as I narrate the contents of these texts. 47

The section of this sarvadevamaya ny¯ asa is not found in the extant portion of the Sv¯ ayambhuvapa˜ ncar¯ atra, but it contains a clear instruction in the seventh chapter (verse 32) to the e↵ect that one performs it after worshipping the three retinues. It is found in the other two texts, the As..ta ¯da´savidh¯ ana and Dev¯ amr.tapa˜ ncar¯ atra. Now we can either imagine that the Dev¯ amr.tapa˜ ncar¯ atra knew the As..ta ¯da´savidh¯ ana and borrowed from there the section in question, or we may postulate that the Sv¯ ayambhuvapa˜ ncar¯ atra itself had it, even though it is not found in our incomplete manuscript. This second option seems more plausible to me, because there appears to be an internal reference to it in the Sv¯ ayambhuvapa˜ ncar¯ atra, as I have just indicated. In any case, we have evidence for this element in all three texts.

Introduction

xxix

The Ritual System these Texts Teach The main deity of both of these texts is a single-headed Vis.n.u. According to the Dev¯ amr.tapa˜ ncar¯ atra (4.8), he can be depicted with two, four, or eight arms, but no detail is provided there. Although this chapter is missing in the Sv¯ ayambhuvapa˜ ncar¯ atra, we can find there a so-called dhy¯anamantra, towards the end of the whole text, which depicts the deity as four-armed and holding conch, discus, and mace, but it is silent about the fourth object.48 This is interesting in the light of the widespread early convention that the fourth object may be a ball or lump of clay representing the earth (see Gail 2009 and Goodall 2011). Nevertheless, both texts list all eight weapons as they are supposed to be installed in the s¯adhaka’s body and need to be worshipped in one of the retinues of the deity. So, it is possible that these texts are mainly concerned with the eight-armed form of Vis.n.u. The eight-armed form of Vis.n.u appears to be well-known already in pre-Gupta times. A number of Kus.¯an.a images of eight-armed Vis.n.us from the Mathur¯ a region have survived (cf. Srinivasan 1997: 247). Besides, an inscription from Nagarjunikonda composed in a mixture of Sanskrit and ¯ ıra King V¯asis.t.h¯ıputra Prakrit and dated in 278 CE in the reign of the Abh¯ Vasus.en.a records re-consecration of an image of As.t.abhujasv¯amin, probably made of udumbara wood (cf. Sircar 1961: 197–204, 1965: 525–526). The identity of As.t.abhujasv¯ amin as Vis.n.u is beyond doubt, because the inscription opens with an obeisance to Pur¯an.apurus.a N¯ar¯ayan.a, and there are also other Vis.n.u-related artefacts obtained from the same site of archaeological excavation, such as a conch with a short inscription that reads bhagavato at.habhujas¯ amisa incised in the same script together with carvings of cakra, chatra and anku´ ˙ sa (see, Indian Archaeology 1959–1960—A review, p. 8). Coming to the early Gupta period, we find a description of Vis.n.u in his eight-armed form in one of the copperplates of Mah¯ar¯aja Bhulun.d.a of Valkh¯ a (cf. Ramesh & Tewari 1990: Grant I, pp. 1–2). This inscription also lists the weapons carried by the deity. We can likewise mention the two Vis.n.u-vikr¯anta images from Nepal which depict the deity in an eight-armed form and bear short identical inscriptions ´ of Saka 389 (467 CE) recording the donation of King M¯anadeva I (cf. 48 This visualisation is very close to the one found in the Ni´sv¯ asaguhya (5.95cd): ´sankhacakragad¯ ˙ ap¯ an.ih. p¯ıtav¯ aso jan¯ ardanah..

xxx

Early Tantric Vais.n.avism

¯ ca ¯ rya: inscription nos. 4 and 5).49 All this tells us that by the Vajra early Gupta period the worship of the eight-armed form of Vis.n.u had become a pan-Indian phenomenon. This tradition of depicting Vis.n.u in the eight-armed form continued over the subsequent centuries, but it is true that the eight-armed form of Vis.n.u is not as popular as four-armed ones. Nevertheless, a good number of images depicting this form, with minor variation in the weapons carried by the deity, can be found in situ and also in museums all over the Indian subcontinent and also in Southeast Asia. Such an image, probably coming from the Gay¯ a region of Bihar but now preserved and displayed in the National Museum, Kathmandu, which can be dated to circa 9th century (cf. Mevissen 2008), is presented in this volume as the frontispiece. This image depicts the deity holding an arrow, a sword, a lump of clay, and a mace in his left arms from top to bottom, and in his right arms in the same order, a shield, a bow, a conch, and a discus. Except for the lump of clay, shield, and arrow, the other five weapons are the same as prescribed in our texts. In place of these three, our texts have lotus, plough, and pestle. But we should remember that, as mentioned above, a lump of clay representing the earth is seen in place of a lotus in early images of both four- and eight-armed forms of Vis.n.u. Visualizations of the eight-armed form of Vis.n.u are found in Var¯ahamihira’s Br.hatsam a (58.33), the Vis.n.upur¯an.a (6.7.82–85), Matsya. hit¯ pur¯ an.a (258.7–8), S¯ atvatasam a (12.223), Vis.n.udharmottara (3.44.12– . hit¯ 13), Agnipur¯ an.a (1.49.16–20) and in other works that follow these texts. The Netratantra, too, prescribes an eight-armed form of Vis.n.u but it is very surprising that there the deity is mounted on a goat and his emblems are di↵erent. According to Ks.emar¯aja, this cult originally comes from the M¯ ay¯ av¯ amanik¯a, a lost Vais.n.ava text. All these visualizations di↵er when it comes to the list of weapons carried by the deity and the order of these weapons, and I have not found all eight weapons of our texts together in one place, neither in other literary or epigraphical sources, nor in actual images. Our texts list conch, mace, discus, sword, bow, lotus, plough, and pestle. The first five can be seen in almost all eight-armed Vis.n.u images and are also common to many of prescriptions for visual49

One of these two images is now kept in the National Museum, Kathmandu, and can actually be seen online: http://www.nationalmuseum.gov.np/gallerydetails.php?gid=11. The other one stands in situ in the area of the famous Pashupati temple.

Introduction

xxxi

ization, but not the other three. In place of a lotus, we normally have a lump of clay (cf. above paragraphs) or a gesture of peace or security (´s¯antida; cf. e.g. Br.hatsam a 58.33). The other two, plough and pestle, . hit¯ are hardly seen in the hands of eight-armed Vis.n.u. In place of these, we find arrow and shield, which make natural pairs with bow and sword respectively. Something close to our list, however, is found in three texts: the first section of the Vis.n.udharmottara, the Mohac¯ ud.ottara, and the Agnipur¯ an.a. Narrating the story of the killing of K¯alanemi, the first text relates that the demon saw Vis.n.u holding plough, pestle, mace, bow, discus, and sword before he fought with him and got killed. It does not specify how many hands the deity had.50 Thus, in this depiction six out of eight weapons found in our texts are mentioned, but the other two, conch and lotus, do not appear. The second text places mace, sword, lotus, the gesture of peace, conch, discus, bow, and shield in the arms of eight-armed Vis.n.u.51 Again, six out of eight weapons found in our texts are mentioned; this time plough and pestle are missing. The third text states that the Trailokyamohana form of Vis.n.u holds in his eight arms a discus, a sword, a pestle, a hook, a conch, a bow, a mace, and a noose. Again, six weapons from our list are attested and two, lotus and plough this time, are replaced with hook and noose.52 In the outermost retinue of the deity’s three-retinue man.d.ala, both of our Pa˜ ncar¯ atras place the eight weapons of the deity. In the second retinue are the twelve names of Vis.n.u starting with Ke´sava. The innermost retinue is in the form of an eight-petalled lotus and installed on its petals are the eight angas ˙ of Vis.n.u in mantra forms: ´siras, ´sikh¯a, kavaca, and astra, together with g¯ ayatr¯ı, s¯ avitr¯ı, netra, and the extraneous pingal¯ ˙ astra. In fact, also the hr.dayamantra, the first member of the well-known set of five ancillary mantras, is an angamantra, ˙ but still the Dev¯ amr.tapa˜ ncar¯ atra (11.21) states that Vis.n.u has eight angas. ˙ To settle this problem, the hr.daya mantra is installed in the centre and equated with Vis.n.u; actually it is more than just Vis.n.u because it is seen as a combination of 50 Vis.n.udharmottara 1.124.12: sajalajaladan¯ılam ayat¯ aks.am . padmapatr¯ . khagapatim avar¯ ud.ham anam | halamusalagad¯ abhih. ´sobhitam . kalpavr.ks.opam¯ . devadevam . varapurus.am ajeyam ´ s a ¯ r ngacakr¯ ˙ a sip¯ a n im k . . 51 Cf. Mohac¯ ud.ottara folio 8r , lines 5–6: gad¯ akhad.g¯ ambujan daks.e turyah. sy¯ at ´sa ¯ntidah. karah. | ´sankhacakradhanuh ˙ ¯ v¯ ame c¯ as.t.abhuje (em.; khet.a v¯ ame c¯ apabhuje . khet.a MS) harau k 52 Agnipur¯ an.a 1.49.19cd–20c trailokyamohanas t¯ arks.ye as.t.ab¯ ahus tu daks.in.e k cakram ˙ sam amake kare | ´sankha´ ˙ sa ¯rngagad¯ ˙ ap¯ a´sa ¯n . . . . khad.gam . ca musalam . anku´ . v¯

xxxii

Early Tantric Vais.n.avism

´ ı and Vis.n.u. This may seem odd when compared with Saiva ´ Sr¯ systems, because there the hr.daya mantra, like other ancillary mantras in the retinue, is neither placed in the centre nor equated with the main deity. In the Jayottaratantra (1.28), the second seed-syllable of the root mantra is identified as the common hr.daya mantra of all four goddesses, and thus, it enjoys a special status. But this mantra is di↵erent from the hr.daya mantra placed among the ancillary mantras, which is not the case in our text.53 We can therefore postulate that there were actually eight anga˙ mantras, not nine, when the netra mantra was not yet included in the group. But when it became included, the hr.daya mantra was pushed to the centre. Musalam

10. H

Halam

ke a

r gam

11. Padman bha

9. r dhara

12. D modara 4.

8. Trivikrama

Padmam

8. Pi gal stram

3. Kavacam 7. V mana

Astram 6. S vitr

H dayam

5. G yatr

7. Netram

1. Ke ava

1. ira

Kha gam

2. N r ya a

2. ikh 6. Madhus dana

3. M dhava

a kham

Gad 5. Vi u

4. Govinda

Cakram

The SP/DP man.d.ala with three retinues 53 Such special hr.daya mantras are found also in an early Buddhist Tantric text, the Ma˜ nju´sriyam¯ ulakalpa. In the second parivarta of this text, together with the root mantra, not only the hr.daya mantra but also the upahr.daya and paramahr.daya mantras are stated (Sˆ astr¯ı 1920: 26, 29). Here the hr.daya mantra is described as ‘applicable for all rites’ (sarvakarmika) which means that it has exceptional status. But none of these hr.dayas is equated with the main deity, and the ancillary mantras are not even mentioned here. So, it is clear that this text follows a di↵erent scheme. Nevertheless, it is worth noticing that this text, too, assigns a special status to the hr.daya mantra.

Introduction

xxxiii

The above combination of these nine mantras as one set deserves our attention. We can see that g¯ ayatr¯ı is grouped together with the original ancillary mantras already in the Ni´sv¯ asamukha (2.3cd–4ab), and by the time of the Ni´sv¯ asaguhya s¯ avitr¯ı is included in this group, and all seven mantras are regarded as one set.54 In the Paus.karap¯arame´svara55 fragments (folio 11), too, these seven are listed together, and the whole set is identified as the set of Yogap¯ıt.ha mantras. But I have not seen anywhere else all these nine together as one group, as they appear in our Pa˜ ncar¯ atras and in the As..t¯ ada´savidh¯ ana. In a separate scheme, we can find the netra mantra grouped together with the original five ancillary mantras: hr.daya, ´siras, ´sikh¯ a, kavaca, and astra. The netra mantra is appended to this set of five mantras, initially as an additional element, but gradually is regarded as a member of the set of ancillary mantras, and s.ad.anga, ˙ ‘six ancillary mantras,’ becomes a clich´e in later literature. In our texts the netra mantra is actually appended in the eighth position only after g¯ ayatr¯ı and s¯ avitr¯ı. Thus, it appears that our texts do not know the set of s.ad.anga ˙ mantras. In the ninth position these texts place an extra astra mantra, the pingal¯ ˙ astra. This mantra, in its extraneous nature and rarity, can be compared with the p¯a´supat¯astra of the K¯alottara tradition (cf. S¯ ardhatri´satik¯ alottara, Chapter 22), even though it belongs to a di↵erent set of mantras there and is not grouped with the five ancillary mantras. Beyond the texts mentioned above, we have very few texts which feature both g¯ ayatr¯ı and s¯ avitr¯ı as two mantras or goddesses.56 For example, in the Matangap¯ ˙ arame´svara (Kriy¯ap¯ada 1.69–70), these are equated with two padas of the vyomavy¯apin mantra, but they are separated from the five ancillary mantras there. G¯ayatr¯ı and s¯avitr¯ı fea54

The entire fifteenth chapter of the Ni´sv¯ asaguhya is concerned with these seven mantras. It refers to them as the mantrasaptaka and each aks.ara of these mantras is raised separately (15.1–220). Appended to this is the clear articulation (procc¯ ara) of mantras raised in that way, including the five brahmamantras raised in the thirteenth chapter. After this articulation, the number of aks.aras in each mantra is counted (15.221–244). 55 For information about the Paus.karap¯ arame´svara and its early ninth-century fragments, see Goodall 1998: xliii-xliv, fnn. 97 and 98, and Sanderson 2001: 4–5, fn. 2. 56 Originally g¯ ayatr¯ı and s¯ avitr¯ı refer to the same Vedic mantra as well as the deity associated with it. If there is any di↵erentiation, it is in visualisation. According to the Sm¯ arta tradition, the deity of this Vedic mantra is visualised as g¯ ayatr¯ı in the morning and as s¯ avitr¯ı in the noon.

xxxiv

Early Tantric Vais.n.avism

ture also in the Kiran.atantra (12.18, 14.6, 36.15) and a citation from the S¯ uks.masv¯ ayambhuva in the Mr.gendrapaddhatit.¯ık¯a.57 N¯ar¯ayan.akan.t.ha in his commentary on Mr.gendratantra (ad Kriy¯ap¯ada 1.5) mentions both of them. The Atharvapari´sis.t.a chapter on the P¯a´supata observance lists two di↵erent mantras as rudrag¯ayatr¯ı and rudras¯avitr¯ı.58 The Ni´sv¯ asak¯ arik¯ a (e.g. IFP T.150, p. 91) mentions both but just in passing.59 The V¯ın.¯ a´sikhatantra (cf. verses 108cd–109) installs both as two goddesses on the left and right sides of Tumburu, treating them as two maids. The same appears true with the eleventh chapter of the Netratantra, which draws on some similar system of Tumburu. The Kubjik¯amata mentions the two goddesses, but only in a ny¯asa (cf. 17.108, 24.21). The Picumata Brahmay¯ amala mentions them in a single line and in an ambiguous way,60 but they appear not to figure otherwise in the ritual system of this Tantra. It is also interesting to note that the feminine presence seems muted ´ ı in this relatively early phase of the Pa˜ ncar¯atra. Even though Goddess Sr¯ is incorporated, she is incorporated in the body of Vis.n.u but remains invisible; there is no image of her worshipped in this system. She is at least symbolically present in the man.d.ala for worship according to the scheme of the Sv¯ ayambhuvapa˜ ncar¯ atra (see fn. 147 below), but in the Dev¯ amr.tapa˜ ncar¯ atra she is not even symbolically present there. Nevertheless, the goddess of earth, Bh¯ u, with whom Vis.n.u is intimately connected, is located in the deity’s throne (cf. Chapter 12 of the text). This situation ´ matches with that in the early pan-Indian Saivasiddh¯ anta. It is only later that Gaur¯ı receives emphasis and later still, and perhaps only in the South, that Manonman¯ı is raised to the status of consort. A detailed ´ discussion on the gradual feminisation of the Saivasiddh¯ anta can be found in Goodall, forthcoming. Equally interesting and significant ritually is the As..t¯ ada´savidh¯ ana, which is found embedded in the Sv¯ ayambhuvapa˜ ncar¯ atra. The core of its ritual system matches with that of our Pa˜ ncar¯atras, but it has some additional components which are also unique and archaic. I will analyse 57

IFP T.1021, p.125. I am grateful to Dominic Goodall for this information. Cf. Bisschop & Griffiths 2003: 330 (Text 2.5–6). See also footnote 79 for their remark that the rudras¯ avitr¯ı mantra seems superfluous. 59 Soma´sambhu has them only in the chapter (4.6) that is concerned with the Sv¯ ayambhuvapa˜ ncar¯ atra. 60 Brahmay¯ amala 80.230cd–231ab: g¯ ayatryom aras¯ avittrim ¯strasamanvit¯ a k . k¯ . sarva´sa kap¯ ale caturo varn.n.a ¯ dras.t.avy¯ a v¯ıravandite | 58

Introduction

xxxv

them just before I summarise that text, after presenting the contents of the Sv¯ ayambhuvapa˜ ncar¯ atra.

Contents of the Sv¯ ayambhuvapa˜ ncar¯ atra Now in the following pages, in order to go some way towards compensating for the lack of an annotated translation, I will present the contents of these texts. As I mentioned while describing the manuscripts, the Dev¯ amr.tapa˜ ncar¯ atra is based on the Sv¯ ayambhuvapa˜ ncar¯ atra, and many of its chapters are identical. So I will occasionally discuss the exposition of some topics as given in both texts when presenting the contents of one of them, in order to avoid giving the same material twice.

Chapter 1: Introduction and Synopsis The first chapter of the Sv¯ ayambhuvapa˜ ncar¯ atra can be called a synoptic chapter programming the whole text in the form of a series of questions. When ¯I´svara was residing on the Mandara mountain, surrounded by V¯ırabhadra and the other gan.as, praised by Indra and other deities, kinnaras and gandharvas, Brahman approached and praised him with his thousand names, with the mantras from all four Vedas: R . g, Yajus., S¯aman, and Atharvan, and said: O great lord! You are the first of all beings, nobody else is the creator of this world. You created the whole world that passes through creation and dissolution. In the world, both forms of N¯ar¯ayan.a, quali´ fied with attributes and free of them, are venerated. I ask you, O Sarva, whether Vis.n.u is sakala or nis.kala, embodied or not embodied, subtle or tangible, manifest or unmanifest, eternal or not eternal, all-pervasive or not all-pervasive, characterised by [mantra] phonemes or devoid of those phonemes? Furthermore, I would like to know how Jan¯ardana, the lord of the world, is installed in this world. Which material is used to cast his image, what are the measurements of his body parts: head, forehead, nose, cheeks, lips, chin, arms, nails, neck, chest, nipple, navel, belly, hips, genital, thighs, knees, shanks, ankles, feet, and toes? How can his vehicle be characterised? How many arms should his images have? How are his weapons? How can Vis.n.u be indicated in the image, which is marked with all auspicious marks? How should his temple be built? How is the

xxxvi

Early Tantric Vais.n.avism

pavilion of sacrifice constructed? How is the pedestal made? How is the tape measuring the inner chamber of the temple (garbhas¯ utra)61 characterised? On which day and weekday, under which star and conjunction is the consecration ceremony held? How are the sections of the pavilion, and how is the sacrificial hollow characterised? How are the banners, flags, and parasols, chowrie, and canopy characterised? How are the various parts of the fire altar made? How should the water-jars, foods and delicacies be? Please tell me about the placing of precious metals, foodgrains, and precious stones at the place of consecration, and also about the characteristics of the consecrating teacher and about the caretakers of the installed image. The Dev¯ amr.tapa˜ ncar¯ atra adopts this chapter in its entirety, only making necessary changes in the opening portion that introduces the interlocutors. The interlocutors of the Sv¯ ayambhuvapa˜ ncar¯ atra are ¯I´svara and Brahman but in the Dev¯ amr.tapa˜ ncar¯ atra they are Brahman and Sanatkum¯ ara. In the first text the teaching takes place on the Mandara mountain, whereas in the latter it takes place on Mount Meru.

Chapter 2: On Vis.n.u’s Nature As he heard the questions of Brahman, ¯I´svara responded: O Brahman! Listen carefully to the details of Vis.n.u’s consecration.62 [This represents] the great knowledge, as the Pa˜ ncar¯atra is the scriptural teaching for the sake of acquisition of worldly things (yoga´s¯astra) and also for one’s liberation [from worldly existence] (moks.a´s¯astra).63 Rudra, Brahman, and Jan¯ ardana are one in reality: these three represent each other and are not di↵erent. So, this ‘Vis.n.u’ pervades everything in the world, moving or standing. He is proclaimed as the supreme deity in all scriptures. He 61 This interpretation of garbhas¯ utra is tentative. Two similar terms, j¯ıvas¯ utra and brahmas¯ utra are attested in Tantric texts in the sense of the central thread(s) used in the course of drawing a man.d.ala, extending or pointing to four directions. 62 This suggests that consecration is the main concern of this text. It is natural, then, that it should be cited in Soma´sambhu’s paddhati. See above, p. xxv. 63 Here moks.a is contrasted with yoga. The same opposition is to be found in the concluding portion of this chapter survived only in the Dev¯ amr.tapa˜ ncar¯ atra version (cf. DP 2.22–23, see below p. xxxviii for a summary). This can be compared with the contrast of s¯ am a (e.g. 3.2, 5.5), which are depicted . khya with yoga in the Bhagavadg¯ıt¯ there as focussed on gnosis and skillful action respectively. Moreover, both of these texts claim that the great knowledge they teach is capable of reconciling the two contrasting paths.

Introduction

xxxvii

is the foundation of everything, the overlord of everyone’s grace or disgrace, from whom all beings spring at the beginning of the eon, and where they all end up at the end. He is called N¯ar¯ayan.a in that he creates men and women and makes them glorious. He rests on S¯am . khya, Yoga, and Pa˜ ncar¯ atra. He alone is the essence of the Vedas, the P¯a´supata system, and of Dharma-related scripture. Whatever scriptural knowledge is told by deities, sages, and humans in the Pur¯an.as, all of them have N¯ar¯ayan.a as their essence.64 What happened, what happens, and what is to happen, all seasons and times, past and present, everything in the world consists of Vis.n.u. He is the R . gveda, Yajurveda, S¯amaveda, and Atharvaveda. He is venerated in all sacrifices: in all types of Pa´subandha, in the Agnis.t.oma, in the great sacrifices of Somasam nas, in . sth¯a, in the Haviryaj˜ the Agnihotra and other Dharma rites. The same reality is expressed by 65 the expressions of sv¯ ah¯ a, vas.at., and aham . brahma. Vis.n.u is the senses, mind, and the fundamental elements also. *** [The text in the Sv¯ ayambhuvapa˜ ncar¯ atra breaks o↵ here as two folios are missing but, since this chapter is identical in the Dev¯ amr.tapa˜ ncar¯ atra up to this point, we can assume that like the first this chapter was simply adopted from the Sv¯ ayambhuvapa˜ ncar¯ atra with little alteration. With this assumption, we can read the missing portion of 14 verses from the Dev¯ amr.tapa˜ ncar¯ atra. These verses are still dealing with the issue of the nature of Vis.n.u, which was raised in various ways in the first chapter.] 64

This echoes the following statement found in the N¯ ar¯ ayan.¯ıya section of the Mah¯ abh¯ arata (12.337.59, 64): s¯ am ncar¯ atram ah. p¯ a´supatam a| . khyam . yogam . pa˜ . ved¯ . tath¯ j˜ na ¯n¯ any et¯ ani r¯ ajars.e viddhi n¯ an¯ amat¯ ani vai k . . . sarves.u ca nr.pa´sres.t.ha j˜ n¯ anes.v etes.u dr.´syate | yath¯ agamam yath¯ a j˜ n a ¯ nam a n¯ ar¯ ayan.ah. prabhuh. k . . nis.t.h¯ S¯ am khya, Yoga, Pa˜ n car¯ a tra, the Vedas, and the P¯ a´supata scripture, know . that all these knowledge systems, O king-sage, are actually di↵erent opinions. . . . In all of these knowledge systems, O best king, in accord with their tradition and in accord with their knowledge, Lord N¯ ar¯ ayan.a is explained as the ultimate reality. 65

A possible reference to aham asmi (Br.had¯ aran.yakopanis.ad I.4.10), one of . brahm¯ the so-called mah¯ av¯ akyas of the Ved¯ anta system. For some reason, the Dev¯ amr.tapa˜ ncar¯ atra replaces it with a similar expression, kham . brahma, which can be found in the end of the V¯ ajasaneya Sam a (40.17). . hit¯

xxxviii

Early Tantric Vais.n.avism

*** Vis.n.u is the deity pervading everywhere, situated in every being; he enjoys being in every body in the form of the witnessing self. The sun and moon, stars as the seizing planets, everywhere in the intermediate space, the nether world, sky and heaven, in every being standing still or moving. Just as the wind, being very subtle, is not seen, though it is everywhere, in the same way Vis.n.u is everywhere but is not seen by people in the world. The non-eternal is the tangible form; that is the bodily image of the god, but the beginningless, bodiless, and beautiful is exactly the subtle that stands as both remote as well as proximate (par¯ apara). Jan¯ ardana stands as the locus and the entity placed over it; for the sake of all beings he assumes both bodily and bodiless forms. Vis.n.u, as he is full of distinctive qualities, is worshipped by all deities as the eternal god, as N¯ar¯ayan.a, Kr.s.n.a, Rudra, and ¯I´s¯ana. The one who 66 So those who are devoted to Visnu but hate ´ is lord Vis.n.u is lord Siva. .. Mah¯ adeva, they go to hell, vast and terrible. Similarly, those proud of worshipping Rudra, if they defame Vis.n.u, fall into a bottomless terrible hell. Those people who always perceive the identity of the two will acquire worldly riches (yoga) and, at the end, will achieve liberation from worldly existence (moks.a). Therefore, one should regularly worship Mah¯adeva and Hari as identified with each other. Thus is he pleased and bestows both worldly acquisitions as well as liberation. Just as the wind and ether ´ are always one, so there is no di↵erence between Ke´sava and Siva. Just as a man su↵ering from an ocular defect sees two moons, in the same way a man su↵ering from ignorance sees Rudra and Ke´sava as separate.

Lost Chapters 3 and 4 The third and fourth chapters of the Sv¯ ayambhuvapa˜ ncar¯ atra are lost, as mentioned before, with the two missing folios. Since the first folio written on both sides contains a total of 40 verses, we can say that nearly 80 verses were there in these two lost folios. If we subtract the 14 verses of the concluding part of the second chapter, we are left with 66 verses. In the Dev¯ amr.tapa˜ ncar¯ atra the third and fourth chapters together have only 43 verses, and it is certain that some verses are missing in the beginning of the next item in the manuscript, the As..t¯ ada´savidh¯ ana. So, it is logical to 66 This is the message of the second half of the N¯ ar¯ ayan.¯ıya; see Mah¯ abh¯ arata 12.328, 330, and 337.

Introduction

xxxix

allocate 43 out of the remaining 66 verses to the third and fourth chapters and assume that another 13 verses belonged to the As..t¯ ada´savidh¯ ana. This presupposes that up to the fourth chapter the Dev¯ amr.tapa˜ ncar¯ atra was adopting the text of the Sv¯ ayambhuvapa˜ ncar¯ atra with little alteration. This is likely because the contents of these two chapters are listed in the programme of the text presented in the first chapter common to both texts.67 But we cannot be certain about the contents of these chapters, particularly because in the Dev¯ amr.tapa˜ ncar¯ atra the fifth chapter is not the same as in the Sv¯ ayambhuvapa˜ ncar¯ atra.68 Nevertheless, I propose that we supply these two chapters from their counterparts in the Dev¯ amr.tapa˜ ncar¯ atra, namely, the third and fourth chapters there.

Interpolation in the Manuscript The immediately following portion of text in the manuscript is part of the As..t¯ ada´savidh¯ ana, a precept for eighteen rituals, which ends with a brief colophon stating ‘Thus, in the Pa˜ ncar¯atra, ends the precept for eighteen rituals.’ Then follows a chapter on construction and ritual preparation of the sacrificial pavilion (y¯ agaman.d.apa). Since the last part of this section is missing, we do not have a colophon, and it is circumstantial evidence that leads us to suppose this portion to be the fifth chapter. We cannot simply take the As..t¯ ada´savidh¯ ana as a sub-section of this chapter, although it is so presented in the manuscript. It is very unlikely that it should be an integral part of the Sv¯ ayambhuvapa˜ ncar¯ atra, as the ritual system taught in it conflicts with what is taught in the seventh chapter of the Sv¯ ayambhuvapa˜ ncar¯ atra. After due reflection, I have therefore separated the text of the As..t¯ ada´savidh¯ ana and placed it after the conclusion of the Sv¯ ayambhuvapa˜ ncar¯ atra. 67 Cf. Sv¯ ayambhuvapa˜ ncar¯ atra 1.12–13 and its identical parallel Dev¯ amr.tapa˜ ncar¯ atra 1.7–8. 68 The fifth chapter of the Sv¯ ayambhuvapa˜ ncar¯ atra appears in the Dev¯ amr.tapa˜ ncar¯ atra as the sixth chapter. In its fifth chapter, the Dev¯ amr.tapa˜ ncar¯ atra presents the rite of mentally associating all prominent deities with the various limbs of Vis.n.u. This matter is not found in the Sv¯ ayambhuvapa˜ ncar¯ atra, although a reference to this rite is there. Everything of this chapter is found instead in the As..ta ¯da´savidh¯ ana (verses 58–69). See also fn. 47 above.

xl

Early Tantric Vais.n.avism

Chapter 5: Consecration of the Pavilion The fifth chapter relates how the sacrificial pavilion is to be built and ritually consecrated. The pavilion should be built close by and in front of the temple where one plans to install Vis.n.u’s image. It should be ritually prepared on a day that is convenient for the consecrator and that falls on the first, second, fifth, thirteenth, or fourteenth of the bright half of a month, or on the full-moon day. Among weekdays, Monday, Wednesday, Thursday, and Friday are considered good for the ritual of consecration. As for asterisms, uttar¯as.¯ad.h¯a, uttar¯abh¯adr¯a, uttar¯aph¯algun¯ı, revat¯ı, a´svin¯ı, rohin.¯ı, hast¯ a, punarvasu, pus.ya, ´sravan.a, and bharan.¯ı are those recommended for the ritual of the installation of Vis.n.u’s image. On a chosen day, having cleaned the ground after removing cha↵, bone, hair, sand, coal, and stone, one should perform the ritual. The teacher performing the ritual should build a nice sacrificial hollow on the ground, four hands wide on both sides, while the pavilion should be ten hands on either side. The hollow must have all the auspicious characteristics, and the pavilion must be built with fresh tree branches. Then the consecrator should bring some sand from a confluence of rivers or some sacred place, and place that in the pavilion, smear its surface, and then build the altars prescribed here. The altar in the east should be rectangular, in the south crescent-moon-shaped, round in the northwest, and lotus-shaped in the north; they are four altogether.69 The archways are made five hands high from the wood of one of these trees: banyan, Ficus Glomerata, wood apple, Butea Frondosa, the holy fig, Mimosa Suma, sandal, and red sandal. As the archways are installed, they should be decorated with white clothes, Ku´sa grasses, cloth strips, and flower garlands. The eastern archway should be installed reciting the first stanza of the R . gveda: agnim ¯ıle purohitam, that in the south reciting the first stanza of the Yajurveda: is.e tvorje tv¯a, that in the west reciting the first stanza of the S¯amaveda: agna ¯a y¯ahi v¯ıtaye, and that in the north reciting the first stanza of the Paippal¯ada version of the Atharvaveda: ´san no dev¯ıh..70 69

In contrast, the Dev¯ amr.tapa˜ ncar¯ atra speaks of five altars in one place, when it relates them with the astral bodies, but in another place speaks of only four and relates them with the three Vedic fires plus the place to heap ashes. See fn. 133 below. 70 These mantras are chosen because the four archways represent the four Vedas. Incidentally, it is worth noticing that the Paippal¯ ada is the recension of the Atharvaveda known to the redactor of our text.

Introduction

xli

The consecrator should then bring waterjars, put some gold inside them, write the svastika sign on their necks with sandalwood paste, decorate them with clothes and ornaments, place them on plates filled with barley or rice grains, and locate them by the archways while reading the ¯ajighra kala´sam stanza.71 He should then cover the rows of pillars in the pavilion with fine cloths and paintings, and decorate them with flowers, wreaths, blossoms, and all types of garlands. He should also fix flags, mirrors, banners, chowries and canopies. After that, he should invite people to celebrate the occasion with great festivity: ringing bells, singing songs and auspicious chants, blowing conches and other instruments, reciting Vedic stanzas, and also, with the ritual proclamation that the day is auspicious, and with repeated cheers (pun.y¯ahajaya´sabdai´s ca). As for banners, they should be of di↵erent colours in di↵erent directions, representing the guardians of the quarters: yellow in the east for Indra, red in the southeast for Agni, dark green (´sy¯ama) in the south for Yama, dark blue (n¯ıl¯ an ˜jana) in the southwest for Nirr.ti, white in the west for Varun.a, dark-brown (kr.s.n.a) in the northwest for V¯ayu, green in the north for Soma, and a banner of all colours in the northeast for ¯I´s¯ana. In the next step, the consecrator demarcates with white and red powders the periphery of the hollow on sacrificial ground, he strews bunches of Ku´sa grass, starting from the hollow platform (vedi), in such a way that the blades point either to the east or north, as he recites the vedy¯ a vedih. sam¯ apyate stanza. Then, reciting the yoge yoge tavastaram . stanza, he creates a good layer of Ku´sa grass strewn there. In this very space he has to o↵er to Vis.n.u a bed made of cotton, silk, and fine clothes, and two beautiful pillows. He should o↵er beautiful rows of lamps, and then food and drink of various kinds: curries and sweets, milk and curd, butter, roots and fruits, delicacies of cakes, and fries dripping with ghee. The Sv¯ ayambhuvapa˜ ncar¯ atra su↵ers from at least one missing folio at this point, which leaves the chapter incomplete, but the Dev¯ amr.tapa˜ ncar¯ atra, which has adopted the whole of this chapter till now, concludes it with three more verses (this chapter is placed there as the sixth). In these three verses, it relates the following things: The consecrator should also o↵er rice-pudding, curd-rice, and gheerice to the deity. Having worshipped the deity in this way, he should place the deity in the centre of the pavilion and o↵er him the celebrated 71

To locate the Vedic stanzas and hymns, see Indices in the back of the book.

xlii

Early Tantric Vais.n.avism

Pun.d.ar¯ıka parasol. After doing all this, he can look forward to the world of Vis.n.u, where all desires are fulfilled.

Chapter 6: Characterisation of the Consecrator Only the last four verses of the sixth chapter have survived, all the rest is lost in the missing folio. It must have had approximately 37 verses, if we subtract the verses at the end of the fifth chapter from the average total of 40 verses contained in one folio. In the four surviving verses, it is recommended that one who is intact in his all limbs, looks beautiful, knows the scripture, and knows the twenty-five tattvas should be chosen as the consecrator. One should not choose a man of ignoble origin as the consecrator, or a pale-faced man, a man with one eye, or a man with a squint, a very white or black man, a man without teeth, or a man who is too fat, or bald, or who has black teeth; a man su↵ering from leprosy or some disease; a crippled, hunch-backed, ulcerated, or noseless man.72

Chapter 7: The Deity inhabits the Pavilion The seventh chapter describes the procedure of incubation of the deity (adhiv¯ asana) in the sacrificial pavilion. First, a chariot to carry the image (brahmaratha) is prepared while reciting the uta v¯ata pit¯asi nah. stanza, and while reciting the sahasra´s¯ırs.¯a purus.ah. stanza the image is carried in that chariot to the pavilion and laid in the bed already prepared (see the end of the fifth chapter above) while reciting the ´sambhav¯aya mantra.73 First, to the accompaniment of the vi´svata´s caks.ur uta stanza, the rite 72 This chapter has a parallel in the tenth chapter of the Dev¯ amr.tapa˜ ncar¯ atra. There the text does not su↵er from any damage or loss of folio. So this should probably be read together with that chapter. It should be borne in mind, however, that the text of the Dev¯ amr.tapa˜ ncar¯ atra has been slightly revised. For example, the Sv¯ ayambhuvapa˜ ncar¯ atra does not mention the followers of other systems among people unworthy to be the consecrator. In the Dev¯ amr.tapa˜ ncar¯ atra, on the other hand, Buddhist and Jaina monks and Kaulas are included among the unworthy ones. 73 The text is corrupt in the manuscript and reads sambhav¯ ayoni mantren.a s.aryy¯ ay¯ am ayeti mantren.a ´sayy¯ ay¯ am . . I have emended that to ´sambhav¯ . , taking ´sambhav¯ aya as a prat¯ıka for VS 16.41: namah. ´sambhav¯ aya. I am aware of the fact that Dev¯ amr.tapa˜ ncar¯ atra 11.24 reads sambhav¯ ayeti g¯ ayatr¯ı and the g¯ ayatr¯ı mantra of this system turns out to be om ayai namah. (see the mantras listed below). . sambhav¯ Nevertheless, I follow the first option, because all mantras immediately preceding and following this mantra are Vedic ones.

Introduction

xliii

of Sakalanis.kala is performed,74 and then in the following way the ritual embodiment (sakal¯ıkaran.a) of the deity is done. Each syllable of the twelve-syllable mantra om . namo bhagavate v¯asudev¯ aya is installed in twelve parts of the image, Touching the relevant part of the image while reciting the syllable: om . in the head, na in the nose, mo in the forehead, bha in the face, ga in the throat, va in the chest, te in the right arm, v¯ a in the left arm, su in the right thigh, de in the left thigh, v¯ a in the right foot, and ya in the left foot.75 Each of these syllables is visualised di↵erently: the syllable om . is visualised as bright as a ring of fire, na as white as cool water, mo as blue as pounded black pigment, bha is to appear like the morning sun, ga like gold, va like the white lotus, te like the flame of fire or sa↵ron, v¯ a like the Kunda flower or snow, su like a streak of lightning, de as yellow as the M¯ahendra banana, v¯ a as white as the ocean of milk, and ya looking like a ruby. According to the text, it is important to meditate on these twelve syllables because those who meditate on these syllables can find that passage which neither deities nor Yogins nor performers of Vedic sacrifices are able to find.76 At this point are given the commonly known five angamantras: ˙ hr.daya, ´siras, ´sikh¯ a, kavaca, and astra, along with four additional ones, g¯ayatr¯ı, s¯ avitr¯ı, netra, and pingal¯ ˙ astra. Here I cite these mantras in due order: 74

Neither here nor in its parallel in the Dev¯ amr.tapa˜ ncar¯ atra (9.40) is it clear what this rite consists of. The rite is to take place when the eyes of the deity incarnated in the image are already opened, but the ritual embodiment of the deity (sakal¯ıkaran.a) is not yet achieved. Does this mean that the deity is not yet sakala but in the state between sakala and nis.kala? Even so, how would the deity be rendered sakalanis.kala? Might the consecrator mentally visualise the deity in this way and worship him while reading the given Vedic stanza? 75 These twelve syllables of the Vais.n.ava mantra are placed in the divine body also in the As..ta ¯da´savidh¯ ana as well as in the Dev¯ amr.tapa˜ ncar¯ atra, but they follow di↵erent schemes. In the first, only two syllables are placed di↵erently: va in the right arm and te in the chest. In the second, however, all other syllables except the first two are placed di↵erently. For a comparison, see p. lxxxii below; see also pp. lxv–lxvi. 76 This statement in praise of the twelve-syllable Vais.n.ava mantra is found with minor variant readings in the Brahmapur¯ an.a (57.31). In the Dev¯ amr.tapa˜ ncar¯ atra, side by side with this verse (11.15), there is another verse (11.16) claiming that meditators upon the twelve-syllable mantra do not return to this world, and that verse can be found in the Vis.n.upur¯ an.a (1.6.40). It is possible, however, that these verses come to our text as freely floating verses or from some independent precept for the cultivation of this mantra. See also fn. 143 below.

xliv

Early Tantric Vais.n.avism (om um . h¯ . vis.n.ave) hr.day¯aya namah. | om . vis.n.ave ´sirase namah. | om abh¯ aya ´sikh¯ayai namah. | . padman¯ om dhruv¯ a ya kavac¯ aya namah. | . om . cakrin.e namah. | om ayai namah. | . sambhav¯ om vijay¯ a yai namah. | . om ay¯ a´sanaye namah. | . netr¯ om aghor¯ a ya namah. | .

This is followed by an instruction for the worship of Vis.n.u in the centre of three consecutive retinues. We have to infer that this man.d.ala is drawn near the newly built image lying there. The above mantras are installed on the petals of an eight-petalled lotus, except for the hr.daya mantra, which is placed in the pericarp. It is equated with the godhead, and worshipped 77 ´ ı, who is represented by the seed-syllable h¯ there together with Sr¯ um . . In the second retinue placed on a twelve-spoked wheel drawn outside this lotus, twelve names of Vis.n.u are worshipped and equated with the syllables of the twelve-syllable mantra: Ke´sava in the east with the syllable om ., N¯ ar¯ ayan.a with na, M¯ adhava with mo, Govinda with bha, Vis.n.u with ga, ´ ıdhara Madhus¯ udana with va, V¯amana with te, Trivikrama with v¯ a, Sr¯ with su, Hr.s.¯ıke´sa with de, Padman¯abha with v¯ a, and D¯amodara with ya. Then in the third retinue, the eight weapons of Vis.n.u are installed: conch in the east, mace in the southeast, discus in the south, sword in the southwest, lotus in the west, plough in the northwest, pestle in the north, and bow in the northeast. At this point, the text enjoins the installation of all deities in the body of Vis.n.u, but no detail is given here.78 Then the fire sacrifice is performed in the Vedic way, while the image is apparently left lying in the bed facing the temple. A R . gveda priest recites 77 As shown in the list above, the hr.daya mantra is damaged: altogether four aks.aras in the beginning are lost. The first syllable must be om . and the following the style of the other mantras, we can say that the remaining three aks.aras must contain some name of Vis.n.u in the dative. For example, vis.n.ave is a plausible candidate. Now when the text speaks of worshipping the hr.daya mantra as the godhead, together ´ ı, this syllable is extra, added to the hr.daya with the seed syllable h¯ um . representing Sr¯ mantra. A similar combination of Vis.n.u’s name with this syllable, om um . h¯ . vis.n.ave namah., can be seen in the As..ta ¯da´savidh¯ ana in the svarab¯ıjany¯ asa (see below, p. lxii). 78 Such detail is found in the twelfth unit of the As..ta ¯da´savidh¯ ana and the fifth chapter of the Dev¯ amr.tapa˜ ncar¯ atra. See also, fn. 47 above.

Introduction

xlv

´ ıs¯ the Sr¯ ukta, P¯ avam¯ anas¯ ukta, N¯asad¯akhya, V¯aks¯ ukta, Vr.s.¯akapis¯ ukta, and Aindr¯ agnas¯ ukta, and makes oblations to the fire in the altar at the eastern corner. A Yajurveda priest recites the Bhadra, Purus.as¯ ukta, ´ adhy¯ ´ Slok¯ aya, Sukriya, Br¯ ahman.a, Pitr.s¯ ukta, and Maitras¯ ukta, and makes oblations to the fire in the southern altar. A S¯amaveda priest recites the following s¯ amans: Devavrata, Purus.agati, Jyes.t.ha, Rathantara, Bh¯arun.d.a, G¯ ayatr¯ı, as well as two further s¯amans, in a row,79 and the V¯amadevya s¯ aman. Then he makes oblations into the fire in the western altar. Similarly, an Atharvaveda priest recites the Atharva´siras, Skambhas¯ ukta, Bhadrakalpas, Paris.ads, Pavitras, Gan.ap¯at.has, N¯ılarudra, and †netrayo †, and makes oblations to the fire in the northern corner.80 The number of oblations should be twelve thousands, or eight, four, three, two, or one. Two long pieces of wood (k¯an.d.a) are moved81 while reciting the mantra agnim¯ urdh¯ a divah.; the tasks of furrowing the altar ground and sprinkling are done while declaiming the agnim utam . d¯ . puro dadhe mantra. Then the life-cycle rites of the fire are performed. The rite of garbh¯ adh¯ ana is performed reciting the sup¯ utap¯ uh. mantra, that of pum savana reciting the svasti na indrah stanza, and that of s¯ımanta . . reciting the ´sam a mah¯ am indram . s¯ . mantra. The rite of j¯atakarman is carried out doing pr¯ an.¯ ay¯ amas, the rite of n¯amakarman is done reciting the sv¯ ah¯ a mantra, and that of nis.kr¯ama reciting the yaj˜ n¯ayaj˜ na mantra. The rite of annapr¯ a´sana is performed reciting the es.a viprair abhis.t.utah. mantra, and c¯ ud.opakaran.a reciting the jyes.t.ha mantra. Then the rite of vratabandha is performed reciting the yathem¯am . v¯acam . mantra, the rite of sam¯ avartana is done reciting the ¯akr.s.n.ena rajas¯a mantra, and that of patn¯ısam . yojana reciting the tryambaka mantra. After these sacraments, s¯ adhakas should perform Agnihotra and other sacrificial acts in a proper manner. All these oblations should be made using the prescribed mantras suffixed with sv¯ ah¯ a. At this point, o↵erings to the mother goddesses, Yaks.as, and the ghostly beings should be made. Meat, turmeric 79

This interpretation is far from certain. The text simply reads g¯ ayatr¯ım . caiva cottare, and I take uttare in this expression as an accusative dual and interpret that as standing for two further s¯ amans. 80 The Ni´sv¯ asaguhya, too, enumerates the hymns of four Vedas to be recited in the Pratis.t.h¯ a ceremony. There the hymns from the R . gveda are all di↵erent, those from the Yajurveda and S¯ amaveda are the same, and those from the Atharvaveda are mixed. 81 Is this a statement about the churning of fire?

xlvi

Early Tantric Vais.n.avism

powder, rice flakes, curd and groats, these make the bh¯ utak¯ ura o↵ering; this should be o↵ered to the ghostly beings. For the fulfillment of all desires, this ceremony of incubation of the deity should be performed for five nights, three nights, or a full day and night.

Chapter 8: Image Installation The ritual of image installation is explained in this chapter. It begins with the deities in the outer periphery, the guardians of the ten quarters. It gives a Vedic mantra together with an invocation verse for each of these deities, which are used to invite and install them in the sanctity. Here I relate the prescribed Vedic mantras and the guardians of the quarters, with their respective quarters in brackets: tr¯ at¯ aram indram agnaye tv¯ a yam¯ aya tv¯ a ayam ¯ gauh . . pa˜ nca nadyah. yaja dev¯ a yuje vahe ¯ı´s¯ av¯ asyam idam . abhi tv¯ a ´su ¯ra namo ’stu sarpebhyah. bh¯ ur bhuvah. svah.

Indra (E) Agni (SE) Yama (S) Nirr.ti (SW) Varun.a (W) V¯ayu (NW) Soma (N) ¯I´s¯ana (NE) Ananta (Nadir) Brahman (Zenith)

The description of Brahman, the guardian of the zenith, is followed by a verse which lists deities, ancestors, demons, mother goddesses, Siddhas, and similar divinities. Perhaps the text is saying that these lesser divinities encircle the deity’s abode, without allocating them to fixed points in the quarters. We have to resort to this guess because the next folio is missing, if not the next two, and the text is discontinuous. When the text resumes, it is speaking about the deposition of various precious stones, minerals, seeds, and so on, together with gold, perhaps representing all metals, in the place in which the image is going to be installed permanently. The text very much reminds us of the seventeenth unit of the As..t¯ ada´savidh¯ ana, but it is not the same. We are told that yellow orpiment should be placed in the eastern corner and red arsenic in the southeast. Similarly, the black mineral (a˜ njana) should be placed in

Introduction

xlvii

the southern corner, pyrite in the southwest, and green sulphate (k¯as¯ısa) in the west. Since text has been lost in the damaged edge, we do not know which minerals are placed in the other three corners. The text further enjoins the placing of the best of all metals and seeds. First of all, two herbs known as R . ddhi (prosperity) and Vr.ddhi (growth) are placed in the middle.82 First, rice seeds along with gold should be placed in the eastern corner. Black and green beans are placed in the southeast corner, barley, wheat, and red lentil seeds are placed in the south, while silver and sesame seeds are placed in the southwest corner. To achieve all desired goals, a piece of crystal should also be placed in the western corner. Here the text is again lost in the damaged edge, and the statement remains incomplete. The same is true for the description of the northwest corner. In the north corner, as the text prescribes, one should deposit white mustard seeds and a ruby, and in the northeast corner an emerald. In the centre, iron and a sapphire should be placed. However, in case of scarcity of these materials, instead of all minerals one should just deposit yellow orpiment. If the various required kinds of seeds are not available, one should place there a grain of rice instead, and a diamond (vajra) if the various required precious stones are not available. The text is again briefly interrupted because of the damaged edge, but it is clear that after the deposition of these materials, the sacrificers assisting the consecrator put the image in a chariot or palanquin, worship it, o↵er gifts in the deity’s name, and bring it into the temple (until now it was lying on a bed outside the temple). Then the consecrator o↵ers the welcome arghya to the deity, sanctifies the pedestal with the deva savitah. stanza, and o↵ers a comfortable seat while reciting a nonVedic mantra which is incompletely preserved.83 The text at this point is somewhat obscure, but it appears that the consecrator next begs the deity to receive the daily worship o↵ered by temple priests on behalf of the hosts. Then the text enumerates the rewards: by worshipping Vis.n.u in this way, a king achieves special achievements, enjoys a long life, and does not face any adversity. The place where Vis.n.u is worshipped shines beautifully, and order is established there. Then inside the temple, a few new rituals are performed. First, the 82

These two herbs appear in this context also in the Ni´sv¯ asaguhya (2.65). I cite the mantra here: om n¯ anavaihr¯ agyai´svaryai [approxi. parame´svara dharmaj˜ mately 22 aks.aras lost] sthir¯ a bhavantu. 83

xlviii

Early Tantric Vais.n.avism

deity is urged with the pr¯arthanamantra.84 The next thing in the text is a non-Vedic sth¯ apanamantra,85 which is followed by dhruve naks.atre, seemingly a prat¯ıka of a Vedic mantra that I am unable to locate. With these two mantras is performed the sthitiny¯asa that ritually fixes the image. The text prescribes that, after this act, food and other o↵erings are made, and people prostrate in front of the deity following the ordained prescriptions. Then the rites of punn¯ama and rudrap¯ada86 as well as the pacificatory rite should be performed. For the sake of the latter, on the fourth day of consecration, by day and night, oblations are made into the sacred fire. Again because of the damage to the edge of this folio, we do not know any other details about this sacrifice. The next rite appears to be that of the deity’s ablution. For this purpose, using perfumed powders, a lotus with pericarp and filaments should be drawn. There should be the worship of the deity associated with the array of nine (navavy¯ uha).87 The guardians of the quarters and the astral bodies should be worshipped situated in all eight directions. Then, in the middle of that lotus, brahmins reciting Vedic hymns should place a jar full of water. With five Vedic mantras dedicated to Varun.a, precious stones, metals, and seeds should be placed in that jar. Likewise, the leaves of the mango and of other trees, as well as perfumes, flowers, and garlands, should also be placed. With the water of that waterjar prepared in this way, the deity should be bathed; he should then be o↵ered flowers 84

Only the last word of this mantra, hk¯ırtiivardhanah., is preserved. I assume that 19 aks.aras were there before this word. 85 This time the mantra is intact: om aya parames.t.hy¯ atmane . namo namah. param¯ namah.. This mantra reminds me of the first of the five upanis.anmantras found in the mantrako´sa chapter of the Brahmar¯ atra section of the Sanatkum¯ arasam a (p. 93): . hit¯ om aya parames.t.hy¯ atmane namah.. . s.aum . namah. param¯ 86 I do not know what kind of rites these two are. 87 I assume that navavy¯ uha is another name for the core of the deity’s man.d.ala in the form of an eight-petalled lotus with the pericarp as the centre. The same lotus was worshipped before as part of the sakal¯ıkaran.a rite carried out outside the temple. Then two retinues of Vis.n.u’s twelve names and eight weapons were worshipped. Now these two retinues are discarded but the guardians of the quarters and the astral bodies are worshipped around the lotus. In some other texts, however, the same designation is used to refer to the set of V¯ asudeva, Sam ar¯ ayan.a, Brahman, Vis.n.u, . kars.ana . , Pradyumna, Aniruddha, N¯ Narasim aha. This set does not appear in the extant Sv¯ ayambhuvapa˜ nca. ha, and Var¯ r¯ atra. In the As..ta ¯da´savidh¯ ana the same group of deities appears in the svarab¯ıjany¯ asa but nowhere in the man.d.ala, and the designation navavy¯ uha is never used. See also p. lv below.

Introduction

xlix

and incense to the accompaniment of the recitation of the Vedic mantras ¯a py¯ayasva and y¯ a te rudra ´siv¯ a tan¯ uh. respectively. From now on, the deity must be visualised in four ways88 and worshipped regularly. Now with the mantra om . aniruddh¯aya moks.¯atmane sv¯ah¯a, the consecrator should worship the deity’s feet, o↵er incense reading the mantra om aya gataprabhave 89 sv¯ ah¯a, and worship him with the mantra om . k¯am¯ . amr.tam¯ urtaye sv¯ ah¯ a. Then together with all brahmins involved in the ritual, everybody should ask the deity for his forgiveness, and the consecrator should recite the g¯ ayatr¯ı mantra 108 times. This g¯ayatr¯ı mantra is most probably the one given in the ninth chapter: n¯ar¯ayan.¯aya vidmahe . . . pracoday¯ at. He should then prepare the caru as ordained previously, and make oblations in the fire with the mah¯avy¯ahr.tis and all those mantras that are used in the daily obligatory ritual. He should make oblations with the root mantra, †the three . . . †,90 the eight-syllable mantra,91 and the twelve-syllable mantra. Then, according to his financial capacity, the host should o↵er the sacrificial fee (daks.in.¯a) to the officiating teacher. An elephant, a horse, or even a village may be given to him. Jewels, gold, a nice bamboo-flute, pairs of clothes, several pieces of silk, linen, and cotton should also be o↵ered to him. Then the consecrator should prostrate himself before the teacher and give him a jewel.92 For one who does so, the reward of that deed cannot be told. Even if a man simply thinks of building and consecrating a temple in this way, he will have the result 88

This might refer to the deity’s four visualisations: three in the morning, noon, and evening, and the fourth as the transcendent form of the deity. Or else, it could be an implicit reference to the Caturvy¯ uha comprised of V¯ asudeva, Sam . kars.an.a, Pradyumna, and Aniruddha. The last of these is mentioned in the mantra immediately following this instruction, which is prescribed for the worship of the deity’s feet. K¯ ama glorified in the mantra prescribed for o↵ering incense should be Pradyumna, but all four are mentioned nowhere in the extant Sv¯ ayambhuvapa˜ ncar¯ atra. However, we should remember that a large portion of text is missing in this very chapter (see p. xlvii). 89 I am not sure about the meaning of this adjective of K¯ ama: gataprabhave. This term as such can be recognised as gataprabhu in the dative. In that case, we need to interpret gataprabhu as ‘the one whose capacity has been lost,’ which might fit for K¯ ama, but the interpretation of prabhu as ‘capacity’ will be a forced one. Another possibility is that gataprabhave is an irregular form intended for gataprabhav¯ aya. Neither of these terms are attested. 90 Text damaged. 91 Although it is not cited in the text, I assume that this mantra is om . namo n¯ ar¯ ayan.a ¯ya. 92 The text is damaged here and thus a piece of information is missing.

l

Early Tantric Vais.n.avism

that all his sins are destroyed that very day. How much more must be gained by a man who actually builds a temple! Therefore, according to financial capacity, everyone should build a house for Vis.n.u.

Chapter 9: Internalisation of Vis.n.u? The ninth chapter is implausibly short. Since the eighth chapter ends towards the end of the first line of the same side as that on which the ninth begins, no folio can be missing. It is in the third line that we find a colophon stating that the ninth chapter has concluded. Since this folio is badly damaged, approximately 68 aks.aras are lost in the opening of this chapter. The first thing we have here is a prescription to visualise Vis.n.u within oneself. Then follows the dhy¯anamantra, which reads like a blend of a verse of visualisation and a g¯ayatr¯ı mantra dedicated to Vis.n.u.93 Then a mantra is prescribed for the use of o↵ering incense,94 and again another g¯ ayatr¯ı mantra together with the formula om . gaccha gaccha punar¯ agaman¯ aya is prescribed for the use of dismissing the deity until the next time. There is again a lacuna and then comes the chapter colophon followed by the post-colophon with the date of copying: Nepal Samvat 147 (1027 CE).

The As.t.¯ ada´ savidh¯ ana The As..t¯ ada´savidh¯ ana is found inserted in the Sv¯ ayambhuvapa˜ ncar¯ atra before the fifth chapter. As mentioned above, I have taken it out from the place in which it appears in the manuscript, and have presented it here after the conclusion of the Sv¯ ayambhuvapa˜ ncar¯ atra. There are good reasons to suppose that the As.t.¯ada´savidh¯ana is a text complete in itself: it is close to the tradition of the Sv¯ ayambhuvapa˜ ncar¯ atra, but it di↵ers in its teachings and seems independent. In the As..t¯ ada´savidh¯ ana, the deity is worshipped in a man.d.ala with four consecutive retinues, while in the Sv¯ ayambhuvapa˜ ncar¯ atra, in the seventh chapter, we find a prescription to worship the deity with only three retinues. Moreover, all except the innermost retinue are formed di↵erently in the As..t¯ ada´savidh¯ ana. 93

The mantra is as follows: p¯ıtav¯ asam ˙ adharam | . caturbhujam . ´sankhacakragad¯ ´sa ¯ntam an.am at k . paramanirv¯ . tan no vis.n.uh. pracoday¯ 94 It reads: om upam . jvala jvala prajvala prajvala om . dh¯ . gr.hn.a gr.hn.a.

Introduction

li

The As..t¯ ada´savidh¯ ana is based on some system that focussed on ten forms of Vis.n.u rather than on twelve. As will be clear below, in the fourth retinue of the man.d.ala prescribed for worship there, ten names of Vis.n.u, beginning with N¯ ar¯ ayan.a, are installed and venerated. Similarly, it prescribes installing in one’s body another set of ten names of Vis.n.u beginning with Ke´sava, a list arrived at by dropping two names from the widely-known set of twelve names (see p. lxv below). rdhvam uttare Madhus dana

Govardhana Vi u ne Musalam

M dhava v yave

r gam

Halam S tyak

Akr ra 4. astram 8. 6. Pi gal S vitr stram 3. Kavacam H dayam 1. ira Balabhadra

S mba

Padmam

Gada

9. Netram

Vaiku ha pa cime

2.

5. G yatr

p rve Ka gam

ikh Aniruddha

Ni a ha Pradyumna

ke a ?

gneye

Gad

a kham

H

N r ya a

Jan rdana

Cakram

nair rte

dak i e Trivikrama

Pau

ar k k a

adha

The AVi man.d.ma alaalawith fourin retinues The four-retinue as described AVi However, the As..t¯ ada´savidh¯ ana has something more significant in the second retinue of this man.d.ala. It installs eight Vr.s.n.i heroes there: Balabhadra, Ni´sat.ha, Pradyumna, Aniruddha, Gada, S¯amba, S¯atyak¯ı, and Akr¯ ura.95 These appear in the Mah¯ abh¯ arata several times, but I have not seen them elsewhere worshipped as a group in this way. It is obvious that the godhead in the centre is implicitly equated with V¯asuveda. Neverthe95

See fn. 125 below for the discussion of a textual problem involving these names.

lii

Early Tantric Vais.n.avism

less, as we see below, the core of the ritual system of the As..t¯ ada´savidh¯ ana contains the same lotus of nine ancillary mantras with the same scheme of one mantra in the pericarp and others on the eight petals: ¯ da´ ¯ na As.t.a savidha Heart mantra in the pericarp as Vis.n.u, alone; Ancillary mantras (first retinue); 8 Vr.s.n.i heroes (second retinue); 8 weapons of Vis.n.u (third retinue); Vis.n.u’s 10 names: N¯ar¯ayan.a etc. (fourth retinue).

¯ yambhuvapan ˜ cara ¯ tra 7 Sva Heart mantra in the pericarp ´ ı; as Vis.n.u, with Sr¯ Ancillary mantras (first retinue); Vis.n.u’s 12 names: Ke´sava etc. (second retinue); 8 weapons of Vis.n.u (third retinue); No fourth retinue. (SP/DP man.d.ala on p. xxxii)

It is uncertain whether the As..t¯ ada´savidh¯ ana had already been incorporated into the Sv¯ ayambhuvapa˜ ncar¯ atra, as it has been in our manuscript, before the composition of the Dev¯ amr.tapa˜ ncar¯ atra. There is very little in the way of parallels between the As..t¯ ada´savidh¯ ana and the extant Dev¯ amr.tapa˜ ncar¯ atra. The fifth chapter of the latter, which relates how all deities are deposited in the body (sarvadevamaya-ny¯asa), matches the twelfth unit of the As..t¯ ada´savidh¯ ana, whereas an exposition of this subject is not found in the Sv¯ ayambhuvapa˜ ncar¯ atra, in which it is merely referred to in a one-line prescription in the seventh chapter. But since both of our Pa˜ ncar¯atras are incomplete, we cannot conclude that the redactor of the Dev¯ amr.tapa˜ ncar¯ atra knew or did not know the As..t¯ ada´savidh¯ ana. Nethertheless, it is possible that the composer of the Dev¯ amr.tapa˜ ncar¯ atra knew a contaminated version of the Sv¯ ayambhuvapa˜ ncar¯ atra like ours. If that was the case, it is possible to say that two conflicting descriptions of the deity’s man.d.ala, one in the As..t¯ ada´savidh¯ ana section and another in the seventh chapter, necessitated revision of that conflation in the Dev¯ amr.tapa˜ ncar¯ atra by throwing o↵ the As..t¯ ada´savidh¯ ana but incorporating some of its components. This is one possibility. On the other hand, there are some details that suggest that the Dev¯ amr.tapa˜ ncar¯ atra did not know the As..t¯ ada´savidh¯ ana. First, both of these texts prescribe astrany¯asa but they do not have the same scheme.96 96

See the relevant section of the As..ta ¯da´savidh¯ ana below, and compare that with the

Introduction

liii

Second, both of these texts prescribe the allocation of Vis.n.u’s names (n¯amany¯ asa), but here again they follow di↵erent schemes and give different numbers of names: the Dev¯ amr.tapa˜ ncar¯ atra has all twelve names, whereas the As..t¯ ada´savidh¯ ana is happy with just ten. We can be sure that this choice of ten names is intentional, not a mistake, for it has another set of ten names in the outermost retinue of the man.d.ala of the deity. Thus, this is a system of tens not twelves. The As..t¯ ada´savidh¯ ana first introduces a series of ny¯asas in its first eleven units. In these ny¯ asas, it contains some remarkable material that reveals the mind of its composer. We have only the last verse of the first angany¯ ˙ asa. But from the Dev¯ amr.tapa˜ ncar¯ atra (11.20–22) we can be sure that this concerns the deposition of eight plus one ancillary mantras found first in the Sv¯ ayambhuvapa˜ ncar¯ atra (7.16). Here in the As..t¯ ada´savidh¯ ana the whole ritual process begins with this ny¯asa, but in the Dev¯ amr.tapa˜ ncar¯ atra it is seen as the final element in the rite of ritual embodiment of the image.97 In the second unit, it contains the Vais.n.ava version of the ´ five brahmamantras styled after the brahmamantras of the Saivas, which ´ were first adopted by the P¯ a´supatas and later by the Saiddh¯antika Saivas. These Vais.n.ava brahmamantras are apparently unique, for they have not been found in any other text.98 In another ny¯asa, it adopts the 38 kal¯as associated with the five brahmamantras. This time it does not invent ´ a new set but simply adopts the available Saiva one, probably from the tradition of the Svacchandatantra, which is also found adopted in other Mantram¯ arga texts. In yet another ny¯ asa, it presents Vijay¯a, Ajit¯a, Apar¯ajit¯a, Jay¯avah¯a, Jambhan¯ı and Stambhan¯ı as a group of ´saktis. I have not found an exact parallel for this set, but the first four names may remind one of the four goddesses of the V¯ ama tradition. According to the V¯ın.¯ a´sikhatantra and the tradition narrated in the eleventh chapter of the Netratantra, they are worshipped around Tumburu, on four petals of a lotus when Tumburu is situated in the pericarp. However, the names of these godsixth chapter of the Dev¯ amr.tapa˜ ncar¯ atra. 97 The ancillary mantras, altogether nine including the hr.daya mantra worshipped in the centre, play a vital role in the ritual system of our Pa˜ ncar¯ atras, as well as in the As..ta ¯da´savidh¯ ana. In all these texts, they are worshipped in the innermost circuit of the man.d.ala, and are installed in the image when it is ritually embodied. 98 See below the section on the brahmamantras.

liv

Early Tantric Vais.n.avism

desses appear in these texts with some di↵erences.99 The ritual system of the V¯ın.¯ a´sikhatantra appears simple: other than Tumburu and the four goddesses, it only has G¯ayatr¯ı and S¯avitr¯ı (cf. verses 94–109). But in the ritual system delineated in the eleventh chapter of the Netratantra, outside the lotus are four d¯ ut¯ıs and four attendants, while Tumburu is flanked by G¯ ayatr¯ı and S¯avitr¯ı, and beneath and above him are placed Anku´ ˙ sa (as Bhairava) and M¯ay¯a (cf. verses 2–11). There among the d¯ ut¯ıs 100 ´ we can find Jambhan¯ı, the next goddess of our set of Saktis. So there is little doubt that the As..t¯ ada´savidh¯ ana is importing these female deities from the V¯ ama tradition. However, we must observe that it is drawing from a system simpler and older than the source of the Netratantra and possibly newer in comparison with that of the V¯ın.¯ a´sikhatantra.101 If we accept that our text is eclectically drawing on the V¯ama tradition, we can review the previous ny¯asas and see whether similar eclectic borrowings might lie behind those too: with the deposition of the brahmamantras, the text aims to exploit the virtues102 of the Siddh¯anta tradition, 99

Jay¯ a is placed first in both of these texts, but in our text she bears a longer name Jay¯ avah¯ a and stands at the end of the row. Jayant¯ı appears in the V¯ın.a ¯´sikhatantra in place of Aji¯ta. 100 Actually Jambhan¯ı and Stambhan¯ı appear together as adjectives in the Ucchus.makalpa from the Atharvapari´sis.t.a (36,1.4: ´sive jat.ile brahmac¯ arin.i stambhani jambhani mohani hum . phat. namah.; see Bisschop & Griffiths 2007: 10) and also in many Buddhist mantras and Dh¯ aran.¯ıs, including the Mah¯ am¯ ay¯ ur¯ıvidy¯ ar¯ aj˜ n¯ı. The same two appear together with Vijay¯ a in some mantras of Vajrav¯ ar¯ ah¯ı found in the S¯ adhanam¯ al¯ a and Siddhaikav¯ıratantra. 101 The V¯ın.a ¯´sikhatantra and the source of the Netratantra di↵er also in their visualisations of Tumburu: the first depicts him with four heads, and the second with five. I conclude that there may once have been di↵erent models of Tumburu-worship available. So it is conceivable that at an earlier stage Tumburu was flanked by Stambhan¯ı and Jambhan¯ı and encircled by the four sisters. Here incidentally I would like to take ¨ rzso ¨ k (2013) has explained, fourheaded representations notice of the fact that, as To come before the five-headed ones in a systematic way, the latter reflecting probably the Saiddh¯ antika idea of five faces for the five brahmamantras. It seems to me that the four-faced Tumburu encircled by four goddesses, all of whom are related to victory (jaya), is a good match for the four-faced Vaikun.t.ha, who is surrounded with four goddesses one of whom is called Jay¯ a. Furthermore, we cannot ignore the fact that the texts teaching this ritual system of Vaikun.t.ha are named after Jay¯ a and that the justification provided by those texts for this naming convention is not particularly compelling. In this context, it is worth noting that, according to the Jay¯ akhyasam a (27.99c-100b), Jayant¯ı, Vijay¯ a, Apar¯ ajit¯ a, and Siddhi are the friends . hit¯ of Jay¯ a. 102 I feel that it is not only a suggestive tool to win the favour of the adherents of

Introduction

lv

and with the deposition of 38 kal¯as the virtues of the Daks.in.a tradition. We can see the same eclectic tendencies at work in the prescription of the deposition of S¯ am . khya/Yoga tattvas and also of Vedic mantras in the fifth and eighth units. Interestingly, this text groups together the four Vy¯ uha deities, N¯ar¯ayan.a as the fifth, and Brahman as the sixth, then Vis.n.u, Narasim . ha, and Var¯ aha (see Unit vii below). The group of these nine deities is known in other texts under the designations navavy¯ uha, navam¯ urti, and also navan¯ayaka,103 but our text does not use such a designation. Likewise, the first five, the four Vy¯ uha deities and N¯ar¯ayan.a, are collectively treated as pa˜ ncam¯ urti (cf. Soma´sambhupaddhati 4.6.35–38) and pa˜ ncatattva (cf. Garud.apur¯ an.a 1.32.1–7). Actually, in this very context Soma´sambhupaddhati (cf. 4.6.38ab) mentions the possibility of performing the ny¯asa following the other alternatives such as as.t.cam¯ urti, but here we have nine entites and none of our texts in their extent form provide any information on these possibilities. All this makes me suspect that there are two sets of deities in this group, one combining the four well-known Vy¯ uha deities with N¯ ar¯ ayan.a and then further with Brahman and the other relating Narasim ha and Var¯ aha to Vis.n.u (see Unit vii below on p. lxii for . more arguments). This second set reminds us of the three-faced form of Vaikun.t.ha. Perhaps the author of this text knows of the three-faced form of Vaikun.t.ha. In its ninth to eleventh units, the As..t¯ ada´savidh¯ ana provides separate schemes for the deposition of Vis.n.u’s twelve names, individual syllables of his twelve-syllable mantra, and his eight weapons. All these three elements can be found in the Dev¯ amr.tapa˜ ncar¯ atra, but in a di↵erent sequence and as part of the rite of sakal¯ıkaran.a.104 Here in the As..t¯ ada´savidh¯ ana, they are treated independently and the rite of sakal¯ıkaran.a is not formally ´ these Saiva schools outside the ritual sphere, but it serves as an auto-suggestion for the ritualist himself which makes him think that with these ny¯ asas he harvests all the fruits of cultivating these tantric traditions. 103 For references, see Rastelli 2007: 200–201. To these references, I add the reference of the Pratis.t.h¯ alaks.an.as¯ arasamuccaya (6.105–106), which lists the nine deities and gives information about their visualizations. See also fn. 88 above. 104 There they appear in the sequence of syllables, names, and weapons. Beyond these three, the rite of sakal¯ıkaran.a in the Dev¯ amr.tapa˜ ncar¯ atra includes the deposition of eight plus one ancillary mantras as the last element of this rite. The same appears true for the Sv¯ ayambhuvapa˜ ncar¯ atra, though it has a shorter scheme for this rite, as it does not include deposition of Vis.n.u’s names and weapons.

lvi

Early Tantric Vais.n.avism

mentioned.105 In the twelfth unit, the As..t¯ ada´savidh¯ ana explains how Vis.n.u houses all deities in his body. This unit is identical with the fifth chapter of the Dev¯ amr.tapa˜ ncar¯ atra, but the timing of teaching this secret is di↵erent in these two texts: the latter tells it immediately after the characterisation of a good image, while the former presents it only after the ritual embodiment of the image. In the thirteenth unit, the rite to put life into the image is related. This is achieved by reading a Vedic hymn. This component is significant because it is not found in either of our other two Pa˜ ncar¯atras, ´ nor in any early tantric material of the Saiva or Vais.n.ava traditions. In other accounts, I suppose, the rite of sakal¯ıkaran.a itself, which is achieved by assembling together mantra-souls and mantric elements, is enough to create the immortal body of the deity. In the fourteenth unit, ablutions are o↵ered to the deity incarnated in the image, his eyes are opened, and he is moved from the bathing place to a specially prepared bed, both in the sacrificial pavilion itself. All these components can be found in detail in the ninth chapter of the Dev¯ amr.tapa˜ ncar¯ atra, but in the Sv¯ ayambhuvapa˜ ncar¯ atra there is only a brief instruction in the beginning of the seventh chapter to the e↵ect that the image is to be brought to the pavilion and, once it is placed in the bed prepared for this purpose, the rite of sakalanis.kala is to be performed. In any case, in both of these Pa˜ ncar¯atras these acts precede the ritual embodiment of the image, whereas in the As..t¯ ada´savidh¯ ana they take place not only after various kinds of ritual embodiment of the image but also after the rite of its animation. The fifteenth unit of our precept deals with the sacraments of the fire and sacrifice in that fire. It enjoins the performance of all sacraments of the fire using the ancillary mantras of its own ritual system, and the whole section is quite brief. However, when the same component is found in the Sv¯ ayambhuvapa˜ ncar¯ atra, it is not so brief, and Vedic mantras are prescribed for all sacraments. This component is not found in the extant Dev¯ amr.tapa˜ ncar¯ atra, which is incomplete towards the end. In the sixteenth unit, the As..t¯ ada´savidh¯ ana teaches the worship of Vis.n.u in the centre of four consecutive retinues, whereas both of our Pa˜ ncar¯ atras are content with just three retinues.106 The next unit tells 105

Practically speaking, however, all ritual acts till the eleventh unit are di↵erent ways of ritual embodiment (sakal¯ıkaran.a) of the deity. 106 For the comparison of these two schemes of worship, see above pp. li–lii.

Introduction

lvii

us the way precious stones, minerals, and seeds are deposited in the place of the image, and then the final unit gives the ritual of consecratory ablution of the installed image. These components can be seen in the eighth chapter of the Sv¯ ayambhuvapa˜ ncar¯ atra, but are not available in what survives of the Dev¯ amr.tapa˜ ncar¯ atra. At this point, let us have an overview of the ritual course of the As..t¯ ada´savidh¯ ana and that of our two Pa˜ ncar¯atras107 side by side so that a comparison can be made. This will reveal that, although they have common components, they do not follow the same course: We can take the first eleven units of the As..t¯ ada´savidh¯ ana as an extended version of the rite of ritual embodiment of the newly built image. This is followed by the mental cultivation of the scheme mapping all prominent divinities in Vis.n.u’s body, and then, animation of the image. Then ablutions are o↵ered to the deity incarnated in that image, his eyes are opened, and he is moved to the sacrificial pavilion. In the pavilion, four archways are ritually installed. In the next step, the image is brought into the temple and installed there. This is followed by a pacificatory rite and the fire sacrifice. Then comes the rite of Vis.n.u’s worship in the centre of four consecutive retinues, followed by the deposition of precious stones, metals, and seeds in the place of the installed image, and finally the o↵ering of the consecratory ablution. In the scheme of the Sv¯ ayambhuvapa˜ ncar¯ atra/Dev¯ amr.tapa˜ ncar¯ atra, first of all, the sacrificial pavilion is constructed and ritually consecrated. Then the newly built image is brought there ceremonially, laid in the bed specially prepared at the time of the construction of the pavilion, and is ritually embodied with mantric elements. The deity is worshipped in the centre of three retinues, and it is meditated upon as consisting of all divinities. After a pacificatory rite, the fire sacrifice is performed, and the image is left in the pavilion at least for one night. After this, the image is brought into the temple and installed there. Finally, after o↵ering the consecratory ablution, the ceremony is ritually concluded. 107 In the Sv¯ ayambhuvapa˜ ncar¯ atra, we can have an idea of its ritual course, even though passages are missing in the middle. In fact, when the real ritual course begins in the fifth chapter, we have at least a few verses from each chapter. This allows us to track the ritual sequence of the text. In the extant part, the Dev¯ amr.tapa˜ ncar¯ atra does not deviate from the course its predecessor follows, rather it provides details of certain rituals which are said to be performed in the extant parts of the Sv¯ ayambhuvapa˜ ncar¯ atra but not described. Therefore I am speaking of the ritual course of our Pa˜ ncar¯ atras as one and the same, though the Dev¯ amr.tapa˜ ncar¯ atra is incomplete towards the end.

lviii

Early Tantric Vais.n.avism

Contents of the As.t.¯ ada´ savidh¯ ana i) angany¯ ˙ asa: In its available form, the As..t¯ ada´savidh¯ ana begins with the statement that the angany¯ ˙ asa is essential. Without installing the ancillary (anga) ˙ [mantras in the image],108 so the text tells us, no s¯adhaka can accomplish the goal. Though all of these angamantras ˙ are absent here, we know them from the context of worship, when they are venerated in the innermost retinue of the deity.109 After this statement the five brahmamantras are given, followed by an instruction about where to place them in di↵erent parts of the body. When it moves further to the next item in the ritual sequence, it identifies that as the third. I suppose therefore that the first item in the list of eighteen ritual steps was angany¯ ˙ asa and what we have now in the beginning of the As..t¯ ada´savidh¯ ana is the conclusion of that step, and that the installation of the brahmamantras is the second. ii) five brahmamantras: The Vais.n.ava version of the five brahmamantras is found only in the As..t¯ ada´savidh¯ ana. It is interesting to note that some of the South Indian P¯an ˜car¯atras give five upanis.anmantras ´ which make the limbs of the deity. This matches the idea that Siva’s body is made of five brahmamantras. However, these upanis.anmantras are elsewhere in the form of n¯amamantras preceded by seed-syllables, and not in the style of P¯a´supata mantras, which is what we find in the As..t¯ ada´savidh¯ ana.110 Since they are unique, I cite these mantras in full, and compare them 108

The text is not always clear about where the mantras are installed: in either or both the image and the consecrator’s body. Logically they could be deposited in both of them, and the Dev¯ amr.tapa˜ ncar¯ atra actually suggests this when it deals with the angany¯ ˙ asa (cf. 6.22ab). However, here in the end of the fourth unit, we find the statement that one endowed with 38 kal¯ as is proclaimed as a ¯c¯ arya. This can be interpreted in favour of the deposition of mantras and mantric elements in the consecrator’s body. However, we cannot deny the possibility that ‘one endowed with the thirty-eight kal¯ as’ can be expanded to ‘one endowed with [the knowledge of] the thirty-eight kal¯ as.’ Furthermore, towards the end of the eighth unit, we find a prescription that certain mantras are to be deposited in horses and weapons. I surmise that the horses mentioned here are those yoked to the chariot supposed to carry the image later. Therefore I assume that all of the ny¯ asas prescribed here are meant for the image. Of course, certain ny¯ asas are deposited in the s¯ adhaka’s body too, but that is not the concern of this text at this point, for the context here is consecration. 109 They are actually nine but counted as eight once the first of them, the hr.daya mantra, is equated with Vis.n.u and placed in the pericarp. For these mantras, see above, pp. xliii–xliv. 110 For these mantras, see T¯ antrik¯ abhidh¯ anako´sa III, s.v. pa˜ ncopanis.anmantra.

Introduction

lix

with the original P¯ a´supata mantras: I. om atha nar¯aya m¯am . nare nare naran¯ . narottama k II. om n¯ aya namo y¯an¯aya namo dharm¯aya [na]mah. . yaj˜ pun.y¯ aya namo vrat¯ aya namo niyam¯aya namo m¯arg¯anus¯arin.e namah. k

III. om alebhyo ’tha k¯alebhyah. k¯alak¯alatarebhya´s ca | . ak¯ sarvatah. [k¯ a](lak¯ alebhyo na)mas te rudrarudrebhyah. k IV. om aya vidmahe hr.s.¯ıke´s¯aya dh¯ımahi tan no . tatsam . yog¯ vis.n.uh. pracoday¯ at k V. bodhaka sarvavidy¯ an¯ am . devad¯anav¯adhipate mah¯apurus.a namo ’stu te k

The first mantra of this set is styled after the second-half of the first P¯ a´supata brahmamantra, which I cite here in the form in which it is found in the Ni´sv¯ asaguhya on the grounds that it is that form that is closest to our mantra:111 bhave bhave n¯atibhave bhavasva m¯am . bhavodbhava. It appears that the first-half of the P¯a´supata mantra has been discarded, unless we suppose that the scribe has made a mistake here. The second mantra above reminds one of its P¯a´supata counterpart because it follows the same pattern but is shorter. In the P¯a´supata mantra, there are eleven epithets ending in the dative, followed by namas, whereas here in the Vais.n.ava version we have only seven epithets.112 The third mantra is of equal length as its P¯a´supata counterpart: it simply replaces ghora of the original with k¯ala and changes the final word, rudrar¯ upebhyah., to rudrarudrebhyah..113 The fourth mantra above adapts the Tatpurus.a mantra114 by replacing tatpurus.a with tatsam . yoga, ¯ . yaka and P¯ Both the Taittir¯ıya Aran a´supatas¯ utra have bhavodbhav¯ aya namah. instead of bhavodbhava in the vocative. 112 For the sake of comparison I cite the P¯ a´supata mantra here: v¯ amadev¯ aya namo jyes.t.h¯ aya namah. ´sres.t.h¯ aya namo rudr¯ aya namah. k¯ al¯ aya namah. kalavikaran.a ¯ya namo balavikaran.a ¯ya namo bal¯ aya namo balapramathan¯ aya namah. sarvabh¯ utadaman¯ aya namo manonman¯ aya namah.. 113 For the sake of comparison, I present the P¯ a´supata mantra here: aghorebhyo ’tha ghorebhyah. ghoraghoratarebhyah. | sarvebhyah. sarva´sarvebhyo namas te astu rudrar¯ upebhyah. k 114 The full Tatpurus.a mantra is this: tatpurus.a ¯ya vidmahe mah¯ adev¯ aya dh¯ımahi tan no rudrah. pracoday¯ at k 111

lx

Early Tantric Vais.n.avism

mah¯ adeva with hr.s.¯ıke´sa, and rudra with vis.n.u. The fifth mantra copies ¯ the I´s¯ ana mantra.115 It is very close in style and wording, but there is a ´ big di↵erence: In the archetype mantra of the P¯a´supatas/Saivas, all designations of the lord are in the nominative, and the mantra says, ‘May such and such a lord be benevolent to me.’ But in this mantra, the sentence is di↵erently constructed and it says, ‘O the greatest person! Salutation to you who are such and such.’ The text instructs the practitioner to install the fifth of these brahmamantras in the head, the fourth in the face, the third in the chest, the second in the private parts, and the first in the feet. This is the way these brahmamantras are installed in the Siddh¯anta tradition of the ´ Mantram¯ argic Saiva system.116 So although the five brahmamantras originally belong to the Atim¯arga of the P¯a´supatas, when our text is imitating these mantras, it appears to have the Siddh¯anta in view. Let me remind the reader here that, in this and some of the subsequent units, our text has apparently incorporated more than one ritual system known at the time of its composition, and so by depositing them in the body of Vis.n.u (and perhaps also on the body of the Vais.n.ava teacher) it implicitly suggests that Vis.n.u is in the centre of all these systems. Perhaps it also intends to invite the adherents of these systems to resort to this Vais.n.ava system. iii) ´ saktiny¯ asa: The next item, the third of the eighteen vidh¯anas, is ´saktiny¯ asa. This time the six ´saktis, namely, Vijay¯a, Ajit¯a, Apar¯ajit¯a, Jay¯ avah¯ a, Jambhan¯ı, and Stambhan¯ı, are installed in the body, starting from the feet up to the head. Again, we are not told whose body this is: is it that of the deity or the consecrator. Presumably this ambiguity has the e↵ect of reinforcing the purpose of this sort of ny¯asa, namely to bring about the identification of the worshipper with the deity worshipped. iv) kal¯ any¯ asa: Next the 38 kal¯as are installed. I present them here in the order in which they appear in the text: Sr.s.t.i, Vr.ddhi, Mati, Laks.m¯ı, Here is the ¯I´sa ¯na mantra as found in the end of the fifth chapter of the P¯ a´supatas¯ utra (s¯ utras 5.42–47) and in the Ni´sv¯ asaguhya (15.226): ¯ı´sa ¯nah. sarvavidy¯ an¯ am ut¯ an¯ am | brahman.o ’dhipatir brahm¯ a ´sivo me astu . ¯ı´svarah. sarvabh¯ sad¯ a´siva k Bisschop (2006: 18), however, in his edition of the s¯ utrap¯ at.ha reads the mantra with one more word, brahm¯ adhipatir, in the beginning of the second-half of the mantra as ¯ . yaka (10.47). According to his notes, all three manuscripts found in the Taittir¯ıya Aran he used for his edition read the mantra with the extra word. 116 Cf. e.g. S¯ ardhatri´satik¯ alottara 2.10cd. 115

Introduction

lxi

Medh¯a, K¯ anti, Svadh¯ a, Sthiti, Raj¯a, Raks.¯a, Rati, P¯aly¯a, K¯am¯a, Tr.s.n.¯a, Mati, Kriy¯ a, R ay¯ a, R¯ atr¯ı, Bhr¯aman.¯ı, Mohan¯ı, M¯ay¯a, Moh¯a, . ddhi, M¯ ´anti, Tam¯a, Tr.s.n.¯ a, Ks.udh¯ a, Mr.tyu, Jar¯a, Bhay¯a, Nivr.tti, Pratis.t.h¯a, S¯ Vidy¯a, T¯ ar¯ a, Sut¯ ar¯ a, Taran.¯ı, Tarayant¯ı, and Sut¯aran.¯ı. This is nearly the same list as is found in the Svacchandatantra (1.54–59ab) and Netratantra (22.26cd–34).117 One endowed with these thirty-eight kal¯as, the text declares, is proclaimed to be an ¯ ac¯ arya, the highest-ranking teacher, who is entitled to initiate and consecrate. Sad¯ a´siva and the five brahmamantras are associated with the 38 kal¯as from earlier times, but they remain unnamed (e.g. S¯ ardhatri´satik¯ alottara 22.2–9). These names of kal¯ as appear first in the Svacchandatantra, and their inclusion in our text may be intended to represent the Daks.in.a tradition rather than that of the Siddh¯anta, because the brahmamantras are there to represent the latter. v) twenty-five tattvas: This time our text simply enumerates the 25 tattvas, without telling us where they are to be installed, in five sets of five tattvas. However, they are not exactly the same as in the S¯am . khya system. Our text makes the first set by grouping the tattvas from purus.a to manas together; the second is made of the five subtle elements, the third set has the gross elements, the fourth has the five sense faculties, but the fifth set is, surprisingly, made of the five manifestations of vital air, the pr¯ an.as. I can hazard a guess that this may represent a variety of the S¯am . khya school, or the school of Yoga. The pr¯an.as are of vital importance for Yogins, and it is possible to argue that the pr¯an.as are responsible for all bodily movements, and it is perhaps for this reason that they have replaced the five organs of action. As at the end of the previous unit, the text says that the knower of these 25 tattvas is proclaimed to be an ¯ac¯arya. vi) aks.arany¯ asa: The next unit allocates all the letters of the Sanskrit alphabet to parts of the body. Starting from the toe and going up to the belly are the sixteen vowels and they are identified as ¯atmatattva. 117 We have Mati, Tr.s.n.a ¯, and M¯ ay¯ a twice in this list. In place of the first Mati, the Svacchandatantra and Netratantra have Dyuti, and in place of the second Tr.s.n.a ¯ they have Nidr¯ a. Thus, they avoid duplication of these two, but still, in these texts as well ´ as in ours, M¯ ay¯ a is duplicated. In these two Saiva works, they are divided into five groups and assigned to the five brahmans, but not in our text. The Ni´sv¯ asak¯ arik¯ a (IFP T.17A, pp. 12–13) also has a list of thirty-eight kal¯ as but that is a di↵erent one and di↵ers from ours midway.

lxii

Early Tantric Vais.n.avism

Starting around the belly, including the hips, up to the lips and in both hands, the stops are placed; they are identified as vidy¯atattva. The rest of the aks.aras—the semivowels, sibilants, ha, and ks.a—are allocated to the cheeks and other parts of the head up to the topknot. These aks.aras are identified as ´sivatattva. This approach of looking at aks.aras in terms of ¯ atma-, vidy¯a-, and ´sivatattvas goes back to the Ni´sv¯asa corpus: Ni´sv¯ asaguhya 9.203–204 (each aks.ara seen that way) and 12.34–44 (all aks.aras grouped into three). Deposition of these three tattvas on the body has a significant place in the V¯ın.¯ a´sikhatantra (cf. verses 60–83); but neither the Ni´sv¯ asaguhya nor the V¯ın.¯ a´sikhatantra appears to be the source of this ny¯ asa in our text, because one groups the letters of the alphabet di↵erently and the other associates those groups di↵erently. vii) svarab¯ıjany¯ asa: In the next step, V¯asudeva, Sam . kars.an.a, Pradyumna, Aniruddha, N¯ar¯ayan.a, Brahman, Vis.n.u, Narasim . ha, and Bh¯ uvar¯ aha are installed from the feet to the head. All these seed-syllables (b¯ıjas) are vowels, which is why this set is named svarab¯ıjany¯asa. However, this name is a little problematic, because the set is not complete with the fifth mantra item and the rest of the mantras in this set do not take vowels as their seed syllables. They have di↵erent b¯ıjamantras or none. It is therefore possible that the last four, or even five, mantra items in this unit were incorporated later, either as additional independent items or as subordinated ones (See also p. lv above for related arguments). 118 v¯ om asudev¯aya na[mah.] | . a (om a ¯ sam . . kars.an.¯aya namah. | om . am . pradyumn¯aya) namah. | om ah aya namah. | . . aniruddh¯ om om n¯ a r¯ a yan a ¯ . . . ya namah. | 119 om . tat sad brahman.e namah. | om um . h¯ . vis.n.ave namah. | om aya namah. | . narasim . h¯ om bh¯ u var¯ a h¯ aya namah. | .

viii) vedany¯ asa: Now a number of Vedic hymns and stanzas are installed in the body. Our text claims that this vedany¯asa is taught 118 See Critical Notes to As..ta ¯da´savidh¯ ana verse 29 on pp. 118–119 for the discussion of textual problems involving these seed syllables. 119 This may remind one of Bhagavadg¯ıt¯ a 17.23ab: om . tat sad iti nirde´so brahman.as trividhah. smr.tah. |

Introduction

lxiii

120 the god of gods, and is the same in all systems: Saiva, ´ ´ by Sambhu, Vais.n.ava, Saura, and Bh¯ agavata.121 Let us now present a table of these Vedic hymns/stanzas and the places in the body to which they are allocated.

´sr¯ıs¯ ukta man.d.alabr¯ ahman.a s¯ uryas¯ aman un nay¯ ami. . . . . . mas¯ aman yenedam . jyotyakarman netran¯ as¯ı p¯avam¯ an¯ı and vir¯ aja tad ev¯ agnis tad ¯ adityah. r.cam acam . v¯ . pr¯ an.ebhyah. sv¯ ah¯ a ¯ı´s¯av¯ asyam . g¯argibr¯ ahman.a kum¯ arab¯ ala? skambhas¯ ukta purus.as¯ ukta purus.am . purus.agati = setus¯ aman varun.as¯ aman jyes.t.has¯ aman tam ¯ı´s¯ anam . v¯amato kankato ˙

topknot head forehead eyes tip of the nose right ear left ear beyond the ears sides of the face teeth tongue palate lips chin cheeks shoulders throat right breast left breast chest navel right arm left arm middle of the right arm

120 ´ Here the vedany¯ asa is said to be taught by Sambhu indeed. This suggests that ´ the source of the vedany¯ asa is Saiva, but we cannot justify this from what we know of ´ ´ Saiva sources. So the reference to Sambhu in this matter is rather intriguing. There ´ is however a possibility that our text borrows the vedany¯ asa from a Saiva text not available to us. In any case, it is obvious that this Vais.n.ava tradition considers the ´ Saiva one older. 121 It is noteworthy that here Bh¯ agavata is distinguished from Vais.n.ava. This suggests that in this text Vais.n.ava refers only to the followers of the Pa˜ ncar¯ atra system.

lxiv kanikranti. . . vid¯ a maghavan indram . naro tr¯ at¯ aram indra indr¯ a y¯ ahi ´suddhavat¯ı [broken] tryambakam . ´s¯ ambhavam . ap¯ am ukta, vikarn.as¯aman . . . . , ´sr¯ıs¯ bh¯ asas¯ aman rudras¯ aman v¯ ata ¯ a v¯ atu bhes.ajam imam v¯ a m . rudras¯ aman vibhr¯ at. br.hats¯ aman lajj¯ a dvija? devavrata, d¯ırgha ajyadoha, palalabh¯ara ¯ akr.s.n.ena, asya v¯ ¯ am yan.vaka janyakarman var¯ ahas¯ aman kadrudr¯ aya n¯ılagr¯ıv¯ a, tarat sa, p¯avam¯an¯ı rathe tis.t.han hiran.yagarbhah.

Early Tantric Vais.n.avism middle of the left arm right hand left hand right-hand fingers left-hand fingers right-hand nails left-hand nails hips/bottom spine testicles penis anus rectum hips right thigh left thigh right knee left knee ankles legs feet right toes left toes right toe-nails left toe-nails soul and limbs horses [of the chariot] weapons

ix) ten names of Vis.n.u: In the next step Vis.n.u’s ten names: Ke´sava, N¯ ar¯ ayan.a, M¯adhava, Govinda, Vis.n.u, Madhus¯ udana, V¯amana, D¯ amodara, Hr.s.¯ıke´sa, and Padman¯abha are placed respectively in the head, face, neck, arms, chest, belly, hips, thighs, knees, and feet. Surprisingly the text is speaking of twelve body parts, not ten and it does not explain how to fit the ten names in twelve places. If we count each item in the dual—arms, hips, and so on—as a single entity, as I have done, we can have ten places to take in the ten names. I may conjecture that when the body parts are independently counted, both arms and feet,

Introduction

lxv

which are far apart and function separately, are counted as two separate parts, and thus the number twelve is reached. As discussed earlier, the man.d.ala prescribed by the As..t¯ ada´savidh¯ ana for Vis.n.u’s worship has another set of Vis.n.u’s names, and again the number is ten. So we can at least be sure that the prescription of ten names for the purpose of ny¯asa here is not the result of textual corruption. In the Dev¯ amr.tapa˜ ncar¯ atra we can find a passage parallel to this (11.12–14), but that passage assigns the matching number of twelve names ´ ıdhara make the set to the twelve body parts. There Trivikrama and Sr¯ complete. It is sure that the As..t¯ ada´savidh¯ ana was not source of this section of the Dev¯ amr.tapa˜ ncar¯ atra. Probably the Dev¯ amr.tapa˜ ncar¯ atra was not here borrowing directly from our version of the Sv¯ ayambhuvapa˜ ncar¯ atra, either. Since in some cases the allocated places too are changed, I present below a comparative table. In this table I present the names in the order in which they appear in the Dev¯ amr.tapa˜ ncar¯ atra and place a serial number before each of these names to indicate the order in which they appear in the As..t¯ ada´savidh¯ ana; two names, marked with XXX, are missing in the latter: names 1. Ke´sava 2. N¯ ar¯ ayan.a 3. M¯ adhava 4. Govinda 5. Vis.n.u 6. Madhus¯ udana 7. V¯ amana X Trivikrama ´ ıdhara X Sr¯ 9. Hr.s.¯ıke´sa 10. Padman¯ abha 8. D¯ amodara

¯ da´ ¯ na as.t.a savidha head face neck arms chest belly hips XXX XXX knees feet thighs

¯ mr.ta deva head face neck arms chest back belly hip right thigh left thigh anus feet

x) twelve syllables of the dv¯ ada´ s¯ aks.aramantra: In the next step, each syllable of the twelve-syllable mantra, om . namo bhagavate v¯asudev¯ aya, is placed on a di↵erent body-part in the following order: head, nose, forehead, face, throat, right arm, chest, left arm, left thigh, right thigh, left foot, and right foot. In this step, exceptionally, the text

lxvi

Early Tantric Vais.n.avism

gives instructions for placing these syllables on body parts and then requires them to be raised again, as if they cannot be left out there because they are needed for the sake of perpetual reciting. This ny¯asa is found also in the Sv¯ ayambhuvapa˜ ncar¯ atra but there this prescription to lift the syllables up again is not found. In the Sv¯ ayambhuvapa˜ ncar¯ atra and Dev¯ amr.tapa˜ ncar¯ atra, the rite of installation of mantra-syllables on the divine body is designated as sakal¯ıkaran.a, which is not the case here. However, as far as the scheme of installation is concerned, the As..t¯ ada´savidh¯ ana is closer to the Sv¯ ayambhuvapa˜ ncar¯ atra, while the Dev¯ amr.tapa˜ ncar¯ atra adopts a completely di↵erent scheme. For details, see below pp. lxxxi– lxxxii. xi) astrany¯ asa: In this unit, eight weapons of Vis.n.u are located in di↵erent segments of the image: the sword is placed in the chest, the bow in the head, the club and plough in the two arms; the discus covers the area of the abdomen, the abdomenal cavity and the back, the conch covers the genitals, the mace covers the anus, hips, thighs, and knees, and the lotus covers the ankles and feet. xii) all deities in Vis.n.u’s body: Next the text explains how Vis.n.u consists of all deities, but does not give any instruction for ny¯asa. I think that the s¯ adhaka is supposed to cultivate these equations in his mind. Brahman is his head, Vanaspati makes his hair, the sun and moon reside in his eyes, the eleven Rudras garland his neck, the asterisms and astral bodies shine as his teeth, dharma and adharma form his lips, Sarasvat¯ı, the goddess of all scriptures, resides in his tongue, and g¯ayatr¯ı and s¯avitr¯ı in his nostrils. Likewise, the directions and sub-directions are situated in his ears, and the hundred Rudras inhabit the middle of his eye-brows. The sages are in his hair-follicles, while Yaks.as, Kinnaras, Gandharvas, Pi´s¯ acas, D¯ anavas, R¯ aks.asas and the like are in his abdomen. All rivers are in his arms and serpents are in his fingers. The multitude of stars is distributed over his nails. In his spine is Mount Meru, and in his heart ´ ı. Further, Yama, the king of Dharma, is in his navel, the Goddess Sr¯ the earth in his hips, Creation in his genitals, the A´svins in his knees, and mountains in his thighs. The hells and the seven lower regions are in his feet, and K¯ al¯ agnirudra is located in his big toe. In this way, Vis.n.u is characterised by all deities, and so all deities have him as their soul. Vis.n.u’s body pervades all that is of speech and everything within and beyond the reach of light, moving or standing. Thus, the relation of Vis.n.u and the world is like that of wind and ether.

Introduction

lxvii

The fifth chapter of the Dev¯ amr.tapa˜ ncar¯ atra is identical to this section. The only thing additional there is a single verse at the end announcing the topic of the next chapter. xiii) animation of the image: This is the thirteenth step in the ritual process. In this step, the teacher in charge of consecration is supposed first to do pr¯ an.¯ ay¯ ama and envisage a body of ten limbs permeating the head, nose, chest, arms, anus, knees, and feet. Then he should make brahmins recite the Hymn of Life (j¯ıvan¯ıs¯ ukta), which gives life to the image, a minimum of seven times on both sides of the image, but the fastening of the image is accomplished only with one thousand or five hundred repetitions.122 xiv) ablution, the opening of the eyes, and the moving of the image to the ritual site: Now the teacher should put waterjars in the middle of the place of ablution and around it: the R . gveda is placed to the east, the Yajurveda to the south, the S¯amaveda to the west, and the Atharvaveda to the north. He should place the throne with the image on top of it,123 reciting the stanza bhadram . karn.ebhih.. Now he should o↵er the following things while reciting the specified Vedic mantras: mantra vr¯ıhaya´s ca tryambakam . kam kato na . k¯an.d.¯ at k¯ an.d.¯ at d¯ırgh¯ ayutv¯ aya ´san no dev¯ır kay¯ a na´s citra g¯ayatr¯ı = tat savitur varen.yam . gandhadv¯ ar¯ am ¯apy¯ ayasva 122

offering unhusked rice Pipal leaves bracelet oil massage mud-pack five-herb124 concoction cow’s urine cow-dung milk

I think that the text may be defective at this point. Usually, as one unit concludes, an announcement in some form tells what is coming next. But this time we have no indication of the conclusion of the thirteenth and of the beginning of the fourteenth ritual step. So at the least this kind of statement is missing. 123 I am compelled to interpret in this way, although the text simply says ‘he should place the throne’, for all ritual actions after this are directed towards the image not the throne. 124 As the text immediately indicates, these five herbs are ´sam¯ı, udumbara, a´svattha, nyagrodha, and pal¯ a´sa.

lxviii mantra dadhikr¯ abn.o tejo ’si devasya tv¯ a

Early Tantric Vais.n.avism offering curd ghee Ku´sa water

Now a paste is prepared of the herbs adhoghan.t.¯a, rath¯ang¯ ˙ a, vis.n.ukr¯ ant¯ a, ´sat¯ avar¯ı, ´sarvar¯ı, ´son.ak¯a, rambh¯a, bh¯ umike´s¯ı, sah¯a, bal¯a, sim aghr¯ı, and that paste is applied on all limbs of the deity. . h¯ı, and vy¯ Then reciting the following mantras ablutions are o↵ered to the deity with various ingredients: mantra y¯ a os.adh¯ıh. agnim ¯ıd.e y¯ a phalin¯ı agnim¯ urdh¯ a ved¯ aham etam . imam me ga nge ˙ . br.haspate †janarakt¯ a paridh¯ am¯anah.† yuv¯ a suv¯ as¯ a ´sr¯ı´s ca te laks.m¯ı´s ca

offering herbal bath flower bath fruit bath milk bath perfume bath river water clothes girdle sacred thread sandal-paste

Now reciting the yu˜ njate mana uta yu˜ njate stanza the consecrator o↵ers to the deity the pigment prepared from the cow’s bile, worships him reciting the agnir devat¯a v¯ato devat¯a stanza, and o↵ers him the incense of bdellium. Then he takes a small stick of gold reciting the stanza hiran.yagarbhah. samavartat¯agre, and marks the eyes on the image as he recites the agnir jyotih. stanza. Now reciting samiddho a˜ njan kr.daram he applies collyrium to the eyelashes, fills up the eyes with honey and ghee reciting the madhuv¯ at¯a r.t¯ayate stanza, and then stirs that filling with a blade of Ku´sa grass. Because the margin is damaged,we do not know the stanza recited for this purpose. Once the eyes of the deity incarnated in the image are open, he is ready to move. So the consecrator lifts up the image while reciting another stanza: ut tis.t.ha brahman.aspate, and brings it into the ritual site (y¯ agasth¯ ana), which appears to be the sacrificial pavilion standing in front of the newly built temple, while reciting the stanza ¯a no bhadr¯ah.. Then

Introduction

lxix

he o↵ers a bed for the deity there as he recites the stanza yaj j¯agrato d¯ uram, and an awning over the image as he recites the stanza k¯an.d.¯at k¯an.d.¯at, and also ornamental archways in four directions of the temple reciting the stanza catv¯ ari ´sr.ng¯ ˙ as trayo asya p¯ad¯ah.. Each of these archways stands on two pillars and has three pinnacles; to the east it is made of Pal¯ a´sa wood, to the west of Udumbara wood, to the west of Peepal wood, and to the north of Nyagrodha wood. Next the consecrator brings the image into the temple while reciting ¯a no bhadr¯ ah., digs a hole and fixes the image reciting dhanvan¯a g¯ah.. Thus the task of installation is complete but still the pacificatory rite has to be performed. For this purpose, first, as he recites es.o ha deva mantra, the fire altars have to be made: rectangular in the east, semi-circular in the south, round in the north-west, and octagonal in the north. Now the consecrating teacher seats himself in the south-east, facing north-east, and begins the fire-sacrifice by taking hold of the fire while reciting the tad ev¯ agnis tad ¯ adityah. stanza. xv) sacraments of the fire: This makes the fifteenth step of the ritual process. The consecrator prepares the place of the fire-altar by furrowing in the ground using the astramantra, sprinkles it with the kavaca´ mantra, installs Sakti there with the hr.dayamantra, and in the womb of ´ the Sakti just installed he puts the sacred fire. After this the sacraments starting with the rite of impregnation are performed and then afterwards oblations are made in the purified sacred fire. xvi) worship of the deity with four retinues: Now the godhead N¯ar¯ayan.a is worshipped at the centre of four concentric retinues. The innermost retinue is made in the form of a lotus. The anusv¯ara in the form of bindu representing Purus.a, the twenty-fifth of the S¯am . khya tattvas, is explained as the heart (hr.daya) of the deity; that is first placed in the pericarp of the lotus, then the ´siras, ´sikh¯a, kavaca, and astra are set respectively in the east, south, west, and north petals. Similarly, the g¯ayatr¯ı, s¯ avitr¯ı, netra and pingal¯ ˙ astra mantras are set one after the other in the south-east, north-east, south-west, and north-west petals. Then comes the second retinue: Balabhadra is worshipped in the east, Ni´sat.ha in the south-east, Pradyumna in the south, Aniruddha in the south-west, Gada in the west, S¯amba in the north-west, S¯atyak¯ı in the north, and Akr¯ ura in the north-east.125 Gada seems to be dancing for 125

I have not seen anywhere else these eight grouped together. All belong to the Vr.s.n.i

lxx

Early Tantric Vais.n.avism

some reason, while the others are not described at all. In the third retinue, Vis.n.u’s eight weapons are placed: the sword in the east, the mace in the south-east, the discus in the south, the conch in the south-west, the lotus in the west, the plough in the north-west, the club in the north, and the bow in the north-east. In the outermost fourth retinue, ten names—and I assume that names imply forms—of the deity are worshipped: N¯ar¯ayan.a in the east, Paun.d.ar¯ık¯ aks.a in the south, Vaikun.t.ha in the west, Madhus¯ udana in the north, Vis.n.u in the northeast, Jan¯ardana in the south-east, Hr.s.¯ıke´sa126 in the south-west, M¯ adhava in the north-west, Trivikrama in the nadir, and Govardhana in the zenith. xvii) deposition of precious stones: In the next step, eight precious stones, eight minerals, and eight kind of seeds are o↵ered to the deity; they are probably placed in specific arrangements. As the text recommends, if the jewels are not all available, then a piece of gold should be o↵ered; if the minerals are not all available, then sulphate of arsenic should be o↵ered, and if not all kinds of seeds are available, then one grain of barley should be o↵ered. xviii) installation: The last ritual step is installation, or rather the final consecration. The text describes this step very briefly and says that the deity should be o↵ered the consecratory ablution while the es.o ha deva stanza is recited.127 These eighteen ritual steps conclude with a brief colophon stating ‘Here ends the precept of eighteen [ritual]s’. As I argued earlier, the As..t¯ ada´savidh¯ ana is complete in itself and has a ritual system similar but di↵erent from that taught in the other two texts. It cannot be left embedded there causing contradiction with what is found in the Sv¯ ayambhuvapa˜ ncar¯ atra, so I have presented it here separately. clan and appear several times in the Mah¯ abh¯ arata (cf. e.g. 1.211.9–11, 1.213.27–29, 14.88.4–6), where they are depicted as commanding heroes. However, Gada is my conjecture: in his place the manuscript actually has Garud.a. It reads garud.a(!) pa´scimato dady¯ at, which makes a hypermetrical p¯ ada. By emending to gadam at, . pa´scimato dady¯ Garud.a, who was out of place in this group of Vr.s.n.i heroes, has been eliminated from the list and at the same time I have solved the metrical problem. See also pp. li–lii above. 126 Hr.s.¯ıke´sa is my tentative conjecture, which fits metrically, for the name allocated to the south-west is lost in a lacuna. 127 Some detail of the consecratory ablution can be found in the Sv¯ ayambhuvapa˜ ncar¯ atra 8.39–43.

Introduction

lxxi

Contents of the Dev¯ amr.tapa˜ ncar¯ atra The Dev¯ amr.tapa˜ ncar¯ atra adopts the first two chapters of the Sv¯ ayambhuvapa˜ ncar¯ atra almost unaltered, so the contents of these two chapters should be read from the previous section on the Sv¯ ayambhuvapa˜ ncar¯ atra. What is altered is just the frame-story that identifies and stages the interlocutors: in the Sv¯ ayambhuvapa˜ ncar¯ atra ¯I´svara was teaching Brahman on the Mandara mountain, but here Brahman is teaching Sanatkum¯ara on Mount Meru. But we may note that the Dev¯ amr.tapa˜ ncar¯ atra does not want to disguise the fact that it is reintroducing the teachings of the Sv¯ ayambhuvapa˜ ncar¯ atra. At the end of the second chapter, Brahman announces that he will explain how an image of Vis.n.u is consecrated following the teaching of Svayambh¯ u, and thus he connects the new text with the Sv¯ ayambhuvapa˜ ncar¯ atra. Now in the following I will narrate the third and fourth chapters of the Dev¯ amr.tapa˜ ncar¯ atra. There is a chance that these two chapters originally come from the Sv¯ ayambhuvapa˜ ncar¯ atra. Although we do not find them now in the Sv¯ ayambhuvapa˜ ncar¯ atra, the programme of that text given in its first chapter lists the contents of these chapters (see p. xxxvi above for related arguments).

Chapter 3: Sculpting Vis.n.u’s Image: Material and Size This chapter is about the sculpting of Vis.n.u’s image (pratim¯alaks.an.a). It first enumerates twenty materials suitable for this purpose, which are: sapphire, dark-blue sapphire, ruby, diamond, emerald, cat’s eye, pearl, coral, copper, white brass, tin, lead, yellow brass, N¯agavr.ks.a,128 silver, gold, iron, stone, clay, and R¯ ajapat.t.a stone. As the text states, images made from these materials are the best if they are three fore-arms in height. They are of middle standard if two hands high and 33 or 25 angulas ˙ high. If they are one to twelve angulas ˙ high they are of low standard. However, an image up to the size of twelve angulas ˙ can be worshipped in a household, not higher than that. Even within that option, an image one angula ˙ in size is just excellent but one that is two angulas ˙ in size causes loss of wealth. That which is three angulas ˙ in size bestows supernatural powers, but one that is four angulas ˙ in size causes 128

This must be a metal or precious stone, not wood, because we find below a separate list of various kinds of wood suitable for the crafting of an image.

lxxii

Early Tantric Vais.n.avism

illnesses. One that is five angulas ˙ in size bestows wealth, but one six angulas ˙ in size causes agitation. Likewise, an image seven angulas ˙ in size fulfills one’s desire for a son, but one that is eight angulas ˙ long causes death. An image which is nine angulas ˙ in size gives cattle, but one that is ten angulas ˙ in size causes destruction of everything. Similarly, an image eleven angulas ˙ in size provides good luck, but one that is twelve angulas ˙ long kills everybody. In order to craft wooden images, according to our text, one should choose sandalwood, or the wood of one of these trees: Tabernaemontana coronaria (tagara), Wood apple, Xanthochymus Pictorius (tam¯ala), Red Eye (? raktalocana), Strychnos nux vomica (tenduka), Deodar, and Mah¯ aris.ika (?). Wooden images of anything from one vitasti, which is generally interpreted as twelve angulas, ˙ to three fore-arms in height are of the lowest standard, those up to the height of six hands are of the middle standard, and those up to nine hands are of the highest standard. The text also gives the proportional measurement of all limbs applicable to the images of the first kind, namely, those made of precious stones and metals. It does not mention the wooden ones in this context.

Chapter 4: What makes a Good Image The fourth chapter tells us what makes a good image. An image should be well-formed and sculpted in the round, plump, and proportionate. It should not be black, extra flat or long, or ill-shaped. It should not have an oblique look gazing down or up, or into the middle. It should not be bent, cracked, with deficient limbs, or without limbs. There should not be any scratch if one wishes success. A cracked image will cause loss of money, and if the image has deficient limbs, it will cause death. If limbless, it will bring devastation to the country; if crooked, it will cause strife; and an image gazing downward will cause famine and drought. Therefore, Vis.n.u, the deity with an eagle in his banner, should be depicted complete with all his limbs, with a calm gaze, looking auspicious, and he should be kept adorned with two pieces of cloth. There should also be installed Narasim . ha, Var¯aha, V¯amana, Trivikrama, and V¯asudeva in his all-encompassing form. These should be installed in the centre and the four sub-directions respectively; this is the pentad of images (pratim¯ apa˜ ncaka).129 129

For a brief discussion on this pentad, see above pp. xxvii–xxviii.

Introduction

lxxiii

The main image of V¯ asudeva should have two or four or eight arms. He should be adorned with a girdle at the waist and an extensive garland of forest-flowers covering his limbs. Choice bracelets should be on his hands, anklets on his feet, and he should wear a single pearl necklace (ek¯aval¯ı) around his neck. The deity should be depicted with curly hair and a diadem on top, with earrings and other ornaments in his ears, and a bright mark on his forehead. He should be shown wearing a serpent for his sacred thread, and another for his belt. A lotus and the mace that destroyed demons should be in his right hands, while a discus and a conch are to be placed in his left hands. A club and a sword are placed in the other two hands to the right and a plough and a bow in the hands to the left.130 A man who builds such a beautiful image is always endowed with all possessions. The fifth chapter of the Dev¯ amr.tapa˜ ncar¯ atra is identical with the twelfth unit of the As..t¯ ada´savidh¯ ana. It teaches the installation of various divinities in the image of Vis.n.u. The only thing added here is a single verse at the end of the chapter that announces the topic of the next chapter. Similarly, the sixth chapter of the Dev¯ amr.tapa˜ ncar¯ atra largely matches the fifth chapter of the Sv¯ ayambhuvapa˜ ncar¯ atra and teaches the ritual preparation of the sacrificial pavilion. That common content is not repeated here, only what has been added in the Dev¯ amr.tapa˜ ncar¯ atra is given. After the ground is cleaned but before the making of the hollow in the sacrificial ground, the consecrator draws a nine-box diagram and worships the sun there in the central box, while the other eight astral bodies are placed around the sun. The Dev¯ amr.tapa˜ ncar¯ atra di↵ers again from the Sv¯ ayambhuvapa˜ ncar¯ atra when it comes to the number and location of fire altars. The Sv¯ ayambhuvapa˜ ncar¯ atra recommends four altars: rectangular in the east, resembling the half-moon in the south, round in the northwest, and in the shape of a lotus in the north. But in the Dev¯ amr.tapa˜ ncar¯ atra they are five, with one extra in the north-east corner. Now the round altar is located in the west, not the northwest.131 Additionally, these five 130 Although it is not stated explicitly, we can guess that the first four weapons are the major ones and the other four are placed in the additional hands when the deity is depicted with eight hands. 131 The Dev¯ amr.tapa˜ ncar¯ atra recommends a bow-shaped altar in the south whereas the Sv¯ ayambhuvapa˜ ncar¯ atra recommends a half-moon-shaped one. But this does not count

lxxiv

Early Tantric Vais.n.avism

altars are associated here with five astral bodies: Soma, Budha, Guru, ´ Sukra, and Ketu respectively. These new elements suggest that at the time of the composition of the Dev¯ amr.tapa˜ ncar¯ atra veneration of the astral bodies had gained in importance. There is no parallel in the extant Sv¯ ayambhuvapa˜ ncar¯ atra for the seventh to ninth chapters of the Dev¯ amr.tapa˜ ncar¯ atra.

Chapter 7: The Pavilion is Divine, not Demonic The seventh chapter begins with Sanatkum¯ara’s curious question: Is this pavilion associated with the Asuras, Pi´s¯acas or R¯aks.asas? How can a sacrifice be made in a pavilion of non-divine nature? Such a pavilion signifies cruelty and is devoid of Dharma.132 In response to this question, Brahman denies the allegations and proposes that the pavilion houses all deities, and says that it is a lofty palace in the middle of the milk ocean; ´ there is the Svetadv¯ ıpa and also the mountains. The text identifies the ¯ altar in the east as the house of the Ahavan¯ ıya fire and explains that the presiding deity of this fire is Brahman. Similarly, it identifies the southern altar as the house of the Daks.in.¯agni fire and associates that with Vis.n.u. It then labels the altar in the northeast as the place of ashes (bhasm¯ ag¯ ara) and associates that with the Yaj˜ napurus.a.133 According to the text, the three gun.as: sattva, rajas, and tamas, preside over the three step-like belts (mekhal¯a) outside the altar. The ‘All-gods’ (vi´svedev¯ah.) are located in the yoni and Kambala and A´svatara are in its mouth. The as a point of di↵erence, because in this kind of contexts describing shapes ardhacandra (half-moon) and dhanus. (bow) are often used as synonyms. 132 What is behind these allegations? In the beginning of the eighth chapter, too, the text states that rituals should be performed using Vedic mantras so that the sacrifice of consecration shuns its demonic nature. Since we are left to surmise, we might pay attention to the fact that the temple structure stands on the v¯ astucakra and its construction might be thought to involve the killing of some living beings. On violent customs involved in the appeasement of the v¯ astu of a house (v¯ astu´samana), see Bakker 2009. Another possibility is that, as shown in the previous chapter, the fire altars are associated with the seizing astral bodies, and since they can influence people negatively, the pavilion as the house of these seizers is seen as non-divine. 133 In the previous chapter, the text was speaking of five fire altars associated with five astral bodies, but here only four of them are mentioned and equated with the three Vedic fires and the Yaj˜ napurus.a, and the fifth is left out. But this fits well with the prescription of the Sv¯ ayambhuvapa˜ ncar¯ atra (5.10–11) that requires the building of four fire altars.

Introduction

lxxv

four Vedic priests in four gates represent the four Vedas and thus all Vedic mantras. The consecrator is V¯ asudeva himself, the teacher of the world, the ultimate being. The archway in the east is related to the eon of Kr.ta and represents the priestly Brahmin class. That in the south is associated with the eon of Tret¯ a and represents the warrior class of Ks.atriyas. The one in the west is associated with the eon of Dv¯ apara and represents the Vai´sya class of farmers and merchants, and the one in the north is related to the eon of ´udras. Here the three classes of ancestors Kali and represents the class of S¯ preside over the three pinnacles of each archway, which are equated with the three worlds. The oceans are symbolically located in water-jars there, the wind-gods in the pair of clothes around the jars, the mountains in the pillars, Yaks.as in flowers and wreaths, Vidy¯adharas in garlands, N¯asatyas in blossoms, Apsarases in chowries, and mankind in the canopy. The pairs of Rati and Medh¯ a and of Tumburu and N¯arada are represented by songs and music, and prominent sages dwelling in the world of Brahman are represented by Vedic chants. The banners stand for Kinnaras, ghostly spirits, guardians of quarters and such like divinities. The moon, stars and asterisms are present in the bunches of Ku´sa grass; in the bed are the serpents; in the beddings are the nine astral bodies; Gandharvas are in the pillow; and Guhyakas are in the side pillows. In the lamp are all heavenly bodies, and all herbs are in the food o↵erings. Dharma stands as the parasol; the eight Vasus stand in the eight directions. Above is heaven with all the gods; below are the lower regions; and in the middle is N¯ar¯ ayan.a, the ultimate and eternal entity. G¯ayatr¯ı is located in the rice-pudding cooked ritually in a special pan (sth¯ al¯ıp¯ aka), s¯ avitr¯ı in the pran.¯ıt¯a-vessel, and ambrosia in the clarified butter. The Brahman priest represents Purus.ottama himself, while the eleven Rudras are represented each by one angula ˙ from the tip of the twelve-angula-long ˙ sacrificial ladle. Likewise, the deities stand by the utterances of sv¯ ah¯ a, and Praj¯apati by the utterances of vas.at.. In Vedic sacrifices, as the text asserts, Brahmins actually o↵er oblations to Yaj˜ navar¯ aha present in the form of the sacred fire. In conformity to this fact, in the sacrificial pavilion people make sacrifices to Vis.n.u present in the form of K¯ al¯ agnirudra fire, who is capable of bestowing liberation. In this way, the deities are present in the sacrificial pavilion in their respective places, and one has to imagine that this pavilion is built on the ´ Svetadv¯ ıpa in the middle of the milk ocean.

lxxvi

Early Tantric Vais.n.avism

Without the officiant knowing this secret, if the pavilion of consecration is built out of his own imagination, then the host as well as the officiating consecrator will go to hell. That ritual will be fruitless, because it will lack Dharma. If performed wrongly, the sacrifice is the worst of enemies: it causes poverty, a✏iction by disease, loss of money, and decay of family. Therefore, with every e↵ort, one should perform the sacrifice assisted by a highly educated priest.

Chapter 8: Consecration of the Pavilion The eighth chapter gives the Vedic mantras to be used in the installation of the pavilion itself, so that the sacrifice shuns its demonic nature and becomes divine.134 First of all, the temple is consecrated while reciting the praty amu˜ nca stanza, then the pavilion while reciting the agna ¯a y¯ahi stanza, and the hollow platform in the sacrificial ground (vedi) while reciting the vedis.ada stanza. The next item might be the soil in that platform but the text is damaged; that is put in place while reciting the sapta mr.d stanza. All fire altars are consecrated while reciting the agnim agn¯ıt stanza, the mekhal¯as with the nejames.a stanza, and the yoni with the ´san no dev¯ıh. stanza. With the r.cam . v¯acam stanza, vigour in the sacrificial priests is achieved, and with the aham . sapta stanza the consecrating teacher is installed as a representative of Vis.n.u. In the next step, reciting the catv¯ ari ´sr.ng¯ ˙ ah. stanza the four archways are consecrated, and a three-spoked pinnacle is placed over each of them while reciting the bh¯ ur bhuvah. svah. stanza. While the abhikrandan stanza is declaimed, water-jars are installed and a pair of clothes is put around each of them while the vasis.t.h¯ a hi stanza is recited. The pillars of the archways are sanctified while reciting the imam . nu somam stanza. Furthermore, with the srakve drapsasya stanza flowers and wreaths are set in place, and with the sutr¯am¯an.am and utt¯ anaparn.e stanzas garlands and blossoms are placed separately. Likewise, with the balavij˜ n¯aya and vi hi sotor stanzas respectively, chowries and canopies are placed, and with the somo dadad stanza Vedic chants are o↵ered. Then with the ubhayam . te stanza both kinds of chants, perhaps those in Sanskrit and in vernaculars as well,135 are o↵ered. All 134

On the demonic nature of the sacrificial pavilion, see fn. 132 above. The Ni´sv¯ asanaya (2.30) states that speech is fourfold and lists Sanskrit, Prakrit, ¯ Apabhram asika. . ´sa, and Anun¯ 135

Introduction

lxxvii

banners are set while the asm¯ akam indrah. stanza is recited, and a layer of Ku´sa grass strewn near the hollow platform is laid while a Vedic mantra is declaimed whose prat¯ıka has now been corrupted beyond recognition: †sajy¯ı´sr.trabh¯ ag¯ as tv¯ a†. The bed for the deity is set with the par¯ıto s.i˜ nca stanza, the beddings are placed with the tubhatt¯ a angirastamah ˙ stanza, and a pair of pillows is . placed with the indr¯ aya s¯ ama g¯ ayata stanza. The lamp is installed with indra jyes.t.ham a bhara, then while the caran vai madhu stanza is re. na ¯ cited, the hollow platform is covered with various food o↵erings, and the parasol is installed with the indrah. ks.atram . stanza. Thereafter the dhanvan¯a g¯ a stanza is recited for the sake of locating the directions ritually, and with the ud u tis.t.ha svadhvara and adh¯a no vi´svasaubhaga stanzas respectively the regions in the earth and beneath it are located, and the middle sphere is located with the anta´s carati stanza. The yad indra pr¯ ag stanza is recited at the moment of placing the ritual pan with the food to be cooked (sth¯al¯ıp¯aka) on the fire, and the p¯ar¯avatasya stanza when placing the Pran.¯ıt¯a vessel. The contents in the plate of the clarified butter to be o↵ered (¯ajy¯anga) ˙ are sanctified ¯ by reading the Ajyadohas¯ aman, and with the aprat¯ıto jayati stanza the Ku´sa figure symbolising the ultimate brahman136 is set. Then by reciting the kr.s.n.o ’si stanza the two wooden ladles named sruc and sruv¯a are sanctified. The adityai vyundanam stanza is recited as he secures the expression of sv¯ ah¯ a, and all that is expressed in the Vedic mantra astu ´sraus.at. . . . including the exclamation vas.at.. After this, while the yaj˜ nena vardhata stanza is declaimed, the Yaj˜ napurus.a is installed; then with the vipram aso stanza the brahmanical fire (vipr¯agni) is gathered and . vipr¯ brought to the pavilion, and with the namaste rudra stanza that fire is installed in the altars as the fire of K¯al¯agnirudra. As the text informs us, this is the divine ceremony of pavilion installation.137 He who performs this will receive the reward of having performed an A´svamedha sacrifice. But he goes to hell, together with his host, if 136 In any kind of fire ritual, a symbolic dummy of Brahman made of fifty stalks of Ku´sa grass is installed near the Pran.¯ıt¯ a vessel to the south of the sacred fire (cf. e.g. Sam sk¯ a rad¯ ıpaka, pp. 27–28). It represents the Brahman priest who is responsible for . supervising the ritual, but our text interprets it as a representation of Purus.ottama, the ultimate person (cf. DP 7.17). 137 Thus the Dev¯ amr.tapa˜ ncar¯ atra depicts even the pavilion, and not merely Vis.n.u’s image, as consisting of all deities and many mantras.

lxxviii

Early Tantric Vais.n.avism

he carries out the consecration ritual without using the Vedic mantras, being deluded by greed. Such an act would ruin his locality, even the whole country, and the king of that country would face great dangers. There would instantly follow all kinds of calamities, and loss of property would ensue.

Chapter 9: Ritual Ablution of the Deity The ninth chapter presents the procedure for ritual ablutions of the deity. First of all, for this purpose, the consecrator chooses some brahmins wellversed in the Vedas as his assistants, and sets a splendid throne just one hand in height in the northeast corner of the temple. He first takes the image around the village mounted on an elephant or a chariot, or carried by brahmins, leading a procession blowing conches and playing instruments. During this procession, o↵erings are made to the image at every step saying ‘Veneration to the bestowers of all things!’ (sarvadebhyo namo ’stu) and at the end the image is carried into the temple complex, while Vedic and other auspicious chantings are going on.138 At the time the image enters the complex, the consecrator recites the rathe tis.t.han nayati stanza, then, while reciting the bhadram . karn.ebhih. stanza, he places the image on the throne prepared a while before. Then the skilled architect (s¯ utradh¯ara), clad in a white robe, takes a water-jar and gives a bath to the image, and gives it the final touch by carving the final line of the eyes, thus producing sight in them. As he does this, the agnir jyotir jyotir agnih. stanza is recited. At this moment, a cow is donated to the consecrator. Then, prouncing the madhuv¯at¯a stanza the consecrator fills the eyes of the image with ghee mixed with honey, picks up a golden stick as he reads the hiran.yagarbhah. stanza, and with that stick, he opens the eyes of the image while reciting the citram . dev¯an¯am . stanza. After this, the consecrator begins the pacificatory rite in order to correct any possible sculpting error in the image. He mixes black sesame ´ ı leaves, mustard seeds, and rice grains in ghee and oblates 108 seeds, Sam¯ times in the fire burning with P¯al¯a´sa wood. He makes oblations further, one hundred times each in the name of the seven worlds from Bh¯ u to 138 This is the pre-installation ablution. This step is described briefly in the As..ta ¯da´savidh¯ ana, but I do not find it included in the Sv¯ ayambhuvapa˜ ncar¯ atra.

Introduction

lxxix

Satya while reciting the seven vy¯ahr.tis.139 Once all obstacles have thus been quelled, the consecrator begins the ritual ablution of the image. For this purpose, he uses water-jars marked with auspicious signs and filled with water and herbs, without holes or black marks at the bottom. He recites the samudram . gaccha stanza and fetches a water-jar full of water and o↵ers the water to the deity while reciting the imam ˙ stanza. He o↵ers some soil from a termite. me gange mount with the m¯ urdh¯ anam . divah. stanza, and washes that reciting again the samudram gaccha stanza. Then he finds the concoction of five herbs: . ´ Sam¯ı, Udumbara, A´svattha, Nyagrodha, and Pal¯a´sa, and o↵ers that to the deity while uttering the yaj˜ n¯ayaj˜ n¯a vo stanza. He brings the five products of the cow and Ku´sa-water, and o↵ers each of them with a separate Vedic mantra: cow-urine with the g¯ayatr¯ı mantra, cow-dung with the gandhadv¯ ar¯ am stanza, milk with the ¯apy¯ayasva stanza, curd with the dadhikr¯ abn.o ak¯ aris.am stanza, ghee with the tejo ’si stanza, and the Ku´sa-water at the end with the devasya tv¯a stanza. Then he o↵ers the deity bathing in water mixed with various fruits: the fruits of Diospyros embryopteris, Emblic Myrobalan, mango, and wood apple, reciting the y¯ a phalin¯ı stanza. Thereafter he mixes these herbs: sahadev¯a, bal¯a, ´satam¯ ul¯ı, ´sat¯ avar¯ı, kum¯ ar¯ı, gud.u ¯c¯ı, sim . h¯ı, and vy¯aghr¯ı in water, and bathes the deity with that water while reciting the y¯a os.adh¯ıh. stanza. He utters the ghr.tavat¯ı stanza, and takes out a portion of ghee and smears it over the limbs of the deity with the drupad¯ad iva mumuc¯anah. stanza. Next the deity’s head is washed with water, while the m¯anastoke stanza is declaimed. This is followed by bathing in perfumed water accompanied by the vis.n.o rar¯ at.am stanza, and finally the deity is bathed in river water while all eight mantras of Vis.n.u are recited.140 The consecrator then visualises the deity dressed in yellow clothes, and requests him to grant his presence, first with Vedic mantras and then with the following eulogy in Anus.t.ubh verses: Obeisance to you who go beyond selfish attachment! You are characterised by the greatness of satisfaction. You take the form of ultimate as well as worldly knowledge. You do not de139 The seven vy¯ ahr.tis match the names of the seven worlds: bh¯ uh., bhuvah., svah., mahar, janah., tapah. and satyam. 140 I have not been able to identify which these mantras are. The Vis.n.us¯ ukta (RV 1.154) might be a candidate, but there are only 6 stanzas in this hymn.

lxxx

Early Tantric Vais.n.avism viate from the glory of brahman, and your impetus transcends the gun.as. You are Purus.a, the one free of attributed qualities, imperishable, and the most ancient. O Lord Vis.n.u, please be present here! O Lord Vis.n.u, please come, do come! O graceful god, please accept your share of the sacrifice! Obeisance to you, O V¯ asudeva!

After all this, the consecrator o↵ers an arghya to the deity while uttering the hiran.mayena p¯atren.a stanza, and then perfumes and garlands him with the idam . vis.n.ur vicakrame stanza. With the br.haspate ati stanza he changes the clothes on the image, and o↵ers the upper garment reciting the ved¯ aham etam . stanza. The sacred thread is o↵ered while the brahma jaj˜ n¯ anam stanza is recited, and the image is covered with flowers accom. panied by the y¯ a os.adh¯ıh. stanza. Then the consecrator puts garlands on the image while reciting the ´sr¯ı´s ca te stanza, burns incense with the dh¯ ur asi stanza, and applies collyrium round the eyes of the image with the samiddho a˜ njan stanza. He makes a mark on the forehead of the image with the pigment of cow’s bile while reciting the yu˜ njanti bradhnam stanza, fans the deity with a chowry while reciting the balavij˜ n¯aya stanza, and shows a mirror with the dev¯a g¯atuvido stanza. After this, he o↵ers the dish of rice, lentils, and sesame seeds cooked together along with ghee while uttering the annapate ’nnasya stanza, and then invites the deity to enter the pavilion while reciting the d¯ırgh¯ayutv¯aya stanza. Pronouncing the ut tis.t.ha brahman.aspate stanza, he lifts the image up and puts it in a special chariot called brahmaratha, which is kept for the sole purpose of carrying deities, and reciting the sahasra´s¯ırs.¯a stanza he brings the image into the pavilion. There again an arghya is o↵ered to the deity, the image is placed on a bed in front of the temple and, with the recitation of the vi´svata´s caks.ur uta stanza, the image is rendered Sakalanis.kala. This rite of Vis.n.u’s ablution is performed by the consecrator and other assisting brahmins together.141

Chapter 10: The Consecrator and his Assistants The tenth chapter characterises the consecrator as well as his assistants. According to the text, brahmins coming from good families and belonging 141

Parts of this chapter are found in the fourteenth section of As..ta ¯da´savidh¯ ana but the schemes there and here are di↵erent.

Introduction

lxxxi

to one of the branches of the four Vedas, well-versed in the Vedas and the ancillary texts, and devoted to Vis.n.u, should be chosen as assistants (sah¯aya) of the consecrator to perform sacrifices; the same people are to be appointed as the caretakers of the image (m¯ urtidhara). As for the consecrator, he should be a brahmin learned in mantras, well-versed in the Pa˜ ncar¯ atra, and intact in all his limbs. He should know the scriptures and look beautiful. He should be of a high descent, skilled in the Vedas, devoted to Vis.n.u, and a knower of the twenty-five tattvas. He should neither be very tall nor dwarfish, neither stout nor thin, neither too white nor black, not impure, and not lacking in Vedic education. He should not be envious, not full of lice, neither too young nor aged. He should not be bald, brown-skinned, monkey-coloured, or overbold. He cannot be a man su↵ering from leprosy or some disease, bad-skinned, one-eyed, or blunt. He should not be practising something other than what is ordained by the Vedas and the scriptures following them. He should not have deficient limbs, and he should not be an adulterer. He cannot be someone ´udra, an impostor, or an outcaste. A devoted to some other deity, a S¯ 142 ¯ Kaula, Devalaka, Saura, Arhata, and Bauddha should be avoided for the worship of Vis.n.u. If appointed as the consecrator of Vis.n.u, a Kaula kills the host, a Devalaka kills his wife, a Saura ruins the village, a naked Jaina burns down the country, and a Bauddha ruins everything: cattle, o↵spring, and wealth. Therefore, with every e↵ort, all those devoted to other deities must be avoided.

Chapter 11: Deposition of Mantras The eleventh chapter is about sakal¯ıkaran.a: deposition of mantric elements in the image and also in the consecrator’s body. This chapter is parallel with the seventh chapter of the Sv¯ ayambhuvapa˜ ncar¯ atra, but di↵ers on a few points. In the very beginning of the process, the Sv¯ ayambhuvapa˜ ncar¯ atra gives an instruction carry out sakal¯ıkaran.a upon the image, touching its relevant parts, whereas the Dev¯ amr.tapa˜ ncar¯ atra enjoins that the consecrator should carry out sakal¯ıkaran.a on his body and touch the image. Although the consecrator installs the twelve syllables of the om asudev¯ aya mantra on di↵erent parts of the image . namo bhagavate v¯ during the sakal¯ıkaran.a process according to both works, each of them 142 For some remarks on the history of this term, see T¯ antrik¯ abhidh¯ anako´sa III, s.v. devalaka.

lxxxii

Early Tantric Vais.n.avism

follow a di↵erent scheme of installation. Here is a table that compares the two: syllables om . na mo bha ga va te v¯ a su de v¯ a ya

¯ yambhuva sva head nose forehead mouth throat chest right hand left hand right shank left shank right foot left foot

¯ mr.ta deva head nose mouth throat arms chest navel back waist thighs shanks feet

After the installation of these syllables in the divine body, the consecrator meditates on each of those syllables separately. Both texts agree on this issue and use the same scheme of visualisation.143 The other component of this rite in the Sv¯ ayambhuvapa˜ ncar¯ atra appears to be the de144 after which position of the hr.daya and the other eight angamantras, ˙ this text begins with the worship of Vis.n.u along with the deities in his three retinues. In the Dev¯ amr.tapa˜ ncar¯ atra, however, we find that, between the installation of the syllables of the twelve-syllable mantra and the angamantras, ˙ the twelve names of Vis.n.u beginning with Ke´sava as well as the eight weapons of the deity are additionally installed. At this point, the text provides some new information about the angamantras ˙ not found in the Sv¯ ayambhuvapa˜ ncar¯ atra or the As..t¯ ada´sa143

At this point both texts extol the meditator. The Sv¯ ayambhuvapa˜ ncar¯ atra uses only one verse for this purpose (see fn. 77 above), while the Dev¯ amr.tapa˜ ncar¯ atra adds a further verse, which can be traced in the Vis.n.upur¯ an.a (1.6.40). This added verse asserts that all heavenly bodies beginning with the sun and moon go and come back every day, but that until today no meditator upon the twelve-syllable mantra has ever returned. 144 The text simply lists these mantras, and only the last item in the list it allocates explicitly to ‘the hands’ (hastayoh.). I assume that this, if not an abbreviation, is the result of eyeskip, and that all angamantras ˙ should in fact be allocated to particular places.

Introduction

lxxxiii

vidh¯ ana. It states that Vis.n.u is proclaimed to be as.t.¯anga, ˙ 145 and hence the angamantras ˙ are installed in the image, in the body of the s¯adhaka, and also in the lotus drawn to worship Vis.n.u. It also identifies the first five of these mantras with the Vy¯ uha deities and N¯ar¯ayan.a, collectively referred as pa˜ ncam¯ urti: hr.daya with V¯asudeva, ´siras with Sam . kars.an.a, ´sikh¯a with Pradyumna, kavaca with Aniruddha, and astra with N¯ar¯ayan.a. Then the text appears to say that Sambhav¯a is g¯ayatr¯ı, it identifies s¯avitr¯ı as Vijay¯ a, netra as a form of light, and pingal¯ ˙ astra as a form of Aghora.146 There is still one more minor but important point on which the Dev¯ amr.ta´ pa˜ ncar¯ atra di↵ers from its predecessor: here the goddess Sr¯ı is not located in the lotus of worship, whereas in the Sv¯ ayambhuvapa˜ ncar¯ atra she is located and worshipped in the pericarp together with the hr.dayamantra, which is defined as V¯ asudeva. Otherwise, the rest of the process of worshipping Vis.n.u in the centre of three concentric retinues matches the account of the Sv¯ ayambhuvapa˜ ncar¯ atra. This chapter concludes at the end of this process.

Chapter 12: The R . gveda, Vis.n.u’s Image, and his Incarnations We have already seen in the previous pages how much both our Pa˜ ncar¯ atras emphasise the Vedic character of the deity and of the cult they are promoting. The Dev¯ amr.tapa˜ ncar¯ atra seems even keener in this regard in its twelfth and thirteenth chapters. Its twelfth chapter begins with Sanatkum¯ ara’s remark that an object installed with the mantras of worship (p¯ uj¯ amantra)—I assume he means non-Vedic mantras—cannot invite trusting devotion. He asks how an image can be divine if it lacks the mantras of R aman. In order to free Sanatkum¯ara from . g, Yajus., and S¯ this worry, Brahman produces the following scheme to equate ten representative mantras from ten Man.d.alas of the R . gveda with various lines and shapes in the image that are held to represent the ten incarnations of Vis.n.u. Finally he equates the last mantra of the R . gveda with some line in the image thought to stand for brahman and to lead to heaven:

145

Not eight but nine of them are there, but the heart mantra is placed in the centre as the essence of Vis.n.u. 146 ´ This suggests an allusion to the Saiva aghor¯ astra.

lxxxiv agnim ¯ıle purohitam kus.umbhakas tad abrav¯ıt avadam ¯ . s tvam . ´sakune gr.n.¯ an¯ a jamadagnin¯a dh¯ aman te vi´svam gant¯ a no yaj˜ nam yo nah. svo aran.o prati caks.va vi caks.va agne y¯ ¯ ahi marutsakh¯a yat te r¯ aja˜ n chr.tam . sam¯ an¯ı vah.

Early Tantric Vais.n.avism Matsya: horizontal line K¯ urma: downward line Var¯aha: upward line Narasim . ha: downward line V¯amana: very slanted line R¯ama: the shape of hands R¯aghava: horizontal line Kr.s.n.a: upward line Buddha: round shape Kalkin: triangular shape brahman: a line to heaven

With all these particular lines and shapes that are homologised with Vedic mantras, the image is regarded as the primaeval Purus.a characterised by all ten Man.d.alas of the R . gveda. At the point of conjunction ´ ı is located. If we want to imitate of these lines and shapes, Goddess Sr¯ the S¯ am . khyas, we can see her as Prakr.ti. Similarly, the goddess of earth, Bh¯ u, with whom Vis.n.u is intimately connected, is located in the deity’s throne.147 Then reciting the svasti no mim¯ıt¯a´svin¯a stanza the image of Garud.a is set. After the main image is put in place, its limbs are attached to it. Although this is not stated, I suppose that this process is symbolically performed, unless the image is too big and must really be assembled piece by piece. But such a process of assembly cannot be applied to all kinds of images, as the following reveals. With the tr¯ın.i j¯ an¯ a pari bh¯ us.anti stanza the left foot of Garud.a is set, and with the es.a stomo maha stanza his right foot is set. Similarly, reading the yu˜ nj¯ath¯am . r¯ asabham . rathe stanza the left wing of Garud.a is set, and the right wing reading the pratis.t.h¯ asi pratis.t.h¯am . stanza. 147

It is noteworthy that in the Dev¯ amr.tapa˜ ncar¯ atra’s prescription of the man.d.ala for the worship of Vis.nu . , the goddess of earth is nowhere explicitly installed. Another ´ ı is incorporated, she is incorporated important point is that, even though Goddess Sr¯ in the body of Vis.n.u but remains invisible; there is no image of her worshipped in this system. In the Dev¯ amr.tapa˜ ncar¯ atra she is not even symbolically present in the man.d.ala for worship; she is at least symbolically present there according to the scheme of the Sv¯ ayambhuvapa˜ ncar¯ atra. See also p. xxxiv above.

Introduction

lxxxv

Chapter 13: Assembling the Image-Limbs Ritually The thirteenth chapter provides a scheme for assembling the limbs into the main body of the image by reciting a Vedic stanza for each of them. Here, too, the same logic is applied: first the full image is installed, and only then, symbolically, individual limbs are attached to the main body. Since this process begins with the putting of the whole image of Vis.n.u in place while reciting the sahasra´s¯ırs.¯a purus.ah., this further reinforces the supposition that the fixing of limbs is to be performed mentally. There is a faint possibility that this chapter is placed here as an appendix which describes something happening in the sculptor’s workshop, not in the temple at this point of time. In the table below, I juxtapose each Vedic stanza with the limb assembled: indr¯ aya g¯ ava ¯ a´siram udvayan tamasah. pari et¯a ars.anti bodhanman¯ a id astu nah. aks.ann am¯ı madanta n¯asad ¯ as¯ın no sad ¯ as¯ıt bhadram . karn.ebhih. g¯ayanti tv¯ a g¯ava´s cid gh¯ a samanyavah. tam u s.t.av¯ ama yam . giram k¯al¯ı kar¯ al¯ı indr¯ asom¯ a tapatam . raks.a tuvigr¯ av¯ a iha rati sam animis.en.a . krandanen¯ ¯a´sum dadhikr¯ a m . . tam u gharmeva madhu jat.hare ap¯ am . pr.s.t.ham asi tan¯ ur asi

head crown forehead eyebrow eyes nose ears cheeks chin lips teeth tongue neck chest shoulder shoulder-blade belly back the bodily form

This chapter is incomplete, for no folio after the 26th is found, but we can surmise that the last part that we have of it must be close to the conclusion, because the ritual process of limb by limb installation or assembling of the image which started at the head has reached the lower region of the body.

lxxxvi

Early Tantric Vais.n.avism

The conclusion of the text as a whole, however, seems to us as if it might still be a long way o↵ when we compare the course of ritual described in this text with that in the Sv¯ ayambhuvapa˜ ncar¯ atra. The ritual sequence in the Sv¯ ayambhuvapa˜ ncar¯ atra is as follows: a) first the sacrificial pavilion is constructed and ritually consecrated, then b) the newly built image is brought to the pavilion and laid in a bed there, where the rite of sakal¯ıkaran.a is performed, c) this is followed by the fire sacrifice and overnight incubation of the image in the pavilion. On the next day, d) the pacificatory rite is performed, and the image is carried into the temple and is installed there, e) the image gets a consecratory ablution, f) then Vis.n.u is internalised and the ceremony is concluded. The last ritual item that is parallel in the Sv¯ ayambhuvapa˜ ncar¯ atra and in what remains of the Dev¯ amr.tapa˜ ncar¯ atra is the rite of sakal¯ıkaran.a. That is covered in the eleventh chapter of the Dev¯ amr.tapa˜ ncar¯ atra, and the rites in the following two chapters can be seen as an extension and enhancement of that rite. Now if we suppose that no other extra content was added in the Dev¯ amr.tapa˜ ncar¯ atra, we must be missing three or four further chapters that would have covered the remaining rites. We would expect the new image to remain in the sacrificial pavilion at least for one night. Only then should it enter the temple and get installed. There should then follow the post-installation rite of the deity’s ablution, and finally the rite of dismissing the invoked deities and the conclusion of the ceremony. But all these are missing from the Dev¯ amr.tapa˜ ncar¯ atra as transmitted to us.

TEXT

; -vAyMBvp—rA/m^; ; ˛TmA@yAy,; [1r 1]m⌃dr-T\

sKAsFn\ ⇤ dvAnA\ ˛BmFòrm^. kArZ\ svBtAnAm⇥Eqd ⇤vgZAEctm^; 1; i⌃d˝Aç {d ⇤vt {, sv {, vFrBd˝AEdEBgZ {,. srEk‡rg⌃Dv {, EsàEvçADr {, -ttm^; 2; ˛EZp(y hr\ ⇤ dv\ lokAnA\ EvjyA(mkm^. (yt^ k)[ ++++++++++ ][2]mcnm^; 3; `yj,sAmATv {-t m⌃/ {v ⇤dsmâv {,. -t(vA nAmsh˙ ⇤Z b }úA vcnmb }vFt^; 4; Bgvn^ svDm‚ aAEdm@yA⌃tgocr. (vmAEd, svBtAnA\ nA⌃y, ktA mh ⇤òr; 5; (vmAEd, sv⇤ dvAnA\ ˙£A ⇤ dvo mh ⇤òr,. (vms⇥jo jg(sv◆ s⇥E£s\hArgocrm^; 6; p⇥QCAEm (vA mhAd ⇤v fv nArAyZA(mkm^. 1 The manuscript begins with : nm, Bgvto vAs⇤ dvAy, (!) • m⌃dr-T\ sKAsFn\ ] em. ; m⌃d-T\ sKAfFnA\ MS • gZAEctm^ ] em. ; gnAEŒt\ MS 2 i⌃d˝Aç {r^ ] em. ; i⌃d˝AdO MS • vFr ] em. ; vFy MS • Esà ] em. ; Es@y MS • -ttm^ ] em. ; -tt { MS 3 ⇤ dv\ ] em. ; ⇤ dv MS • mcnm^ ] em. ; mcn\ MS 4 `yj,sAmA ] em. ; `yjsAm MS 4 b }vFt^ ] em. ; b }vFt^, MS 5 Bgvn^ svDm‚ ] em. ; BgvA svDm‚A MS • gocr ] em. ; gocr\ MS 5 (vmAEd, ] em. ; (vyAEd MS • nA⌃y, ] em. ; nA(yo MS • mh ⇤òr ] conj. ; mh ⇤òr\ MS 6 (vmAEd, ] em. ; (vyAEd MS 6 (vms⇥jo jg(sv◆ ] conj. ; (vys⇥>jny(sv MSpc ; (vys⇥>jny MSac • s\hAr ] em. ; shAr MS 7 nArAyZA(mkm^ ] conj. ; nArAynA(mk MS

-vAyMBvp—rA/ ⇤

4

˛TmA@yAy,

[sgZ\ EngZ\ c {v ] [3]íy\ lok ⇤ ˛p>yt ⇤; 7; Ek\ mEtñ r^ - amEtñ ajAto jAt m^ - ev c. Ek\ n s#m\ EvjAnFyAE(k\ vA -TlEmhoQyt ⇤; 8; Ek\ Ev Z, skl, ˛oƒ, Ek\ vA En kl - m^ - ev vA. Ek\ vA&yƒ\ EvjAnFyAïƒ!pF c Ek\ Bv ⇤t^; 9; Ek\ En(y\ c aEn(y\ vA a"ro _n"r-tTA. svg, Ek\ n Ev‚ ⇤y, Ek\ vAsvgt, -m⇥t,; 10; Ek\ (n v)[ZA(mko Ev Z, Ek\ vA vZ][4]EvvEjt,. kT\ c -TA=yt ⇤ lok ⇤ loknATo jnAdn,; 11; ˛Et§A t-y ⇤ dv-y ôotEmQCAEm tÇvt,. k ⇤n d˝&y ⇤Z kt&yA t-y v { ˛EtmA fBA; 12; Ek\ ˛mAZ\ t Ev‚ ⇤ymΩ˛(yΩyo-tTA. frFr\ Ek\˛mAZ\ t kt&y\ v { Zv {jn {,; 13; 7cd–13 Dev¯ amr.tapa˜ ncar¯ atra 1.2cd–8:

Ek\ n mEtrmEtvA ajAto jAt ev vA; sgZ\ t kT\ ˛oƒ\ EngZ\ vA jgãro. Ek\ n s#m\ EvjAnFyAE(k\ vA -TlEmhoQyt ⇤; Ek\ Ev Z, skl, ˛oƒ, Ek\ n En kl - m^ - ev vA. Ekm&yƒ\ EvjAnFyAïƒ!p\ t Ek\ Bv ⇤t^; Ek\ n En(ymEn(y\ vA"ro _n"r ev vA. svg, Ek\ n Ev‚ ⇤y, Ek\ vAsvgt, -m⇥t,; Ek\ n vZA(mko Ev Z, Ek\ vA vZEvvEjt,. kT\ vA -TA=yt ⇤ Ev ZloknAT jgãro; ˛Et§A t-y ⇤ dv-y ôotEmQCAEm tÇvt,. kE-m⌃d˝&y ⇤ t kt&yA t-y v { ˛EtmA fBA; Ek\ ˛mAZ\ t Ev‚ ⇤ymΩ˛(yΩyo-tTA. frFr\ Ek\˛mAZ\ t kt&y\ l"ZAE⌃vtm^; 8 mEtñ r^ - amEtñ ajAto ] em. ; mEÅ— r^ - amEt— a>yto MS • jAtm ⇤v ] MSpc ; jAm ⇤v MSac • n ] conj. ; t MS • yAE(k\ vA -TlEmho ] em. ; yA EkMvA -TmEyho MS 9 Ev Z, skl, ˛oƒ, ] conj. ; Ev Z skl\ ˛oƒ\ MS • m^ - ev vA ] conj. ; m ⇤v dA, MS • &yƒ\ ] em. ; &yƒ MS • &yƒ!pF c ] MSpc ; &yƒ! c MS 10 En(y\ c aEn(y\ vA ] conj. ; EZ(yoñ aEn(yMvA MS • _n"r-tTA ] em. ; n"ro-tTA MS • svg, Ek\ n Ev‚ ⇤y, ] em. ; svg\ Ek\ t Ev‚ ⇤y ⇤ MS 11 nATo jnAdn, ] em. ; ZAT jnAàZ\ MS 12 tÇvt, ] em. ; t(vt\ MS • kt&yA t-y ] em. ; kt&y\ t-yA MS 13 Z\ t Ev‚ ⇤ymΩ˛(yΩ ] em. ; Z ⇤ t Ev‚ ⇤yA\ aΩ˛(yg MS • ˛mAZ\ t kt&y\ ] em. ; ˛mAn\ t kÅ&y MS • v {jn {, ] conj. ; v ⇤jn {, MS

5

-vAyMBvp—rA/ ⇤

˛TmA@yAy,

mkV-y ˛mAZ\ t Efrñ {v kT\ B(v ⇤t^) . llAV\ t pn-t-y kt&y\ (Ek\)[˛mAZt,; 14; nAEsk\ t] [5]kT\ kAy◆ kT\ n ⇤/mTAEdf ⇤t^. g⌫Xyo, Ek\ ˛mAZ\ t kZ◆ c {v kT\ Bv ⇤t^; 15; kTmo§O t kt&y\ Ecbk ⇤ Ek\ n l"Zm^. bAhjtAm^. bAlkAΩArpAqAZAn^ s\„(y km - m^ - aArB ⇤t^; 1 at, ] em. ; at MS • sm⌃tt, ] conj. ; sm⌃tt\ MS 2 pŒ ] em. ; pQC MS • p—MyA\ t ] em. ; p—My⌃t MS 4 et ⇤ sOMyg }hA, ˛oƒA, ] em. ; et(soMyg }hA ˛oƒA MS • E/qÅ ] em. ; t⇥sÅ MS • (yA\ aEò⌃yA\ roEhZFq ] conj. ; (yA aEò⌃yo roEhEZf MS 5 h-t ⇤ pnvsO cAEp p y ⇤Z ôvZ ⇤n ] em. ; h-t pnsoñAEp py ⇤n ôvn ⇤n MS • Br⌫yA\ c {v ] MSpc ; + + +— {v MSac • n"/ ⇤ ] em. ; n"/\ MS 6 vEjtAm^ ] em. ; vEjt\ MS • pAqAZAn^ ] em. ; pAfAnA\ MS

-vAyMBvp—rA/ ⇤

11

* p—mA@yAy,

bAlkAΩArpAqAZAn^ „(vA km smArB ⇤t^; 6; cth-tsmA(y)[ƒA\ v ⇤dF\ kyAt^[5]s]foBnAm^. m⌫Xp-y ˛mAZ\ t dfh-tsmAytm^; 7; svl"Zs\yƒA\ v ⇤dF\ kyAEíc"Z,. m⌫XpmAd˝fAKAEB, kAry ⇤EíEDpvkm^; 8; ndFs\gmtFT ⇤q bAlkA\ t/ dApy ⇤t^. upEl=y tto BEm\ kAry ⇤(k⌫Xl"Zm^; 9; ct koZsmAyƒ\ pvk⌫X\ t dApy ⇤t^. dE"Z ⇤ cADc⌃d˝\ t vAz⌫yA\ EdEf vtlm^; 10; pîAkAr\ t [aOdFQyA\ -TApy ⇤Œ] [6]sm⌃tt,. torZ\ sM˛v#yAEm p—h-tsmAytm^; 11; ⌃yg }oDo ’ dMbrñ {v EbSv, plAf - m^ - ev c. aò(Tñ fmF c {v c⌃dnmBy\ tTA; 12; 7–11ab Dev¯ amr.tapa˜ ncar¯ atra 6.10–15ab: cth-tsmAyƒA\ svl"Zs\ytAm^. pvoÅ-

r⁄vA\ c {v v ⇤dF\ Ev-tArfoEBtAm^; m⌫Xp\ cAd˝fAKAEB, kAry ⇤EíEDpvkm^. ctr˙\ ctíAr\ rMy\ íAdfh-tkm^; -tMB {, qoXfEByƒ\ EnE[Cd˝\ svt, fBm^. upEl=y tto BEm\ kAry ⇤(k⌫Xl"Zm^; pvk⌫X\ ct koZ\ som-TAn ⇤ ⌃ys ⇤à˝vm^. dE"Z ⇤ DnqAkAr\ bD-TAn ⇤ ˛kSpy ⇤t^; pEñm ⇤ vtl\ kyAãz-TAn ⇤ Env ⇤fy ⇤t^. uÅr ⇤ p¨jAkAr\ fá-TAn ⇤ ˛Et§y ⇤t^; IfAn ⇤ cá!p\ t k ⇤t-TAn ⇤ ⌃ys ⇤à˝vm^.

11cd–15 Dev¯ amr.tapa˜ ncar¯ atra 6.17cd–22: torZAn^ sM˛v#yAEm p—h-tsmAEôtAn^; ⌃yg }oDo ’ dMbrñ {v EbSv, pAlAf ev c. aò(Tñ fmF c {v c⌃dnAvByO t-

TA; cth-tA Bv ⇤(p”F sADh-t\ E/flkm^. h-t\ h-tA⌃tr ⇤ kyAE‡Kn ⇤à-tmA/km^; f v-/smAyƒA⌃dBmE£smE⌃vtAn^. udkMBs˙`dAmA⌃torZA⌃t/ dApy ⇤t^; aE`nmFl ⇤ ’ Et m⌃/ ⇤Z dApy ⇤(pvtorZm^. iq ⇤ (voj ⇤ ’ Et m⌃/ ⇤Z dE"Z ⇤ torZ\ ⌃ys ⇤t^; a`n aA yAEh m⌃/ ⇤Z dçA(pEñmtorZm^. f‡o ⇤ dvFEt m⌃/ ⇤Z ⌃ys ⇤dÅrtorZm^; 7 cth-t ] em. ; cth-tA MS • foBnAm^ ] em. , soBZ\ MS • ˛mAZ\ ] em. ; ˛mAn\ MS • smAytm^ ] em. , mAyt\ MS 8 yƒA\ v ⇤dF\ ] em. ; yƒ\ v ⇤dF MS • mAd˝fAKAEB, ] em. ; màfAKAEB, MS 9 kAry ⇤t^ ] em. ; kAry ⇤ MS 10 vtlm^ ] em. ; vtl\ MS 11 pîA ] em. ; pª MS 12 ñ {v EbSv, ] em. ; ñ {v\ EbSv MS • aò-Tñ ] em. ; aò-T-y MS • c⌃dnmBy\ tTA ] conj. ; c⌃d˝n\mByo-tTA MS

-vAyMBvp—rA/ ⇤

12

* p—mA@yAy,

[f ]v-/smAyƒAn^ dBcFrsmE⌃vtAn^. p pmAlAEctA\ñ {v torZ\ t/ dApy ⇤t^; 13; aE`nmF(l ⇤ ’ Et)[m⌃/ ⇤Z] [*7v1] (dAp)y ⇤(pvtorZm^. iq ⇤(v ⇤n t m⌃/ ⇤Z dE"Z\ torZ\ ⌃ys ⇤t^; 14; a`n aA yAhFEt m⌃/ ⇤Z pEñmAyA\ t dApy ⇤t^. f‡o ⇤ dvFEt m⌃/ ⇤Z dçAdÅrtorZm^; 15; klfA\ñ smAdAy ⇤ hmgBsmE⌃vtAn^. ò ⇤tc⌃dnp¨ ⇤n k⌫W ⇤ -vE-tkBEqtAn^; 16; yvfAElfrAv {ñ v-/Al\k⇥tEvg }hAn^. aAEjG } klfm⌃/ ⇤[Z t ⇤n {v klfAn^] [2]⌃ys ⇤t^; 17; dklk {Eñ/pV {, s\v ⇤˜ -tMBmAElkAm^. p p˙`dAmmAlAEB, p•v {ñopfoEBtAm^; 18; 14 RV 1.1.1: aE`nmFl ⇤ proEhtm^ • VS 1.1: iq ⇤ (voj ⇤ (vA 15 SV 1.1: a`n aA yAEh vFty ⇤ • AVP 1.1.1: f‡o ⇤ dvFrEB£y ⇤ 17 VS 8.42: aAEjG } klf\

mEh

16–20 Dev¯ amr.tapa˜ ncar¯ atra 6.23–27:

klfA\ñ smAdAy ⇤ hmgBsmE⌃vtAn^. ElÿA\ñ⌃dnp¨ ⇤n k⌫W ⇤ -vE-tkBEqtAn^; yvfAElfrAv {ñ v-/Al\k⇥tEvg }hAn^. aAEjG } klfm⌃/ ⇤Z t ⇤n {v klfAn^ ⌃ys ⇤t^; dklEc/p” {ñ v ⇤£y ⇤(-tMBmAElkAm^. p p˙`dAmmAlAEB, p•v {ñopfoEBtAm^; @vjAdfptAkAEB, cAmr {ñ EvtAnk {,. G⌫VAfNdEnnAd {ñ g ⇤ymΩlvAEdEB,; f≤vAEd/EnGoq {v ⇤d@vEnsmE⌃vt {,. p⌫yAhjyfNd {ñ kAry ⇤t mho(svm^; 13 yƒAn^ ] conj. ; yƒA MS • smE⌃vtAn^ ] conj. ; smE⌃vtA MS • mAlAEctA\ñ {v ] conj. ; mAlAEctA c {v MS • torZAn^ ] em ⇤⌃d^ ; torZn^ MS 14 torZm^ ] em. ; soBZ\ MS • iq ⇤(v ⇤n ] em. ; iq(v ⇤n t MS 15 a`n aA yAhFEt ] em. ( unmetrical ); aE`n yAEhEB MS • f‡o ] em. ; f⇥‡o MS. An unfinished aks.ara is canceled after dçA. 15 f‡o ] em. ; f⇥‡o MS • dçAdÅr ] MSpc ; dçA⌃ddÅr MSac 16 klfA\ñ ] conj. ; klqAEB MS • smE⌃vtAn^ ] em. ; smE⌃vt\ MS • ò ⇤tc⌃dnp¨ ⇤n ] em. ; s ⇤tc⌃d˝Zp¨ ⇤Z\ MS • BEqtAn^ ] em. ; rEqt, MS 17 Evg }hAn^ ] em. ; Evg }hA MS • aAEjG } klf ] em. ; aAEjG }A kf\ MS. This p¯ ada is unmetrical. • mAElkAm^ ] em. ; mAElkA MS 18 pV {, s\v ⇤˜ ] conj. ; pV {ñv ⇤£ MS • mAElkAm^ ] em. ; mAElkA MS • foEBtAm^ ] conj. ; foEBt\ MS

-vAyMBvp—rA/ ⇤

13

* p—mA@yAy,

@vjAdfptAkAEB, cAmr {ñ EvtAnk {,. G⌫VAfNdEvEc/ {ñ g ⇤ymΩlvAck {,; 19; f≤vAEd/EnGoq {v ⇤d@vEnsmE⌃vt {,. p⌫yAhjyfNd {ñ kAry ⇤t mho(svm^; 20; ptAkAEn EvEc/AEZ @vjmAlop(fo)[EBtAm^] . [3]ptAkA\ pFEtkA\ c {v pv - Em⌃d˝Ay dApy ⇤t^; 21; aA`n ⇤⇧yA\ rƒvZABA\ yAMyAyA\ [yAmvEZkAm^. nFlA“nsvZA◆ t n { (yA\ t ˛dApy ⇤t^; 22; vAz⌫yA\ EstvZABA\ k⇥ ZA\ vAy&ygocr ⇤. hErtAmÅr ⇤ c {v IfA⌃yA\ svvEZkAm^; 23; ò ⇤trƒkcZ ⇤n aAElK ⇤í ⇤Edm⌫Xlm^. 24 VS 19.17: v ⇤çA v ⇤Ed, smA=yt ⇤ 21–23 Dev¯ amr.tapa˜ ncar¯ atra 6.28–30:

@vjd⌫X˛mAZAEn qoXfAΩlEv-trA,. p—h-tAEn svAEZ p pDpAEct\ bElm^; ptAkA\ pFEtkA\ c {v pv - m^ - i⌃d˝Ay dApy ⇤t^. aA`n ⇤⇧yA\ rƒvZA◆ t yAMyAyA\ [yAmvEZkAm^; nFlA“nEnBAkArA\ n { (yA\ t ˛dApy ⇤t^. vAz⌫yA\ EstvZABA\ pFtA\ vAy&ygocr ⇤; hErtA\ coÅr ⇤ dçAdFfA⌃yA\ svvEZkAm^. ò ⇤tA\ t nAgrAjAy ò ⇤tA\ b }úEZ dApy ⇤t^; 24 –26 Dev¯ amr.tapa˜ ncar¯ atra 6.31–33: ò ⇤trƒkcZ ⇤n aAElK ⇤í ⇤Edm⌫Xlm^. v ⇤çA v ⇤dFEt m⌃/ ⇤Z v ⇤dFmAElHy m⌃/Evt^; pvAg }AnÅrAg }An^ vA kfA\-tFy Evc"Z,. yog ⇤ yog ⇤ ’ Et m⌃/ ⇤Z kAry ⇤(-trZ\ bD,; f⇧yA t/ ˛kt&yA tElp”A\fk {-tTA. g⌫Xk ⇤ ⇤ í EvEc/ ⇤ c -TApy ⇤⇤ íEdm@yt,; 19 df ] em. ; dp MS • EvtAnk {, ] em. ; EvtAnk\ MS • G⌫VA ] em. ; G⌫W MS • vAck {, ] em. ; vAck { MS 20 EnGoq {v ⇤d ] em. ; EnGoq {v ⇤d MS • p⌫yAh ] em. ; pjy ⇤ MS • sdA ] conj. ; -yvA, MS • Colophon: q§o _@yAy, ] em. ; q§mo @yA MS

; sÿmA@yAy,; at, pr\ ˛v#yAEm aEDvAsnmÅmm^. ut vAt ⇤n m⌃/ ⇤Z kyAd^ b }úrT\ fBm^; 1; sh˙fFqA pzq, m⌫Xp\ t ˛v ⇤fy ⇤t^. fMBvAy ⇤Et m⌃/ ⇤Z f⇧yAyA\ t Env ⇤fy ⇤t^; 2; Evòt, c["Er(y ⇤v\ kyA(s][3]klEn klm^. pnr ⇤v\ ˛v#yAEm sklFkrZmÅmm^; 3; ˛EtmAyA\ Evf ⇤q ⇤(Z) kyAŒAlBn\ bD,. : nmo Bgvt ⇤ vAs⇤ dvAy. m⌃/Emd\ ˛oƒm^. 4 1 RV 10.186.2: ut vAt EptAEs n, 2 RV 10.90.1: sh˙fFqA pzq, nm, f\BvAy c 3 RV 10.81.3 = VS 17.19: Evòtñ"zt

• VS 16.41:

1cd–3ab Dev¯ amr.tapa˜ ncar¯ atra 9.38–40ab:

uEŧ ⇤Et smà⇥(y k⇥(vA b }úrT\ fBm^. sh˙fFqA pzq, m⌫Xp\ t ˛v ⇤fy ⇤t^; tt, ˛v ⇤fy ⇤⇤ åvmG◆ dçA(˛y◊t,. fMBvAy ⇤Et m⌃/ ⇤Z f⇧yAyA\ t Env ⇤fy ⇤t^; Evòtñ"Er(y ⇤v\ kyA(sklEn klm^. 2cd–8 Soma´sambhupaddhati 4.6.8–10ab:

mEt!p\ c f⇧yAyA\ Env ⇤[y tdn⌃trm^. íAdfA"rm⌃/-y k⇥(vA ⌃yAsmn⌃trm^; K≥AdFnA\ c Ev⌃yAs\ m⌃/F kyAŒ tçTA. EfroG }AZllAV ⇤òA-yk⌫W„dy ⇤q c; d"dovAmdovAmd"jøAE¿q ámAt^.

3cd–4 Dev¯ amr.tapa˜ ncar¯ atra 11.1–2: aTAcn\ ˛v#yAEm sklFkrZ\ fBm^. -TApk ⇤n

t kt&y\ -vd ⇤⇤ h ⌃yAsmÅmm^; ˛EtmAyA\ Evf ⇤q ⇤Z kyAdAlBn\ bD,. íAdfA"rm⌃/\ t mlm⌃/mdA„tm^. : nmo Bgvt ⇤ vAs⇤ dvAy; 1 at, ] em. ; at MS • ˛v#yAEm ] em. ; ˛v#yAEm, MS 1 ut vAt ⇤n ] em. ; u vAt ⇤n t MS • rT\ ] em. ; rT MS 2 m⌫Xp\ ] em. ; m⌫Xl\ MS • fMBvAy ⇤Et m⌃/ ⇤Z f⇧yAyA\ ] conj. ; sMBvAyoEn n ⇤Å ⇤Z q⇧yAyA\ MS 3 The last p¯ada ~ is unmetrical. 4 kyAŒAlBn\ ] conj. ; kyAŒAln\ MS • nmo Bgvt ⇤ vAs⇤ dvAy ] em. ; nmo, Bgvt { vAs⇤ dvAy, MS

-vAyMBvp—rA/ ⇤

17

sÿmA@yAy,

:kArAEdykArA⌃tAn^ Ev⌃ys ⇤dnpvf,. :kAr\ Ev⌃ys ⇤⌃mE nkAr\ nAEskA⌃tr ⇤; 5; mokAr-t llAV ⇤ _y\ (BkAr\ vdn ⇤) ⌃y[s ⇤t^. gkAr\ k⌫Wd ⇤][4] (f ⇤ t) vkAr\ „dy ⇤ ⌃ys ⇤t^; 6; t ⇤kAr\ t Bj ⇤ ⇤ dy\ dE"Z ⇤n mhAblm^. vAm ⇤ Bj ⇤ vAkAr\ t EcE⌃tt&y\ Evc"Z; 7; skAr\ vAmj\GAyA\ ⇤ dkAr\ dE"Z ⇤ tTA. vAkAr\ vAmpAd ⇤ t ykAr\ dE"Z ⇤ ⌃ys ⇤t^; 8; @yAn\ t ⇤ s\˛v#yAEm svpAp˛Zodnm^. :kAr\ t sdA @yAy ⇤>>vAlA(mA)[lAsm˛Bm^; 9; 5cd–8 Dev¯ amr.tapa˜ ncar¯ atra 11.3–5: :kAr\ Ev⌃ys ⇤⌃mE nkAr\ nAEskopEr. mokAr\

t mK ⇤ ⌃y-y BkAr\ k⌫Wgocr ⇤; gkAr\ bAh⇤ df ⇤ t vkAr\ „dy ⇤ ⌃ys ⇤t^. t ⇤kAr\ nAEB⇤ df ⇤ t vAkAr\ p⇥§d ⇤ft,; skAr\ kEVm@y ⇤ t ⇤ dkAr\ cozm@yt,. vAkAr\ j\Gyo⌃y-y ykAr\ pAdyo⌃ys ⇤t^; 9 cd–14 Dev¯ amr.tapa˜ ncar¯ atra 11.6cd–12ab: :kAr\ t sdA @yAy ⇤>>vAlAmAlAsm˛Bm^; nkAr\ f vZAB\ nkAr\ fFtl\ jlm^. mokAr\ Ec⌃ty ⇤(˛A‚o EB‡A“nsm˛Bm^; tzZAEd(ys\kAf\ BkAr\ Ec⌃ty ⇤(sdA. gkAr\ t pn-t ⇤qA\ ⇤ hmvZ◆ t Ec⌃ty ⇤t^; fà-PEVks\kAf\ vkAr\ @yAnmÅmm^. t ⇤kAr\ Ec⌃ty ⇤(˛A‚o Ed&yk⌃dsm˛Bm^; ⇤ hmk⌃dEnBAkAr\ vAkAr\ vzZAlym^. Evç(p“EnBAkAr\ skAr\ t/ Ec⌃ty ⇤t^; ⇤ dkAr\ t pn@yAy ⇤⌃mhAnFlsm˛Bm^. "FrodAZvs\kAf\ vAkArmm⇥tA"rm^; ykAr\ t sdA @yAy ⇤(pîrAgsm˛Bm^. 9–14 Soma´sambhupaddhati 4.6.10cd–12ab:

>vAlAmAlAklò ⇤tm ⇤ckAEd(ys˛BAn^; pFtf AzZò ⇤tEvç(pFtEstAzZAn^. ao¨ArAEdykArA⌃tAn^ vZAn^ íAdf Ev⌃ys ⇤t^; -vyMBpA—rA/ ⇤ c svm ⇤tddFErtm^. 5 Ev⌃ys ⇤dn ] em. ; Ev\⌃ys ⇤çn MS • :kAr\ Ev⌃ys ⇤n^ ] em. ; :kAr Ev⌃ys ⇤ MS • nkAr\ nAEskA⌃tr ⇤ ] em. ; nkArA‡AEskA⌃tr ⇤t^ MS 6 llAV ⇤ _y\ ] conj. ;

llAV ⇤y MS 7 t ⇤kAr\ ] em. ; t {kAr\ MS • vAkAr\ t ] conj. ; vAkAr\ MS • t&y\ Evc"Z ] em. ; t&y Evc"Z\ MS 8 r\ vAmj\GAyA\ ] em. ; rA\ vAmjGAyA\ MS • t ] em. ; t, MS • dE"Z ⇤ ] em. ; dE"n\ MS 9 @yAn\ t ⇤ ] conj. ; @yAn\ MS • ˛Zodnm^ ] em. ; ˛codn\ MS • :kAr\ ] conj. ; ao\kArA\ MS

-vAyMBvp—rA/ ⇤

18

sÿmA@yAy,

nkAr\ f v] [5]ZAB\ nkAr\ fFtl\ jlm^. mokAr\ nFlEnB\ p“\ EB‡A“nsm˛Bm^; 10; tzZAEd(ys\kAf\ BkAr\ Ec⌃ty ⇤(sdA. gkAr\ t pn-t ⇤qA\ ⇤ hmvZ◆ EvEc⌃ty ⇤t^; 11; f pîEnBAkAr\ vkAr\ @yAnmÅmm^. t ⇤kAr-y t Ev‚ ⇤y\ >yoEt, k\kmsE‡Bm^; 12; Ehmk⌃dEnB\ (c {)[v vAkAr\ vzZAlym^] . [*Yv 1]Evç(p“EnBAkAr\ skAr\ t/ Ec⌃ty ⇤t^; 13; @yAy ⇤⇤ åkArbFj\ t pFt\ mAh ⇤⌃d˝sE‡Bm^. "FrodAZvsAd⇥[y\ vAkAr\ vzZAlym^. ykAr\ t sdA @yAy ⇤(pîrAgsm˛Bm^; 14; n tA\ gEt\ yAE⌃t srA n yogF n c yAE‚kA,. yA\ gEt\ yAE⌃t lok ⇤ _E-mn^ íAdfA"r[Ec⌃tkA,; 15; : h\ Ev Zv ⇤ „dyAy] [2] nm,. : Ev Zv ⇤ Efrs ⇤ nm,. : pînABAy EfKAy { nm,. : D }vAy kvcAy nm,. 15 Dev¯ amr.tapa˜ ncar¯ atra 11.15: n tA\ gEt\ yAE⌃t srA n yogF n c yAE‚kA,.

yA\ gEt\ yAE⌃t yog ⇤ _E-mn^ íAdfA"rEc⌃tkA,; Brahmapur¯ an.a 57.31 n tA\ gEt\ srA yAE⌃t yoEgno n {v sompA,. yA\ gEt\ yAE⌃t Bo Ev˛A íAdfA"rt(prA,; 10 nkAr\ ] em. ; nkArA\ MS • mokAr\ ] em. ; mokArA\ MS • p“\ ] em. ; pj\ MS. This p¯ ada is unmetrical. • EB‡A“n ] em. ; EB‡A“n MS 11 BkAr\ ] em. ; BkAr MS • gkAr\ t pn ] em. ; gkArA⌃t kr MS • vZ◆ ] em. ; võ MS 12 pî ] em. ; pª MS • Ev‚ ⇤y\ >yoEt, ] em. ; Ev‚ ⇤yAm^ joEt MS

13 En-

B\ ] em. ; EnB MS • Evç(p“ ] em. ; Evçp“ MS 14 @yAy ⇤d^ ] em. ; @yAy ⇤ MS • pFt\ ] em. ; pFt MS • "FrodAZv ] em. ; "FrodA‡v MS • sAd⇥[y\ ] conj. ; sA†˙\ MS • ykAr\ t ] em. ; ykArA⌃t MS • pî ] em. ; pª MS 15 n tA\ gEt\ yAE⌃t n srA Z ] em. ; n tA gtF yAE⌃t n srA Z MS • yAE‚kA, ] MSpc ; yAE‚kAkA, MSac • yA\ ] em. ; yA MS 16 Ev Zv ⇤ ] em. ; v { Zv ⇤ MS • pî ] em. ; p(m MS • kvcAy ] em. ; kvcAy { MS

-vAyMBvp—rA/ ⇤

19

sÿmA@yAy,

: cEáZ ⇤ nm, – a-/m^. : sMBvAy { nm, – gAy/F. : EvjyAy { nm, – sAEv/F. : n ⇤/AyAfny ⇤ >yoEt!pAy nm, – n ⇤/m^. : aGorAy nm, – EpΩlA-/m^. h-tyo,. 16. [yjn\] [3]s\˛v#yAEm { dv\ nArAyZ-y t. E/EBrAvrZ {, kAy◆ dlB\ ssrAs{ r,; 17; m@y ⇤ cá\ ˛Et§A=y íAdfAr\ sfoBnm^. t⌃m@y ⇤ kml\ ˛oƒ\ p/A£kskEZkm^; 18; svA(mA sklo ⇤ dv, Ed&ymAlAsnAtn,. EôyA m@y ⇤ t „dy\ h\kA(r ⇤Z) t pjy ⇤t^; 19; Efr, pvdl ⇤ d[çAåE"Z ⇤ t Ef] [4]KA\ ⌃ys ⇤t^. 17 Dev¯ amr.tapa˜ ncar¯ atra 11.25:

yjn\ s\˛v#yAEm Ed&y\ nArAyZ-y t. E/EBrAvrZ {, kAy◆ dlB\ t srAs{ r,; As.t.a ¯da´savidh¯ ana verse 102:

yjn\ s\˛v#yAEm { dv\ nArAyZ\ -m⇥tm^. ctrAvrZ {, kAymEct\ ssrAs{ r,; 18cd–21 Dev¯ amr.tapa˜ ncar¯ atra 11.26–30ab:

m@y ⇤ cá\ ˛Et§A=y ar {íAdfEBytm^. t⌃m@y ⇤ kml\ ˛oƒm£p/\ skEZkm^; m@y ⇤ t „dy\ p>y h\kAr ⇤Z mhA(mnA; Efr, pvdl ⇤ dçAåE"Z ⇤ t EfKA\ ⌃ys ⇤t^. pEñm ⇤ kvc\ dçAd-/\ v { coÅr\ ⌃ys ⇤t^; aA`n ⇤⇧yA\ EdEf gAy/F\ sAEv/FmFfgocr ⇤. n ⇤/ ⇤ c {v t n { (yA\ vAy&yA\ EdEf EpΩlm^; etãùtr\ HyAt\ gBAvrZmÅmm^. 18cd–21 cf. Soma´sambhupaddhati 4.6.27cd–31a:

Ev⌃y-y cAEdtñá\ íAdfAr\ sBA-vrm^; t-y m@y ⇤ pnd ⇤y\ pîm£dl\ tt,. „⌃m⌃/\ kEZkAyA\ c Efr, pvdl ⇤ tt,. EfKA\ c dE"Z ⇤ p/ ⇤ pEñm ⇤ kvc\ ⌃ys ⇤t^; a-/mÅrto ⌃y-y gAy/FmE`np/k ⇤; sAEv/FmFfp/ ⇤ c n ⇤/\ c n { t ⇤ dl ⇤. ttñ vAyp/ ⇤ c EpΩlA-/\ EvEnE"p ⇤t^; gBAvrZEm(yƒm^ 16 cEáZ ⇤ ] conj. ; aA cEáZ ⇤ MS • a-/m^ ] em. ; aA-/, MS • yAfny ⇤ >yo-

Et ] conj. ; yAfnFy { &yot MS • n ⇤/m^ ] conj. ; n ⇤/ MS • aGorAy ] em. ; aGorAy, MS • -/\ h-tyo, ] conj. ; -/ h-tAyo, MS 17 { dv\ ] em. ; ⇤ dv\ MS • Z-y t ] em. ; n-y t, MS 18 ˛Et§A=y íAdfAr\ ] em. ; ˛Et§AyA\ íAdfAn MS • t⌃m@y ⇤ ] em. ; t(m@y ⇤ MS 19 sklo ⇤ dv, ] em. ; sklA ⇤ dv MS 20 Efr, ] em. ; Efr MS • dçAd-/\ ] em. ; dçAv-/\ MS

-vAyMBvp—rA/ ⇤

20

sÿmA@yAy,

pEñm ⇤ kvc\ dçAd-/\ c {voÅr ⇤Z t; 20; gAyÏyA`n ⇤yEd`BAg ⇤ sAEv/FmFòr ⇤ -vym^. n ⇤/\ c {v t n { (yA\ EpΩlA-/\ t vAyv ⇤. gùAãùtr\ gù\ gBAvrZmÅmm^; 21; EítFy\ s\˛v#yAEm Ev ZmEt◆ ˛pjy ⇤t^. íAdfAr ⇤ tTA cá ⇤ ⌃ys ⇤d^ (íA)dfmty,; 22; (k ⇤) [fv\ t ar ⇤ pv ⇤ :kA] [5]⇤ rZ t pjy ⇤t^; EítFy ⇤ t nkAr ⇤Z ‚ ⇤y\ nArAyZ\ tTA; 23; t⇥tFy ⇤ mADv\ p>y mokAr ⇤Z mhA(mn,. BkArA"rd ⇤v ⇤n goEv⌃d\ t ctTk ⇤; 24; p—m ⇤ t gkAr ⇤Z Ev Z\ c {v ˛pjy ⇤t^. vkArA"rd ⇤v ⇤n q§ ⇤ v { mDsdnm^; 25; 22–28 Dev¯ amr.tapa˜ ncar¯ atra 11.30cd–37:

EítFy\ s\˛v#yAEm sv⇤ dvmy\ -ttm^; íAdfAr ⇤ tTA cá ⇤ arAg ⇤ }q Evc"Z,. íAdfA"rs\yƒA ⌃ys ⇤d^ íAdfmty,; k ⇤fv\ t ar ⇤ pv ⇤ :kAr ⇤Z t pjy ⇤t^. EítFy\ t nkAr ⇤Z ⇤ dv\ nArAyZ\ yj ⇤t^; t⇥tFy ⇤ mADv\ p>y mokAr ⇤Z mhA(mnA. BkArA"rd ⇤v ⇤n goEv⌃d\ t ctTkm^; p—m ⇤ c gkAr ⇤Z Ev Z\ c {v t pjy ⇤t^. vkArA"rs\yƒ\ q§ ⇤ t mDsdnm^; E/Evám\ tt, pñAÅ ⇤kAr ⇤Z t pjy ⇤t^. vAmn\ cA£m ⇤ c {v vAkAr ⇤Z t pjy ⇤t^; ôFDr\ nvm ⇤ p>y skAr ⇤Z mhA(mnA. dfm ⇤ t „qFk ⇤f\ ⇤ dkArA"rpEjtm^; ekAdf ⇤ pînAB\ vAkAr ⇤Z Enyojy ⇤t^. íAdf ⇤ t ykAr ⇤Z yƒ\ dAmodr\ ⌃ys ⇤t^; 22cd–31 cf. Soma´sambhupaddhati 4.6.31b–33cd:

aDnAvrZA⌃trm^. íAdfAr ⇤ c cá ⇤ _E-mn^ k ⇤fvAçA⌃yTAámm^; ˛ZvAç {yTAkArmƒpv {, -vnAmEB,. ˛AgAEdtñ Ev⌃y-y K≥\ gdAmn⌃trm^; cá\ f≤\ c pî\ c hl\ c msl\ tt,. fAΩ◆ c Ev⌃ys ⇤d ⇤v\ t⇥tFyAvrZ\ Bv ⇤t^; 21 Ed`BAg ⇤ ] em. ; Ed`BAg\ MS • n { (yA\ ] em. ; n {Er(yA\ MS • vAyv ⇤ ] em. ; vAyv ⇤t^ MS 22 EítFy\ ] em. ; EítFyA\ MS • Em, Ev ZmEt◆ ] em. ; Em, Ev ZmEÅ MS • ⌃ys ⇤d^ íA ] em. ; ⌃ys ⇤ íA MS 23 EítFy ⇤ t n ] em. ; EítFy t Z MS • ‚ ⇤y\ nArAyZ\ tTA ] em. ; ‚ ⇤yA\ nArAyZ⌃tTA, MS 24 t⇥tFy ⇤ ] em. ; t⇥tFy\ MS • p>y ] conj. ; p>y\ MS • ctTk ⇤ ] em. ; ctTk {, MS 25 p—m ⇤ ] em. ; p—m\ MS • Ev Z\ ] em. ; Ev Z MS • d ⇤v ⇤n ] em. ; d ⇤v ⇤Z MS

-vAyMBvp—rA/ ⇤

21

sÿmA@yAy,

sÿm ⇤ vAmn\ c {v t ⇤kAr ⇤Z t pjy ⇤[t^. r ⌃y[*X 1]s ⇤]íAkArbFj ⇤n a£m ⇤ t E/Evámm^; 26; ôFDr\ nvm ⇤ c {v skAr ⇤Z t pjy ⇤t^. dfm ⇤ t „qFk ⇤f\ ⇤ dkAr ⇤Z t pjy ⇤t^; 27; ekAdf ⇤ t vAkAr\ pînAB\ ˛B\ Evd,. íAdf ⇤ t ykAr ⇤Z nAflA dAmodr, -m⇥t,; 28; EítFyAvrZ\ HyAt\ t⇥tFy ⇤ _-/AEZ Ev⌃ys ⇤t^. f≤\ c {v ⌃ys ⇤(pvmA`n ⇤⇧yA[⌃t gdA\ ⌃ys ⇤t^; 29; [2]d](E")Z ⇤n

Bv ⇤Œá\ K≥\ n { (ygocr ⇤.

pî\ pEñmto EvçAíAy&yA\ t hl\ ⌃ys ⇤t^; 30; msl\ coÅr ⇤ dçAdFfA⌃yA\ fAΩ ’ Ev⌃ys ⇤t^. etãùtr\ yAg\ dlB\ prm\ pdm^; 31; sv⇤ dvmy\ k⇥(vA ⌃yAs\ v { ZvmÅmm^. ˛AsAdAEBmK\ ⇤ dv\ f⇧yAyA\ t Env ⇤fy ⇤t^; 32; aE`nkAy◆ tt, k⇥(vA p(jA)[v ⇤d[3]˛]coEdtA. 29–31ab Dev¯ amr.tapa˜ ncar¯ atra 11.38cd–40:

EítFyAvrZ\ HyAt\ t⇥tFy ⇤ _-/AEZ Ev⌃ys ⇤t^; K≥\ c {v ⌃ys ⇤(pv ⇤ aA`n ⇤⇧yA\ t gdA\ ⌃ys ⇤t^. dE"Z ⇤ Ev⌃ys ⇤Œá\ f\K\ En Etgocr ⇤; pî\ pEñmto ⌃y-y vAy&yA\ t hl\ ⌃ys ⇤t^. msl\ coÅr ⇤ dçAdFfA⌃yA\ f⇥EΩ Ev⌃ys ⇤t^; 26 c {v ] em. ; c {v, MS • E/Evámm^ ] em. ; E/Evám, MS

27 nvm ⇤ c {v ] em. ; nvm— {v MS 28 vAkAr\ ] em. ; vAkAr MS • ˛B\ ] em. ; ˛B MS • t ykAr ⇤Z ] em. ; BkAr ⇤Z MS • r, -m⇥t, ] em. ; r\ -m⇥t\ MS 29 Z\ HyAt\ t⇥tFy ⇤ _-/AEZ ] em. ; Z HyAtA\ t⇥tFy ⇤ n-/AEt MS • f≤\ ] em. ; f≤ MS • ⌃ys ⇤(pvmA`n ⇤⇧yA ] em. ; ⌃ys ⇤ p®◆ aA`n ⇤yA MS 30 Bv ⇤c^ ] em. ; Bv ⇤ MS • n { (ygocr ⇤ ] em. ; n {Er(ygocr ⇤t^ MS • pî\ ] em. ; pî MS • EvçAd^ ] em. ; EvçA MS • hl\ ] em. ; hl MS 31 msl\ coÅr ⇤ ] em. ; msl co⌃tto MS • ⌃yA\ fAΩ ’ ] conj. ; ⌃yA sArAΩ ’ MS • etd^ ] em. ; et MS • yAg\ dlB\ ] em. ; yAgA\ dlB\ MS 32 dAEBmK\ ⇤ dv\ ] em. ; dEBmK ⇤ dv MS 33 kAy◆ ] em. ; kAy MS • bAü⇥c\ pvk⌫X ⇤ ] em. ;

bAü⇥k\ pvk⌫X\ MS • yAMy ⇤ _@vy◆ tT {v ] em. ; yAm ⇤ @vy-tT {v MS

-vAyMBvp—rA/ ⇤

22

sÿmA@yAy,

bAü⇥c\ pvk⌫X ⇤ t yAMy ⇤ _@vy◆ tT {v c; 33; pEñm ⇤ c {v C⌃dogmÅr ⇤Z aTvZm^. ôFsƒ\ pAvmAn\ c nAsdAHy\ svAEckm^; 34; v⇥qAkEp\ c e ⇤⌃d˝A`n\ bAü⇥c, pvto yj ⇤t^. zd˝\ pzqsƒ\ c öokA@yAy\ sfEáym^; 35; b }AúZ\ Ept⇥m {/\ cA@vydE"Zto yj ⇤t^. ⇤ dvv }t\ pzqgEt\ >y ⇤§sAm rT⌃tr[m^; 36; BAz[4]⌫X]AEn c sAmAEn gAy/F\ c {v coÅr ⇤. vAmd ⇤&y\ c QC⌃dog, sdA pEñmto yj ⇤t^; 37; aTvEfrs\ c {v -kMBsƒ\ tT {v c. Bd˝kSpA, pErqd, pEv/An^ gZpAWkAn^; 38; nFlzd˝\ †n ⇤/yo† _Tv uÅrto yj ⇤t^. 34 ôFsƒm^ RV khila. 2.6 ; pAvmAnm^ RV 9.67.21–27 ; nAsdAHym^ RV 10.129 ; vAEckm^ RV 10.125 35 v⇥qAkEp RV 10.86 ; e ⇤⌃d˝A`n RV 8.40 ; zd˝sƒm^ or zd˝A@yAy, VS 16 ; pzqsƒm^ VS 31 ; öokA@yAy, No definite identification but VS 20 is a candidate. ; fEáym^ TA 4.1–42 36 b }AúZ\ According to Baudh¯ayanaGS, TB ¯ G¯ 2.8.8.67–69 ; EpÏy@yAy, VS 35 ; m {/m⌃/, VS 39.5 ; ⇤ dvv }tm^ SV A. a. 212–214 ; pzqgEtsAm SV 1.594 ; >y ⇤§sAm SV 1.31 = RV 6.7.1 (according to Haradatta), SV 1.33 = RV 1.115.1 (according to Govinda); cf. Olivelle 2000: 712 ; rT⌃trsAm RV ¯ G¯ ¯ G¯ 7.32.22 = SV 1.233; A. a. 76 37 BAz⌫XsAm SV A. a. pari. 7 ; gAy/F RV 3.62.10 sung in S¯ aman manner ; vAmd ⇤&ysAm RV 4.31.1 = SV 1.169 38 aTvEfr, MuU 1.1, according to Nandapan.d.ita’s Commentary on Vis.n.usmr.ti 56.22, cf. Olivelle 2000: 711 ; -kMBsƒm^ AV 10.7.1–44 39 nFlzd˝m^ = nFlzd˝opEnqd^ ( see Lubin 2007 ) 34 aTvZm^ ] em. ; aTvZ, MS • nAsdAHy\ ] em. ; nAsAdAs\My MS 35 e ⇤⌃d˝A`n\ bAü⇥c, ] em. ; e ⇤⌃d˝A bAü⇥c MS • yj ⇤t^ ] em. ; y‚ ⇤t^ MS • Bd˝\ ] em. ; Bd˝ MS • öokA@yAy\ sfEáym^ ] conj. ; aöokAy, spEd˝ym^ MS 36 Ept⇥m {/\ cA@vyd ] conj. ; p/m ⇤Å~ñ@vyd MS 36 The third p¯ada is unmetrical. • >y ⇤§sAm ] conj. ; [j ⇤ ] §AmAq MS • rT⌃t ] em. ; rTA⌃t MS 37 sAmAEn ] em. ; sAmAEZ MS • gAy/F\ ] em. ; gAy/F MS 37 vAmd ⇤&y\ c ] em. ; vAmd ⇤vAy MS • -kMB ] em. ; -k⌃d MS • c. ] em. ; c, MS 38 kSpA, pErqd, ] em. ; kSpo pErEqqd, MS • pEv/An^ ] conj. ; pEv/A MS 39 fdMby ] em. ; ldMbyA MS • mAg-tT {v ] em. ; mAg tT {v MS

23

-vAyMBvp—rA/ ⇤

sÿmA@yAy,

fmFplAfdMby apAmAg-tT {v c; 39; íAdf {v sh˙AEZ a£O c(vAEr (c {)[v vA] . (íyO) /FEZ f[tA⌃y ⇤[5]v](m)T◆ c {k\ t homy ⇤t^; 40; aE`nmD ⇤Et m⌃/ ⇤Z kA⌫XyoñAln\ tTA. u• ⇤KAytA\ ] conj. ; p>ytA MS 29 yEnzp ] em. ; yEnzp MS • pAlyt ⇤ p⇥LvF\ Dn ] conj. ; pAlytA p⇥LvF DnA MS 30 yE-mn^ p>yt ⇤ t(-TAn ⇤ ] conj. ; yso p>yyt ⇤ -TAn\ MS • The second p¯ ada is unmetrical. • p>ytA\ yjmAnAT◆ ] conj. ; p>ytA yjmAnAy, MS 31 ˛ATn ] em. ; ˛ATZ MS 32 : nmo nm, ] em. ; : mo nm MS • m ⇤§+A(mn ⇤ ] em. ; m ⇤§A(mn ⇤ MS 33 mnnA ] conj. ; v {nA MS • ⌃yAs\ kzt ⇤ ] conj. ; ⌃yAs,; "zt ⇤ MS 34 : nmo ] conj. ; : mo MS • ˛EZp(y yTAEvED ] em. ; ˛Enp(y yTAEvED, MS 35 mhorA/ ⇤Z ] conj. ; mvArA/ ⇤Z MS • The third p¯ada is a bad ma-vipul¯ a. 37 kyA•"ZlE"tm^ ] em. ; kyAl"ZlE"t, MS 37 t⌃m@y ⇤ ] em. ; t(m@y ⇤ MS • (pî\ ] em. ; (p(m\ MS 38 kt&yA ] em. ; kÅ&y\ MS • &yhsmE⌃vtA ] em. ; &yEhsmE⌃vt, MS

-vAyMBvp—rA/ ⇤

30

a£mA@yAy,

lokpAlA g }hA, p>yA, svEd" smE⌃vtA,; 38; pjEy(vA tt, pñAt^[2]++++++++. · · · · · · · · · · · · · · · · · · · · · · · · · · · · · ·; 39; [v ⇤dvA](d)rt {, fA⌃t {, Ev ZfA-/EvfArd {,. pîm@y ⇤ t Ev⌃y-y klf\ Ed&y!EpZm^; 40; sfMbr\ dFEÿm⌃t\ toypZ◆ gZAE⌃vtm^. al\ t v ⇤dk {m⌃/ {, p—EBvzZd {vt {,; 41; t/ r◊AEn ⇤ dyAEn DAtbFjAEn c {v Eh. aAm }AEdp•v {ñ {v sg⌃D {, p pdAmk {,; 42; an ⇤n (Ev)[EDnA] [3]++++++++++. · · · · · · · · · · · · · · · · · · · · · · · · · · · · · ·; 43; tt, s\‹AEpt\ ⇤ dv\ p p\ Dp\ c dApy ⇤t^. aA =yAy-v ⇤Et m⌃/ ⇤Z yA t ⇤ zd˝ EfvA tn,. ctEvD\ @yAns\yƒ\ ⇤ dvmvr MS 11 Env⇥EÅ ] em. ; EnEvEt MS • fAE⌃tEvçA ] em. ; fAE⌃tEvçA MS • tArA ] em. ; trA MS • y⌃tF stArZF ] em. ; yE⌃t -vtArZF MS • E/\f(klop ⇤t ] em. ; E/R^fklAp ⇤t MS • y, smdA„t, ] em. ; y smdA„tA, MS

p—rA/ ⇤

39

a£AdfEvDAnm^

tArA stArA trZF tAry⌃tF stArZF. a£E/\f(klop ⇤t aAcAy, smdA„t,; 11; p—m⌃yAs,; pzq, ˛k⇥EtbEàrh\kAro mnñ ⇤Et p—. fNd, -pfo rso !p\ g⌃Dñ ⇤Et p—. p⇥ET&yAp-t ⇤jo vAyrAkAfñ ⇤Et p—. c", ôo/\ G }AZ\ [(vg }snA c ⇤][5] Et p—. ˛AZo _pAn, smAn udAno &yAnñ ⇤Et p—. p—Ev\fEttÇv‚ aAcAy, smdA„t,; 12; svA"rmy\ q§\ ⇤ dvd ⇤v ⇤ syojy ⇤t^. akAr\ dE"ZAΩ§ ⇤ aAkAr\ vAmto ⌃ys ⇤t^; 13; ikAr\ dE"Z ⇤ gSP ⇤ IkAr\ vAmto ⌃ys ⇤t^. ukAr\ nlk ⇤ c {v UkAr\ vAmt, E-Ttm^; 14; kA[r\ dE"Z ⇤ jAnO ⌘][*4v1] kAr\ vAmjAnk ⇤. ⇠kAr\ Uzk ⇤ c {v ⇡kAr\ vAmt, E-Ttm^; 15; ekAr\ gù ⇤ Ev⌃y-y e ⇤kAr\ ElΩ ev Eh. aokAr\ dE"Z ⇤ k‘A\ aOkAr\ vAm - m^ - ev c; 16; Eb⌃dyƒmkAr\ t nAEB-TAn ⇤ Enyojy ⇤t^. jWr ⇤Et pnñ {v a, y“Ft Evf ⇤qt,; 17; 12 ˛k⇥EtbEàrh\kAro mnñ ⇤Et p— ] em. ; ˛k⇥Et,. bEà ah\kAr p—m, MS • fNd, -pfo rso !p\ ] em. ; fNd pq rs !p MS • p— ] em. ; p—m, MS • p⇥ET&yAp-t ⇤jo ] em. ; p⇥ETvF aAp t ⇤j MS • c", ôo/\ G }AZ\ ] em. ; c" ôo/ G }AnA MS • p— ] em. ; p—m, MS • ˛AZo _pAn, ] em. ; ˛AZo ˛AZ MS • udAno &yAn ] em. ; udAn &yAZ MS • aAcAy, ] em. ; aAcAy MS 13 akAr\ ] em. ; akAr MS • aAkAr\ ] em. ; aAkAr MS 14 r\ dE"Z ⇤ gSP ⇤ ] em. ; r dE"ZAΩ§ ⇤ MS •

IkAr\ ] em. ; ikAr MS • ukAr\ nlk ⇤ c {v ] em. ; ukAr nlko— {v MS • E-Ttm^ ] em. ; E-Tt, MS 15 Uzk ⇤ c {v ] em. ; Uzkñ {v MS • E-Ttm^ ] em. ; E-Tt, MS 16 gù ⇤ Ev⌃y-y ] em. ; gù Ev\⌃y-y MS • Eh ] em. ; Eh, MS • aokAr\ ] em. ; aokAr ⇤ MS • aOkAr\ ] em. ; aokAr MS 17 yƒmkAr\ t ] em. ; yƒ akAr-t MS

40

p—rA/ ⇤

a£AdfEvDAnm^

qoXfñ {v bFjAEn -vr⌃yAsmnÅmm^. aA(mtÇvEmEt HyAt\ EB[ +++++++ ] ; 18; [2]kkArmrEs

Ev⌃y-y KkAr\ dE"Z ⇤ tTA.

gkAr\ -tnvAm ⇤ t GkAr\ k⌫Wto ⌃ys ⇤t^; 19; RkAr\ Ecbk ⇤ dçAå ⇤vd ⇤v ⇤ yTAámm^. ckAr\ p⇥§v\f ⇤ t CkAr\ svp⇥[§k ⇤] ; 20; jkAr\ kEVkAyA\ t JkAr\ kE"dE"Z ⇤. ⇢kAr\ vAmk"AyA\ bFjAEn ámfo ⌃ys ⇤t^; 21; h-t ⇤ t dE"Z ⇤ _Ω§ ⇤ VkAr\ t/ Ec⌃ty ⇤t^. [WkAr\ ++++ ] [3]c XkAr\ tElkA-tTA; 22; YkAr\ bAhm@y ⇤ t ZkAr\ vAm aΩk ⇤. tkAr\ h-tp⇥§ ⇤ c TkAr\ dfn ⇤ tTA; 23; dkAr\ tAlk ⇤ c {v DkAr\ bAhm@yt,. nkAr\ dE"ZA[Ω ⇤] c pkArmo§yo⌃ys ⇤t^; 24; PkAr\ dfn ⇤ k⇥(vA g }FvAyA\ t bkAr ’ v {. BkAr\ tAlk ⇤ c {v mkAr\ GE⌫VkA-tTA; 25; Ev[çA]tÇvEmEt ˛oƒ\ k(t)[&y\ t EvjAntA] . [4]ykAr\

dE"Z ⇤ g⌫X ⇤ rkAr\ vAmg⌫Xk ⇤; 26;

nAEskAyA\ lkAr\ t vkAr\ t llAVk ⇤. fkAr\ dE"Z ⇤ n ⇤/ ⇤ qkAr\ vAmlocn ⇤; 27; skAr\ dE"Z ⇤ kZ ⇤ hkAr\ vAmto ⌃ys ⇤(t^) . 18 qoXfñ {v bFjAEn ] conj. ; qoYkñ {v bFjAEZ MS • mnÅmm^ ] mntmm^ MS • HyAt\ ] conj. ; HyAtA MS • a, ] em. ; a MS 21 JkAr\ ] em. ; JkAr MS • k"AyA\ ] em. ; y"AyA MS 22 h-t ⇤ t ] em. ; h-t ⇤-t MS • _Ω§ ⇤ ] em. ; g§ ⇤

MS • tElkA-tTA ] conj. ; tEl\kA-tTA MS 23 p⇥§ ⇤ ] conj. ; p⇥§O MS • dfn ⇤ tTA ] em. ; dfZ ⇤-tTA MS 24 m@yt, ] em. ; m@yt, MS 25 dfn ⇤ ] em. ; dfZ ⇤ MS • v { ] em. ; v {, MS • BkAr\ ] em. ; BkAr MS 27 fkAr\ ] em. ; qkAr MS • n ⇤/ ⇤ qkAr\ ] em. ; n ⇤/\. skAr MS

41

p—rA/ ⇤

a£AdfEvDAnm^

Efrsñ EfKAyA\ t "kAr\ t/ Ec⌃ty ⇤t^. EfvtÇvEmd\ ˛oƒ\ yTAHyAt\ t fMBnA; 28; sÿm\ -vrbFj ⇤n; : a vAs⇤ dvAy n(m,. ) [: aA s\kqZAy nm,. : a\ ˛çflAy] [5]nm,. : a, aEnzàAy nm,. : : nArAyZAy nm,. : t(sd^ b }úZ ⇤ nm,. : h\ Ev Zv ⇤ nm,. : nrEs\hAy nm,. : BvrAhAy nm,; aApAdm-tk\ yAv⌃⌃ys ⇤ìFjAEn svt,; 29; v ⇤d⌃yAsA£m\ ˛oƒ\ kt&y\ t EvjAntA; ôFsƒ\ EfKAyA\ (⌃y)-y EfrEs m⌫Xlb }AúZm^; 30; sysAm llAV ⇤ t u‡yA(Em)[ c n ⇤/yo,. (*5r1)++m]sAm

nAsAg ⇤ }; 31;

y ⇤n ⇤d\ dE"Z ⇤ kZ ⇤ >yo(ykm c vAmk ⇤. n ⇤/nAsF tdÅr ⇤; 32; pAvmAnFEvrAjAy ⇤§sAm nAB ⇤Et.

tmFfAn\ dE"Z ⇤ bAh\ vAmto vAmbAhk ⇤; 37; k¨to bAhm@y ⇤ t vAmm@y ⇤ kEná⌃tm^. EvdA mGvn^ h-t ⇤ t vAm ⇤ kr ⇤ i⌃d˝nArym^; 38; /AtArEm⌃d˝ aΩSyA\ i⌃d˝AyAEh Ec/BAno vAmAΩSyA\ c. fàv(yo nKAg ⇤ } t [ ++++[3]+ ] KA\ sdA. ÏyMbk\ kEVml ⇤ t fAMBv\ p⇥§v\fyo,. ´ 14.6.6 35 IfAvA-ym^ VS 40.1 ; gAEgb }AúZm^ SB 36 -kMBsƒm^ AV 10.7.1-44 ; pzqsƒm^ RV 10.90 ; pzqv }t ⇤ SV 1.222, 1.609 ; pzqgEt, ( s ⇤tqAm ) SV 1.594 37 vzZsAm SV >y ⇤§sAm SV 1.31 = RV 6.7.1 (according to Haradatta), SV 1.33 = RV 1.115.1 (according to Govinda); cf. Olivelle 2000: 712 tmFfAn\ RV.1.89.5 38 k\kto n k\kt RV 1.191.1 kEná⌃tm^ = kEná\Et hErrA RV 9.95.1 SV 1.530 ¯ G¯ EvdA mGvn^ ( mhAnAflF ) RV Khila. 5.4.1 = SV A. a. pari. 1 i⌃d˝nArym^ = i⌃d˝\ nro n ⇤mEDtA hv⌃t ⇤ RV 7.27.1 = SV 1.318 38 prose. /AtArEm⌃d˝ RV 6.47.11 i⌃d˝AyAEh Ec/BAno RV 1.3.4 39 fàv(y, = eto E⌃v⌃d˝ -tvAm fàm^ RV 8.95.7–9 = SV 1.350 ÏyMbk\ yjAmh ⇤ RV 7.59.12 ; fAMBvm^ = nm, fMBvAy ; VS 16.41 ; apA\ p⇥§mEs VS 11.29, 13.53 ; EvkZ = b }úsAm = EvB }AX^ b⇥ht^ RV 10.170.1; SV ¯ A. G¯ a. 257 35 IfAvA-y\ t ] conj. ; IqAvAd⌃t MS • Ecbk ⇤ gAEgb }AúZm^ ] conj. ; Ecbk ⇤ gApc ac `gEtúZ\ MS ; Ec ( b ) ++++úZ\ MS • kpolyo, ] em. ; kyolyo MS • The first p¯ ada is unmetrical, but in an even p¯ ada it would have alright. The last p¯ ada, too, is unmetrical. 36 kMbjA Eíj unidentified ¯ G¯ ¯ G¯ 43 ⇤ dvv }tm^ SV A. a. 212–14 ; aA>ydohsAm RV 6.7.1 = SV 1.67; A. a. 43–45 ; pllBArm^ ?? 44 aAk⇥ Z ⇤n RV 1.35.2=VS 33.43 a-y vAm-y RV 1.164.1 ; y⌫v¯ G¯ km^ = i⌃d˝EmãAETno b⇥ht^ RV 1.7.1 = SV 1.198; A. a. 122 45 ˛ kA&ymfn ⇤v = vrAhsAm SV 1.524 kd˝d˝Ay ˛c ⇤ts ⇤ RV 1.43.1-9 ; nFlg }FvA VS 16.56-57 tr(s m⌃dF SV 1.500 = RV 9.58.1–4 pAvmAnF ?? 39 apA\ p⇥§ ⇤Et ] conj. ; aD, p⇥E£ MS • EvkZ◆ v⇥qZ ⇤ ] em. ; Evkõ v⇥qn ⇤ MS 40 BAs\ ] em. ; BAq MS • gù ⇤ zd˝\ ] em. ; gù zd˝ MS • vAt aA vAt B ⇤ ] em. ; vAto vA cAt B { MS 41 vAEmEt ] em. ; vAm ⇤Et MS 42 EvB }AX^b⇥h(sAm\ ] em. ; EvB }AmV^. b⇥h-sAm MS 43 v }t\ ] em. ; v }t ⇤ MS • crZO ] conj. ; trZO MS • The third p¯ ada is unmetrical. 44 aAk⇥ ZA ] conj. ; aAk⇥ Z MS • y⌫vkmΩly, ] conj. ; y ⇤⌫ykm◆⌃gly, MS • j⌃ykmAEZ vAmAΩ ] conj. ; j ⇤⌃ykmEZ vAmAg MS • Except the second, all other p¯adas are unmetrical. 45 vrAh ] MSpc ; v written in secunda manus in the bottom margin, the original is damaged by worm but still partly visible) • sAm ] em. ; sAm\ MS • kd˝d˝A ] em. ; zd˝d˝A MS • The first and third p¯ adas are unmetrical. 46 aA(mAΩ ⇤ _ΩA ] conj. ; aªEΩ aΩA MS • r⌃D ⇤ }q ] em. ; r⌃D ⇤ }q MS • rT⌃tr\ ] em. ; rTA⌃tr\ MS • vAEjq ] em. ; vAEjn ⇤q c MS • The last p¯ada is unmetrical.

44

p—rA/ ⇤

a£AdfEvDAnm^

rT⌃tr\ rTm@y ⇤ rT ⇤ (Et)[§E‡[5]Et](vA)Ejq Ev⌃ys ⇤t^; 46; Ehr⌫ygBm-/ ⇤q sdfn ⇤q c Ev⌃ys ⇤t^. f {v ⇤ c v { Zv ⇤ c {v sOr ⇤ BAgvt ⇤ tTA. v ⇤d⌃yAsEmEt ˛oƒ\ ⇤ dvd ⇤v ⇤n fMBnA; 47; v ⇤d⌃yAso _£m,; nvm\ nAmt, ˛oƒ\ ⇤ dvd ⇤vjg(pt ⇤,. k ⇤fv\ t Efr ⇤ ⌃y-y mK ⇤ nArAyZ\ ⌃ys ⇤t^. g }FvAyA\ mADv\ c {v goEv⌃d\ bAh ’ Ev⌃ys ⇤t^; 48; „dy ⇤ Ev⌃ys ⇤Eí Z\ j(W)[r ⇤ mDs[6]dnm^] . kEVyoEtm MS • l"Z\ cAEp ] em. ; l"Z⌫yAEp MS • cyA t ] conj. ; (scyA MS • l"Z\ cAEp ] em. ; l"Z⌫yAEp MS • sEmào a“⌃k⇥drm^ ] em. ; sEmDo ajn\ k⇥tr\ MS 91 vAtA tAyt ⇤ ] conj. ; vAcA cAyA⌃t MS • pry ⇤⌃mD ] em. ; pry ⇤mD MS • The same line appears before 90cd in this reading: mDvAcA (scAyA pry ⇤(mDsE=psA; , but considering it superfluous, I cancel the line there. 91 kfAg ⇤ }Z t ] conj. ; ksok ⇤Z aA MS 92 Z pt ⇤ !p\ c {v ] em. ; Z pty ⇤ p— {v MS • aA no ] em. ; a no MS 93 y>jAg }to drm{ pEt f⇧yA\ ] em. ; yyAyto dm⇤ pEt sùAs^ MS • dApy ⇤t^ ] MSpc ; ˛dApy ⇤t^ MSac • dçAÅ-y ] em. ; dçAt-y MS • The first p¯ada is unmetrical. 94 ZAñAzdfnA, ] em. ; ZAñAzdfnAt^ MS • The first p¯ ada is unmetrical. 95 pv ⇤ plAf\ ] em. ; pv ⇤Z plAfF MS • The first p¯ada is unmetrical. 95 aò(T\ ] em. ; aò(T MS • EvçAn^ ] em. ; Ev⌃çA MS 96 aA no ] em. ; a no MS • ˛AsAd-TAn\ ] em. ; ˛AsAd\-TAn MS • D⌃vnA ] em. ; D(vAZA MS • gB◆ ] em. ; gt\ MS

52

p—rA/ ⇤

a£AdfEvDAnm^

D⌃vnA gA iEt m⌃/ ⇤Z gt◆ c {v t kAry ⇤t^. 96. -TAEpt-y t ⇤ dv-y tto Eh fAE⌃tmArB ⇤t^. eqo h ⇤ dv (i)[Et m⌃/ ⇤Z aE`nk⌫X\ t] [2]kAry ⇤t^; 97; pv ⇤Z ctr˙\ -yAdDc⌃d˝\ t dE"Z ⇤. vAz⌫yA\ vtl\ k⌫Xm£koZ\ tToÅr ⇤; 98; aA`n ⇤⇧yA\ EdfABAg ⇤ aAcAy aE`nkAy◆ ˛ArB ⇤t^. td ⇤vAE`n-tdAEd(y-tEdEt m⌃/ ⇤Z aE`n\ c {vAlB ⇤t^. 99. p—dfmo _E`ns\-kAr,; a-/ ⇤Z l ⇤Kn\ kyA(kvc ⇤nAjAtv ⇤dsm^. gBADAnAEdk\ k⇥(vA hot&y\ cAnpvf,; 101; yAg, qoXfmo Eíj; ⌦; 96 RV 1.89.1: aA no Bd˝A

˛Edfo _n

• D⌃vnA gA, RV.6.75.2

97 VS 32.4: eqo h ⇤ dv,

99 VS 32.1: td ⇤vAE`n-tdAEd(y-tíAy-td c⌃d˝mA,

100–101 S¯ ardhatri´sat¯ık¯ alottara 6.1cd–3ab:

a-/ ⇤Zo• ⇤Kn\ kyAímZAjAtv ⇤dsm^; gBADAnAEdk\ k⇥(vA En k⇥Et\ cA=ypEñmAm^. 97 ⇤ dv-y ] conj. ; ⇤ dv ⇤q\ MS • tto Eh fAE⌃tmA ] conj. ; •

The third p¯ ada is unmetrical.

tto fAE⌃t˛A MS

98 ctr˙\ -yAd^ ] conj. ; ctr˙\ MS • vtl\

k⌫Xm£koZ\ ] conj. ; ctl\. a£koZ ⇤ MS 99 aA`n ⇤⇧yA\ EdfA ] em. ; aA`n ⇤yAE⌃dfA MS • kAy◆ ] em. ; kAy MS 99 td ⇤vAE`n-tdAEd(y-tEdEt ] em. ; td {vAE`ntdAEd(y ⇤EdEt MS • aE`n\ c {vA ] conj. ; aE`n c {v cA MS 100 mo _E`ns\-kAr, ] conj. ; mmE`ns\-kAr\ MS 100 -m⇥tm^ ] -m⇥t, MS 100 ⌃yAs\ tto ] em. ; ÍyAs⌃tto MS • The second p¯ada is unmetrical. 101 p ⇤>jAtv ⇤dsm^ ] em. ; p ⇤jAtv ⇤ds, MS • k\ k⇥(vA ] em. ; kn^ k⇥(vA MS 102 yAg, qoXfmo ] em. ; yAg qoXfm\ MS • Eíj ] em. ; Eíj, MS • yjn\ ] em. ; yjn^ MS • { dv\ nArAyZ\ ] conj. ; ⇤ dv\ nArAy MS • ctrA ] em. ; ctEBrA MS • kAy◆ dlB\ ] em. ; kAyA◆ v [ l ] B\ MS

p—rA/ ⇤

53

a£AdfEvDAnm^

yjn\ s\˛v#yAEm { dv\ nArAyZ\ -m⇥tm^. ctrAvrZ {, kAy◆ dlB\ ssrAs{ r,; 102; p—Ev\fEtm\ pzq\ mkAr\ Eb⌃d!EpZm^. „dy\ t-y smAHyAt\ kEZkAyA\ t Ev⌃ys ⇤t^; 103; Efr\ pvdl ⇤ dçAd^ [dE"Z ⇤ t EfKA\ ⌃ys ⇤t^] . ⌧pEñm ⇤ kvc\ dçAd-/\ v { coÅr\ ⌃ys ⇤t^; 104; [4]gAyÏyA`n ⇤yp/ ⇤

t sAEv/FfAngocr ⇤. n ⇤/\ c {v t n { (yA\ EpΩlA-/\ t vAyv ⇤. gùAãùtr\ gù\ gBAvrZmÅmm^; 105; EítFy\ s\˛v#yAEm sv⇤ d[v]nm-k⇥tm^. blBd˝\ ⌃ys ⇤(pv aA`n ⇤⇧yA\ EnfW\ ⌃ys ⇤t^; 106; 102 Sv¯ ayambhuvapa˜ ncar¯ atra 7.17: +++ s\˛v#yAEm { dv\ nArAyZ-y t. E/EBrAvrZ {, kAy◆ dlB\ ssrAs{ r,; 102–105 Dev¯ amr.tapa˜ ncar¯ atra 11.25–30ab:

yjn\ s\˛v#yAEm Ed&y\ nArAyZ-y t. E/EBrAvrZ {, kAy◆ dlB\ t srAs{ r,; m@y ⇤ cá\ ˛Et§A=y ar {íAdfEBytm^. t⌃m@y ⇤ kml\ ˛oƒm£p/\ skEZkm^; bAùtñtr˙\ t ctíAr\ sfoBnm^. m@y ⇤ t „dy\ p>y h\kAr ⇤Z mhA(mnA; Efr, pvdl ⇤ dçAåE"Z ⇤ t EfKA\ ⌃ys ⇤t^. pEñm ⇤ kvc\ dçAd-/\ v { coÅr\ ⌃ys ⇤t^; aA`n ⇤⇧yA\ EdEf gAy/F\ sAEv/FmFfgocr ⇤. n ⇤/ ⇤c {v t n { (yA\ vAy&yA\ EdEf EpΩlm^; etãùtr\ HyAt\ gBAvrZmÅmm^. 106–108 Dev¯ amr.tapa˜ ncar¯ atra 11.30cd–38ab:

EítFy\ s\˛v#yAEm sv⇤ dvmy\ -ttm^; íAdfAr ⇤ tTA cá ⇤ arAg ⇤ }q Evc"Z,. íAdfA"rs\yƒA ⌃ys ⇤d^ íAdfmty,; k ⇤fv\ t ar ⇤ pv ⇤ :kAr ⇤Z t pjy ⇤t^. EítFy\ t nkAr ⇤Z ⇤ dv\ nArAyZ\ yj ⇤t^; t⇥tFy ⇤ mADv\ p>y mokAr ⇤Z mhA(mnA. 103 Ev\fEtm\ pzq\ ] em. ; EvR^fEtmMpzq {, MS • HyAt\ kEZkA ] em. ; HyAtA(kEõkA MS • The first and third p¯adas are unmetrical. 104 dl ⇤ dçA ] em. ; dl⌃dçA MS • cd ] omitted in the Sv¯ ayambhuvapa˜ ncar¯ atra, supplied from the Dev¯ amr.tapa˜ ncar¯ atra. 105 ÏyA`n ⇤y ] em. ; ÏyAE`n MS • gocr ⇤ ] em. ; gocr ⇤t^ MS • n ⇤/\ c {v t n { (yA\ ] em. ; n ⇤/A— {v t n {Er(yA\ MS • gùAd^ ] em. ; gùA MS • mÅmm^ ] em. ; mnÅm\ MS 106 s\˛v#yAEm sv⇤ dv ] em. ; sM˛vHyAEm sv{ dv MS • Bd˝\ ] em. ; Bd˝ MS • aA`n ⇤⇧yA\ EnfW\ ] em. ; aA`n ⇤yA EnqX\ MS

p—rA/ ⇤

54

a£AdfEvDAnm^

dE"Z ⇤ c {v ˛çflmEnzà\ t n { t ⇤. gd\ pEñmto dçA‡A‘⌃t\ (kOf)[lo(kVm^; 107; sAMb\ dçAŒ vA][5]y&yA\ sA((ykF\) coÅr ⇤Z t. aár\ t mhAvFy◆ dçAdFfAngocr ⇤; 108; EítFyAvrZm^; ⌦; t⇥tFy ⇤ _-/AEZ Ev⌃ys ⇤Eí ZorcAcn-y c. K≥\ c {v ⌃ys ⇤(pv aA`n ⇤⇧yA\ t gdA\ ⌃ys ⇤t^; 109; dE"Z ⇤n ⌃ys ⇤Œá\ f≤\ n { (ygocr ⇤. pî\ pEñmto ⌃y-y vAy&yA\ t hl\ ⌃ys ⇤t^. msl\ coÅrto dçAd^ [IfA⌃yA\ fAΩm ⇤v c] ; 110; [6]t⇥tFyAvrZm^; ⌦; ctT◆ s\˛v#yAEm nAmAnA\ s\˛kFtnm^. pv ⇤ nArAyZo ⇤ dv, pO⌫XrFkA"-t dE"Z ⇤; 111; BkArA"rd ⇤v ⇤n goEv⌃d\ t ctTkm^; p—m ⇤ c gkAr ⇤Z Ev Z\ c {v t pjy ⇤t^. vkArA"rs\yƒ\ q§ ⇤ t mDsdnm^; E/Evám\ tt, pñAÅ ⇤kAr ⇤Z t pjy ⇤t^. vAmn\ cA£m ⇤ c {v vAkAr ⇤Z t pjy ⇤t^; ôFDr\ nvm ⇤ p>y skAr ⇤Z mhA(mnA. dfm ⇤ t „qFk ⇤f\ ⇤ dkArA"rpEjtm^; ekAdf ⇤ pînAB\ vAkAr ⇤Z Enyojy ⇤t^. íAdf ⇤ t ykAr ⇤Z yƒ\ dAmodr\ ⌃ys ⇤t^; pî ⇤ t yjn\ k⇥(vA pñAŒá\ ˛pjy ⇤t^. 109–110 Dev¯ amr.tapa˜ ncar¯ atra 11.38cd–40: EítFyAvrZ\ HyAt\ t⇥tFy ⇤ _-/AEZ Ev⌃ys ⇤t^; K≥\ c {v ⌃ys ⇤(pv ⇤ aA`n ⇤⇧yA\ t gdA\ ⌃ys ⇤t^. dE"Z ⇤ Ev⌃ys ⇤Œá\ f\K\ En Etgocr ⇤; pî\ pEñmto ⌃y-y vAy&yA\ t hl\ ⌃ys ⇤t^. msl\ coÅr ⇤ dçAdFfA⌃yA\ f⇥EΩ Ev⌃ys ⇤t^; 107 ˛çflmEn ] em. ; ˛çfl ⇤ aEn MS • n { t ⇤ ] em. ; n {Ert ⇤ MS • gd\ ] conj. ; gzX MS • dçA‡A‘⌃t\ ] em. ; dçAt^ nA‘t\ ? MS 108 sA(ykF\ co ] em. ; sA [ ⌫X ] kF vo MS • aár\ t ] em. ; aár-t MS • gocr ⇤ ] em. ; gocr ⇤t^ MS 108 vrZm^ ] em. ; vrZ MS 109 rcAcn-y ] conj. ; rŒn-y MS • K≥\ ] em. ; g≥\ MS • ⌃ys ⇤(pv ] em. ; ⌃ys ⇤(pv◆ MS • aA`n ⇤⇧yA\ t gd\ ] em. ; aA`n ⇤yA\ t gj\ MS 110 ⌃ys ⇤Œá\ f≤\ ] em. ; ⌃ys ⇤cá\ s\K MS • n { (ygocr ⇤] pc em. ; n {Er(ygocr ⇤t^ MS 110 pî\ ] em. ; pî MS • pEñmto ] MS ; pEñto MSac • vAy&yA\ t ] em. ; vAy&y⌃t MS 111 T◆ s\˛v#yAEm ] em. ; T s\˛vHyAEm MS • nAmAnA\ s\ ] conj. ; nAmnAmA MS 111 nArAyZo ⇤ dv, ] conj. ; ZArAyn\ ⇤ dv\ MS • "-t dE"Z ⇤ ] conj. ; "⌃t dE"Z ⇤, MS • The last p¯ada is unmetrical.

55

p—rA/ ⇤

a£AdfEvDAnm^

pEñm ⇤n t v {k⌫W uÅr ⇤ mDsdn,. IfA⌃yA\ r"t ⇤ Ev ZrA`n ⇤⇧yA\ t jnAdn,; 112; [n { (yA\ t „qFk ⇤fo [*7r1] vA]y&yA\ c {v mADv,. aD-tAEÅ~Evámo [d ⇤v U]@v ⇤ govDno r"t ⇤. EdfAs EvEdfAs c; 113; sÿdfm\ r◊rcn\ ˛v#yAMynpvf,. i⌃d˝nFl\ mhAnFl\ mrgt\ pîrAgm^; 114; sykAE⌃t c⌃d˝kAE⌃t aE`nkAE⌃t tT {v c. jlA˙v\ cA£m\ b }ún^ DAtnA\ kTyAMyhm^; 115; hErtAl\ mn,EflA\ DAt (mAE")k[h ⇤mk ⇤. ] ++[2]†EvE⌃d†smAyƒ\ †nAgvF\g\†

pArd\ c Evf ⇤qt,; 116;

tto dçAìFjAnA\ kTyAEm t ⇤.

DA⌃yAçA fZpy⌃tA v }FEh⌃y-tADBAgt,; 117; alAB ⇤ svr◊AnA\ kA—n\ c ˛dApy ⇤t^. alAB ⇤ svDAtnA\ hErtAl\ t dApy ⇤t^. alAB ⇤ svbFjAnA\ yvm ⇤k\ t dApy ⇤t^; 118; 118

Sarvaj˜ na ¯nottara 19.110cd–111: alABA(svDAtnA\ hErtAl\ t dApy ⇤t^; alA-

BA(svbFjAnA\ yvbFj\ ˛dApy ⇤t^. alABA(svr◊AnA\ vÖm ⇤k\ t dApy ⇤t^; 112 v {k⌫W ] conj. ; v {k⌫W\ MS • sdn, ] conj. ; sdn\ MS 112 ⌃yA\ r"t ⇤ Ev ZrA`n ⇤⇧yA\ t ] conj. ; ⌃yA r"t ⇤ Ev Z a`n ⇤yA⌃t MS • jnAdn, ] conj. ; jnAàZ\ MS 113 y&yA\ c {v mADv, ] em. ; y&y— {v mADv\ MS 113 -tAEÅ~Evámo ⇤ dv ] em. ; -/AEtEvám [ ⌃d ⇤v\ ] MS • govDno ] em. ; govàZ\ MS • The third and fourth p¯ adas are unmetrical. 114 Mynpvf, ] em. ; mnpvf, MS • nFl\ mhAnFl\ ] em. ; nFl mhAnFl MS • pîrAgm^ ] conj. ; pîrAg tT {v c MS • The first and last p¯ adas are unmetrical. 115 kAE⌃t aE`nkAE⌃t ] conj. ; kAE⌃tñ aE`nkAE⌃t MS • b }ún^ DAtnA\ ] em. ; b }úA DAtnA MS • The third p¯ ada is unmetrical. 116 tAl\ mn,EflA\ DAt\ ] em. ; tAlA mnEQClAñ {v DAt MS • pArd\ ] em. ; pArd\ MS • The first p¯ ada is unmetrical. 117 nA\ kTyAEm t ⇤ ] em. ; nA kTyAEm t ⇤, MS • v }FEh⌃y-tA ] em. ; v }FEh⌃y-tA MS 118 t dApy ⇤t^ ] em. ; t ˛dApy ⇤t^ MS • t dApy ⇤t^ ] em. ; t ˛dApy ⇤t^ MS

56

p—rA/ ⇤

a£AdfEvDAnm^

-TApn\ cA£Adfmm^; -TAEpt-y t ⇤ dv[-yAEBq ⇤k\ smA][3]rB ⇤t^; eqo h ⇤ dv iEt m⌃/ ⇤Z aEBq ⇤k\ t dApy ⇤t^; 119; iEt p—rA/ ⇤ _£AdfEvDAn\ smAÿm^; •;

119 VS 32.4: eqo h ⇤ dv ˛Edf 119 cA£Adfmm^ ] em. ; cA£EvD\ MS • t dApy ⇤t^ ] em. ; t ˛dApy ⇤t^ MS • ⇤ dv ] conj. ; ⇤ dv ⇤f MS (unmetrical) is unmetrical.

•

smAÿm^ ] em. ; smAÿ, MS • The third p¯ada

; ⇤ dvAm⇥tp—rA/m^; ; ˛TmpVl, ; [1v ]m ⇤zp⇥§ ⇤

sKAsFn\ b }úA ’ lokEptAmhm^. sn(kmAr, ôFm⌃t\ p⇥QCt ⇤ prm ⇤òrm^; 1; sn(kmAr uvAc. ko _y\ Ev Z, pro ⇤ dv, svfA-/ ⇤q p’t ⇤. Ek\ n mEtrmEtvA ajAto jAt ev vA; 2; sgZ\ t kT\ ˛oƒ\ EngZ\ vA jgãro. Ek\ n s#m\ EvjAnFyAE(k\ vA -TlEmhoQyt ⇤; 3; Ek\ Ev Z, skl, ˛oƒ, Ek\ n En kl - m^ - ev vA. Ekm&yƒ\ EvjAnFyAïƒ!p\ t Ek\ Bv ⇤t^; 4; Ek\ n En(ymEn(y\ vA"ro _n"r ev vA. 2ab Sv¯ ayambhuvapa˜ ncar¯ atra 2.6ab:

s v { Ev Z, pro ⇤ dv, svfA-/ ⇤q p’t ⇤. 2cd–6 Sv¯ ayambhuvapa˜ ncar¯ atra 1.8–11:

Ek\ mEtñ r^ - amEtñ ajAto jAt m^ - ev c. Ek\ n s#m\ EvjAnFyAE(k\ vA -TlEmhoQyt ⇤; Ek\ Ev Z, skl, ˛oƒ, Ek\ vA En kl m^ - ev vA. Ek\ vA&yƒ\ EvjAnFyAïƒ!pF c Ek\ Bv ⇤t^; Ek\ En(y\ c aEn(y\ vA a"ro _n"r-tTA. svg, Ek\ n Ev‚ ⇤y, Ek\ vAsvgt, -m⇥t,; Ek\ (n v) +++ ++ + + ++EvvEjt,. kT\ c -TA=yt ⇤ lok ⇤ loknATo jnAdn,; 1 The manuscript begins with : nmo Bgvt ⇤ vAs⇤ dvAy; • sn(kmAr, ] MSpc ; s(kmAr, MSac 2 Ev Z, ] em. ; Ev Z MS • Ek\ n ] em. ; Ek⌃t MS • mEtvA ] em. ; mEtTA MS 3 jgãro ] conj. ; jãz\ MS • Ek\ n ] em. ; Ek⌃t MS •

s#m\ ] em. ; s#m MS 4 Ek\ n ] em. ; Ek⌃t MS • Ekm&yƒ\ ] em. ; Ek\m&yƒ\ MS 5 Ek\ n ] em. ; Ek⌃t MS • En(ym ] MSpc ; En(y\m MSac • _n"r ] conj. ; _⌃y\"r MS

⇤ dvAm⇥tp—rA/ ⇤

60

˛TmpVl,

svg, Ek\ n Ev‚ ⇤y, Ek\ vAsvgt, -m⇥t,; 5; Ek\ n vZA(mko Ev Z, Ek\ vA vZEvvEjt,. kT\ vA -TA=yt ⇤ Ev ZloknAT jgãro; 6; ˛Et§A t-y ⇤ dv-y ôotEmQCAEm tÇvt,. r [2 ]kE-m⌃d˝&y ⇤ t kt&yA t-y v { ˛EtmA fBA; 7; Ek\ ˛mAZ\ t Ev‚ ⇤ymΩ˛(yΩyo-tTA. frFr\ Ek\˛mAZ\ t kt&y\ l"ZAE⌃vtm^; 8; mkV-y ˛mAZ\ t Efrsñ kT\ Bv ⇤t^. llAV\ t pn-t-y kt&y\ Ek\˛mAZt,. 9; nAEskA t kT\ kAyA n ⇤/ ⇤ c {v EptAmh. g⌫Xyo, Ek\ ˛mAZ\ t kZyoñ kT\ Bv ⇤t^; 10; ao§O Ek\ t-y kt&yO Ecbk ⇤ Ek\ n l"Zm^. bAhoñ {v tdΩSyA g }FvAyA, Ek\˛mAZt,; 11; 7–17ab Sv¯ ayambhuvapa˜ ncar¯ atra 1.12–22:

˛Et§A t-y ⇤ dv-y ôotEmQCAEm tÇvt,. k ⇤n d˝&y ⇤Z kt&yA t-y v { ˛EtmA fBA; Ek\ ˛mAZ\ t Ev‚ ⇤ymΩ˛(yΩyo-tTA. frFr\ Ek\ ˛mAZ\ t kt&y\ v { Zv\ jn {,; mkV-y ˛mAZ\ t Efrñ {v kT\ Bv ⇤t^. llAV\ t pn-t-y kt&y\ Ek\++++; +++ + kT\ kAy◆ kT\ n ⇤/mTAEdf ⇤t^. g⌫Xyo, Ek\ ˛mAZ\ t kZ◆ c {v kT\ Bv ⇤t^; kTmo§O

t kt&y\ Ecbk ⇤ Ek\ n l"Zm^. bAhjg\gz\ MS 7 t-y ] em. ; k-yA MS • fBA ] MS ; fBA, MS 10 EptAmh ] em. ; EptAmhA MS 11 ao§O ] em. ; ao§o MS • kt&yO Ecbk ⇤ Ek\ n ] em. ; kt&yo Ecpk ⇤ Ek⌃t MS 11 tdΩSyA ] em. ; tdgSyA MS • ˛mAZt, ] em. ; ˛mAnt, MS pc

ac

⇤ dvAm⇥tp—rA/ ⇤

61

˛TmpVl,

v"ñ {v kT\ kAy◆ -tnkO nAEBm⌫Xlm^. jWr-y ˛mAZ\ t k‘A v { ElΩm ⇤v c; 12; UzkO jAnnF c {v nlkO gSPpAdyo,. aΩSyA\ nK - m^ - aΩ§ ⇤ svl"Zm ⇤v c; 13; vAhn\ n pn-t-y kt&y\ t jnAdn ⇤. aAyDAEn c ⇤ dv-y kF[2v ]d⇥fAEn t kAry ⇤t^; 14; svl"Zs\p‡\ kT\ Ev Z\ EvEnEdf ⇤t^. ˛AsAd\ t kT\ kAy◆ foBn\ yAgm⌫Xpm^; 15; EpE⌫XkA t kT\ kAyA gBs/\ kT\ Bv ⇤t^. g }hn"/l`n\ c EtETvAr\ tT {v c; 16; m⌫Xp-y EvBAg\ t v ⇤çAñ {v t l"Zm^. -TApk-y t Ec°AEn y ⇤ c mEtDrA, -m⇥tA,. et(sv◆ smAs ⇤n Bgv⌃vƒmhEs; ⌦; 17; iEt ⇤ dvAm⇥t ⇤ p—rA/ ⇤ ˛Tm, pVl,; ⌦;

17c-f Sv¯ ayambhuvapa˜ ncar¯ atra 1.25: ++++++++ + c mEtDrA, -m⇥tA,.

etÅt, m ⇤ _f ⇤q\ t ˛Et§AyAgmÅmm^. yTAtLymsE⌃d`D\ b }Eh tÇv\ s⇤ ròr; 13 aΩSyA\ ] em. ; aΩSyA MS • Ω§ ⇤ ] conj. ; Ω§o MS

15 s\p‡\ kT\ Ev Z\

Ev ] conj. ; s\p\n, kT\ Ev ZEv MS • ˛AsAd\ ] conj. ; ˛AsAd ⇤ MS • kAy◆ ] em. ; kAy, MS 17 v ⇤çAñ {v ] em. ; v ⇤dAñ {v MS • DrA, ] em. ; prA, MS 17 et(sv◆ smAs ⇤n ] em. ; et(smAs ⇤n MS • v⌃vƒ ] em. ; v(vƒ MS

; EítFypVl, ;

b }úovAc. f⇥Z Ev˛ mhA˛A‚ svfA-/EvfArd. p—rA/\ mhA‚An\ ydƒ\ prm ⇤E§nA; 1; prA íAdfsAh˙F s\EhtA c myoà⇥tA. tdh\ s\˛v#yAEm f⇥Z v ⇤kmnA mn ⇤; 2; ekA mEt-/yo ⇤ dvA b }úEv Zmh ⇤òrA,. mEt⇤ rkA -m⇥tA v(s nAnABAvo n Evçt ⇤; 3; yo zd˝, s [3r ]-vy\ b }úA yo b }úA s jnAdn,. t ⇤n {v &yAEpt\ sv◆ jgd ⇤tŒrAcrm^; 4; aADAr, svBtAnA\ Eng }hAng }h ⇤ E-Tt,. y-y svAEZ BtAEn ˛BvE⌃t ygAgm ⇤. 1

Sv¯ ayambhuvapa˜ ncar¯ atra 2.1cd–2ab: f⇥Z b }ún^ ˛y◊ ⇤n Ev Zo, -TApnmÅmm^.

p—rA/\ mhA‚An\ svfA-/ ⇤q coÅmm^. 3ab Matsyapur¯ an.a 3.16:

gZ ⇤jnAdn ⇤. t⇥ZgSmltAv⇥" ⇤ pvtAg ⇤ } tT {v c; 15; yTA svgto vAyrEts#mo n d⇥[yt ⇤. tTA svgto Ev Zlok ⇤ lok {n d⇥[yt ⇤; 16; aEn(y\ -TlEm(yAh, s lok ⇤ mEtEvg }h,. anAEdEn kl, ôFmAn^ s#m ev prApr,; 17; aADArñ tTAD ⇤yo EíDAv-To jnAdn,. svBtEhtATAy EíDA sklEn kl,; 18; [4r ]Ev

Z\ l"Zs\p‡\ sv⇤ dv {-t p>yt ⇤.

⇤ dv\ nArAyZ\ k⇥ Z\ zd˝mFfAnm&yym^. s ev BgvAn^ Ev Z, s ev BgvAE Cv,; 19; y ⇤ c Ev ZprA lokA mhAd ⇤v\ EíqE⌃t c. t ⇤ yAE⌃t nrk\ Gormn⌃t\ dAzZ\ mht^; 20; Ev ZEn⌃dAprA y ⇤ t zd˝ArADngEvtA,. ptE⌃t nrk ⇤ Gor ⇤ EnrAlMb ⇤ sdAzZ ⇤; 21; zd˝k ⇤fvyoEn(ym ⇤k\ p[yE⌃t y ⇤ nrA,. t ⇤qA\ yog\ c mo"\ c En(ym ⇤v vdAMyhm^; 22; t-mAâ≈A mhAd ⇤v\ pjy ⇤(stt\ hErm^. 13 jnAdn, ] em. ; jnAàn MS 15 t>jnAdn ⇤ ] em. ; t>jnAàn ⇤ MSpc ; t>jnA\àn ⇤ MSac • t⇥Z ] MSpc ; t⇥Z ⇤ MSac 16 s#mo ] em. ; s#m ⇤ MS • pc Ev Zlok ⇤ lok {n ] em. ; Ev Z•ok lok ⇤‡ MS 17 Evg }h, ] MS ; Evg }h MSac ac • anAEdEn ] em. ; anAEdEn MS 18 jnAdn, ] em. ; jnADn, MS 19 BgvAn^ ] em. ; Bgv\ MS 22 y ⇤ ] em. ; yo MS

⇤ dvAm⇥tp—rA/ ⇤

65

EítFypVl,

t ⇤qA\ yog\ c mo"\ c ˛FtA(mA s ˛yQCEt; 23; vAy - r^ - aMbryoEn(y\ yTA c {k\ Enr⌃trm^; uByor⌃tr\ nAE-t k ⇤fv-y Efv-y c; 24; yTA c EtEmrA⌃DAnA\ c⌃d˝O íO dfn\ Bv ⇤t^. a‚AnEtEmrA⌃DAnA\ zd˝k ⇤fvdfnm^; 25; t-y ⇤ dvAEtd ⇤v-y Ev[4v ] ZorâtkmZ,. -TApn\ s\˛v#yAEm yTAgFt\ -vy\BvA; 26; iEt ⇤ dvAm⇥t ⇤ p—rA/ ⇤ EítFy, pVl,;

24 yoEn(y\ ] em. ; yoE‡(y MS • Enr⌃trm^ ] em. ; EnrÅr\ MS • r⌃tr\ ] em. ;

rÅr\ MS 25 c⌃d˝O íO dfn\ ] conj. ; c⌃d˝ íO dfno MS • k ⇤fvdfnm^ ] em. ; k ⇤fdfn\ MS 26 -vy\BvA ] em. ; -vy\Bk { MS

; t⇥tFypVl, ; b }úovAc. at, pr\ ˛v#yAEm ˛EtmAl"Z\ fBm^. yE-m⌃d˝&y ⇤ t kt&y\ ⇤ dv\ nArAyZ\ fBm^; 1; i⌃d˝nFl\ mhAnFl\ pîrAgmy\ tTA. vÖ\ mrkt\ v {Xy◆ mOEƒk\ c ˛vAlkm^; 2; tAm }k\ vAEp kA\-y\ c /pfFfkr {(ykm^. nAgv⇥"my\ rO=y\ kA—n\ lohj\ tTA; 3; f {lj\ pAETv\ vAEp rAjp”my\ tTA. Ev\fEtñ {v d˝&yAEZ EnEd£A ’ ˛EtmATt,; 4; t ⇤qA\ t l"Z\ Ed&y\ ˛mAZ\ c vdAMyhm^. aΩlAEn c Ev˛ ⇤⌃d˝ ekEíE/ct-tTA; 5; p— qV^ sÿm\ c {v a£m\ nvm\ tTA. df {kAdfEm(yAhíAdf ⇤ h-tm ⇤v c; 6; et(k⌃ysEm(y[5r ]ƒ\ m@ym\ c EnboDt. aΩlAEn /yE-/\f(p—Ev\fAΩlAEn c; 7; Eíh-tA t smEå£A m@ymA ˛EtmA fBA. 1b }úo ] em. ; b }ú u MS • nArAyZ\ ] em. ; nArAyn\ MS 2 i⌃d˝nFl\ ] em. ; i⌃d˝nFl MS • vÖ\ mrkt\ v {Xy◆ ] em. ; vÖ mrkt v {Xy MS. This p¯ada is unmetrical. 3 r {(ykm^ ] em. ; En(yk\ MS • my\ ] conj. ; t ⇤y\ MS • rO=y\ ] em. ; ro=y\ MS 4 pAETv\ ] MSpc ; pAETk\ MSac • my\ ] em. ; sy\ MS 4 d˝&yAEZ ] em. ; d˝&yAEn MS 5 aΩlAEn ] em. ; aΩSyAEn MS 6 a£m\ ] em. ; a£\ MS • h-

íAdf ⇤ ] em. ; híAdf ⇤ MS 7 et(k⌃ys ] MSpc ; et(k⌃yAs MSac • m@ym\ c EnboDt ] em. ; m@ymA c EnboDt, MS 8 smEå£A ] em. ; smEå£A MS

⇤ dvAm⇥tp—rA/ ⇤

67

t⇥tFypVl,

E/h-tA c {v kt&yA svl"Zs\ytA; 8; i(y ⇤tA, ˛EtmA, HyAtA >y ⇤§m@ymk⌃ysA,. ekAΩlA(smAryA at U@v◆ n pjy ⇤t^. ekAΩlA Bv ⇤QC ⇤ ~ §A ïΩlA DnnAfnF; 10; ÏyΩlA t mtA EsEà ’ rogdA ctrΩlA. p—AΩlA t DndA uí ⇤gA t qXΩlA; 11; sÿAΩlA p/kAmA m⇥(ydA£AΩlA bD. nvAΩlA c gov⇥Eà, svnAfA dfAΩlA; 12; ekAdfA t sOBA`yA svhA íAdfAΩlA. dArvA\ t ˛v#yAEm yA c p⌫yA mhFtl ⇤; 13; c⌃dn\ tgr\ EbSv\ tmAl\ rƒlo[5v ]cnm^. t ⇤⌃dk\ ⇤ dvdAz\ c mhAErEqkm ⇤v c; 14; i(y£O ˛EtmA, ˛oƒA dAzjAnA\ vdAMyhm^. EvtE-t\ pvmAry ⇤§A ˛kFEttA; 16; r◊jA lohjA c {v f {ljA pAETvA tTA. t-yAΩAEn ˛v#yAEm ˛(yΩAEn yTAámm^; 17; 9 i(y ⇤tA, ˛EtmA, ] em. ; it ⇤tA, ˛EtmA MS • k⌃ysA, ] em. ; k\⌃ysA MS • fAΩlA ] em. ; fAΩlA, MS 11 ÏyΩlA t ] conj. ; (yΩlo Å MS • p—AΩlA

t DndA ] em. ; p\—AΩlA t DnnA MS • uí ⇤gA t ] conj. ; uí ⇤g⌃t MS lA ] em. ; sÿAΩSy MS • bD ] conj. ; bD, MS

12 sÿAΩ-

13 sOBA`yA ] em. ; sOBA`y MS

• dArvA\ ] em. ; dArvA MS • p⌫yA ] em. ; p⌫yA MS 14 locnm^ ] em. ; locn MS • tmAl\ ] conj. ; tTA. v ⇤ MS • dAz\ c ] conj. ; dA!— MS 15 ˛EtmA, ] em. ; ˛EtmA MS • mAr>vlm^. &yAly‚opvFt\ t kEVs/smAytm^; 10; gdA dE"Zh-t ⇤ t dAnvAnA\ "y\krF. vAm ⇤ c {v t dAt&y\ cá\ / {lo»BqZm^; 11; dE"Z ⇤ c kr ⇤ pî\ f\K\ vAmkr ⇤ E-Ttm^. ms[8r ]l\ K≥mAdAy hl\ DnEq s\ytm^; 12; eBF !p {, smAyƒA kt&yA ˛EtmA fBA. y imA\ -TApy ⇤•ok ⇤ s yƒo nA/ s\fy,; 13; iEt ⇤ dvAm⇥t ⇤ p—rA/ ⇤ ctT, pVl,;

8 kVkAEn ] em. ; kVAkAEn MS

9 dAt&y\ ] conj. ; dAt&y MS 10 sm>>vlm^ ] em. ; st ⇤>vlm^ MS 12 K≥mAdAy ] em. ; f⇥ΩmAdAy MS • DnEq ] em. ; dnEq MS 13 eBF ] em. ; eEB MS • yƒo ] em. ; yƒ ⇤ MS

; p—mpVl, ; b }úovAc. at, pr\ ˛v#yAEm sv⇤ dvmyo hEr,. c⌃d˝AEd(yO -m⇥tO t-y c"íysmAEôtO; 1; b }úA⌫X\ c Efr-t-y k ⇤f ⇤yoEt ] em. ; >y ⇤Et MS 16 mD, ] em. ; mD MS • nArAyZo ⇤ dv, ] em. ; nArAyZ ⇤ dv MS 17 sr-vtF. ] em. ; sr-vtF, MS • qoÅm, ] em. ; qoÅm\ MS

83

⇤ dvAm⇥tp—rA/ ⇤

sÿmpVl,

aA>y ⇤ c {vAm⇥t\ ˛oƒ\ b }úA c pzqoÅm,; 17; ˙cAg { }kAdfA zd˝A, ˙cAg }Ad^ íAdfAΩlAt^. -vAhAkAr ⇤ E-TtA ⇤ dvA vq÷Ar ⇤ ˛jApEt,; 18; yAg ⇤ y‚vrAh\ t m⌃/ ⇤q hnt ⇤ EíjA,. Ev Z\ kAlAE`nzd˝A[14r ]Hy\ yj ⇤⌃mo"Pl˛dm^; 19; etå ⇤vgZ\ ˛oƒ\ -v ⇤ -v ⇤ -TAn ⇤ E-Tt\ mn ⇤. -vnAmm⌃/ {vOqX⌃t {m≈ƒA nA/ s\fy,; 20; "FrodAZvm@y-T\ ò ⇤tíFp\ ˛kSyy ⇤t^. aEvEd(vA im\ y-t ˛Et§Am⌫Xp\ EDyA; 21; ktA kArEytA c {v t¯y\ nrk\ v }j ⇤t^. sA EáyA En PlA ˛oƒA Bv ⇤àmEvvEjtA; 22; dAErd˝\ &yAEDj\ d,KmTnAf\ kl"ym^. ky£\ kzt ⇤ sv◆ nAE-t yAgsmo Erp,. t-mA(sv˛y◊ ⇤n Evf ⇤q‚ ⇤n kAry ⇤t^; 23; iEt ⇤ dvAm⇥t ⇤ p—rA/ ⇤ sÿm, pVl,;

18 ˙cAg }Ad^ ] conj. ; ˙cAg }A\ MS

19 yAg ⇤ ] conj. ; yog ⇤ MS • Ev Z\ ] em. ; Ev Z

MS • zd˝AHy\ ] em. ; zd˝AHy MS

20 etå ⇤v ] conj. ; etd ⇤v MS • m⌃/ {-

vOqX⌃t { ] conj. ; m⌃/ {vApX⌃t { MSm≈ƒA ] conj. ; m≈oƒA MS 21 íFp\ ˛kSyy ⇤t^ ] em. ; dFp\ ˛Spy ⇤t^ MS • aEvEd(vA ] em. ; avEd(vA MSpc ; aAvEd(vA MSac • im\ ] conj. ; imA\ MS • m⌫Xp\ EDyA ] conj. ; m⌫XpA EDyA\ MS 22 t¯y\ ] em. ; tíy\ MS • En PlA ] em. ; En Pl\ MS • EvvEjtA ] em. ; Evv>jtA MS 23 dAErd˝\ ] em. ; dErd˝\ MS • nAf\ ] em. ; nAf MS • yAgsmo ] em. ; yAg\smo MS • t-mA(sv ] MSpc ; t-mA(ssv MSac

; a£mpVl, ; b }úovAc. f⇥Z v(s ˛v#yAEm m⌫Xp-yA[14v ]Ep -TApnm^. y {m⌃/ {ñAsr\ m— ⇤çAgo ⇤ dvmyo Bv ⇤t^; 1; ˛(ym— ⇤Et m⌃/ ⇤Z ˛AsAd\ t ˛Et§y ⇤t^. (a`n aA) yAEh m⌃/ ⇤Z m⌫Xp\ t smAlB ⇤t^; 2; v ⇤Edqd ⇤Et m⌃/ ⇤Z v ⇤Ed\ s\-TApy ⇤d^ Eíj,. sÿ (m⇥EdEt) m⌃/ ⇤Z sÿ (- - -) kA ⌃ys ⇤t^; 3; aE`nm`nFEt m⌃/ ⇤Z k⌫XA⌃y ⇤v ˛Et§y ⇤t^. n ⇤jm ⇤⇤ qEt m⌃/ ⇤Z m ⇤KlA\ -TApy ⇤ìD,; 4; f‡o ⇤ dvFEt m⌃/ ⇤Z yoEn\ cAEp ˛Et§y ⇤t^. c\ vAc ⇤Et m⌃/ ⇤Z ( E(vjA\) kOtk\ ⌃ys ⇤t^; 5; ah\ (sÿ ⇤)Et m⌃/ ⇤Z k ⇤fv(v ⇤ gzBv ⇤t^. c(vAEr f⇥ΩA m⌃/ ⇤Z torZ\ -TApy ⇤d^ Eíj,; 6; BBv, -v ⇤Et m⌃/ ⇤Z E/fl\ c ˛Et§[15r ]y ⇤t^. ´ 2 AVS.10.6.15: · · · ˛(ym—t f\Bvm^ ?; SV 1.1 a`n aA yAEh vFty ⇤ 3 RV 1.140.1: v ⇤Edqd ⇤ E˛yDAmAy sçt ⇤ 4 TB 3.3.7.3: aE`nm`nFEÅ~E-/, • RVKhila 10.184.1: n ⇤jm ⇤q prA pt 5 RV 10.9.4: f‡o ⇤ dvFrEB£y aApo Bv⌃t , VS 36.1: c\ vAc\ ˛pç ⇤ 6 RV 10.49.9: ah\ sÿ ˙vto DAry\ v⇥qA , RV.4.58.3: c(vAEr f⇥ΩA-/yo a-y pAdA, 7 Vs 36.3: BBv, -v-t(sEvtv⇤ r⌫y\ , RV 9.86.11: aEBá⌃dn^ klf\

vA>yqEt

1b }úo ] em. ; b }ú u MS • v(s ] em. ; v(s\ MS • m— ⇤çAgo ⇤ dvmyo ] conj. ; m—çAg\ ⇤ dvmy\ MS 2 m— ⇤Et ] conj. ; m—EÅ MS 3 -TApy ⇤d^ ] em. ; -TApy ⇤ MS 5 ⇤ dvFEt ] conj. ; ⇤ dvF MS • yoEn\ ] em. ; yoEn MS • E(vjA\ ] conj. ; EtjA\ MS 6 k ⇤fv(v ⇤ ] conj. ; k ⇤fv(v\ MS • f⇥ΩA ] conj. ; f⇥Ω MS • -TApy ⇤d^ ] em. ; -TApy ⇤ MS

⇤ dvAm⇥tpA—rA/ ⇤

85

a£mpVl,

aEBá⌃d ⇤Et m⌃/ ⇤Z klfAn^ -TApy ⇤Åt,; 7; vEs§A hFEt m⌃/ ⇤Z v-/y`mO t Ev⌃ys ⇤t^. im\ n somsAm ⇤n -tMB\ t pErkSpy ⇤t^; 8; ˙À ⇤ d˝=s ⇤Et m⌃/ ⇤Z p p˙`dAmk\ ⌃ys ⇤t^. s/AmAZ ⇤Et m⌃/ ⇤Z mAlAs/\ ˛kSpy ⇤t^; 9; uÅAnpZ ⇤ m⌃/ ⇤Z p•vA\-t/ Ev⌃ys ⇤t^. blEv‚Ay m⌃/ ⇤Z bAl&yjnk\ ⌃ys ⇤t^; 10; Ev Eh sotoErEt m⌃/ ⇤Z EvtAn\ co@vto ⌃ys ⇤t^. somo ddã⌃Dv ⇤Et v ⇤dgA⌃Dvk\ ⌃ys ⇤t^; 11; uBy\ t ⇤ vy-yo∆A uBy\ -TApy ⇤d^ @vEnm^. a-mAkEm⌃d˝ - m⌃/ ⇤Z ptAkA\ c ˛Et§y ⇤t^; 12; †s>yFf⇥/BAgA-(v ⇤Et†

m⌃/ ⇤Z -tr[15v ]Z\ -TApy ⇤ìD,.

prFto Eq— - sAm ⇤n py¨\ -TApy ⇤Åt,; 13; tBÅA aEΩr-(v ⇤Et tEl\ t/ ˛Et§y ⇤t^. i⌃d˝Ay sAm gAy ⇤Et g⌫XkO t tto ⌃ys ⇤t^; 14; 8 RV 1.26.1: vEs§A Eh Emy ⇤@y , RV I.179.5: im\ n sommE⌃tt,

9 RV 9.73.1 ˙À ⇤

d˝=s-y Dmt, , RV 10.63.10 s/AmAZ\ p⇥ETvF\ çAmn ⇤hsm^ 10 RV 10.145.2 uÅAnpZ ⇤ sBg ⇤ , RV 10.103.5 blEv‚Ay -TvFrA, ˛vFrA, 11 RV 10.86.1 Ev Eh sotors⇥"t , RV 10.85.41 somo ddã⌃DvAy 12 RV 2.9.5 uBy\ t ⇤ n "Fyt ⇤ vy-ym^ , VS 10.103.11: a-mAkEm⌃d˝, sm⇥t ⇤q @vj ⇤q 13 RV.9.107.1: prFto Eq—tA stm^ 14 RV 8.43.18: tyoEt\ c -TApy ⇤t^. cr⌃v { mD m⌃/ ⇤Z n {v ⇤ç {v ⇤Ed ’ pry ⇤t^; 15; i⌃d˝QC/ ⇤Et m⌃/ ⇤Z QCÅ~\ s\-TApy ⇤d^ Eíj,. D⌃vnA g ⇤Et m⌃/ ⇤Z m⌫Xp ⇤ c EdfA ⌃ys ⇤t^; 16; ud Et§ ⇤Et m⌃/ ⇤Z -vEdfA\ c ˛Et§y ⇤t^. aDA no EvòsOB ⇤Et BEmpAtAlkAn^ ⌃ys ⇤t^; 17; a⌃tñrEt m⌃/ ⇤Z a⌃tEr"\ ˛Et§y ⇤t^. yEd⌃d˝ ˛Agd?sAflA -TAlFpAk\ t -TApy ⇤t^; 18; pArAvt-y m⌃/ ⇤Z ˛[16r ]ZFtA\ c ˛Et§y ⇤t^. aA>ydoh ⇤Et sAm ⇤n aA>yAΩAn^ sADy ⇤d^ Eíj,; 19; a˛tFto jy(yAh b }úAZ\ pErkSpy ⇤t^. k⇥ ZoEs c ⇤Et m⌃/ ⇤Z -TApy ⇤⇤ dv ˙?˙vO; 20; aEd(y { &y⌃dn\ c {v -vAhAkAr\ ˛Et§y ⇤t^. a-t ôOqV^ c m⌃/oƒ\ vq÷Ar\ ⌃ys ⇤ìD,; 21; y‚ ⇤n vDt j ⇤Et y‚p\s\ ˛Et§y ⇤t^. 15 RV 6.46.5: i⌃d˝ >y ⇤§\ n aA Br , AB 7.15.5: cr⌃v { mD 16 RV Khila 10.142.5: i⌃d˝, "/\ ddAt , RV.6.75.2: D⌃vnA gA, 17 RV 8.23.5 = VS 11.41: ud Et§ -v@vr , RV 1.42.6: aDA no EvòsOBg 18 VS 3.7: a⌃tñrEt , RV 8.4.1: yEd⌃d˝ ¯ G¯ ˛AgpAgdk^ 19 RV 8.34.18: pArAvt-y rAEtq , RV 6.7.1 = SV 1.67; A. a. 43–45: aA>ydohsAm 20 RV 4.50.9: a˛tFto jyEt , VS 2.1: k⇥ Zo-yAKr ⇤§o 21 VS 2.2: ´ 1.6.11.1: tA vA etA, p— &yA„tyo Bv⌃(yo ôAvy a-t aEd(y { &y⌃dnmEs , SB ôOqX^ yj y ⇤ yjAmh ⇤ vOqEXEt 22 RV 2.2.1: y‚ ⇤n vDt jAtv ⇤dsm^ , RV 8.11.6:

Ev˛\ Ev˛Aso _vs ⇤

15 >yoEt\ c ] em. ; >y ⇤Etñ MS • cr⌃v { mD ] em. ; cr(v {mD MS 16 -TApy ⇤d^ ] em. ; -TApy ⇤ MS • D⌃vnA g ⇤Et ] em. ; D\⌃vnA g {Et MS 17 ud Et§ ⇤Et ] em. ; uíE-t§ ⇤Et MS • -vEdfA\ ] conj. ; #mEdfA MS • BEmpAtAlkAn^ ] conj. ; B\m⇥tpAlkA MS 18 ˛Agd?sAflA ] conj. ; yA\Ωd?sAm MS 19 pArAvt-y ] conj. ;

pArAtAñ MS • aA>ydoh ⇤Et ] em. ; aAydOh ⇤Et MS • aA>yAΩAn^ sADy ⇤d^ ] em. ; aAyΩA\ sA@yt ⇤ MS 20 k⇥ ZoEs ] conj. ; k⇥ ZoEs\ MS 21 aEd(y { &y⌃dn\ c {v ] em. ; aAEd(y { &y⌃dnñ {v MS • ôOqª⌃/oƒ\ vq÷Ar\ ] conj. ; ôoyím⌃/Oƒ\ v£kAr\MS 22 vDt j ⇤Et ] conj. ( unmetrical ); vDt-(v ⇤Et MS • p\s\ ] em. ; p\s MS • Ev˛A`nF\-t ] em. ; Ev˛A`nF-t MS

87

⇤ dvAm⇥tpA—rA/ ⇤

a£mpVl,

Ev˛\ Ev˛ ⇤Et m⌃/ ⇤Z Ev˛A`nF\-t smhy ⇤t^; 22; nm-t ⇤ zd˝ m⌃/ ⇤Z k⌫X ⇤ kAlAE`n ’ Ev⌃ys ⇤t^. eq ⇤ dvmyo yAg, m⌫Xp-TApn\ prm^; 23; aòm ⇤DPl\ v(s ktA ˛AŸoEt bEàmAn^. am⌃/oƒA\ t y, kyA(˛Et§A\ loBmoEht,; 24; s [16v ]yAEt nrk\ Gor\ ktA kArEytA c y,. -TAnB }\fo rA£~B }\f-t/ rA‚ ⇤ mhâym^. aEcr ⇤Z Bv ⇤(svm(pAt\ t Dn"ym^; 25; iEt ⇤ dvAm⇥t ⇤ p—rA/ ⇤ _£m, pVl,;

23 VS 16.1: nm-t ⇤ zd˝ m⌃yv ⇤ 23 k\X ⇤ ] em. ; k\XA\ MS • myo yAg, m⌫Xp ] em. ; my yAg\ m⌫X MS 24 ƒA\ t y, kyA(˛Et§A\ loB ] em. ; ƒ t y, kyA(pEt§A loh MS 25 -TA-

nB }\fo ] em. ; -TAn\B }\f\ MS • B }\f-t/ ] em. ; B }\f\ t/ MS • m(pAt\ t ] conj. ; mçog⌃t MS

; nvmpVl, ; b }úovAc. at, pr\ ˛v#yAEm ‹AnkmEvEDyTA. -TApk, sshAy-t b }AúZ {v ⇤dpArg {,; 1; m⌫Xp-y EvBAg ⇤n IfAn ⇤n yTAámm^. h-tmA/˛mAZ ⇤n Bd˝pFW\ tto ⌃ys ⇤t^; 2; g }Am\ ˛dE"Z\ k⇥(vA f≤vAEd/En-vn {,. bEl\ c svto dçA(sv⇤ dy\ t/ smAdAy mDnA sh s\ytm^. 5 RV 6.75.6: rT ⇤ Et§‡yEt vAEjn, , RV 1.89.8 = VS 25.21: Bd˝\ kZ ⇤EB, f⇥ZyAm 7 VS 3.9: aE`n>yoEt>yoEtrE`n, 8 RV 1.90.6: mD vAtA tAyt ⇤ 1b }úo ] em. ; b }úA u MS • yoEtr ⇤qA\ MS 7 EáymAZ ⇤ ] em. ; EáyAmAZ ⇤ MS aA>y\ t/ ] em. ; aA>yÅ/ MS • s\ytm^ ] em. ;

⇤ dvAm⇥tp—rA/ ⇤

89

nvmpVl,

mD vAt ⇤Et m⌃/ ⇤Z pry ⇤⌃mDsEpqF; 8; Ehr⌫ygBm⌃/ ⇤Z svZ-y flAkyA. Ec/\ ⇤ dvAnA\ m⌃/ ⇤Z n ⇤/ ⇤ u⌃mFly ⇤ìD,; 9; Bv ⇤⌃⌃ynAEtErƒ ⇤n ˛EtmAΩ ⇤q bEàmA[17v ]n^. doq˛fmnF\ fAE⌃t\ -TApk-t smArB ⇤t^; 10; EtlAn^ k⇥ ZAn^ G⇥tAy\ g⇥¢FyAå ⇤v-y (v ⇤Et kfodkm^. p—g&y ⇤n ‹Apy ⇤t^; 22 As.t.a ¯da´savidh¯ ana 84ab:

yA PElnF yAPlA iEt m⌃/ ⇤Z Plodk ⇤n ‹Apy ⇤t^. 22cd–23ab Garud.apur¯ an.a 1.48.44c-45b:

shd ⇤vF blA c {v ftmlF ftAvrF; kmArF c gXcF c Es\hF &yAG }F tT {v c. 17 fMydMbrmò(T\ ] em. ; sMy ⇤dbrmò(T MS • plAfjm^ ] em. ; plfj\ MS 18 sEp, ] em. ; sE=p MS 19 gAyÏyA ˛Tm\ ] em. ; g }Ay(yA ˛T\ MS 20 y("Fr\ ] em. ; y("r\ MS • áANZo ’ Et v { ] áAŸoEt v ⇤MS 21 The first p¯ada is unmetrical. • g&ymdA ] MSpc ; g&y\mdA MSac • Et⌃dkAmlk ⇤ ] em. ; t ⇤⌃dkAmlkA MS 22 yA, PElnFEt ] em. ; yA PlnFEt MS • blA ] em. ; brA MS 23 Es\hF ] em. ; Eq=pF MS • m>jl {, ] em. ; mΩl\ MS

⇤ dvAm⇥tp—rA/ ⇤

91

nvmpVl,

yA aoqDFEt m⌃/ ⇤Z ‹AnmoqEDm>jl {,; 23; G⇥tvtFEt m⌃/ ⇤Z aA>yBAg\ t kAry ⇤t^. d˝pdAEd pn-t ⇤qA\ kyAdítn\ bD,; 24; Efr,‹An\ tt, kyA⌃mAn-tokAEBmE⌃/tm^. Ev Zo rrAVm⌃/ ⇤Z dçAã⌃Dodk\ fBm^; 25; tto nçdk ⇤n {v "Aly ⇤QCBvAErZA. g⇥ZAno v { ZvA⌃m⌃/An£Av ⇤v mhAmn ⇤; 26; aAvAhy ⇤Åto ⇤ dv\ pFtAMbrDr\ hErm^. sAE‡@ykrZATAy m⌃/ {v ⇤dsmâv {,; 27; nm-t ⇤ (yƒsΩAy s\toqEvBvA(mn ⇤. ‚An[19r ]Ev‚An!pAy b }út ⇤jonpAEyn ⇤; 28; gZAEtáA⌃tv ⇤gAy pzqAyAgZA(mn ⇤. a"rAy prAZAy Ev Zo sE‡Ehto Bv; 29; eù ⇤Eh BgvE⌃v Zo lokAng }hkArk. 23 RV 10.97.1: yA aoqDF, pvA jAtA, 24 RV 6.70.1: G⇥tvtF BvnAnAmEBEôyA , VS 20.20: d˝pdAEdv mmcAn, 25 RV 1.114.8 = VS 16.16: mAn-tok ⇤ tny ⇤ mA 25 VS 5.21: Ev Zo rrAVmEs 26 I have not been able to identify these eight mantras. I can think of Ev Zsƒ (RV 1.154), but there are only 6 stanzas. 23 As.t.a ¯da´savidh¯ ana 83ab: yA aoqDFEt m⌃/ ⇤Z ‹AnmoqEDm>jl {,. 27–29 Agnipur¯ an.a I.60.24–26:

sAE‡@ykrZ‡Am „dy\ -p⇥[y v { jp ⇤t^. sƒ⌃t pOzq\ @yAyn^ id\ gùmn\ jp ⇤t^; nm-t ⇤-t s⇤ rfAy s⌃toqEvBvA(mn ⇤. ‚AnEv‚An!pAy b }út ⇤jonyAEyn ⇤; gZAEtáA⌃tv ⇤fAy pzqAy mhA(mn ⇤. a"yAy prAZAy Ev Zo sE‡Ehto Bv; 25 kyA⌃mAn ] em. ; kyA(mAn MS 26 dk ⇤n {v ] em. ; dk ⇤n ⇤v MS • g⇥ZAno v { ZvA ] em. ; g⇥ZAZo v ⇤ ZvA MS • n£Av ⇤v ] em. ; n£ ev MS 27 Dr\ ] em. ; Dr MS • sAE‡@y ] em. ; sA‡ {@y MS • m⌃/ {v ⇤d ] em. ; m⌃/ ⇤⇤ íd MS pc ac 28 s\toq ] em. ; s\Åos MS • Ev‚An ] MS ; EvEv‚An MS • npAEyn ⇤ ] conj. ; npAEln ⇤ MS 29 yAgZA(mn ⇤ ] em. ; ygZA(mn ⇤ MS 30 vE⌃v Zo lokA ] em. ; vAE⌃v Zo•okA MS • kArk ] em. ; kArk, MS • vAs⇤ dvAy pc ac t ⇤ ] em. ; vAs⇤ dvy v ⇤ MS ; vAs⇤ dy v ⇤ MS

⇤ dvAm⇥tp—rA/ ⇤

92

nvmpVl,

y‚BAg\ ˛g⇥¢F v vAs⇤ dvAy t ⇤ nm,; 30; Ehr⌫my ⇤n pA/ ⇤Z a]y◆ dçA(˛y◊t,. g⌃DmASyopvFt\ t id\ Ev ZEvcám ⇤; 31; b⇥h-pt ⇤ ’ Et m⌃/ ⇤Z vAs\ t pErDApy ⇤t^. v ⇤dAhEmEt m⌃/ ⇤Z soÅrFy\ t kAry ⇤t^; 32; b }ú j‚ ⇤Et m⌃/ ⇤Z upvFt\ tto ⌃ys ⇤t^. yA aoqDFEt m⌃/ ⇤Z p { prvEkr ⇤Åt,; 33; ôFñ t ⇤ l#mF - m⌃/ ⇤Z mASy\ t pErDApy ⇤t^. DrEs [19v ]c ⇤Et m⌃/ ⇤Z Dp\ dçAÅ g`glm^; 34; sEmào a“ - m⌃/ ⇤Z a“n\ t ˛dApy ⇤t^. y“E⌃t b } - m⌃/ ⇤Z rocnA\ t/ dApy ⇤t^; 35; 31 VS 40.17: Ehr⌫my ⇤n pA/ ⇤Z s(y-yAEpEht\ mKm^

31 RV 1.22.17: id\ Ev ZEv-

cám ⇤ / ⇤DAndD ⇤ pdm^ 32 RV 2.23.15/VS 26.3: b⇥h-pt ⇤ aEt ydyo • VS 31.18: v ⇤dAhm ⇤t\ pzq\ mhA⌃t\ 33 VS 13.3: b }ú j‚An\ ˛Tm\ pr-tAt^ , RV 10.97.1: yA aoqDF, pvA jAtA, 34 VS 31.22: ôFñ t ⇤ l#mFñ , VS 1.8: DrEs Dv Dv⌃t\ 35 VS 29.1: sEmào a“⌃k⇥drm^ , RV 1.6.1: y“E⌃t b }msrm^ 31cd–34 As.t.a ¯da´savidh¯ ana 85–89ab:

v ⇤dAhm ⇤t\ pzqEmEt m⌃/ ⇤Z g⌃Dodk ⇤n ‹Apy ⇤t^. im\ m ⇤ gΩ ⇤ ymn ⇤ ’ Et m⌃/ ⇤Z nçodk ⇤n ‹Apy ⇤t^; b⇥h-pEtm⌃/ ⇤Z v-/\ c {v ˛dApy ⇤t^. †jnrƒA pErDAmAn,† m ⇤Kl\ pErdApy ⇤t^; yvA svA iEt m⌃/ ⇤Z y‚opvFt\ c dApy ⇤t^. ôFñ t ⇤ l#mFñ ⇤Et m⌃/ ⇤Z c⌃dn\ c ˛dApy ⇤t^; y“t ⇤ mn ut y“t ⇤ go(rocn\ dA) +++. ++⇤ dvtA vAto ⇤ dvtA iEt m⌃/ ⇤Z ⇤ dv\ c {v t pjy ⇤t^; DrEs Dv Dv⌃t\ Dp\ dçAÅ gΩlm^. 35ab As.t.a ¯da´savidh¯ ana 90cd: sEmào a“⌃k⇥drm“n\ t ˛dApy ⇤t^. 31 pA/ ⇤Z a]y◆ ] em. ; ˛A/ ⇤n aG }◆ MS • dçA ] MSpc ; ççA MSac • mASyopvF-

t\ t ] em. ; mASy˛Evt⌃t MS • Evcám ⇤ ] em. ; EŒcám {, MS 32 pErDA ] conj. ; ˛ErDA MS 33 j‚ ⇤Et ] conj. ; y‚ ⇤Et MS • rvEkr ⇤Åt, ] em. ; rvEkr\tt, MSpc ; rv. Ekr\tt, MSac 34 l#mF ] conj. ; l"F\ MS • mASy\ t pErDAp ] em. ; mASy-t pErdAp MS 34 DrEs c ⇤Et ] em. ( unmetrical ); DrEsñ ⇤Et MS • dçAÅ ] em. ; dçA\t MS 35 a“ ] em. ; a— MS • rocnA\ t/ ] em. ; rocZAÅ⌃/ MS

93

⇤ dvAm⇥tp—rA/ ⇤

nvmpVl,

blEv‚Ay m⌃/ ⇤Z bAl&yjnk\ dd ⇤t^. ⇤ dvA gAE(vEt m⌃/ ⇤Z dpZ\ c ˛dApy ⇤t^; 36; a‡pt ⇤ ’ Et m⌃/ ⇤Z sG⇥t\ k⇥sr\ dd ⇤t^. tto Enm⌃/y ⇤⇤ åv\ dFGAy(v ⇤Et mE⌃/tm^; 37; uEŧ ⇤Et smà⇥(y k⇥(vA b }úrT\ fBm^. sh˙fFqm⌃/ ⇤Z m⌫Xp\ t ˛v ⇤fy ⇤t^; 38; tt, ˛v ⇤fy ⇤⇤ åvmG◆ dçA(˛y◊t,. ˛AsAdAEBmK\ ⇤ dv\ f⇧yAyA\ t Env ⇤fy ⇤t^; 39; Evòtñ"Er(y ⇤v\ kyA(sklEn klm^. ‹A[20r ]nkm - m^ - id\ ˛oƒ\ k ⇤fv-y jg(pt ⇤,. b }AúZA, -TApkñ {v ky, kvFt y◊t,; 40; iEt ⇤ dvAm⇥t ⇤ p—rA/ ⇤ nvm, pVl,;

36 RV 10.103.5 blEv‚Ay -TvFrA, ˛vFrA, , VS 8.21 ⇤ dvA gAtEvdo gAt\ EvÇvA gAtEmt 37 VS 11.83: a‡pt ⇤ _‡-y no ⇤ dEh , AV 2.41: dFGAy(vAy b⇥ht ⇤ 38 RV 1.40.1: uEŧ b }úZ-pt ⇤ , RV 10.90.1: sh˙fFqA pzq,

40 RV 10.81.3: Evòtñ"zt

EvòtomK, 38 As.t.a ¯da´savidh¯ ana 92cd–93ab:

uEŧ b }úZ-pt ⇤ !p\ c {v smàr ⇤t^; aA no Bd˝A ’ Et m⌃/ ⇤Z yAg-TAn\ ˛v ⇤fy ⇤t^. 38cd Sv¯ ayambhuvapa˜ ncar¯ atra7.2a: sh˙fFqA pzq, m⌫Xp\ t ˛v ⇤fy ⇤t^. 36 m⌃/ ⇤Z ] em. ; m⌃/Ah MS • Et m⌃/ ⇤Z ] MSpc ; Et MSac • dpZ\ c ] conj. ; d=pn\ MS 37 Enm⌃/y ⇤ ] conj. ; Enm(sy ⇤ MS 38 smà⇥(y ] em. ; smâ⇥(y MS • ˛v ⇤fy ⇤t^ ] em. ; Evf ⇤qy ⇤t^ MS 39 y ⇤⇤ åv ] em. ; y ⇤t^ ⇤ dv MS -TApk ] em. ; b }úZA -TAkA MS

40 b }AúZA,

; dfmpVl, ; b }úovAc. aTAt, s\˛v#yAEm

E(vjAnA\ t l"Zm^. bü⇥c\ b }AúZ\ c {v vAEjm@yE⌃dn\ tTA; 1; C⌃dogo _TvZñ {v v ⇤dv ⇤dAΩpArgA,. v { ZvA, kls\p‡A et ⇤ mEtDrA, -m⇥tA,; 2; -TApk-y gZAn^ v#y ⇤ f⇥Z v ⇤kAg }mAns,. b }AúZ\ m⌃/Evdq\ p—rA/EvfArdm^; 3; s\pZ◆ svgA/ ⇤q fA-/‚\ E˛ydfnm^. mhAEBjns\p‡\ v { Zv\ v ⇤dpArgm^; 4; !ptÇv[20v ]smAyƒmEBEqƒ\ fEcv }tm^. p—Ev\fEttÇv‚mAcAy◆ smdA„tm^; 5; vjnAy ˛v#yAEm y {-t km n kAry ⇤t^. nAEtdFGo n †-vñ nAEt-Tl, k⇥f-tTA; 6; nAEtgOro n k⇥ Z-t nAfEcv ⇤dvEjt,. 4 –7 Sv¯ ayambhuvapa˜ ncar¯ atra 6.*1–*4: · · · broken q fA-/‚, E˛ydfn,. p—Ev\f-

EttÇv‚, -TApk, smdA„t,; vjnFyA\-t v#yAEm y {-t km n kAry ⇤t^. lklFf\ EvvZ-T\ kAZ\ Epc”locnm^; nAEtgOr\ n k⇥ Z\ c End⌃t\ -Tld ⇤hkm^. KSvAV\ [yAmd⌃t\ c kE§n\ &yAEDpFEXtm^; KoX\ kNj\ v }ZF kAZo +++++++; B`nnAs\ ˛y◊ ⇤n v>jy ⇤(-TApk, sdA; 1 b }úo ] em. ; b }ú u MS 2 s\p‡A ] em. ; sA\p\nA MS 3 gZAn^ ] em. ; gZA MS • Evdq\ ] conj. ; Ev⌃dq\ MS 4 s\pZ◆ ] em. ; s\põ MS • dfnm^ ] em. ; dfn\ MS 4v { Zv\ v ⇤dpArgm^ ] em. ; v ⇤ ZvA\ v ⇤dpArgAn^ MS 6 vjnAy ˛v#yAEm ] em. ; v>jnAy ⇤ ˛v"AEm MS • y {-t km n ] em. ; y ⇤q km Z MS • dFGo ] em. ; dFG MS

⇤ dvAm⇥tp—rA/ ⇤

95

dfmpVl,

m(srF ykrñ {v nAEtbAlo n v⇥àk,; 7; KSvAVF EpΩlñ {v kEplo EnBy-tTA. k§F c &yAEDtñ {v dñmA kAZk⌫WkO; 8; ôEt-m⇥EtEáyAbAù\ hFnAΩ\ k⌫Xgolkm^. a⌃yd ⇤vrt\ fd˝\ pAq⌫X\ jAEthFnkm^; 9; kOl\ ⇤ dvlk\ sOrmAht\ c {v bOàkm^. et ⇤ v { Ev ZyAg ⇤ _/ vjnFyA, ˛y◊[21r ]t,; 10; kOlo hnEt ktAr\ BAyA◆ ⇤ dvlko hn ⇤t^. sOrñ hnt ⇤ g }Am\ n`no rA£~\ c End⇤ ht^; 11; bOàñ kzt ⇤ sv◆ pfp/Dn"ym^. t-mAç◊ ⇤n Ev˛ ⇤⌃d˝ a⌃yd ⇤vrtA\-(yj ⇤t^; 12; iEt ⇤ dvAm⇥t ⇤ p—rA/ ⇤ dfm, pVl,;

8 kAZ ] em. ; kAf MS 9 rt\ fd˝\ ] em. ; tr\ fd˝ MS • pAq⌫X\ ] conj. ; q⌫XF MS 10 mAht\ c {v ] em. ; mAht—v MS • Ev ZyAg ⇤ _/ ] em. ; Ev ZyAgo

/ MS 11 ktAr\ ] MSpc ; kAÅAr\ MSac • ⇤ dvlko ] em. ; ⇤ dvko MS em. ; ˛s MS 12 rtA\-(yj ⇤t^ ] em. ; rtA-(yj ⇤t^ MS

12 pf ]

; ekAdfmpVl, ;

b }úovAc. aTAcn\ ˛v#yAEm sklFkrZ\ fBm^. -TApk ⇤n t kt&y\ -vd ⇤⇤ h ⌃yAsmÅmm^; 1; ˛EtmAyA\ Evf ⇤q ⇤Z kyAdAlBn\ bD,. íAdfA"rm⌃/\ t mlm⌃/mdA„tm^. : nmo Bgvt ⇤ vAs⇤ dvAy; 2; :kAr\ Ev⌃ys ⇤⌃mE nkAr\ nAEskop[21v ]Er. mokAr\ t mK ⇤ ⌃y-y BkAr\ k⌫Wgocr ⇤; 3; gkAr\ bAh⇤ df ⇤ t vkAr\ „dy ⇤ ⌃ys ⇤t^. t ⇤kAr\ nAEBd ⇤f ⇤ t vAkAr\ p⇥§d ⇤ft,; 4; skAr\ kEVm@y ⇤ t ⇤ dkAr\ cozm@yt,. 1 –5 Sv¯ ayambhuvapa˜ ncar¯ atra 7.3cd–8: pnr ⇤v\ ˛v#yAEm sklFkrZmÅmm^; ˛Et-

mAyA\ Evf ⇤q ⇤[Z] kyAŒAlBn\ bD,. : nmo Bgvt ⇤ vAs⇤ dvAy; m⌃/Emd\ ˛oƒm^; :kArAEdykArA⌃tAn^ Ev⌃ys ⇤dnpvf,. :kAr\ Ev⌃ys ⇤⌃mE nkAr\ nAEskA⌃tr ⇤; mokAr-t llAV ⇤ _y\ (BkAr\ vdn ⇤) ⌃y++. +++ +++ ( f ⇤ t ) vkAr\ „dy ⇤ ⌃ys ⇤t^; t ⇤kAr\ t Bj ⇤ ⇤ dy\ dE"Z ⇤n mhAblm^. vAm ⇤ Bj ⇤ vAkAr\ t EcE⌃tt&y\ Evc"Z; 7; skAr\ vAmj\GAyA\ ⇤ dkAr\ dE"Z ⇤ tTA. vAkAr\ vAmpAd ⇤ t ykAr\ dE"Z ⇤ ⌃ys ⇤t^; 2cd–5 Soma´sambhupaddhati 4.6.8–10ab:

mEt!p\ c f⇧yAyA\ Env ⇤[y tdn⌃trm^. íAdfA"rm⌃/-y k⇥(vA ⌃yAsmn⌃trm^; K≥AdFnA\ c Ev⌃yAs\ m⌃/F kyAŒ tçTA. EfroG }AZllAV ⇤òA-yk⌫W„dy ⇤q c; d"dovAmdovAmd"jøAE¿q ámAt^. 1b }úo ] em. ; b }ú u MS • kt&y\ ] em. ; kÅ&y MS MS

2 m⌃/\ ] em. ; m⌃/ ⇤\

5 kEV ] em. ; kAEV MS • yo⌃ys ⇤t^ ] em. ; yo⌃ys ⇤t^ MS

⇤ dvAm⇥tp—rA/ ⇤

97

ekAdfmpVl,

vAkAr\ j\Gyo⌃y-y ykAr\ pAdyo⌃ys ⇤t^; 5; ev\ ⌃yAsEvED\ k⇥(vA aAcAy, ssmAEht,. :kAr\ t sdA @yAy ⇤>>vAlAmAlAsm˛Bm^; 6; nkAr\ f vZAB\ "r⌃t\ fFtl\ jlm^. mokAr\ Ec⌃ty ⇤(˛A‚o EB‡A“nsm˛Bm^; 7; tzZAEd(ys\kAf\ BkAr\ Ec⌃ty ⇤(sdA. gkAr\ t pn-t ⇤qA\ ⇤ hmvZ◆ t Ec⌃ty ⇤t^; 8; f[22r ]à-PEVks\kAf\ vkAr\ @yAnmÅmm^. t ⇤kAr\ Ec⌃ty ⇤(˛A‚o Ed&yk⌃dsm˛Bm^; 9; ⇤ hmk⌃dEnBAkAr\ vAkAr\ vzZAlym^. Evç(p“EnBAkAr\ skAr\ t/ Ec⌃ty ⇤t^; 10; ⇤ dkAr\ t pn@yAy ⇤⌃mhAnFlsm˛Bm^. "FrodAZvs\kAf\ vAkArmm⇥tA"rm^; 11; ykAr\ t sdA @yAy ⇤(pîrAgsm˛Bm^. 6cd–12ab Sv¯ ayambhuvapa˜ ncar¯ atra 7.9cd–14:

:kAr\ t sdA @yAy ⇤>>vAlA ( mA ) +++++; ++++++ZAB\ nkAr\ fFtl\ jlm^. mokAr\ nFlEnB\ p“\ EB‡A“nsm˛Bm^; tzZAEd(ys\kAf\ BkAr\ Ec⌃ty ⇤(sdA. gkAr\ t kr-t ⇤qA\ ⇤ hmvZ◆ EvEc⌃ty ⇤t^; f pîEnBAkAr\ vkAr\ @yAnmÅmm^. t ⇤kAr-y t Ev‚ ⇤y\ >yoEt, k\mmsE‡Bm^; Ehmk⌃dEnB\ ( c { ) +++++++++. Evç(p“EnBAkAr\ skAr\ t/ Ec⌃ty ⇤t^; @yAy ⇤⇤ åkArbFj\ t pFt\ mAh ⇤⌃d˝sE‡Bm^. "FrodAZvsAd⇥[y\ vAkAr\ vzZAlym^. ykAr\ t sdA @yAy ⇤(pîrAgsm˛Bm^; 6–12ab Soma´sambhupaddhati 4.6.10cd–12ab:

>vAlAmAlAklò ⇤tm ⇤ckAEd(ys˛BAn^; pFtf AzZò ⇤tEvç(pFtEstAzZAn^. ao¨ArAEdykArA⌃tAn^ vZAn^ íAdf Ev⌃ys ⇤t^; -vyMBpA—rA/ ⇤ c svm ⇤tddFErtm^. 7 r\ f vZAB\ ] em. ; r f vZAB MS • "r⌃t\ ] conj. ; "kAr⌃t\ MS • jlm^ ] em. ; jl MS • y ⇤(˛A‚o ] em. ; y ⇤˛A‚o MS • EB‡A“n ] em. ; EB\nAjn MS 9 Ed&y ] em. ; Ed⌃y MS

10 t/ Ec⌃ty ⇤t^ ] em. ; t\t EcÅy ⇤t^ MS

@yAy ⇤ ] em. ; pn@yAy ⇤ MS • rmm⇥tA ] em. ; r\mm⇥tA MS

11 pn-

98

⇤ dvAm⇥tp—rA/ ⇤

ekAdfmpVl,

k ⇤fv\ EfrEs ⌃y-y mK ⇤ nArAyZ\ tTA; 12; g }FvAyA\ mADv\ c {v goEv⌃d\ bAhk ⇤ ⌃ys ⇤t^. „dy ⇤ Ev⌃ys ⇤Eí Z\ p⇥§ ⇤ c mDsdnm^; 13; jWr ⇤ c {v k‘A\ v { vAmn\ c E/Evámm^. ôFDr\ c „qFk ⇤f\ j\G ⇤ t pErkSpy ⇤t^. [22v ]gù ⇤

c pînAB\ t pAdO dAmodr\ ⌃ys ⇤t^; 14;

n tA\ gEt\ yAE⌃t srA n yogF n c yAE‚kA,. yA\ gEt\ yAE⌃t yog ⇤ _E-mn^ íAdfA"rEc⌃tkA,; 15; g(vAg(vA Envt⌃t ⇤ c⌃d˝syAdyo g }hA,. açAEp n Envt⌃t ⇤ íAdfA"rEc⌃tkA,; 16; a-/⌃yAs\ ˛v#yAEm sv⇤ d⇤ hq Ev⌃ys ⇤t^. 12cd–14 As.t.a ¯da´savidh¯ ana 48-50: k ⇤fv\ t Efr ⇤ ⌃y-y mK ⇤ nArAyZ\ ⌃ys ⇤t^. g }FvAyA\ mADv\ c {v goEv⌃d\ bAh ’ Ev⌃ys ⇤t^; „dy ⇤ Ev⌃ys ⇤Eí Z\ jW+ +++++. kEVy mokAr ⇤Z mhA(mn,. BkArA"rd ⇤v ⇤f\ goEv⌃d\ t ctTk ⇤; p—m ⇤ t gkAr ⇤Z Ev Z\ c {v ˛pjy ⇤t^. vkArA"rd ⇤v ⇤f\ q§ ⇤ v { mDsdnm^; 25 yjn\ ] em. ; y‚n MS • v#yAEm ] em. ; v"AEm MS • yZ-y ] em. ; yn-y MS • kAy◆ ] em. ; kArA MS 26 ar {íA ] em. ; ar {íA MS • t⌃m@y ⇤] em. ; t(m@y ⇤ MS • p/\ ] em. ; p/ MS 27 r˙\ t ctíAr\ ] em. ; r˙ t ctíAr\ MS 28 dçAd^ ] em. ; dçA MS • EfKA\ ] em. ; EfKA MS 29 gAy/F\ ] em. ; gAy/F MS • sAEv/FmFf ] conj. ; sAEv/F IfAn MS • c {v t ] em. ; c { t MS • n { (yA\ vAy&yA\ ] em. ; n { (yA\ vAy&yA MS 30 d ⇤vmy\ ] em. ; d ⇤vtm\ MS

⇤ dvAm⇥tp—rA/ ⇤

101

ekAdfmpVl,

íAdfAr ⇤ tTA cá ⇤ arAg ⇤ }q Evc"[24r ]Z,. íAdfA"rs\yƒA ⌃ys ⇤d^ íAdfmty,; 31; k ⇤fv\ t ar ⇤ pv ⇤ :kAr ⇤Z t pjy ⇤t^. EítFy\ t nkAr ⇤Z ⇤ dv\ nArAyZ\ yj ⇤t^; 32; t⇥tFy ⇤ mADv\ p>y mokAr ⇤Z mhA(mnA. BkArA"rd ⇤v ⇤n goEv⌃d\ t ctTkm^; 33; p—m ⇤ c gkAr ⇤Z Ev Z\ c {v t pjy ⇤t^. vkArA"rs\yƒ\ q§ ⇤ t mDsdnm^; 34; E/Evám\ tt, pñAÅ ⇤kAr ⇤Z t pjy ⇤t^. vAmn\ cA£m ⇤ c {v vAkAr ⇤Z t pjy ⇤t^; 35; ôFDr\ nvm ⇤ p>y skAr ⇤Z mhA(mnA. dfm ⇤ t „qFk ⇤f\ ⇤ dkArA"rpEjtm^; 36; ekAdf ⇤ pînAB\ vAkA[24v ]⇤ rZ Enyojy ⇤t^. íAdf ⇤ t ykAr ⇤Z yƒ\ dAmodr\ ⌃ys ⇤t^; 37; pî ⇤ t yjn\ k⇥(vA pñAŒá\ ˛pjy ⇤t^. EítFyAvrZ\ HyAt\ t⇥tFy ⇤ _-/AEZ Ev⌃ys ⇤t^; 38; K≥\ c {v ⌃ys ⇤(pv ⇤ aA`n ⇤⇧yA\ t gdA\ ⌃ys ⇤t^. sÿm ⇤ vAmn\ c {v t ⇤kAr ⇤Z t pjy ⇤+. ++ íAkArbFj ⇤n a£m ⇤ t E/Evámm^; ôFDr\ nvm ⇤ c {v skAr ⇤Z t pjy ⇤t^. dfm ⇤ t „qFk ⇤f\ ⇤ dkAr ⇤Z t pjy ⇤t^; ekAdf ⇤ t vAkAr\ pînAB\ ˛B\ Evd,. íAdf ⇤ t ykAr ⇤Z nAflA dAmodr, -m⇥t,; 38cd–40 Sv¯ ayambhuvapa˜ ncar¯ atra 7.29–31ab:

EítFyAvrZ\ HyAt\ t⇥tFy ⇤ _-/AEZ Ev⌃ys ⇤t^. f≤\ c {v ⌃ys ⇤(pvmA`n ⇤⇧yA+++++; + ( E" ) Z ⇤n Bv ⇤Œá\ K≥\ n { (ygocr ⇤. pî\ pEñmto EvçAíAy&yA\ t hl\ ⌃ys ⇤t^; msl\ coÅr ⇤ dçAdFfA⌃yA\ fAΩ ’ Ev⌃ys ⇤t^. 31 arAg ⇤ }q ] MSpc ; aArAg ⇤ }q MSac • ⌃ys ⇤d^ ] em. ; nAs ⇤t^ MS 32 ⇤ dv\ nArAyZ\ ] em. ; ⇤ dv nArAyn MS 33 p>y ] em. ; p>y { MS 34 pjy ⇤t^ ] em. ; pjy ⇤t MS 35 E/Evám\ ] em. ; E/Evá\ MS • pñAt^ ] em. ; p\cAt^ MS • vAmn\ cA£m ⇤ ] em. ; vAbn co£m ⇤ MS • vAkAr ⇤Z ] em. ; vA\kAr ⇤n MS 37 dAmodr\ ] em. ; dAmOdr\ MS 39 ⇧yA\ t gdA\ ] em. ; ⇧yA t gdA MS • f\K\ En Et ] em. ; f\K n { Et MS

102

⇤ dvAm⇥tp—rA/ ⇤

dE"Z ⇤ Ev⌃ys ⇤Œá\ f\K\ En Etgocr ⇤; 39; pî\ pEñmto ⌃y-y vAy&yA\ t hl\ ⌃ys ⇤t^. msl\ coÅr ⇤ dçAdFfA⌃yA\ f⇥EΩ Ev⌃ys ⇤t^; 40; iEt ⇤ dvAm⇥t ⇤ p—rA/ ⇤ ekAdfm, pVl,;

40 vAy&yA\ ] em. ; vAy yA\ MS

ekAdfmpVl,

; íAdfmpVl, ; sn(kmAr uvAc; aôà ⇤yEmd\ ⌃yAs\ pjAm⌃/ {, ˛EtE§tm^. r kT\ BvEt ⇤ dv(vm⇥`yj[25 ],sAmvEjt ⇤; 1; b }úovAc. aT s\-TApn\ v#y ⇤ ˛EtmA\ cAEDvAEstAm^. dfA(mA pzq, HyAt, v ⇤doƒm⌫XlA(mk,; 2; aE`nmFl ⇤

cA (v ⇤kA m(-yr ⇤KA t EtygA.

kq\Bk-td ˛Ah, km⇤ rKA aDogtA; 3; aAvd\-(vEmEt ˛oƒA vArAhF coDtomKA. g⇥ZAnA jmd`nFEt nArEs\hF aDomKA; 4; DAm⌃t ⇤ EvòEm(yAhrEtk ⇤krvAmnA. g⌃tA no y‚Em(yƒA rAmr ⇤KA BjAk⇥Et,; 5; yo n, -vo arZo ˛AhE-tygA rAGvAE(mkA. aE`nmFl ⇤ proEht\ , RV 1.191.16: kq\Bk-tdb }vFt^ 4 RV 2.43.3: aAvd\-(v\ fkn ⇤ Bd˝mA vd , RV 3.62.18: g⇥ZAnA jmdE`nnA 5 RV 4.58.11: DAm⌃t ⇤ Evòm^ , RV 5.87.9: g⌃tA no y‚m^ 6 RV 6.75.19: yo n, -vo arZo yñ En˜o · · · , RV 7.104.25: ˛Et c#v Ev c#v 3 RV 1.1.1:

1 Emd\ ] em. ; Emd MS • m⌃/ {, ] em. ; m⌃/ ⇤, MS 2 b }úo ] em. ; b }ú u MS 2 pn\ v#y ⇤ ] em. ; pn v" ⇤ MS • vAEstAm^ ] conj. ; vAEstA MS 2 q, HyAt, v ⇤doƒ ] em. ; q HyAt\ v ⇤doƒ\ MS • (mk, ] em. ; (mk\ MS 3 m(-yr ⇤KA t EtygA ] em. ; m(sr ⇤KA y tFygA MS • (mk, ] em. ; (mk\ MSpc ; (mk, MSac • k-td ] em. ; k-tdA MS 4 jmd`nF ] em. ; yAud`nF MS 5 DAm⌃t ⇤ ] em. ; DAmÅ ⇤ MS • rEtk ⇤krvAmnA ] conj. ; rEVkokrcAmnA MS • Em(yƒA ] em. ; Em(y∆A MS 6 n, -vo ] em. ; n -to MS • ˛Ah ] em. ; ˛h MS • c#v ] em. ; c#vA MS

⇤ dvAm⇥tp—rA/ ⇤

104

íAdfmpVl,

˛Et c#v Ev c#v ⇤Et k⇥ Zr ⇤KA tto@vgA; 6; aA[25v ]`n ⇤ yAhFEt y(˛oƒA bàr ⇤KA t vtlA. yÅ ⇤ rAj C ⇤KA E/koZgA; 7; ⇥ t\ ˛Ah, kSkFr smAnF v, EfKA coƒA b }úr ⇤KA Edv\gtA. s⇤ rK\ t smEå£\ pzq\ m⌫XlA(mkm^; 8; t(s\yog ⇤ t ôF\ kyAâd˝pFW ⇤ mhF E-TtA. km◆ s\v(sr\ pî\ smd˝AçZvAE⌃vtm^; 9; -vE-t no mFEt m⌃/ ⇤Z gz(m⌃t\ t -TApy ⇤t^. /FEZ jAn ⇤Et m⌃/ ⇤Z vAmpAd\ Enyojy ⇤t^; 10; eq -tom ⇤Et m⌃/ ⇤Z d"pAd\ ˛Et§y ⇤t^. y“ATAEmEt m⌃/ ⇤Z vAmp"\ t s⌃Dy ⇤t^; 11; ˛Et§AsFEt sAm ⇤n s&yp"\ t d^ - aAlB ⇤t^. sA\v(srEmd\ [26r ]pFW\ p—AΩ\ gAzX\ Bv ⇤t^; 12; iEt ⇤ dvAm⇥t ⇤ p—rA/ ⇤ íAdfm, pVl,;

7 RV 8.103.14: aA`n ⇤ yAEh mz(sKA , RV 9.114.4: yÅ ⇤ rAj\ QC ⇥ t\ hEv, 8 RV 10.191.4: smAnF v, 10 RV 5.51.11: -vE-t no EmmFtAEònA Bg, · · · , RV 1.95.3:

/FEZ jAnA pEr Bq⌃(y-y 11 RV 7.24.5: eq -tomo mh ug }Ay vAh ⇤ , RV 8.85.7: y“ATA\ rAsB\ rT ⇤ 12 AB 3.8.3: ˛Et§AEs ˛Et§A\ gQC ˛Et§A\ mA gmy.

7 aA`n ⇤ ] em. ; aA`n MS • bà ] em. ; bOD MS • rAj C ⇥ t\ ˛Ah, kSkF ] em. ; rAj\ EQCt\ ˛AhvSkF MS 8 v, EfKA coƒA ] em. ; v EfKA\ to∆A MS •

mEå£\ ] em. ; mEå§\ MS 9 kyAd^ ] em. ; kyAt^ MS • smd˝Aç ] em. ; smd˝AdA MS 10 gz(m⌃t\ ] em. ; gz(mÅ\ MS • pAd\ ] em. ; pAd MS 11 -tom ⇤Et ] em. ; -tm ⇤Et MS • y“ATAEmEt ] conj. ; y“ATAEmEt MS 12 sA\v(sr ] em. ; s\v(sr MS • gAzX\ ] em. ; gzX\ MS

; /yodfmpVl, ; sh˙fFqyAcAyA\ Ev ZmEt◆ tto ⌃ys ⇤t^. i⌃d˝Ay gAv aAfFEt EfromEt◆ t s⌃Dy ⇤t^; 1; uíy\ t ⇤Et m⌃/ ⇤Z mkVmEt◆ t s⌃Dy ⇤t^. etA aqE⌃t m⌃/ ⇤Z llAVmEt◆ t s⌃Dy ⇤t^; 2; boD⌃mn ⇤Et m⌃/ ⇤Z B }vomEt◆ c s⌃Dy ⇤t^. a"‡mFEt m⌃/ ⇤Z aE"mEt◆ c s⌃Dy ⇤t^; 3; nAsdAsFEt m⌃/ ⇤Z nAEskAmEt ’ s⌃Dy ⇤t^. Bd˝\ kZ ⇤ ’ Et m⌃/ ⇤Z kZmEt◆ c s⌃Dy ⇤t^; 4; gAyE⌃t (v ⇤Et m⌃/ ⇤Z g⌫Xm[26v ]Et◆ c s⌃Dy ⇤t^. gAvEñä ⇤Et m⌃/ ⇤Z EcbkmEt◆ c s⌃Dy ⇤t^; 5; tm £vAm y\ EgEr(yo§mEt◆ c s⌃Dy ⇤t^. kAlF krAlF - m⌃/ ⇤Z dfnmEt◆ c s⌃Dy ⇤t^; 6; i⌃d˝Asom ⇤Et m⌃/ ⇤Z EjüAmEt◆ c s⌃Dy ⇤t^. 1 RV 10.90.1/VS 31.1: sh˙fFqA pzq, , RV 8.69.6: i⌃d˝Ay gAv aAEfrm^ 2 RV 1.50.10: uíy\ tms, pEr , RV 4.58.5: etA aqE⌃t 3 SV 1.140: boD⌃mnA id-t n, , RV 1.82.2: a"‡mFmd⌃t 4 RV 10.129.1: nAsdAsF‡o sdAsFt^ , RV 1.89.8: Bd˝\ kZ ⇤EB, f⇥ZyAm ⇤ dvA, 5 RV 1.10.1: gAyE⌃t (vA , RV 8.20.21: gAvEñäA sm⌃yv, 6 RV 8.95.6: tm £vAm y\ Egrm^ , Mun.d.akopanis.ad 1.2.4: kAlF krAlF c mnojvA 7 RV 7.104.1: i⌃d˝AsomA tpt\ r" uNjtm^ , RV 8.17.8 or 5.2.12: tEvg }Fv, 1 gAv aAfFEt ] conj. ; gAvo sF(y-yA MS 2 t ⇤Et ] em. ; tFEt MS • The second and last p¯ adas are unmetrical. 3 boD⌃mn ⇤Et ] conj. ; boDmAnFEt MS 4 c s⌃Dy ⇤t^ ] em. ; c s⌃Dy ⇤(y ⇤t^ MS 5 gAvEñä ⇤Et ] em. ; gAv\EvçoEt MS • The last p¯ ada is unmetrical. 6 EgEr(yo§ ] em. ; EgEr(yo£ MS • kAlF krAlF ] conj. ; kArArF klAlF MS • The first and last p¯adas are unmetrical. 7 i⌃d˝Asom ⇤Et ] em. ; c\d˝Asom ⇤Et MS • tEvg }Fv ⇤Et ] em. ; tEvg }Av ⇤Et MS

⇤ dvAm⇥tp—rA/ ⇤

106

/yodfmpVl,

tEvg }Fv ⇤Et m⌃/ ⇤Z g }FvAmEt◆ c s⌃Dy ⇤t^; 7; ih rtFEt m⌃/ ⇤Z v"omEt◆ c Ev⌃ys ⇤t^. s\á⌃d˝n ⇤Et m⌃/ ⇤Z -k⌃DmEt◆ c s⌃Dy ⇤t^; 8; aAf\ dDFEt m⌃/ ⇤Z a\smEt◆ c s⌃Dy ⇤t^. Gm ⇤v mD jWr ⇤ jWrmEt◆ c s⌃Dy ⇤t^; 9; apA\ p⇥§ ⇤Et m⌃/ ⇤Z p⇥§mEt◆ c s⌃Dy ⇤t^. a`n ⇤-tnrsFEt m⌃/ ⇤Z tnm[Et◆ c s⌃Dy ⇤t^] ; 10;

Folios after 26 are lost.

8 VS 8.51: ih rEt , RV 10.103.2/VS 17.34: s\á⌃d˝n ⇤nAEnEmq ⇤Z

dEDáA\ tm n £vAm, , RV 10.106.8: Gm ⇤v mD jWr ⇤ sn ⇤z p⇥§mEs yoEnr`n ⇤, 10 VS 5.1: a`n ⇤-tnrEs Ev Zv ⇤

9 RV 4.39.1: aAf\ 10 VS 11.29: apA\

8 ih rtFEt ] MSpc ; iEh rtFEt MSac 8 s\á⌃d˝n ⇤Et ] em. ; s\á\d˝m ⇤Et MS 9 a\s ] em. ; a\g MS • The last p¯ada is unmetrical. 10 p⇥§ ] em. ; p⇥£ MS • a`n ⇤-tn ] em. ; -tn MS • tn ] conj. ; -tn MS • unmetrical.

The third p¯ ada is

CRITICAL NOTES

These notes comprise miscellaneous observations, mainly those related to text-critical problems, and remarks on specific grammatical and lexical entities belonging to the Ai´sa register of Sanskrit. Sometimes I comment upon scribal errors, and I occasionally note what kinds of changes are introduced when one text borrows from the other. Metrical irregularities, too, are sometimes discussed.

¯ yambhuvapan ˜ cara ¯ tra Sva 1.2 v¯ırabhadr¯ adibhih . ] V¯ıryabhadra, as the manuscript reads, might be authorial. A teacher of this name appears in Somendra’s introduction (verse 12) to the Avad¯anakalpalat¯a. However, I have changed it to V¯ırabhadra, because I have not seen ´ the name of Siva’s gan.a spelled this way. 1.4a r.gyajuh am¯ atharvais ] The manuscript reads yaju instead . s¯ of yajuh., and in the light of the use of Ai´sa Sanskrit in the text it may not be necessary to emend it. Nevertheless, I have done so because metrically the regular word with a visarga fits better. 1.5 mahe´ svara ] Here I emend the manuscript reading mahe´svaram . to the vocative mahe´svara. In 1.6b, however, I keep mahe´svarah. as in the manuscript and interpret it literally, not as a name. 1.6c tvam asr.jo jagat sarvam . ] The original manuscript reading was tvayasr.jjanaya, which has been corrected to tvayasr.jjanayat sarvam . . My conjecture is economical and closer to the reading but an alternative conjecture could be tyay¯a sr.s.t.am . jagat sarvam .. 1.7c Here in the lacuna I supply sagun.am . nirgun.am . caiva which fits well logically. The Dev¯amr.tapa˜ ncar¯atra, too, contains a line speaking of the same idea but in a di↵erent wording covering two p¯adas: sagun.am . tu katham . proktam . nirgun.am . v¯a jagadguro.

110

Early Tantric Vais.n.avism

1.8ab Hiatus-bridgers are occasionally used in this text. But sometimes even within a p¯ada hiatus is left untreated. This halfverse has two di↵erent hiatus-bridgers: r and m. The first one disappears in the Dev¯amr.tapa˜ ncar¯atra (1.4), when the line is rewritten there. For a discussion on hiatus and hiatus-bridgers, see introduction, p. xxii. 1.8c, 10c, 16b kim . tu in all . nu ] The manuscript reads kim three places but still I have conjectured kim . nu, because this matches in meaning with kim v¯ a of the Dev¯ amr.tapa˜ ncar¯atra . parallel, and also because once in 1.11a and again in 1.19a (= Dev¯amr.tapa˜ ncar¯atra 1.14a) kim . nu is attested. 1.10a kim a ] The Dev¯amr.tapa˜ ncar¯atra, . nityam . ca anityam . v¯ too, has these words in the neuter. Therefore, I conjecture this reading. One can think of an alternative reading: kim . nityo v¯a anityo v¯a, keeping the original hiatus intact. The manuscript reads kim . n.ityo´s ca anityam v¯a. 1.12 kena dravyen.a of the Sv¯ayambhuvapa˜ ncar¯atra is changed into kasmin dravye in Dev¯amr.tapa˜ ncar¯atra 1.7c. The first expression goes well with casting: ‘with what material is the image cast?’ and the other with carving/sculpting: ‘on what material is the image carved/sculpted?’ 1.13 vais.n.avair janaih. of this passage is changed to laks.an.a¯nvitam in Dev¯amr.tapa˜ ncar¯atra 1.8d. 1.14ab, 15bd : The Dev¯amr.tapa˜ ncar¯atra corrects the case-endings and tries to rewrite these p¯adas to make them syntactically and logically smoother: makut.asya pram¯an.am . tu ´sira´s caiva katham . bhavet becomes makut.asya pram¯an.am . tu ´sirasa´s ca katham bhavet (Dev¯ a mr tapa˜ n car¯ a tra 1.8d) and katham . . . netram ath¯adi´set | gan.d.ayoh. kim . pram¯an.am . tu karn.am . caiva katham bhavet becomes netre caiva pit¯ a maha | gan d ayoh . .. . kim . pram¯an.am ncar¯atra . tu karn.ayo´s ca katham . bhavet (Dev¯amr.tapa˜ 1.10bd). 1.16-20 In this portion, there are too many Ai´sa forms. The redactor of the Dev¯amr.tapa˜ ncar¯atra has tried unsuccessfully to correct them.

Critical Notes

111

1.16a katham os.t.hau tu kartavyam . ] This reading is changed to os.t.hau kim ncar¯atra 1.11a, . tasya kartavyau in Dev¯amr.tapa˜ which is syntactically better but still problematic. 1.19b: vaktavyam ardanam ] The accusative ending is . tu jan¯ peculiar. Is it used in the sense of a dative? Note that its parallel in Dev¯amr.tapa˜ ncar¯atra 1.14b has the locative ending, therefore the accusative ending could possibly be emended. The locative would be acceptable assuming that it were used as a dative, and it is also conceivable that a final -e should have been misread as an anusv¯ara. Nevertheless, I keep the manuscript reading as it is, relying on the possibility that one can understand jan¯ardanam as an adjective to v¯ahanam. Such constructions are not rare in Ai´sa Sanskrit. 1.19cd The syntax of this half verse is odd, even though we understand the meaning. It is missing in the Dev¯amr.tapa˜ ncar¯atra version. 1.22a kim a naks.atrayogena ] This p¯ada is rewritten in . v¯ Dev¯amr.tapa˜ ncar¯atra 16c as grahanaks.atralagnam . ca. In this rewritten text, lagna takes the place of yoga. 1.23–24 These two verses are not found in the Dev¯amr.tapa˜ ncar¯atra. Does this mean that the topics enumerated in these verses are not included in that text? 1.25cf These two p¯adas, yath¯atathyam asam uhi tattvam . digdham . br¯ . sure´svara, are replaced in the Dev¯amr.tapa˜ ncar¯atra by etat sarvam . sam¯asena bhagavan vaktum arhasi. The reason for this is unclear. 2.1c, 4c brahman ] The manuscript reads brahma (1c) and brahm¯a (4c) inconsistently. As the context requires a vocative, in both places I adopt brahman. 2.2d Only the first syllable of this p¯ada, bh¯a, is visible in the manuscript. Therefore, the reading I suggest is merely a speculation.

112

Early Tantric Vais.n.avism

2.3d yath¯ akhy¯ atam sambhun¯ a ] ‘As it is explained by . tu ´ ´ Sambhu’: It is possible that this is a reference to some earlier text from which the Sv¯ayambhuvapa˜ ncar¯atra has borrowed. In the Dev¯amr.tapa˜ ncar¯atra this p¯ada has been revised to yad uktam . parames.t.hin¯a: ‘which is taught by Parames.t.hin.’ I take this as an honest statement about the source of the Dev¯amr.tapa˜ ncar¯atra. For a discussion on the mutual relation of the two texts, see introduction, p. xxvi. 2.4ab ekam¯ urtim an¯ıy¯ a brahmavis.n svaram ] . vij¯ . umahe´ The manuscript reads m¯ urti and mahe´svarah.. Because of the verb vij¯an¯ıy¯ah. I opt for accusative ending. In the Dev¯amr.tapa˜ ncar¯atra, however, the whole line has been revised in this way: ek¯a m¯ urtis trayo bh¯ag¯ah. brahmavis.n.umahe´svar¯ah. (2.3ab). 2.6d nigrah¯ anugrahe´ svarah . ] It is possible to take this word as a compound, but most probably this is a result of double sandhi between nigrah¯anugrahe and ¯ı´svarah.. Cf., e.g., Svacchandatantra 10.1128: nigrah¯anugrahe ratah.. 2.7cd This line contains t¯ani unnecessarily but not yasmin/yatra which is desired here. The Dev¯amr.tapa˜ ncar¯atra has the same reading (2.5cd). 2.8 Note that naran¯ar¯ıprakart¯arah. is treated as a nominative singular, possibly for metrical reasons. 2.9 In this verse, all accusative endings are justified, governed by sam¯a´sritah. in p¯ada b. But the last item is in the locative plural: dharma´s¯astres.u, which also can be justified. But this locative ending may have prompted the Dev¯amr.tapa˜ ncar¯atra (2.7) to put all items in this verse in the locative. 2.12–13 These verses are syntactically peculiar: words in the instrumental are alternated with words in the locative, and used with locative meaning. This is a Middle-Indic feature. Probably all items in the instrumental ending are to be treated as means for agnihotr¯adidharma. 2.12b Note that r is used as a hiatus-bridger.

Critical Notes

113

2.13ab somasam amah¯ ayaj˜ naih nca. sth¯ . ] The Sv¯ayambhuvapa˜ r¯atra takes somasam . sth¯a correctly as an a¯-stem but the Dev¯amr.tapa˜ ncar¯atra changes it into an a-stem: somasam . sthair mah¯ayaj˜ naih. (2.11). 2.13d sarvam¯ atre jan¯ ardanah . ] I am reading the text in this way with minimum emendation to the manuscript reading sarvam¯atre jan¯arddhanam. The expression sarvam¯atre sounds a little odd, but not inconceivable, in the sense of ‘in all and everything.’ This expression is attested but in a compound in Madhva’s Mah¯abh¯aratat¯atparyanirn.aya (17.239): vip¯at.it¯atmakaupin¯adi(!)sarvam¯atrakah.. It perhaps is a vernacular expression that has been Sanskritised. One could nonetheless consider reading sarvam¯atram . jan¯ardanam, treating jan¯ardanam as an adjective (see the note to 1.19b above for a similar case) and rendering the whole p¯ada with ‘everything here belongs to Vis.n.u.’ 2.14 In the Dev¯amr.tapa˜ ncar¯atra, aham . brahma is changed to kham . brahma. See Introduction, fn. 62. 2.15 Only the first word of this verse, vis.n.u , is preserved, but it is possible to read at least this verse from the Dev¯amr.tapa˜ ncar¯atra (2.13). The rest of the second chapter is lost in missing folios, and so are the third and fourth chapters. See introduction, pp. xxxv–xxxvi. 5.1 samantatah . . This . ] The manuscript reads samuntatam emendation is based on the Dev¯amr.tapa˜ ncar¯atra parallel (6.1). The same expression is found a little further down in 5.11 also. But it is possible that samunnatam was originally intended. 5.2,5 Note that nominative and instrumental endings have been used where locative endings would have been desirable. 5.7-8 The scheme is a bit more elaborate in the Dev¯amr.tapa˜ ncar¯atra (cf. 6.10-12) 5.9 t¯ırthes.u b¯ aluk¯ am apayet ] I take this as an ellip. tatra d¯ tical sentence and consider that it is possible to understand this

114

Early Tantric Vais.n.avism

sentence properly by supplying gr.h¯ıtv¯a, ‘having fetched,’ after b¯aluk¯am .. 5.12 nyagrodhodumbara´ s ] This is perhaps a case of double sandhi, but the possibility of the use of dumbara as a stem alternative to udumbara cannot be ruled out. 5.15 The manuscript reading agni y¯ah¯ıti is metrical whereas the emended reading, agna ¯a y¯ah¯ıti, which restores the right prat¯ıka, is unmetrical. So it is possible that the author himself had read the prat¯ıka wrongly and applied double sandhi to produce agn¯a y¯ah¯ıti. Moreover, the context shows that the mantra intended here is the first stanza of the S¯amaveda, and therefore, though closer to the manuscript reading, a¯gne y¯ahi marutsakh¯a (RV 8.103.14) is out of the question. 5.17c ¯ ajighra kala´ samantren . a ] Though unmetrical, I adopt this reading attested in the Dev¯amr.tapa˜ ncar¯atra version. To make the metre right, kala´sa- is pronounced rapidly in such a way that it sounds as if it had two syllables. This phenomenon is found not only in Ai´sa but also in epico-puranic Sanskrit. The manuscript reading a¯jighr¯a ka´sam . mantren.a preserves the prat¯ıka defectively, and is not metrically good either; a bad ma-vipul¯a. 5.24 vedy¯ a ved¯ıti ] The manuscript reads y¯avad¯a caiva. My conjecture comes from the parallel in the Dev¯amr.tapa˜ ncar¯atra (6.32), but I must say that the manuscript reading is remarkably di↵erent. 5.26 (vedi)-m-agratah . ] The manuscript reads (veda )m¯agratah.. My emendation is the closest to what the manuscript has; I reject the reading in the Dev¯amr.tapa˜ ncar¯atra parallel: vedimadhyatah. (6.34). 5.28b dh¯ upitv¯ a ] I take this as a metri causa shortening of dh¯ upayitv¯a: ‘after fumigating/making fragrant,’ and infer that flavouring and seasoning of curries and other delicacies is intended.

Critical Notes

115

5.29a lehyapey¯ adi ] The manuscript reads lehyayesy¯adi . One might think of emending it to lehyacos.y¯adi . But the manuscript reading is a good bit closer to the adopted reading. We can see another example of the scribe, apparently, adding an -s-, in the next entry in the critical apparatus: skandam¯ ula for skandam¯ ula ). Occasionally lehya and peya are found together in similar contexts and compounds. 6.3-4 Though the metre would allow me to change both words in 3b to the accusative, I retain them unchanged, because I see that the same problem of the use of the nominative ending instead of the accusative is present in 4a and the metre does not allow to correct that. 7.1 uta v¯ atena mantren . a ] The manuscript reading u v¯atena tu mantren.a is metrical but the prat¯ıka of the Vedic mantra is wrong. I suspect that in the course of transmission of the text tu may have crept in when the ta of uta got dropped accidentally. The Dev¯amr.tapa˜ ncar¯atra parallel (9.38), however, has a di↵erent mantra in this p¯ada. 7.3d This p¯ada is hypermetrical with one extra syllable, but it is hardly felt since it involves three short syllables in a row that can be pronounced very fast. See also the note to 5.17d above. 7.4 After 4ab, the text reads the twelve-syllable mantra, and states very economically, using just three words, that this is the mantra. In its parallel in the Dev¯amr.tapa˜ ncar¯atra, however, a half-verse is written for this purpose. 7.6 (gak¯ aram se ] This reconstruction is based on log. kan . t.hade)´ ical speculation, because there is no exact parallel in the Dev¯amr.tapa˜ ncar¯atra. The Sv¯ayambhuvapa˜ ncar¯atra and Dev¯amr.tapa˜ ncar¯atra di↵er in their scheme for installation of the twelve syllables of the mantra. See Introduction, pp. xl–xli (footnote 72) and lxxviii–lxxix. 7.10a mok¯ aram njam . n¯ılanibham . pu˜ . ] Because of one extra syllable, this p¯ada is unmetrical. Nevertheless, as noted earlier under 5.17a and 7.3d, it is possible to pronounce the short

116

Early Tantric Vais.n.avism

syllables in the middle fast with the e↵ect that the metre sounds alright. 7.11 punas tes.¯ am . , is in . ] The manuscript reading, karas tes.a¯m itself meaningful but does not fit the context. I have therefore adopted the reading of the Dev¯amr.tapa˜ ncar¯atra parallel. 7.12 vak¯ aram anam uttamam ] An odd syntactical con. dhy¯ struction. It sounds like vak¯aram . qualifies dhy¯anam. Perhaps, it would be a good idea to emend it to yield a compound instead: vak¯aradhy¯anam. 7.16 h¯ um sriye ] My emendation is based on SP 7.19: ´sriy¯a mad. ´ hye tu hr.dayam um ujayet. The Dev¯amr.tapa˜ nca. h¯ . k¯aren.a tu p¯ ´ r¯atra includes h¯ um in the mantra but Sr¯ı is not mentioned; see introduction, p. xxxiv. 7.33↵ The portion beginning from this point and continuing untill the end of the chapter is designated as vedapracodit¯a p¯ uj¯a in the text. This portion is not found in the extant Dev¯amr.tapa˜ ncar¯atra, but it is possible that this portion was presented there as a separate chapter towards the end of the text, which is now lost to us. 7.35d ´ slok¯ adhy¯ ayam sukriyam ] This is the best guess I . sa´ have come up with based on the manuscript reading, a´slok¯aya sapudriyam. 7.36a pitr.maitram . ] This is a diagnostic conjecture; the manuscript reads putramettra´s. I cannot find a Vedic mantra or hymn associated with putra. 7.36c This p¯ada listing two s¯amans is unmetrical, but as in 5.17a, 7.3d, and 7.10c it is possible to read the short syllables standing in a row rapidly so that the p¯ada sounds metrical. 7.38 skambhas¯ uktam uktam ., . ] The manuscript reads skandas¯ and this might be authorial. The same name is spelled kambas¯ ukta in the As..t¯ada´savidh¯ana manuscript; see, apparatus to As..t¯ada´savidh¯ana verse 36. This spelling may be the result of Middle-Indic influence. In many modern Indian languages, including Tamil, we find cognate words derived from skambha,

Critical Notes

117

for example, Nep. khamb¯a, Hindi khambe, and Tamil kampa, which stand very close to the spelling in our manuscript. For more words derived from skambha, see Turner 1999, s.v. skambha. 7.40c The manuscript reads dv¯ayo. In view of the possibility of pr.s.t.ham¯atr¯as somewhere in the transmission, this is pointing towards dvayau, a form which perhaps grates on the ear somewhat, but may be acceptable in our text; some Tantric texts certainly use it. 7.45c I am unable to identify with certainty which Vedic mantra is being indicated as the jyes.t.hamantra. But I guess that jyes.t.hamantra is intended to be the same as jyes.t.has¯aman (SV 1.31 = RV 6.7.1 or SV 1.33 = RV 1.115.1). For alternative identifications of the jyes.t.has¯aman, see Olivelle 2000: 712. 7.48[sarvam a ]heti ] This reconstruction is based merely on . sv¯ my logical speculation. 7.49 dadhi ca saktavah . ] The manuscript reads dadhi sayuktava. My emendation is based on palaeographical resemblance and the definition of bh¯ utak¯ ura found in the Prayogama˜ njar¯ı (10:15): p¯ urn.am ni´ s ¯ a y¯ a h palalam sal¯ a jam bh¯ u tagrahebhyo . . . . dadhi saktumi´sram | dady¯ad dvit¯ıye ’hani bh¯ utak¯ uram . balim . samantram krama´ s o di´ s a ¯ su k . 8.2 deve´ sa ] The manuscript reading vede ca, though meaningful, does not fit the context. 8.12 daityagandharva-r-¯ apsa(r¯ ah . ) ] This metri causa formation uses a hiatus-bridger and shortens the final member of the compound apsaras to apsara. 8.21c This p¯ada is unmetrical but as in many other cases in this text, it is possible to read it as if it were metrically normal by uttering the short syllables in a row rapidly. 8.26c deva savita mantren . a ] This p¯ada is unmetrical, though it has eight syllables as needed.

118

Early Tantric Vais.n.avism

8.28a This verse exhibits elliptical and bad syntactical style. Moreover, the third p¯ada of this verse reappears as the third p¯ada of verse 8.30. It perhaps results from a scribal error. 8.41 We are not told which the five Vedic mantras of Varun.a are. 8.46a br¯ ahman ah . apurogam¯ . sarve ] I am treating this as the first p¯ada of a new verse even though it is unmetrical. ¯ da´ ¯ na As.t.a savidha 4 ak¯ alebhyo ’tha k¯ alebhyah alak¯ alatarebhya´ s ca ] . k¯ Although the manuscript readings k¯alebhyo and k¯alak¯al¯antarebhya´s ca are somehow meaningful, I emend them ´ following the pattern in the P¯a´supata/Saiva Aghora mantra which has been imitated in this mantra. As in this mantra, ca is present at the end of the line in the Maitr¯ayan.¯ı Sam . hit¯a version of the Aghora mantra, but not in the Taittir¯ıya ¯ . yaka version; see Bisschop 2006: 11. Aran 6 bodhaka sarvavidy¯ an¯ am . ] The manuscript reading rodhaka sarvavidy¯an¯am , though meaningful, does not logically fit the . context. 13b devadeve suyojayet ] The form suyojayet, though grammatically fine, is not common. One could consider reading devadeves.u yojayet and interpreting the locative plural as honorific, but that, too, would be unusual in this context. The same form appears again later in this text: sth¯ane sth¯ane suyojayet (verse 73b). There, too, one could consider reading sth¯anesth¯anes.u yojayet, which is not impossible. Nevertheless, I find it more reasonable to read suyojayet as one word at both places. 22c Following the sequence of the mantra-syllables to be deposited in the body, we can say that t.hak¯aram . must feature in the lacuna here. The part of the hand where this syllable is placed could be the fingers. 29 (om a sam aya namah aya) . ¯ . kars.an .¯ . | om . am . pradyumn¯ namah ] As in the previous line, a shorter text, (om a ¯ sam . .. kars.an.a¯ya) namah., would have been enough to fill the physical

Critical Notes

119

gap. But I have assumed a larger omission (some text might after all have been written in the margin). Logically this is needed to make the set of Vy¯ uha deities complete. This suppletion is furhtermore supported by the evidence of the vis.n.upratis.t.h¯a chapter (cf. 4.6.35–38) of the Soma´sambhupaddhati, which, as the text itself states, is based on the Sv¯ayambhuvapa˜ ncar¯atra. This passage speaks of a pa˜ ncam¯ urtiny¯asa that brings the four Vy¯ uha deities and N¯ara¯ayan.a together. The same passage instructs the worshipper to make the mantras of these five deities by combining certain seed-syllables with their names. Although here too the text is not free of errors, and the seedsyllables of Sam . kars.an.a and Pradyumna have been corrupted, we can repair the text (cf. Brunner 1998: 306) and know that om . a, om . a¯, om . am . , om . ah., and om . om . are the seed-syllables of these deities. In this e↵ort, we can have some support from a passage (1.32.1–7) from the Garud.apur¯an.a that prescribes the worship of the same five deities with the same mantras. There too, the seed-syllables are each read with an anusv¯ara and the second seed-syllable of Aniruddha is missing. Nevertheless, we can put all these di↵erent versions together and arrive at the solution suggested above. The seed-syllables of the first four are known also in the tradition of the S¯atvatasam . hit¯a (cf. e.g. 4.36–46). Ala´singa ˙ Bhat.t.a clearly spells these out at the end of his commentary on the fourth chapter of the text. In the available edition of the text, a and a¯ are each read with an anusv¯ara, which I think is unnecessary, because, if we add an anusv¯ara for each the seed-syllables of the first and third deities will be the same. 32 The manuscript, by mistake, reads daks.in.e twice, but I have emended it in the second occurrence to v¯amake. 35 g¯ argibr¯ ahman . am ] The manuscript reads g¯aggatihman.am, but I have emended this keeping in view the fact that the Man.d.alabr¯ahman.a is mentioned in verse 30. ´ 47 Note the claim of this verse: the vedany¯asa is found in Saiva, Vais.n.ava, Saura, and Bh¯agavata traditions.

120

Early Tantric Vais.n.avism

50b This p¯ada is metrically very bad: it has ten syllables instead of eight and does not fit the metre. 72a dvisaptakam . ca paks.es.u ] This conjecture stays close to the manuscript reading: dvisaptakarmapaks.is.u, but why ‘twice seven times’ and why paks.es.u is here is difficult to answer. 72b This p¯ada has too many extra syllables. 73b sth¯ ane sth¯ ane suyojayet ]

See the note to 13b above.

76, 93 k¯ an at k¯ an a’ti ] We have enough examples of metri .d .¯ .d .¯ causa dropping of the final consonant or a full syllable, but here it seems that the final consonant of the preceding word and the initial vowel of the following are both dropped. The same is found in verse 92: a¯ no bhadr¯a’ti, and also in verse 94: catv¯ari ´sr.ng¯ ˙ a’ti. 77b nimardanam ] As an alternative, the manuscript reading nimatthanam . could perhaps be corrected to nimanthanam . and interpreted in the same sense of massaging or rubbing of the body. But the combination of ni-math is not attested in this sense, which is why I have emended it to a word derived from ni-mr.d, ‘to rub o↵.’ 81 The Garud.apur¯an.a passage cited in the apparatus is not exactly parallel to our passage but many of the herbs mentioned in our passage are also mentioned there. The reconstruction in p¯ada a is uncertain. 84–85, 96ab, 99 These lines are hardly metrical. I have therefore treated them as prose. Verse 99 contains an irregular sandhi: a¯c¯arya agni instead of a¯c¯aryo ’gni . 88cd The mantra-prat¯ıka is too long; agnir devat¯a would have been enough for the indication of the desired mantra here. 89d gr.hn . ed ] This is a metrically conditioned Ai´sa formation. 105ab For metrical reasons, nominative endings are used here instead of accusative ones.

Critical Notes

121

107 gadam . ] The manuscript reads garud.a(!) but this is unmetrical and does not fit the context. The text is describing a retinue of the members, actually the warrior heroes, of the Vr.s.n.i clan. So, I have adopted the name of Gada, which is close to the manuscript reading. This emendation also corrects the metre. The names of Vr.s.n.i heroes, including Gada, Ni´sat.ha and others, can be found in many places in the Mah¯abh¯arata, for example in 12.130.28. However, I have not found these eight as a group anywhere else. See also Introduction, p. xlviii. 111b n¯ am¯ an¯ am . ] tion.

This is a metrically conditioned Ai´sa forma-

113a Though the name of Hr.s.¯ıke´sa is randomly chosen to fill in the lacuna, it is one of the best-known names of Vis.n.u and it fits metrically. The set of ten names of Vis.n.u with which we are concerned here is not known to me from any other source. 117a n¯ agav¯ım . gam . ] I have no idea what this cruxed word means, but, as the context is suggesting, this should be a mineral. ¯ mr.tapan ˜ cara ¯ tra Deva 1.1 The prologue that situates the interlocutors and relates the commencement of their dialogue is abbreviated here, compressed into one verse, whereas in the Sv¯ayambhuvapa˜ ncar¯atra it covers six and a half verses. brahm¯ a’ ] It appears that brahm¯an.am . is shortened to brahm¯a’ for the sake of the metre. 1.4b The redactor of the Dev¯amr.tapa˜ ncar¯atra generally alters the text to remove hiatus-bridgers, but this time he apparently forgot to do so. 1.4d vyaktar¯ upam nca. tu kim . bhavet ] The Sv¯ayambhuvapa˜ r¯atra reading vyaktar¯ up¯ı tu kim bhavet (1.9d) is better than its . revision here. Perhaps the secondary reading is simply a result of corruption.

122

Early Tantric Vais.n.avism

1.5a kim a ] The text still retains neuter . nu nityam anityam . v¯ endings as in the Sv¯ayambhuvapa˜ ncar¯atra (1.10a: kim . nityam . ca anityam v¯ a ). The reading however is slightly improved here . as the hiatus inside this p¯ada has been removed. 1.6d lokan¯ atha jagadguro ] In the Sv¯ayambhuvapa˜ ncar¯atra parallel (1.11d), lokan¯atho jan¯ardanah. refers to Vis.nu . , but here both words are in the vocative case and they refer to Brahman, the narrator of the new text. 2.1-2 These two verses replace three opening verses of the second chapter of the Sv¯ayambhuvapa˜ ncar¯atra; the rest of the Sv¯ayambhuvapa˜ ncar¯atra (2.4–15a) until it breaks o↵ is found nearly verbatim in the Dev¯amr.tapa˜ ncar¯atra as 2.3–13a. 2.5ab This line is almost hanging here, as the first half of this verse has been moved to Chapter One. Here is the original verse from the Sv¯ayambhuvapa˜ ncar¯atra(2.6): sarvair vis.n.uh. paro devah. sarva´sa¯stres.u pat.hyate | a¯dh¯arah. sarvabh¯ ut¯an¯am . nigrah¯anugrahe´svarah. k 2.24 v¯ ayu-r-ambarayor nityam a caikam . yath¯ . ] The same simile of the unity of wind and sky/ether is used in the As..t¯ada´savidh¯ana (Verse 69). Furthermore, the same phrase v¯ayu-r-ambarayor yath¯a is used there, whreas it does not fit well syntactically. Here yath¯a caikam . is used in the sense of yath¯a caikyam . . 2.25 candrau dvau dar´ sanam . ] The manuscript reads candra dvau dar´sano, and my conjecture is based closely on that reading; dvicandradar´sanam . would be a grammatically and metrically correct expression conveying the intended sense. 3.2c This p¯ada is unmetrical but would sound normal if its three consecutive short syllables were uttered swiftly. 3.4d The choice of nirdis..t¯ a’ in place of nirdis.t.a¯ni is an instance of the metri causa elimination of the final syllable. 3.5d ekadvitricatuh . ] This word as a compound is odd, it could perhaps be emended to ekam . dvis tri´s catuh. but this too is equally odd.

Critical Notes

123

3.11a tryangul¯ ˙ a tu mat¯ a siddhi’ ] tryangule ˙ tu mat¯a siddhih. is a possible emendation, but it appears that this construction with nominative tryangul¯ ˙ a is authorial. For, similar constructions are seen in the preceding and following verses. We perhaps have to interpret siddhi’ in the sense of siddhid¯a. Verse 3.12c has an exactly parallel construction. 3.14d mah¯ aris.ikam ] The context reveals that this should be a tree but I am unable to identify it. I have kept the manuscript’s orthography. 3.17c tasy¯ ang¯ ˙ ani ] This looks like a simple case of double sandhi, but it is equally possible that this results from a gender confusion. 3.18a mastakasya ] Though meaningful, the manuscript reading, saptakasya, does not fit the context. 3.29 pratim¯ ake´ savam . ] Apparently this is an inverted compound: pratim¯ake´sava in the sense of ke´savapratim¯a. 4.3ab, 4e, 6e, 8d Apparently a hiatus inside a p¯ada or between two of them is not a problem for the composer/transmitter of the Dev¯amr.tapa˜ ncar¯atra, and these serve as a few examples. But in verse 6.7 (see below), in an exceptional case, the Dev¯amr.tapa˜ ncar¯atra contains a hiatus-bridger when its predecessor has managed to avoid it. 4.13cd sa yukto n¯ atra sam sayah .´ . ] The manuscript reads yukte in place of yukto. An alternative emendation could be yunkte. ˙ The whole statement is elliptical and what is intended is not clear. If we read it in the light of Sv¯ayambhuvapa˜ ncar¯atra 2.3, which contrasts yoga with moks.a, the meaning will be that anybody who installs an image of Vis.n.u will acquire riches. Otherwise, we have to say that such a person will be united with Vis.n.u. 5.3a naks.atr¯ a´ s ca ] The natural neuter plural form, naks.atr¯an.i, is used in the As..t¯ada´savidh¯ana parallel. 5.5 bhruvor madhye ´ sivam . ] In the As..t¯ada´savidh¯ana parallel ´ ´ shown in the apparatus, Siva’s place is given to the Satarudras.

124

Early Tantric Vais.n.avism

5.13cd This line of the Dev¯amr.tapa˜ ncar¯atra obviously indicates that the following sixth chapter is derived from the Sv¯ayambhuvapa˜ ncar¯atra. 6.3 tr.t¯ıy¯a and da´sam¯ı are not included in the list found in the Sv¯ayambhuvapa˜ ncar¯atra parallel. 6.7d The Dev¯amr.tapa˜ ncar¯atra passage uses a hiatus-bridger and reads sam hr tya karma-m-¯ arabhet, whereas the parallel in its . . predecessor does not contain a hiatus: hr.tv¯a karma sam¯arabhet. 6.8–9, 13–14 These verses are not found in the Sv¯ayambhuvapa˜ ncar¯atra. This component is new and accommodates the grahacakray¯aga. 6.15cd–16, 18 Although these verses are not found in the Sv¯ayambhuvapa˜ ncar¯atra, they appear to fit the context there. It is therefore possible that these once formed part of the Sv¯ayambhuvapa˜ ncar¯atra. 6.29d p¯ıt¯ am ncar¯atra parallel assigns a . ] The Sv¯ayambhuvapa˜ black flag to the north-west, but here it is yellow. 7.15 naivedyaus.adhayah . ] This is clearly a case of double sandhi. There is another example of the same a little further down in 7.18a: sruc¯agraik¯ada´sa¯ rudr¯ah.. 7.19 hunate ] Even in Tantric texts this irregular formation is rarely seen. vis.n al¯ agnirudr¯ akhyam . um . k¯ . ] I think that this identification of Vis.n.u as K¯al¯agnirudra is unique. 7.21 aviditv¯ a im¯ am . ] tus within a p¯ada.

Note that this is another example of hia-

8.2b, 4b, 5b, 7b, 12d, 14b, 18b, 19b, 21b, 22b An interesting verbal form pratis.t.hayet, obviously a denominative optative derived from the noun pratis.t.h¯a, is found in these verses. No other occurrences of this irregular form are found in Pa˜car¯atra ´ texts, but in Saiva texts it is fairly common, and it can also be found occasionally elsewhere. In any case, the regular form pratis.t.h¯apayet would have made these verses in our text unmetrical.

Critical Notes

125

8.3d Even after checking the original manuscript I am unable to read this p¯ada properly; the surface of the folio has been badly damaged by rubbing. 8.11a This p¯ada containing a prat¯ıka is unmetrical. 8.13a I am unable to emend the cruxed part and find out which Vedic mantra is meant. The last word mantren.a was added later in the margin, and even without it the p¯ada is complete. 8.15, 23 In both of these places anusv¯aras are dropped for metrical reasons. 8.17b svardi´ s¯ am . ] The manuscript reads ks.madi´s¯a and I have conjectured svardi´sa¯m . . The Vedic mantra prescribed in this context suggests that this line is concerned with the zenith. This is what we logically expect after ‘the directions’ and before ‘the nether worlds’ in the downward direction. Though bh¯ umip¯at¯alak¯an, meaning the nether worlds, in the next line (17d) is a conjecture, the Vedic mantra prescribed there, adh¯a no vi´svasaubhaga, is definitely meant for the downward direction. It should also be noted that the next item in the list is antariks.a, the intermediate region. 8.22a yaj˜ nena vardhata jeti ] Although it is unmetrical, I have conjectured this reading on the basis of the reading of the Vedic mantra (RV 2.2.1) intended here. We have a number of unmetrical p¯adas that are unmetrical because they incorporate a prat¯ıka. The manuscript reading yaj˜ nena vardhatastveti is metrical but incorrect. I would have accepted the manuscript reading if it had contained a space-fillers such as ca, v¯a, hi, or the like. 9.9c This p¯ada containing a prat¯ıka is unmetrical. 9.12 satyaparya’tam ] The correct form behind this reading is satyaparyantam but for metrical reasons the -n- has been dropped. 9.15 dady¯ a’bhimantritam ] Here the final consonant of the preceding word as well as the initial vowel of the following word

126

Early Tantric Vais.n.avism

are dropped. This phenomenon can be seen in As..t¯ada´savidh¯ana 76 and 93 also. 9.34c As in DP 8.22a, I have rejected a metrical reading of the manuscript in order to have the cited prat¯ıka in its correct form. 10.1–2 A confusion of syntax is evident. The first verse starts with a description of the caretaker temple priests in the accusative, but the second verse ends with the same in the nominative. In the following three verses, too, the use of active and passive voices is confused. 10.11 An exclusively Vedic usage, hanati, surfaces here. This verse also uses its optative hanet, and the middle indicative hanate. 11.22–24 These verses equating di↵erent mantric entities with various divinities or with other mantric entities are found neither in the Sv¯ayambhuvapa˜ ncar¯atra nor in the As..t¯ada´savidh¯ana. 11.40 ´ sr.ngi ˙ ] I keep this reading as found in the manuscript, because, like ´sa¯rnga, ˙ it can be taken as a synonym of dhanus. 12.5 atikekarav¯ aman¯ a ] The constituted text is a conjecture based on the manuscript reading: at.ikokarac¯aman¯a. There is no doubt that V¯amana should feature here. It is also certain that this half-verse, like the preceding and following ones, is speaking about a line or lines of a particular shape. Therefore, I emend the manuscript reading to atikekarav¯aman¯a, thinking that atikekara can somehow be interpreted as ‘very slanted.’ 12.12b tu d-¯ alabhet ] One might think of emending this reading to tad¯alabhet. But since tu is used in the same situation in the preceding line, I retain tu and assume the following -d- to be a rare hiatus-bridger. 13.2b, 2d, 4b, 5d, 6d, 9d, 10c All these p¯adas are unmetrical in one way or the other, because they must all contrive to fit in a constraint to fit a fixed entity, either the name of a body-part or a prat¯ıka. 13.4bc In p¯ada b an anusv¯ara is dropped for metrical reasons, and in p¯ada c a double sandhi is practiced for the same reason.

WORKS CONSULTED

Abbreviations AVi As.t.a ¯da´savidh¯ ana AVP Atharvaveda in Paippal¯ ada Recension ´ ´ AVS Atharvaveda in Saunaka Recension BY Brahmay¯ amala DP Dev¯ amr.tapa˜ ncar¯ atra NAK National Archives of Kathmandu NGMPP Nepal-German Manuscript Preservation Project RV R . gveda SP Sv¯ ayambhuvapa˜ ncar¯ atra SV S¯ amaveda ´ Satapatha-br¯ ´ SB ahman.a ow. overwritten IFI Institut Fran¸cais d’Indologie (misnomer used in old publications) IFP Institut Fran¸cais de Pondich´ery/French Institute of Pondicherry VS V¯ ajasaneyi-sam a . hit¯

Primary Material including Manuscripts ¯ n.a. Agni Puranam By Shrimanmaharshi Vedavyas. Gurumandal Series No. agnipura XVII. Calcutta: 5, Clive Row, 1957. ¯ da-sam ¯ . Paippal¯ atharvaveda paippala ada-sam a of the Atharvaveda. Vol. hita . hit¯ ume one, consisting of the first fifteen K¯ an.d.as, ed. Dipak Bhattacharya. Calcutta: Asiatic Society of Bengal, 1997. ´ ¯ in Saunaka atharvavedasam recension, ed. R. Roth and W.D. Whitney. Berlin . hita 1856. ¯ nakalpalata ¯ of Ks.emendra. Volume I, ed. P. Vaidya. Darbhanga: Mithila Inavada stitute, 1959.

128

Early Tantric Vais.n.avism

¯ gamapra ¯ ma ¯ n.ya of Y¯ a amun¯ ac¯ arya, ed. M. Narasimhachary. Gaekwad’s Oriental Series No. 160. Baroda: Oriental Institute, 1976. ¯ with the Commentary (bha ¯ s.ya) of S¯ r.gvedasam ayan.a. ed. N. S. Sontakke and . hita C. G. Kashikar (vols. 1-3 editors not mentioned). Poona: Vaidika Sam . ´sodhana Man.d.ala, 1933, 1936, 1941, 1946 and 1951. kiran.atantra with the commentary of Bhat.t.a R¯ amakan.t.ha. Bhat.t.a R¯ amakan.t.has Commentary on the Kiran.atantra. Volume I: chapters 1–6. Critical edition and annotated translation, ed. Dominic Goodall. Publications du d´epartement d’indologie 86.1. Pondicherry: IFP/EFEO, 1998. ¯ matatantra. The Kubjik¯ kubjika amatatantra. Kul¯ alik¯ amn¯ aya version. Critical Edition, ed. T. Goudriaan, and J. A. Schoterman. Orientalia Rheno-traiectina 30. Leiden: Brill, 1988. ¯ n.a. Garud.apur¯ garud.apura an.am mahars.ivedavy¯ asapran.¯ıtam. editor and publisher, J¯ıv¯ ananda Vidy¯ as¯ agara. Calcutta: Sarasvat¯ı Press, 1890. ¯ khyasam ¯ . Critically Edited with an introduction in Sanskrit, Indices etc. by jaya . hita Embar Krishnamacharya. With a Foreword in English by B. Bhattacharyya. Gaekwad’s Oriental Series No. 54. Baroda: Oriental Institute, 1931. —–. NAK MS No. 1–49, NGMPP Reel No. B 29/3. Palm-leaf manuscript. Newari Script, copied A.D. 1294. jayottaratantra. NAK MS 4/82, NGMPP Reel No. A 1306/24. Palm-leaf manuscript, Newari script, copied A.D. 1383. The verse and chapter numeration used in the text and footnotes is that of my own edition in progress. ¯ ran.yaka with the commentary (bha ¯ s.ya) of Bhat.t.abh¯ taittir¯iya a askarami´sra, ed. A. Mahadeva Sastri and K. Rangacarya with an introduction by T. N. Dharm¯ adhik¯ ari. Delhi: Motilal Banarsidass, 1985 (reprint of Mysore Government Oriental Library Series Nos. 26, 27, and 29, published 1900–2). ¯ saka ¯ rika ¯ . Institut Fran¸cais de Pondich´ery, Transcript 17A and 150. Paper tranni´ s va scripts in Devan¯ agar¯ı. E-text prepared principally by S.A.S Sarma and Nibedita Rout. ¯ satattvasam ¯ . NAK MS No. 1–227, NGMPP Reel No. A 41/14. Palm–leaf, ni´ s va . hita Folios 117, early Nepalese (Kut.il¯ a) script. There are two apographs available, both in Devanagari and on paper: NAK MS 5–2401, NGMPP Reel No. A 159/18, and Welcome Institute for the History of medicine, London, Sanskrit MS I.33. The verse and chapter numeration used in the text and footnotes is that of Goodall’s edition in progress. netratantra with the commentary (uddyota) by Ks.emar¯ aja, ed. by Madhus¯ udan Kaul. KSTS 46, 59. Bombay, 1926, 1939. ¯´ ¯ tra with the commentary (pan ˜ ca ¯ rthabha ¯ s.ya) of Kaun.d.inya, ed. Ananpa supatasu thakrishna Sastri. Trivandrum Sanskrit Series No. CXLIII. Trivandrum: The Oriental Manuscript Library of the University of Travancore, 1940. ¯ rame´ paus.karapa svara. Cambridge University Library MS Add. 1049. Palm-leaf, early Nepalese (Kut.il¯ a) script, copied A.D. 819.

Works Consulted

129

¯ laks.an.asa ¯ rasamuccaya of Vairocana, ed. Damodar Sharma and Babukrpratis.t.ha ishna Sarma. Two volumes. Kathmandu: Nepal National Archives, 1966 and 1968. ˜ ja¯ı of Ravi, ed. C. K. Raman Nambiar with K. Achyutha Pothuval. prayogaman Trippunithura: Sanskrit College, 1953–54. ¯ of Var¯ br.hatsam ahamihira, ed. H. Kern. Calcutta: Baptist mission press, 1865. . hita ¯ ran.yakopanis.ad. See Limaye & Vadekar. 1958. br.hada ¯ n.a. Sanskrit Indices and Text of the Brahmapur¯ brahmapura an.a, ed. Peter Schreiner and Renate S¨ ohnen. Wiesbaden: Otto Harrassowitz, 1987. ¯ mala (Picumata). NAK MS 3–370, NGMPP Reel No. A 42/6. Palm-leaf, brahmaya Folios 358, Early Newari script, copied 1052 A.D. The verse and Chapter numeration used in footnotes is that of Shaman Hatley’s transcription. ¯ . The Bhagavadg¯ıt¯ bhagavadg¯ıta a being reprint of relevant parts of Bh¯ıs.maparvan from Bhandarkar Oriental Research Institutes’s Edition of the Mah¯ abh¯ arata, critically ed. S. K. Belvalkar. Poona: Bhandarkar Oriental Research Institute, 1945. ˆ ˜ ju´ ¯ lakalpa. Aryamanjusrˆ man sriyamu ım¯ ulakalpa, ed. T. Gan.apati Sˆ astr¯ı, 2 volumes, Trivandrum Sanskrit Series, 70 and 76. Trivandrum: Government of Travancore, 1920 and 1922. ˙ ¯ rame´ ¯ gama, kriya ¯ pa ¯ da, carya ¯ pa ¯ da and yogapa ¯ da, with the matangap a svara ¯ pa ¯ da 11:12b, ed. N. R. commentary (-vr.tti) of Bhat.t.a R¯ amakan.t.ha up to kriya Bhatt. Publications de l’IFI No. 65. Pondicherry: IFI, 1982. ¯ ¯ n.a, Anand¯ matsyapura a´srama Sam avali 54. editor not mentioned. . skr.ta Granth¯ ¯ Poona: Anand¯ a´srama Press, 1907. ¯ s.ya of Medh¯ manusmr.ti with the manubha atithi, ed. Gang¯ an¯ atha Jh¯ a. Bibliotheca Indica No. 256. 2 Vols. Calcutta: Royal Asiatic Society of Bengal, 1932 and 1939. ¯ bha ¯ rata, ed. V. Sukthankar, with the cooperation of S. K. Belvalkar, A. B. maha Gajendragadkar, V. Kane, R. D. Karmarkar, P. L. Vaidya, S. Winternitz, R. Zimmerman, and other scholars and illustrated by Shrimant Balasaheb Pant Pratinidhi. (Since 1943 ed. S. Belvalkar). 19 Vols. Poona: Bhandarkar Oriental Research Institute, 1927–1959. ¯ ¯ bha ¯ ratata ¯ tparyanirn.aya of Anandat¯ maha ırtha with the Commentary of Jan¯ ardana Bhat.t.a, ed. Uddhava Ain¯ apure and V¯ asudeva Ain¯ apure. Gan.apata Kr.s.n.a ¯j¯ı Press, 1891. ¯ ma ¯ yu ¯ r¯ı vidya ¯ ra ¯ jn ˜¯ı. ed. Shuyo Takubo. Tokyo: Sankibo Book Press, 1972. I maha have used an text input by Ryugen Tanemura in 2003 who has corrected typographical errors silently. ¯ gama [= mr.gendratantra], kriya ¯ pa ¯ da and incomplete carya ¯ pa ¯ da mr.gendra with the commentary (-vr.tti) of Bhat.t.a N¯ ar¯ ayan.akan.t.ha, ed. N. R. Bhatt. Publications de l’IFI No. 23. Pondicherry: IFI, 1962. ¯ urottara . NAK MS No. 4–1622, NGMPP Reel No. B 27/18. Palm-leaf mohacu manuscript. Newari Script.

130

Early Tantric Vais.n.avism

¯ jn ˜ avalkyasmr.ti of Yog¯ı´svara Y¯ ya aj˜ navalkya with the commentary (v¯ıramitrodaya) ´astr¯ı Khiste ¯ ks.ara ¯ ) of Vij˜ of Mitra Mi´sra and (mita na ¯ne´svara, ed. N¯ ar¯ ayan.a S¯ ´ and Jagann¯ atha S¯ astr¯ı Ho´singa. ˙ Chowkhambha Sanskrit Series No. 62. Benares: Chowkhambha Sanskrit Series Office, 1929. ´akh¯ ¯ jasaneyi-sam ¯ in the M¯ va adhyandina and the K¯ an.va-S¯ a with the Commen. hita tary of Mah¯ıdhara. ed. Albrecht Weber. Berlin 1852 / Reprint: Varanasi 1972. ¯ sudevakalpa. Kaiser Library, Kathmandu, MS no. 420, NGMPP Reel No. C 44/6. va Palm-leaf, Newari script, copied A.D. 1252. The verse and chapter numeration used in the text and footnotes is that of my own edition in progress. ¯ n.a. The Critical Edition of the Vis.n.upur¯ vis.n.upura an.a. Volume I: 1–3 am . ´sas. ed. M. M. Pathak. Vadodara: Oriental Institute, 1997. ´ vis.n.udharmottara. First Khan.d.a. ed. Madhus¯ udana Sarman and M¯ adhava Pras¯ ada ´ Sarman, Bombay: Venkat ˙ . e´svara Press, 1912–1913 (Vikrama Year 1969). vis.n.udharmottara. Vis.n.udharmottarapur¯ an.a Critically edited with Introduction, Notes, etc. Third Khan.d.a (Text, Critical Notes etc.), ed. Priyabala Shah. Gaekwads Oriental Series 130. Vadodara: Oriental Institute, 1994. ´ ¯´ v¯ın.a sikhatantra. The V¯ın.a ¯´sikhatantra: A Saiva Tantra of the Left Current, ed. and transl. Teun Goudriaan. Delhi: Motilal Banarsidass, 1985. ´ ¯ hman.a. The C satapatha-bra ¸ atapatha-Brˆ ahman.a in the Mˆ adhyandina-C ¸a ˆkhˆ a with Extracts from the Commentaries of Sˆ ayan.a, Harisvˆ amin, and Dvivedaganga. ˙ ed. Albrecht Weber. Reprint of the Edition: Berlin 1855. Leipzig: Otto Harrassowitz, 1924. ´ ¯m ¯ yanagr.hyasu ¯ tra, ed. S. R. Sehgal. Second revised edition. Delhi: Sri Satsa . kha guru, 1987. ¯ rad¯ıpaka. sam sam arad¯ıpakasya prathamo bh¯ agah.. parvat¯ıyanity¯ anandena . ska . sk¯ viracitah. ´sodhita´s ca. Benares: Vidya Vilas Press, 1914. ¯ rasam ¯ of the P¯ sanatkua an ˜car¯ atra. ed. V. Krishnamacharya. Madras: The Adyar . hita Library and Research Centre, 1969. ¯ tvatasam ¯ with Ala´singas sa ˙ Commentary. ed. Vrajavallabha Dviveda. Sarasvat¯ı . hita Bhavana Library Rare Text Publication Series No. 6. Benares: Sampurnanand Sanskrit University, 2001. ¯ dhanama ¯ la ¯ , ed. Benoytosh Bhattacharya. 2 vols. Gaekwad’s Oriental Series Nos. 26 sa and 41. Baroda: Oriental Institute, 1968 (reprinted from first editions of 1925 and 1928). ¯ mavidha ¯ na Bra ¯ hman.a with Ved¯ sa arthaprak¯ a´sa of S¯ ayan.a and Pad¯ artham¯ atravivr.ti of Bharatasv¯ amin. Second edition. ed. B. R. Sharma. Tirupati: Kendriya Sanskrit Vidyapeetha, 1980. ¯ mavedasam ¯ of the Kauthuma School, with Padap¯ sa at.ha and the Commentaries . hita of M¯ adhava, Bharatasv¯ amin and S¯ ayan.a. ed. B. R. Sharma. 2 vols. Harvard Oriental Series, 57 & 58. Cambridge & London: Harvard University Press, 2000 & 2002.

Works Consulted

131

¯ rdhatri´ ¯ lottara ¯ gama with commentary (-vr.tti) of Bhat.t.a R¯ sa satika amakan.t.ha, ed. N. R. Bhatt. Publications de l’IFI No. 61. Pondicherry: IFI, 1979. siddhaikav¯ıratantra. Chapter 1, ed. Mikky¯ o Seiten Kyeky¯ ukai (Nobuo Otsuka), in Taisho Daigaku Sogo-Bukkyo-Kenkyujo-Kiyo. Vol.15, pp. 366–349 (pp.(1)–(18)), 1995. I have used an etext prepared by Ryugen Tanemura in 2003. ´ agamasoma´ sambhupaddhati, ed. K. M. Subrahman.ya´sa ¯strin. Devak¯ ot.t.ai: Siv¯ siddh¯ antaparip¯ alanasangha, ˙ 1931. See also Brunner 1963, 1968, 1977. svacchandatantra with the commentary (-uddyota) of R¯ aj¯ anaka Ks.emar¯ aja. ed. ´astr¯ı. KSTS 31, 38, 44, 48, 51, 53, and 56. Bombay 1921–35. Madhus¯ udan Kaul S¯

Secondary Material Bakker, Hans. 2009. ‘Purus.amedha, Manasarapurus.a, V¯ astupurus.a: The Image of Man in the Sacrificial Context,’ in Journal of Indological Studies (Kyoto), Nos. 20–21, pp. 1–23. Bisschop, Peter. 2006. “The S¯ utrap¯ at.ha of the P¯ a´supatas¯ utra,” in Indo-Iranian Journal 49, pp. 1–21. Bisschop, Peter & Griffiths, Arlo. 2007. “The Practice involving the Ucchus.mas (atharvapari´sis.t.a 36),” in Studien zur Indologie und Iranistik, Band 24. Bremen: Hempen Verlag. Brunner, H´el`ene, ed. and trans. 1963, 1968, 1977, 1998. Soma´sambhupaddhati. 4 vols: Premi`ere Partie. Le rituel quotidien dans la tradition ´siva¨ıte de l’Inde du Sud selon Soma´sambhu; Deuxi`eme Partie. Rituel Occasionnels dans la tradition ´siva¨ıte de l’Inde du Sud selon Soma´sambhu I : Pavitr¯ arohan.a, Damanap¯ uj¯ a et Pr¯ aya´scitta; and Troisi`eme Partie. Rituels occasionels dans la tradition ´siva¨ıte de l’Inde du Sud selon Soma´sambhu II : d¯ıks.a ¯, abhis.eka, vratoddh¯ ara, antyes.t.i, ´sr¯ addha; and Rituels dans la tradition siva¨ıte selon Soma´sambhu. Quatri`eme partie : rituels optionnels : pratis.t.h¯ a. Publications de l’IFI No. 25. Pondicherry: IFI. Gail, Adalbert J. 2009. “The Earth and the Lotus: A contribution to Vis.n.u’s iconography in India,” in Pandanus ’09: Nature in Literature, Art, Myth and Ritual. Volume 3, No. 1, pp. 83–91, + 11 pictures. Prague: Charles University in Prague. Goodall, Dominic. 1998. See kiran.atantra. ´ Goodall, Dominic. 2004. The Par¯ akhyatantra. A scripture of the Saiva Siddh¯ anta. A critical edition and annotated translation. Collection Indologie 98. Pondicherry: IFP/EFEO. Goodall, Dominic. 2011. “Textes sanskrits indiens et inscriptions du Cambodge,” in ´ Annuaire de l’Ecole pratique des hautes ´etudes, Sciences religieuses 119 (2010– 2011), pp. 97–98. Goodall, Dominic, in collaboration with Alexis Sanderson and Harunaga Isaacson, with contributions of Nirajan Kafle, Diwakar Acharya & others, 2015

132

Early Tantric Vais.n.avism

´ (forthcoming). The Ni´sv¯ asatattvasam a. The Earliest Surviving Saiva Tantra. . hit¯ Volume 1. A Critical Edition and Annotated Translation of the M¯ ulas¯ utra, Uttaras¯ utra & Nayas¯ utra. Collection Indologie 128 / Early Tantra Series 1. Pondicherry: EFEO/IFP. ´ Goodall, Dominic. forthcoming. “Introduction” to Saiva Rites of Expiation. A ¯ ya´ first edition and translation of Triocana´siva’s twelfth-century pra scittasamuccaya (with a transcription of a manuscript transmitting Hr.daya´siva’s Pr¯ aya´scittasamuccaya). ed. and trans. by R. Sathyanarayanan. Collection Indologie. Pondicherry: EFEO/IFP. ¯ matatantra. Goudriaan and Schoterman. 1988. See kubjika Indian Archaeology 1959–1960—A Review. Edited by A. Shosh. New Delhi: Department of Archaeology, 1959. Limaye, V. P. & Vadekar, R. D. 1958. Eighteen Principal Upanis.sads. Poona: Vaidika Sam . ´sodhana Man.d.ala. Lubin, Timothy. 2007. “The N¯ılarudropanis.ad and the Paippal¯ adasam a: A Critical . hit¯ Edition with Translation of the Upanis.ad and N¯ ar¯ ayan.a’s D¯ıpik¯ a,” in Arlo Griffiths and Annette Schmiedchen, The Atharvaveda and its Paippal¯ adasam a. . hit¯ Geisteskultur Indiens, Texte und Studien, Band 11, pp. 81–139. Aachen: Shaker Verlag. Mevissen, Gerd J.R. 2008. “A Solitary Vis.n.u Sculpture from Bihar in the National Museum of Nepal, Kathmandu,” in Claudine Bautze-Picron, Religion and Art: New Issues in Indian Iconography and Iconology. Volume 1 of the proceedings of the 18th conference of the British Association of South Asian Archaeologists, London, 2005, pp. 125–139. London: The British Academy, British Association of South Asia Studies. Oberlies, Thomas. 2003. A Grammar of Epic Sanskrit. Indian Philology and South Asian Studies, 5. Berlin/New York: Walter de Grunter. ¯ Olivelle, Patrick. 2000. Dharmas¯ utras: The Law Codes of Apastamba, Gautama, Baudh¯ ayana, and Vasis.t.ha. Annotated Text and Translation. Delhi: Motilal Banarsidass. Rasteli, Marion. 2007. The “P¯ an ˜car¯ atra Passages,” in Agnipur¯ an.a in Dominic Goodall & Andr´e Padoux, Tantric Studies in Memory of H´ele`n Brunner. Collec´ tion Indologie 106, pp. 187–229. Pondicherry: French Institute/Ecole Fran¸caise d’Extrˆeme-Orient. Rajan, K. V. Soundara. 1967–1968. “Kaustubha Pr¯ as¯ ada–New Light on Jay¯ akhya Tantra,” in Journal of The Oriental Institute, Baroda. Volume XVII, pp. 71–85. Ramesh, K.V. & Tewari, S.P. 1990. A Copper-plate Hoard of the Gupta Period From Bagh, Madhya Pradesh. New Delhi: Archaeological Survey of India. ´ Sanderson, Alexis. 2001. ‘History through Textual Criticism in the Study of Saivism, the P¯ an ˜car¯ atra and the Buddhist Yogin¯ıtantras’ in F. Grimal (ed.): Les Sources et le Temps/Sources and Time. A colloquium: Pondicherry, 11–13 January 1997, pp. 1–47. Publications du d´epartement d’indologie 91. Pondicherry: IFP/EFEO.

Works Consulted

133

´ ´ Sanderson, Alexis. 2009. “The Saiva Age: The Rise and Dominance of Saivism during the Early Medieval Period,” in Shingo Einoo (ed.): Genesis and Development of Tantrism, pp. 41–349. Tokyo: Institute of Oriental Culture, University of Tokyo. Sircar, D.C. 1961. “Nagarjunikonda Inscription of the time of Abhira Vasushena, Year 30,” in Epigraphia Indica. Volume 34: 1961–1962 edited by D.C. Sircar. Delhi: Epigraphical Survey of India, pp. 197–204. Sircar, D.C. 1965. Select Inscriptions bearing on Indian History and Civilization. Volume 1: From the Sixth Century B.C. to Sixth Century A.D. Second Edition. Calcutta: University of Calcutta. Srinivasan, Doris. 1979. “Early Vais.n.ava Imagery: Caturvy¯ uha and Variant Forms,” in Archives of Asian Art. Volume 32, pp. 39–54. Srinivasan, Doris Meth. 1997. Many Heads, Arms and Eyes: Origin, Meaning and Form of Multiplicity. Studies in Asian Art and Archaeology, vol. 20. Leiden/NewYork/K¨ oln: Brill. ¯ ntrika ¯ bhidha ¯ nako´ ta sa I & II. A Dictionary of Technical Terms from Hindu Tantric Literature. sous la direction de H. Brunner, G. Oberhammer et ¨ A. Padoux. Osterreichische Akademie der Wissenschaften, PhilosophischHistorische Klasse, Sitzungsberichte, 681, & 714. Beitr¨ age zur Kultur- und Geis¨ tesgeschichte Asiens 35 & 44. Wien: Verlag der Osterreichischen Akademie der Wissenschaften, I: 2000, II: 2004. ¯ ntrika ¯ bhidha ¯ nako´ ta sa III. A Dictionary of Technical Terms from Hindu Tantric Literature. sous la direction de H. Brunner, G. Oberhammer et A. Padoux. Direction ´editoriale du troisieme volume: Dominic Goodall et Marion Rastelli. ¨ Osterreichische Akademie der Wissenschaften, Philosophisch-Historische Klasse, Sitzungsberichte, 839. Beitr¨ age zur Kultur- und Geistesgeschichte Asiens 76. ¨ Wien: Verlag der Osterreichischen Akademie der Wissenschaften, 2013. ¨ rzso ¨ k, Judit. 1999. ‘The Doctrine of Magic Female Spirits.’ A critical edition of To selected chapters of the Siddhayoge´svar¯ımata(tantra) with annotated translation and analysis. (Unpublished doctoral thesis submitted to the University of Oxford.) ¨ rzso ¨ k, Judit. 2013. “The Heads of the Godhead: The Number of Heads/Faces of To ´ Yogin¯ıs and Bhairavas in Early Saiva Tantras,” in Indo-Iranian Journal 56.2, pp. 135–155. Turner, R. L. 1999. A Comparative Dictionary of the Indo-Aryan Languages. First Published from Oxford University Press, London. The Main Volume, 1962–1966. Three Supplementary Volumes, 1969–1985. First Indian Edition, 1999. Delhi: Motilal Banarsidass. ¯ ca ¯ rya, Dhanavajra. 1973. Licchavik¯ Vajra alak¯ a Abhilekha. [Inscriptions of the Licchavi Period.] Kathmandu: Centre for Nepal and Asian Studies. ¯ ca ¯ rya, Dhanavajra & Malla, Kamal P. 1985. Gop¯ Vajra alar¯ ajavam ¯val¯ı. Nepal Re. ´sa search Centre Publications 9. Wiesbaden: Franz Steiner Verlag.

FACSIMILES

137 Facsimiles

Upper: folio *7 verso, left-half

Lower: folio 6 recto, left-half

1`HvghMi`B+goBțƠpBbKg 138

Upper: folio *7 verso, right-half

Lower: folio 6 recto, right-half

139 Facsimiles

Upper: folio 6 verso, left-half

Lower: folio 5 verso, left-half

1`HvghMi`B+goBțƠpBbKg 140

Upper: folio 6 verso, right-half

Lower: folio 5 verso, right-half

141 Facsimiles

Upper: folio 5 recto, left-half

Lower: folio 4 verso, left-half

1`HvghMi`B+goBțƠpBbKg 142

Upper: folio 5 recto, right-half

Lower: folio 4 verso, right-half

143 Facsimiles

Upper: folio 4 recto, left-half

Lower: verso of the second-last folio, left-half

1`HvghMi`B+goBțƠpBbKg 144

Upper: folio 4 recto, right-half

Lower: verso of the second-last folio, right-half

145 Facsimiles

Upper: recto of the second-last folio, left-half

Lower: recto of the last folio, left-half

1`HvghMi`B+goBțƠpBbKg 146

Upper: recto of the second-last folio, right-half

Lower: recto of the last folio, right-half

147 Facsimiles

Verso of the last folio, left-half

1`HvghMi`B+goBțƠpBbKg 148

Verso of the last folio, right-half

149 Facsimiles

Recto of the first folio, left-half

1`HvghMi`B+goBțƠpBbKg 150

Recto of the first folio, right-half

151 Facsimiles

Upper: verso of the first folio, left-half

Lower: folio *X verso, left-half

1`HvghMi`B+goBțƠpBbKg 152

Upper: verso of the first folio, right-half

Lower: folio *X verso, right-half

153 Facsimiles

Upper: folio *X recto, left-half

Lower: folio *Y verso, left-half

1`HvghMi`B+goBțƠpBbKg 154

Upper: folio *X recto, right-half

Lower: folio *Y verso, right-half

155 Facsimiles

Upper: folio *Y recto, left-half

Lower: folio *7 recto, left-half

1`HvghMi`B+goBțƠpBbKg 156

Upper: folio *Y recto, right-half

Lower: folio *7 recto, right-half

Facsimiles

157

158

1`HvghMi`B+goțƠpBbK

Facsimiles

159

160

1`HvghMi`B+goțƠpBbK

Facsimiles

161

162

1`HvghMi`B+goțƠpBbK

Facsimiles

163

164

1`HvghMi`B+goțƠpBbK

Facsimiles

165

166

1`HvghMi`B+goțƠpBbK

Facsimiles

167

168

1`HvghMi`B+goțƠpBbK

Facsimiles

169

170

1`HvghMi`B+goțƠpBbK

Facsimiles

171

172

1`HvghMi`B+goțƠpBbK

Facsimiles

173

174

1`HvghMi`B+goțƠpBbK

Facsimiles

175

176

1`HvghMi`B+goțƠpBbK

Facsimiles

177

178

1`HvghMi`B+goțƠpBbK

Facsimiles

179

180

1`HvghMi`B+goțƠpBbK

Facsimiles

181

182

1`HvghMi`B+goțƠpBbK

Facsimiles

183

184

1`HvghMi`B+goțƠpBbK

INDICES

Index of P¯ adas in all three Texts

The various kinds of brackets used in the body of the text are not shown in this index. a\smEt◆ c s⌃Dy ⇤t^ DP 13:9b a, y“Ft Evf ⇤qt, AVi 17d akAr\ dE"ZAΩ§ ⇤ AVi 13c ak⇥ Zml {EnE[Cd {, ˝ DP 9:14c aár\ t mhAvFy◆ AVi 108c a"‡mFEt m⌃/ ⇤Z DP 13:3c a"rAy prAZAy DP 9:29c a"ro _n"r-tTA SP 1:10b _"ro _n"r ev vA DP 1:5b aE"mEt◆ c s⌃Dy ⇤t^ DP 13:3d a`n aA yAEh m⌃/ ⇤Z DP 6:21a, 8:2c a`n aA yAhFEt m⌃/ ⇤Z SP 5:15a aE`n\ dtEmEt -m⇥tm^ SP 7:41d aE`nkAE⌃t tT {v c AVi 115b aE`nkAy◆ tt, k⇥(vA SP 7:33a aE`nk⌫X\ t kAry ⇤t^ AVi 97d aE`nk⌫XEvBAg\ t SP 1:23c aE`nm`nFEt m⌃/ ⇤Z DP 8:4a aE`nmFl ⇤ cA (v ⇤kA DP 12:3a aE`nmFl ⇤ ’ Et m⌃/ ⇤Z SP 5:14a, 5:83c; DP 6:20a aE`n>yoEtm⌃/ ⇤Z AVi 90a aE`n>yotFEt m⌃/ ⇤Z DP 9:7a aE`nmD ⇤Et m⌃/ ⇤Z SP 7:41a aE`n£omAEdEBmK {, SP 2:12d; DP 2:10d aE`nho/AEZ kmAEZ SP 7:47c aE`nho/AEdDm ⇤q SP 2:13c a`n ⇤-tnrsFEt m⌃/ ⇤Z DP 13:10c ag }tñAEp kt&y\ SP 5:1c ag }tñ {v kt&y\ DP 6:1c

aΩ⌃yAsEvED\ kz AVi 1d aΩ⌃yAso EvDFyt ⇤ DP 11:22b aΩ˛(yΩ ’ l ⇤py ⇤t^ AVi 82b aΩ˛(yΩyo-tTA SP 1:13b; DP 1:8b aΩhFnA n Es@yE⌃t AVi 1a aΩlAEn c Ev˛ ⇤⌃d˝ DP 3:5c aΩlAEn /yE-/\ft^ DP 3:7c aΩSy, pErkFEttA, DP 3:26b aΩSyA\ nK - m^ - aΩ§ ⇤ SP 1:18c; DP 1:13c aEcr ⇤Z Bv ⇤(svm^ DP 8:25e ajAto jAt ev vA DP 1:2d ajAto jAt m^ - ev c SP 1:8b a‚AnEtEmrA⌃DAnA\ DP 2:25c a“n\ t ˛dAt&y\ SP 8:14c a“n\ t ˛dApy ⇤t^ DP 9:35b a“n\ t ˛dApy ⇤t^ AVi 90d at U@v◆ n pjy ⇤t^ DP 3:10b at, pr\ ˛v#yAEm AVi 58a, 5:1a, 7:1a, 8:1a; DP 3:1a, 4:1a, 5:1a, 6:1a, 9:1a aEtk ⇤krvAmnA DP 12:5b aEts#mo n d⇥[yt ⇤ DP 2:16b atFtAnAgt\ c {v SP 2:11c; DP 2:9c aT s\-TApn\ v#y ⇤ DP 12:2a aTv◆ coÅr ⇤ ⌃ys ⇤t^ AVi 74b aTvEfrs\ c {v SP 7:38a aTvAEΩrs-tTA DP 7:6b aTAt, s\˛v#yAEm DP 10:1a aTAcn\ ˛v#yAEm DP 11:1a

188 aEd(y { &y⌃dn\ c {v DP 8:21a ad {v ⇤q kT\ yAg\ DP 7:1c açAEp n Envt⌃t ⇤ DP 11:16c aD, pAtAl ’ s\E-Ttm^ DP 7:16b aDño˘◆ c m@yt, DP 4:2d aD-tAEÅ~Evámo ⇤ dv, AVi 113c aDA no EvòsOB ⇤Et DP 8:17c aEDvAsnmÅmm^ SP 7:1b aDoG⌫VA rTAΩA c AVi 81a aDod⇥E£-t kAry ⇤t^ DP 4:4f anΩ\ vág\ tTA DP 4:3b anΩA ⇤ dfnAfAy DP 4:4c an⌃t\ dAzZ\ mht^ DP 2:20d anAEdEn kl, ôFmAn^ DP 2:17c aEn(y\ -TlEm(yAh, DP 2:17a aEnzà\ t n { t ⇤ AVi 107b _Enzà, kvc\ Bv ⇤t^ DP 11:24b an ⇤n EvEDnA. . . SP 8:43a an ⇤n {v EvDAn ⇤n SP 8:55a an ⇤n {v EvBAg ⇤n DP 3:26a a⌃tEr"\ ˛Et§y ⇤t^ DP 8:18b a⌃tEr" ⇤ c pAtAl ⇤ DP 2:14c a⌃tñrEt m⌃/ ⇤Z DP 8:18a a‡pt ⇤ ’ Et m⌃/ ⇤Z DP 9:37a a⌃yd ⇤vrt\ fd˝\ DP 10:9c a⌃yd ⇤vrtA\-(yj ⇤t^ DP 10:12d apA\ p⇥§ ⇤Et m⌃/ ⇤Z DP 13:10a apA\ p⇥§ ⇤Et ôFsƒ\ AVi 39e apAno j⌃mBnAEB, VKP 16a apAmAg-tT {v c SP 7:39d a˛tFto jy(yAh DP 8:20a aEB (vA f⇤ rEt m⌃/\ SP 9c aEBá⌃d ⇤Et m⌃/ ⇤Z DP 8:7c aEBEqƒ\ fEcv }tm^ DP 10:5b aEBq ⇤k\ t dApy ⇤t^ AVi 119d aEBq ⇤k\ smArB ⇤t^ AVi 119b am⌃/oƒA\ t y, kyAt^ DP 8:24c arAg ⇤ }q Evc"Z, DP 11:31b ar {íAdfEBytm^ DP 11:26b

Early Tantric Vais.n.avism aG◆ dçA(˛y◊t, DP 9:39b a]y◆ dçA(˛y◊t, DP 9:31b a]y◆ dçA(smAEht, SP 8:26d - aT◆ c {k\ t homy ⇤t^ SP 7:40d aTnAf\ kl"ym^ DP 7:23b aDc⌃d˝\ t dE"Z ⇤ AVi 98b al\ t v ⇤dk {m⌃/ {, SP 8:41c alAB ⇤ svDAtnA\ AVi 118c, SP 8:22c alAB ⇤ svbFjAnA\ AVi 118e, SP 8:23a alAB ⇤ svr◊AnA\ AVi 118a, SP 8:23c aEvEd(vA im\ y-t DP 7:21c af ⇤q\ vA¡y\ sv◆ AVi 69a; DP 5:12a aôà ⇤yEmd\ ⌃yAs\ DP 12:1a aò\ vA g }Amm ⇤v vA SP 8:51d aò(T\ pEñm ⇤ EvçAt^ AVi 95c aò(Tñ fmF c {v SP 5:12c; DP 6:17e aòm ⇤DPl\ v(s DP 8:24a aEò⌃yA\ roEhZFq c SP 5:4d aEò⌃yo rohZFq c DP 6:5d a£koZ\ tToÅr ⇤ AVi 98d a£E/\f(klop ⇤t, AVi 11e a£p/\ skEZkm^ DP 11:26d a£m\ nvm\ tTA DP 3:6b a£m ⇤ t E/Evámm^ SP 7:26d a£A"r\ tto h(vA SP 8:50c a£AΩo Ev ZzQyt ⇤ DP 11:21d a£Av ⇤v mhAmn ⇤ DP 9:26d a£O c(vAEr c {v vA SP 7:40b asrFmEdtAEn c SP 5:30a a-t ôOqV^ c m⌃/oƒ\ DP 8:21c a-/\ c {voÅr ⇤Z t SP 7:20d a-/\ v { coÅr\ ⌃ys ⇤t^ AVi 104d a-/\ v { coÅr\ ⌃ys ⇤t^ DP 11:28d a-/⌃yAs\ ˛v#yAEm AVi 55a; DP 11:17a a-/ ⇤Z l ⇤Kn\ kyAt^ AVi 100a a-/o nArAyZ, ˛oƒ, DP 11:24c a-mAkEm⌃d˝ - m⌃/ ⇤Z DP 8:12c ah\ b }ú tdQyt ⇤ SP 2:14b ah\ sÿ ⇤Et m⌃/ ⇤Z DP 8:6a

P¯ ada Index ahorA/\ tT {v c SP 7:50b aA no Bd˝A ’ Et m⌃/ ⇤Z AVi 92c aA =yAy-v ⇤Et m⌃/ ⇤Z SP 8:44a aAkAr\ vAmto ⌃ys ⇤t^ AVi 13d aAkAr ⇤Z smE⌃vtm^ AVi 53b aAkAf ⇤ EdEv gocr ⇤ DP 2:14d aAk⇥ ZA-yvAmFEt AVi 44a aAk⇥ Z ⇤n ⇤Et homy ⇤t^ SP 7:46d aAgAr\ Ev Z{ dvtm^ DP 7:4b aA`n ⇤ yAhFEt y(˛oƒA DP 12:7a aA`n ⇤⇧yA\ EnfW\ ⌃ys ⇤t^ AVi 106d aA`n ⇤⇧yA\ t gdA\ ⌃ys ⇤t^ DP 11:39b aA`n ⇤⇧yA\ t jnAdn, AVi 112d aA`n ⇤⇧yA\ t ˛dApy ⇤t^ SP 8:18b aA`n ⇤⇧yA\ t mn,EflAm^ SP 8:14b aA`n ⇤⇧yA\ EdEf gAy/F\ DP 11:29a aA`n ⇤⇧yA\ rƒvZA◆ t DP 6:28c aA`n ⇤⇧yA\ rƒvZABA\ SP 5:22a aA`n ⇤⇧yA⌃t gdA\ ⌃ys ⇤t^ SP 7:29d aAcAy◆ smdA„tm^ DP 10:5d aAcAy, smdA„t, AVi 11f, 5:12b aAcAy, ssmAEht, DP 11:6b aAcAyñ pr\ b }ú DP 7:7a aAEjG } klf\ m⌃/ ⇤Z SP 5:17c aAEjG } klfm⌃/ ⇤Z DP 6:23c aA>y\ t/ smAdAy DP 9:8a aA>ydohpllBAr ⇤ c AVi 43c aA>ydoh ⇤Et sAm ⇤n DP 8:19c aA>yBAg\ t kAry ⇤t^ DP 9:24b aA>yAΩAn^ sADy ⇤d^ Eíj, DP 8:19d aA>y ⇤ c {vAm⇥t\ ˛oƒ\ DP 7:17c aAtp/ ⇤ E-Tto Dmo DP 7:15c aA(mtÇvEmEt HyAt\ AVi 18c aA(mAΩ ⇤ _ΩA⌃trA— {v AVi 46a aAEdm@yA⌃tgocr SP 1:5b aADAnmE`nho/\ c DP 2:11c aADAr, svBtAnA\ SP 2:6c; DP 2:5a aADArñ tTAD ⇤yo DP 2:18a aAny ⇤t jnAdnm^ DP 9:5b aApAdm-tk\ yAvt^ AVi 8c, 29a

189 aA=yAy-v iEt "Fr\ AVi 79c aA=yAy-v ⇤Et y("Fr\ DP 9:20a aAm }EbSvPlAEn c DP 9:21d aAm }AEdp•v {ñ {v SP 8:42c aAyDAEn c ⇤ dv-y SP 1:20a; DP 1:14c aAht\ c {v bOàkm^ DP 10:10b aAElK ⇤í ⇤Edm⌫Xlm^ SP 5:24b; DP 6:31b aAvd\-(vEmEt ˛oƒA DP 12:4a aAvAhy ⇤Åto ⇤ dv\ DP 9:27a aAf\ dDFEt m⌃/ ⇤Z DP 13:9a aAsr\ Ek\ n { pfAc\ DP 7:1a aAhr ⇤ÅAm }BAjn ⇤ DP 9:18b ikAr\ dE"Z ⇤ gSP ⇤ AVi 14a iEt dfnpE≠q AVi 34b iEt m⌃/ ⇤Z Enmdnm^ AVi 77b i(y£O ˛EtmA, ˛oƒA, DP 3:15a i(y ⇤tA, ˛EtmA, HyAtA, DP 3:9a id\ Ev ZEvcám ⇤ DP 9:31d i⌃d˝ >y ⇤§ ⇤Et m⌃/ ⇤Z DP 8:15a i⌃d˝, srpEtñ {v SP 8:2a i⌃d˝QC/ ⇤Et m⌃/ ⇤Z DP 8:16a i⌃d˝nFl\ mhAnFl\ AVi 114c; DP 3:2a i⌃d˝nFlmy\ c {v SP 8:22a i⌃d˝-yAvAhn\ Bv ⇤t^ SP 8:2f i⌃d˝AEd EvEnv ⇤fy ⇤t^ SP 8:17d i⌃d˝Aç {d ⇤vt {, sv {, SP 1:2a i⌃d˝Ay gAv aAfFEt DP 13:1c i⌃d˝Ay sAm gAy ⇤Et DP 8:14c i⌃d˝AyA\ hErtAl\ t SP 8:14a i⌃d˝Asom ⇤Et m⌃/ ⇤Z DP 13:7a iE⌃d˝yAEZ mnñ {v SP 2:14c; DP 2:12c i⌃d˝o b }úA smAEôtO DP 5:5b im\ n somsAm ⇤n DP 8:8c im\ m ⇤ gΩ ⇤ ymn ⇤ DP 9:15c iq ⇤ (voj ⇤ ’ Et m⌃/ ⇤Z DP 6:20c iq ⇤(v ⇤n t m⌃/ ⇤Z SP 5:14c ih rtFEt m⌃/ ⇤Z DP 13:8a IkAr\ vAmto ⌃ys ⇤t^ AVi 14b IfAnAy ˛kSpy ⇤t^ SP 9d IfAn ⇤ cá!p\ t DP 6:14e

190 IfAn ⇤n yTAámm^ DP 9:2b IfA⌃yA\ t yTAámm^ SP 8:21d IfA⌃yA\ r"t ⇤ Ev Z, AVi 112c IfA⌃yA\ fAΩ ’ Ev⌃ys ⇤t^ SP 7:31b IfA⌃yA\ fAΩm ⇤v c AVi 110f IfA⌃yA\ f⇥EΩ Ev⌃ys ⇤t^ DP 11:40d IfA⌃yA\ svvEZkAm^ SP 5:23d; DP 6:30b IfAvA-y\ t ao§yo, AVi 35a IfAvA-yEmd\ m⌃/\ SP 8e Iòr, ˛(yBAqt SP 2:1b ukAr\ nlk ⇤ c {v AVi 14c ut vAt ⇤n t m⌃/ ⇤Z SP 7:1c uÅmAn^ svDAtnA\ SP 8:17a uÅr ⇤ p¨jAkAr\ DP 6:14c uÅr ⇤ mDsdn, AVi 112b uÅr ⇤Z aTvZm^ SP 7:34b uÅAnpZ ⇤ m⌃/ ⇤Z DP 8:10a uEŧ b }úZ-pt ⇤ AVi 92a uEŧ ⇤Et smà⇥(y DP 9:38a u(pAt\ t Dn"ym^ DP 8:25f udkMBs˙`dAmAn^ DP 6:19c udr ⇤q smAEôtA, AVi 64b; DP 5:7b ud Et§ ⇤Et m⌃/ ⇤Z DP 8:17a udMbr\ c dE"Z ⇤ AVi 95b uíy\ t ⇤Et m⌃/ ⇤Z DP 13:2a uí ⇤gA t qXΩlA DP 3:11d u‡yAEm c n ⇤/yo, AVi 31b upEl=y tto BEm\ SP 5:9c; DP 6:12c upvFt\ tto ⌃ys ⇤t^ DP 9:33b uBy\ t ⇤ vy-yo∆A DP 8:12a uBy\ -TApy ⇤d^ @vEnm^ DP 8:12b uByor⌃tr\ nAE-t DP 2:24c u•nAEn EvEc/AEn SP 5:29c u• ⇤KAyoEt, k\mmsE‡Bm^ SP 7:12d >yoEtr ⇤KA\ ˛kSpy ⇤t^ DP 9:7b >yo(ykm c vAmk ⇤ AVi 32b >vAlAmAlAsm˛Bm^ SP 7:9d; DP 11:6d JkAr\ kE"dE"Z ⇤ AVi 21b ⇢kAr\ vAmk"AyA\ AVi 21c VkAr\ t/ Ec⌃ty ⇤t^ AVi 22b WkAr\ ++++ c AVi 22c XkAr\ tElkA-tTA AVi 22d YkAr\ bAhm@y ⇤ t AVi 23a ZkAr\ vAm aΩk ⇤ AVi 23b tkAr\ h-tp⇥§ ⇤ c AVi 23c tt, ˛v ⇤fy ⇤⇤ åvm^ DP 9:39a tt, fAE⌃t\ c kAry ⇤t^ SP 8:34d tt, ‹An\ smArB ⇤t^ DP 9:14d tto doq, ˛fAMyEt DP 9:13b tto nçdk ⇤n {v DP 9:26a tto Enm⌃/y ⇤⇤ åv\ DP 9:37c tto r◊AEn d˝&yAEZ SP 8:13a tto Eh fAE⌃tmArB ⇤t^ AVi 97b t(˛mAZ ⇤n kvFt DP 3:21c

196 t/ pjA c kt&yA SP 8:38a t/ r◊AEn ⇤ dyAEn SP 8:42a t/ rA‚ ⇤ mhâym^ DP 8:25d t(s\yog ⇤ t ôF\ kyAt^ DP 12:9a tTA svgto Ev Z, DP 2:16c tdDO crZO -m⇥tO DP 3:28d tdh\ s\˛v#yAEm DP 2:2c td ⇤vAE`n-tdAEd(y AVi 34a td ⇤vAE° ˛mQyt ⇤ SP 8:56b t¯y\ nrk\ v }j ⇤t^ DP 7:22b tnmEt◆ c s⌃Dy ⇤t^ DP 13:10d t⌃m@y ⇤ kml\ ˛oƒm^ DP 11:26c t⌃m@y ⇤ t ⌃ys ⇤(pî\ SP 8:37c tmh\ kFtEy yAEm SP 8:4c tmh\ sM˛v#yAEm SP 2:3c tmAl\ rƒlocnm^ DP 3:14b tmFfAn\ dE"Z ⇤ bAh\ AVi 37c tm £vAm y\ EgEr - DP 13:6a tzZAEd(ys\kAf\ SP 7:11a; DP 11:8a t-mA(sv˛y◊ ⇤n AVi 1c; DP 6:15c, 7:23e t-mAd ⇤v Enyojy ⇤t^ VK 334b t-mAâ≈A mhAd ⇤v\ DP 2:23a t-mAç◊ ⇤n Ev˛ ⇤⌃d˝ DP 10:12c t-mAEíÅAnsAr ⇤Z SP 8:57e t-m { En(y\ nmo nm, SP 8:3d, 8:7d t-m { r^ - i⌃d˝Ay v { nm, SP 8:2d t-m { svA(mn ⇤ nm, SP 8:11d t-y kAygt\ pAp\ SP 8:56a t-y ⇤ dvAEtd ⇤v-y DP 2:26a t-y yAg\ ˛v#yAEm DP 5:13c t-y v { ˛EtmA fBA SP 1:12d; DP 1:7d t-yAΩAEn ˛v#yAEm DP 3:17c t-yAD ⇤n t kvFt DP 3:19a tAd⇥f {v t m ⇤KlA DP 6:16b tAEn c {v ˛lFy⌃t ⇤ SP 2:7c; DP 2:5e tAm }k\ vAEp kA\-y\ c DP 3:3a tAry⌃tF stArZF AVi 11d tArA stArA trZF AVi 11c tArAgZsmAEôtA, AVi 65b; DP 5:8b

Early Tantric Vais.n.avism tAlk ⇤ t smAHyAtO DP 5:5a EtETvAr\ tT {v c SP 1:22b; DP 1:16d Et⌃dkAmlk ⇤ c {v DP 9:21c EtygA rAGvAE(mkA DP 12:6b Etlk\ t sm>>vlm^ DP 4:10b EtlAn^ k⇥ ZAn^ G⇥tAjAtv ⇤dsm^ AVi 101b ˛ZMy grv ⇤ pñAt^ SP 8:53a ˛EZp(y yTAEvED SP 8:34b ˛EZp(y hr\ ⇤ dv\ SP 1:3a ˛ZFt-TA sr-vtF DP 7:17b ˛ZFtA\ c ˛Et§y ⇤t^ DP 8:19b ˛Et c#v Ev c#v ⇤Et DP 12:6c ˛EtpŒ EítFyA c DP 6:3a ˛EtpŒ EítFyAyA\ SP 5:2c ˛EtmA\ cAEDvAEstAm^ DP 12:2b ˛EtmAk ⇤fv\ fBm^ DP 3:29d ˛EtmAΩ ⇤q bEàmAn^ DP 9:10b ˛EtmAp—k\ -m⇥tm^ DP 4:6f ˛EtmAyA\ Evf ⇤q ⇤Z SP 7:4a; DP 11:2a ˛EtmAl"Z\ fBm^ DP 3:1b ˛Et§A t-y ⇤ dv-y SP 1:12a; DP 1:7a ˛Et§A\ loBmoEht, DP 8:24d ˛Et§Am⌫Xp\ EDyA DP 7:21d ˛Et§AyAgkmEZ SP 5:4b; DP 6:5b ˛Et§AyAgmÅmm^ SP 1:25d ˛Et§AsFEt sAm ⇤n DP 12:12a ˛(yΩAEn yTAámm^ DP 3:17d ˛(ym— ⇤Et m⌃/ ⇤Z DP 8:2a

203 ˛(yh\ pErvtnAt^ SP 8:28b ˛(y ⇤k {k\ t t(sv◆ DP 9:12c ˛Tm\ pAdy`m ⇤ t AVi 7e ˛çflñ EfKA ‚ ⇤yA - DP 11:24a ˛Bv, svBtAnA\ SP 8:11c ˛BvE⌃t ygAgm ⇤ SP 2:7b; DP 2:5d ˛mAZ\ c vdAMyhm^ DP 3:5b ˛v#yAMynpvf, AVi 114b ˛v ⇤Eft ⇤ c ⇤ dv ⇤f ⇤ SP 8:26c ˛fA⌃t ⇤E vh doq ⇤q DP 9:13c ˛AZAyAm {EvdbDA, SP 7:43d ˛AsAd\ t kT\ kAy◆ SP 1:21a; DP 1:15c ˛AsAd\ t ˛Et§y ⇤t^ DP 8:2b ˛AsAd\ prm\ prm^ DP 7:2d ˛AsAd-y sm⌃tt, SP 5:1d; DP 6:1d ˛AsAdAEBmK\ ⇤ dv\ SP 7:32c; DP 9:39c ˛AsAd ⇤ EvEvD ⇤n t SP 8:56d ˛FtA(mA s ˛yQCEt DP 2:23d PkAr\ dfn ⇤ k⇥(vA AVi 25a Pl\ ˛Aÿ\ n s\fy, SP 8:57d Pl\ ˛AsAdEv-trm^ SP 8:55b Pl‹An\ t kAry ⇤t^ DP 9:22b PlAEn EvEvDAEn c DP 6:34b blBd˝\ ⌃ys ⇤(pv AVi 106c blEv‚Ay m⌃/ ⇤Z DP 8:10c, 9:36a bEl\ c svto dçAt^ DP 9:3c bElm/ ˛kSpy ⇤t^ SP 7:48d bü⇥c\ b }AúZ\ c {v DP 10:1c bAl&yjnk\ dd ⇤t^ DP 9:36b bAl&yjnk\ ⌃ys ⇤t^ DP 8:10d bAlkA\ t/ dApy ⇤t^ SP 5:9b bAlkAΩArpAqAZAn^ SP 5:6c; DP 6:7c bAhyt ⇤ -TAn\ SP 8:30a yE-m⌃d˝&y ⇤ t kt&y\ DP 3:1c y-y n { (ygocr ⇤ SP 8:19b y-y svAEZ BtAEn SP 2:7a; DP 2:5c yA aoqDFEt m⌃/ ⇤Z AVi 83a; DP 9:23c, 9:33c yA c p⌫yA mhFtl ⇤ DP 3:13d yA t-y Pls\pEÅ, SP 8:54c yA t ⇤ zd˝ EfvA tn, SP 8:44b yA\ gEt\ yAE⌃t yog ⇤ _E-mn^ DP 11:15c yA\ gEt\ yAE⌃t lok ⇤ _E-mn^ SP 7:15c yA, PElnFEt m⌃/ ⇤Z DP 9:22a yAgo ⇤ dvmyo Bv ⇤t^ DP 8:1d yAg, qoXfmo Eíj AVi 102 yAgm⌫Xpm@y ⇤ t DP 6:36a yAgm⌫XpmÅmm^ SP 5:1b; DP 6:1b yAg-TAn\ ˛v ⇤fy ⇤t^ AVi 92d yAg ⇤ y‚vrAh\ t DP 7:19a yAd⇥f\ t Bv ⇤(k⌫X\ DP 6:16a yAEn fA-/AEZ p’⌃t ⇤ SP 2:10c; DP 2:8c yAMyAyA\ c EvEnE"p ⇤t^ SP 8:18d yAMyAyA\ t sm⌃tt, SP 8:14d yAMyAyA\ [yAmvEZkAm^ SP 5:22b; DP 6:28d yAMy ⇤ t s\E-Tt\ / ⇤tA - DP 7:8a yAMy ⇤ _@vy◆ tT {v c SP 7:33d yAvŒ íAdfAΩlA DP 3:9d yƒ\ dAmodr\ ⌃ys ⇤t^ DP 11:37d y“E⌃t b } - m⌃/ ⇤Z DP 9:35c y“ATAEmEt m⌃/ ⇤Z DP 12:11c y ⇤ c mEtDrA, -m⇥tA, SP 1:25b; DP 1:17d y ⇤ c Ev ZprA lokA, DP 2:20a y ⇤n ⇤d\ dE"Z ⇤ kZ ⇤ AVi 32a

P¯ ada Index

207

y {m⌃/ {ñAsr\ m— ⇤t^ DP 8:1c ⇤ rck ⇤n (yj ⇤d^ b }ún^ AVi 70c y {-t km n kAry ⇤t^ SP 6:2b; DP 10:6b rogdA ctrΩlA DP 3:11b yo n, -vo arZo ˛Ah, DP 12:6a rocnA\ t/ dApy ⇤t^ DP 9:35d yo b }úA s jnAdn, SP 2:5b; DP 2:4b lklFf\ EvvZ-T\ SP 6:2c yo zd˝, s -vy\ b }úA SP 2:5a; DP 2:4a l"Z\ cAEp dApy ⇤t^ AVi 90b yogfA-/Emd\ ˛oƒ\ SP 2:3a l"Z ⇤ EáymAZ ⇤ t DP 9:7c yog ⇤ yog ⇤ ’ Et m⌃/ ⇤Z SP 5:25c; DP llAV\ t-y foBnm^ DP 3:18d 6:32c llAV\ t pn-t-y SP 1:14c; DP 1:9c yoEn\ cAEp ˛Et§y ⇤t^ DP 8:5b llAVmEt◆ t s⌃Dy ⇤t^ DP 13:2d yonF, kvFt foBnA, DP 6:16d llAV ⇤ c {v kt&y\ DP 4:10a rkAr\ vAmg⌫Xk ⇤ AVi 26d lAjA dED c sƒv, SP 7:49b rƒo vE°mhAbl, SP 8:3b lAvZAlÅkAEn c SP 5:30d rjA r"A rEt, pASyA AVi 9c ElΩ ⇤ c v⇥qZ ⇤ c {v AVi 56c; DP 11:19a r◊\ c {v svZ◆ c SP 8:52a ElΩ ⇤ s⇥E£, smAEôtA AVi 66d; DP 5:9d r◊\ dçAt^ . . . SP 8:53b ElÿA\ñ⌃dnp¨ ⇤n DP 6:22c r◊jA lohjA c {v DP 3:17a l ⇤ùp ⇤yAEdd ⇤yAEn SP 5:29a r◊Ev⌃yAsm ⇤v c SP 1:24d loknAT jgãro DP 1:6d r◊AdFEn EvEnE"p ⇤t^ SP 8:13d loknATo jnAdn, SP 1:11d rT⌃tr\ rTm@y ⇤ AVi 46c lokpAlA g }hA, p>yA, SP 8:38c rT⌃tr\ vƒ ⇤~ ⌃ys ⇤t^ AVi 33c lokpAlAEdd ⇤vtA, DP 7:13b rT ⇤ Et§ ⇤Et m⌃/ ⇤Z DP 9:5a lokAnA \ EvjyA(mkm ^ SP 1:3b rT ⇤Et§E‡Et vAEjq Ev⌃ys ⇤t^ AVi 46d lokAn g }hkArk DP 9:30b rMy\ íAdfh-tkm^ DP 6:11d lokAlok\ crAcrm^ AVi 69b; DP 5:12b rA"s\ m⌫XpADmm^ DP 7:1b lok ⇤ lok {n d⇥[yt ⇤ DP 2:16d rA"sA mAtrñ {v SP 8:12c vkAr\ t llAVk ⇤ AVi 27b rA"sAñ gZA, sv ⇤ AVi 64a vkAr\ @yAnmÅmm^ SP 7:12b; DP 11:9b rA"sAñ gZA, EsàA DP 5:7a rAjp”my\ tTA DP 3:4b vkAr\ BjdE"Z ⇤ AVi 52b rAjA Evf ⇤qmAŸoEt SP 8:29a vkAr\ „dy ⇤ ⌃ys ⇤t^ SP 7:6d; DP 11:4b rAmr ⇤KA BjAk⇥Et, DP 12:5d vkArA"rd ⇤v ⇤n SP 7:25c zd˝\ pzqsƒ\ c SP 7:35c vkArA"rs\yƒ\ DP 11:34c zd˝k ⇤fvdfnm^ DP 2:25d vƒ&y\ t jnAdnm^ SP 1:19b zd˝k ⇤fvyoEn(ym^ DP 2:22a vád⇥E£\ n kt&ym^ DP 4:2c zd˝mFfAnm&yym^ DP 2:19d váF Evg }hkAErkA DP 4:4d zd˝A, k⌫W ⇤ smAEôtA, AVi 59d v"ñ {v kT\ kAy◆ SP 1:17a; DP 1:12a zd˝ArADngEvtA, DP 2:21b v"-Tl ’ sm‡tm^ DP 3:22d !p\ c {v smàr ⇤t^ AVi 92b v"-y t EvBAg ⇤n DP 3:27a !ptÇvsmAyƒm^ DP 10:5a v"Ad^ EígZt, kAyO DP 3:25a !=y\ c {v EtlAn^ dçAt^ SP 8:19a v"AD ⇤n t kt&yO DP 3:23a ⇤ rKk\ vA n kt&y\ DP 4:3c v"omEt◆ c Ev⌃ys ⇤t^ DP 13:8b

208 vÖ\ mrkt\ v {Xy◆ DP 3:2c vÖm ⇤k\ t dApy ⇤t^ SP 8:23d vÖh-to mhAbl, SP 8:2b vnmAlAΩBEqtm^ DP 4:7d vrAhsAm nKAg ⇤ } AVi 45a vzZsAm ⌃ys ⇤à⇥Ed AVi 37a vzZAy nmo nm, SP 8:6d vjnAy ˛v#yAEm DP 10:6a vjnFyA\-t v#yAEm SP 6:2a vjnFyA, ˛y◊t, DP 10:10d vjy ⇤(-TApk, sdA SP 6:4d vq÷Ar\ ⌃ys ⇤ìD, DP 8:21d vq÷Ar ⇤ ˛jApEt, DP 7:18d vEs§A hFEt m⌃/ ⇤Z DP 8:8a v-/\ c {v ˛dApy ⇤t^ AVi 86b v-/y`msmAyƒ\ DP 4:5c v-/y`mAEn ⇤ dyAEn SP 8:52c v-/y`mO t Ev⌃ys ⇤t^ DP 8:8b v-/Al\k⇥tEvg }hAn^ SP 5:17b; DP 6:23b v-/ ⇤ yƒA mzãZA, DP 7:10b vAkAr\ j\Gyo⌃y-y DP 11:5c vAkAr\ p⇥§d ⇤ft, DP 11:4d vAkAr\ vzZAlym^ SP 7:13b, 7:14d; DP 11:10b vAkAr\ vAmpAd ⇤ t SP 7:8c; AVi 54a vAkArmm⇥tA"rm^ DP 11:11d vAkAr ⇤Z t pjy ⇤t^ DP 11:35d vAkAr ⇤Z Enyojy ⇤t^ DP 11:37b vAEjm@yE⌃dn\ tTA DP 10:1d vAt aA vAt B ⇤qjm^ AVi 40d vAmto vAmbAhk ⇤ AVi 37d vAmd ⇤&y\ c QC⌃dog, SP 7:37c vAmn\ c E/Evámm^ DP 4:6b, 11:14b vAmn\ cA£m ⇤ c {v DP 11:35c vAmp"\ t s⌃Dy ⇤t^ DP 12:11d vAmpAd\ Enyojy ⇤t^ DP 12:10d vAmm@y ⇤ kEná⌃tm^ AVi 38b vAm-tn ⇤ pzqgEtm^ AVi 36d vAm ⇤ kr ⇤ i⌃d˝nArym^ AVi 38d vAm ⇤ c {v t dAt&y\ DP 4:11c

Early Tantric Vais.n.avism vAm ⇤ Bj ⇤ vkAr\ t AVi 53a vAm ⇤ Bj ⇤ vAkAr\ t SP 7:7c vAy&yA\ c {v mADv, AVi 113b vAy&yA\ t hl\ ⌃ys ⇤t^ AVi 110d; DP 11:40b vAy&yA\ t hl\ ⌃ys ⇤t^ SP 7:30 vAy&yA\ EdEf EpΩlm^ DP 11:29d vAy - r^ - aMbryoEn(y\ DP 2:24a vAy - r^ - aMbryoyTA AVi 69d vArAhF coDtomKA DP 12:4b vAz⌫yA\ EdEf - m^ - aAEô(y SP 8:15c vAz⌫yA\ EdEf vtlm^ SP 5:10d vAz⌫yA\ íApr\ nAm DP 7:8c vAz⌫yA\ vtl\ k⌫X\ AVi 98c vAz⌫yA\ EstvZABA\ SP 5:23a; DP 6:29c vAs\ t pErDApy ⇤t^ DP 9:32b vAs⇤ dv\ jgãzm^ DP 4:6d vAs⇤ dv-t „dy ⇤ DP 11:22c vAs⇤ dvAy t ⇤ nm, DP 9:30d vAs⇤ dvo jgãz, DP 7:7b vAhn\ n kT⌃t-y SP 1:19a vAhn\ n pn-t-y DP 1:14a Ev Eh sotoErEt m⌃/ ⇤Z DP 8:11a Ev\fEtñ {v d˝&yAEZ DP 3:4c Ev\fm ⇤n t BAg ⇤n DP 3:26c EvkZ◆ v⇥qZ ⇤ ⌃ys ⇤t^ AVi 39f EvtE-t\ pvmAryFf⇥/BAgA-(v ⇤Et ? m⌃/ ⇤Z DP 8:13a sÇv\ rj-tmñ {v DP 7:5a sdA pEñmto yj ⇤t^ SP 7:37d sn(kmAr, ôFm⌃t\ DP 1:1c s\toqEvBvA(mn ⇤ DP 9:28b sÿ++mkA ⌃ys ⇤t^ DP 8:3d sÿEjüo mhAvFy, SP 8:3c sÿdfm\ r◊rcn\ AVi 114a sÿm\ -vrbFj ⇤n AVi 5:29 sÿm⇥EdEtm⌃/ ⇤Z DP 8:3c sÿm ⇤ vAmn\ c {v SP 7:26a sÿAΩlA p/kAmA DP 3:12a smAnF v, EfKA coƒA DP 12:8a smAvtn i(y ⇤v\ SP 7:46c smAsAç-t Ec⌃ty ⇤t^ SP 8:55d

211 sEmào a“ - m⌃/ ⇤Z DP 9:35a sEmào a“⌃k⇥dr\ AVi 90c smd˝\ gQC m⌃/ ⇤Z DP 9:16c smd˝\ gQC -vAh ⇤Et DP 9:15a smd˝A jWr ⇤ E-TtA, AVi 63b smd˝A jWr ⇤ -m⇥tA, DP 5:6b smd˝AçZvAE⌃vtm^ DP 12:9d sMBvAy { ’ Et gAy/F DP 11:24c s\pZ◆ svgA/ ⇤q DP 10:4a s\pZ, svgA/ ⇤q SP 6:1a sv◆ nArAyZA(mkm^ SP 2:10d; DP 2:8d sv◆ Ev Zmy\ jgt^ SP 2:11d; DP 2:9d sv◆ -vAh ⇤Et m⌃/ ⇤Z SP 7:48a svkAmPl˛dm^ DP 6:37b svkAmsm⇥ày ⇤ SP 8:19d svkAmsm⇥èT◆ SP 7:50c svg, Ek\ n Ev‚ ⇤y, SP 1:10c; DP 1:5c svt ⇤jomyo ⇤ dvo SP 8:3a sv/ vAy⇤ dvtA SP 8:7b svEd" smE⌃vtA, SP 8:38d sv⇤ d vnm-k⇥tm^ AVi 106b sv⇤ dyt ⇤ DP 2:19b sv⇤ d⇤ hq Ev⌃ys ⇤t^ AVi 55b; DP 11:17b svnAfA dfAΩlA DP 3:12d svpAp˛Zodnm^ SP 7:9b sv˛AZA(mko ⇤ dv, SP 8:7a svBtgt, E-Tt, DP 5:13b svBtEhtATAy DP 2:18c svBt ⇤ vvE-Tt, DP 2:13b svmA/ ⇤ jnAdn, SP 2:13d svm ⇤t>jnAdn ⇤ DP 2:11d, 2:15b svr⌃D ⇤ }q Ev⌃ys ⇤t^ AVi 46b

212 svl"Zm ⇤v c SP 1:18d; DP 1:13d svl"Zs\yƒ\ DP 3:29c svl"Zs\yƒA\ SP 5:8a svl"Zs\ytA DP 3:8d svl"Zs\ytAm^ DP 6:10b svl"Zs\p‡\ DP 1:15 svl"Zs\pZ◆ SP 1:20c svEvçAEDpo ⇤ dv, SP 8:9a svfA-/\ mK ⇤ t-y DP 5:4a svfA-/myF ⇤ dvF AVi 61a svfA-/EvfArd DP 2:1b svfA-/ ⇤q coÅmm^ SP 2:2b svfA-/ ⇤q p’t ⇤ SP 2:6b; DP 1:2b svhA íAdfAΩlA DP 3:13b svA, ôo/ ⇤ &yvE-TtA, AVi 62b svA"rmy\ q§\ AVi 13a svA(mA sklo ⇤ dv, SP 7:19a svAvyvs\pZ◆ DP 4:5a sv ⇤ ⇤ dvA-tdA(mkA, AVi 68d s&yp"\ t d^ - aAlB ⇤t^ DP 12:12b ssvZ◆ -vh-t ⇤n SP 8:13c shd ⇤vA blA c {v DP 9:22c sh˙fFqm⌃/ ⇤Z DP 9:38c sh˙fFqyAcAyA\ DP 13:1a sh˙fFqA pzq, SP 7:2a sA\v(srEmd\ pFW\ DP 12:12c sA EáyA En PlA ˛oƒA DP 7:22c sA t >y ⇤§A ˛kFEttA DP 3:16d sA"FBto jnAdn, DP 2:13d sA(ykF\ coÅr ⇤Z t AVi 108b sAE‡@ykrZATAy DP 9:27c sAmv ⇤dmTvZm^ DP 2:10b sAmv ⇤do - r^ - aTvZ, SP 2:12b sAMb\ dçAŒ vAy&yA\ AVi 108a sADh-t\ E/flkm^ DP 6:18b sAEv/F EvjyA -m⇥tA DP 11:24d sAEv/FmFòr ⇤ -vym^ SP 7:21b sAEv/FmFfgocr ⇤ DP 11:29b sAEv/FfAngocr ⇤ AVi 105b Es\hF &yAG }F tT {v c AVi 81e, DP 9:23b

Early Tantric Vais.n.avism EsàcArZp‡gA, SP 8:12b EsàEvçADr {, -ttm^ SP 1:2d EsàATkA"top ⇤tAn^ DP 9:11c sFm⌃t\ c pn-t ⇤qA\ SP 7:43a skAr\ kEVm@y ⇤ t DP 11:5a skAr\ t/ Ec⌃ty ⇤t^ SP 7:13d; DP 11:10d skAr\ vAmj\GAyA\ AVi 53c, SP 7:8a skAr ⇤Z t pjy ⇤t^ SP 7:27b skAr ⇤Z mhA(mnA DP 11:36b sg⌃D {, p pdAmk {, SP 8:42d s/AmAZ ⇤Et m⌃/ ⇤Z DP 8:9c sdfn ⇤q c Ev⌃ys ⇤t^ AVi 47b spt ⇤(y`n ⇤, SP 7:42a srEk‡rg⌃Dv {, SP 1:2c s⇤ rK\ t smEå£\ DP 12:8c svZ-y flAkyA DP 9:9b sfMbr\ dFEÿm⌃t\ SP 8:41a sEö£\ c {v kt&y\ DP 4:1c ssmAÿ\ g⇥h\ y-y SP 8:57c s#m ev prApr, DP 2:17d s/DAro mhA˛A‚, DP 9:6a s/ ⇤Z sEmt\ ámAt^ DP 6:9d sykAE⌃t c⌃d˝kAE⌃t AVi 115a sy ñ {v t c⌃d˝ñ DP 2:14a sysAm llAV ⇤ t AVi 31a syAvrZmAdAy DP 6:9c s⇥E£v⇥EàmEtl#mF, AVi 9a s⇥E£s\hArgocrm^ SP 1:6d soÅrFy\ t kAry ⇤t^ DP 9:32d somrAjAy v { nm, SP 8:8d somrAj ⇤ Enyojy ⇤t^ SP 8:8f soms\-TAmhAy‚ {, SP 2:13a soms\-T {mhAy‚ {, DP 2:11a som-TAn ⇤ ⌃ys ⇤à˝vm^ DP 6:13b somo ddã⌃Dv ⇤Et DP 8:11c somo b⇥h-pEtñ {v SP 5:3c; DP 6:4c somo rAjA ˛kFEtt, SP 8b soMyAyA\ sqpA\ñ {v SP 8:21a

P¯ ada Index sOMyd⇥E£\ fB ⇤"Zm^ DP 4:5b sOMy ⇤ kElyg\ nAm DP 7:9a sOrñ hnt ⇤ g }Am\ DP 10:11c sOr ⇤ BAgvt ⇤ tTA AVi 47d -k⌃DmEt◆ c s⌃Dy ⇤t^ DP 13:8d -k⌃DO t-yAEp foBnO DP 3:23b -kMBsƒ\ tT {v c SP 7:38b -tnkO nAEBm⌫Xlm^ SP 1:17b; DP 1:12b -tnkO pErkFEttO DP 3:24b -tMB\ t pErkSpy ⇤t^ DP 8:8d -tMB-TA, pvtA y"A, DP 7:10c -tMB {, qoXfEByƒ\ DP 6:12a -trZ\ -TApy ⇤ìD, DP 8:13b -t(vA nAmsh˙ ⇤Z SP 1:4c -TAnB }\fo rA£~B }\f, DP 8:25c -TAn ⇤ -TAn ⇤ syojy ⇤t^ AVi 73b -TApk, smdA„t, SP 6:1d -TApk, sshAy-t DP 9:1c -TApk-t smArB ⇤t^ DP 9:10d -TApk-y gZAn^ v#y ⇤ DP 10:3a -TApk-y c Ec°AEn SP 1:25a; DP 1:17c -TApk ⇤n t kt&y\ DP 11:1c -TApn\ cA£Adfmm^ AVi 119 -TApn\ t yTAámm^ SP 8:1b -TApn\ s\˛v#yAEm DP 2:26c -TApy ⇤Œ sm⌃tt, SP 5:11b -TApy ⇤⇤ dv ˙?˙vO DP 8:20d -TApy ⇤⇤ íEd - m^ - ag }t, SP 5:26d -TApy ⇤⇤ íEdm@yt, DP 6:33d -TAEpt-y t ⇤ dv-y AVi 97a -TAEpt-y t ⇤ dv ⇤f-y AVi 119a -TAlFpAk\ t -TApy ⇤t^ DP 8:18d -TAlFpAk ⇤ t gAy/F DP 7:17a -TAvr ⇤ jΩm ⇤ c {v DP 2:15a ‹Ankm - m^ - id\ ˛oƒ\ DP 9:40c ‹AnkmEvEDyTA DP 9:1b ‹AnmoqEDm>jl {, AVi 83b; DP 9:23d ‹Apy ⇤(klf\ g⇥ù DP 9:6c

213 E‹`D\ -vAd c Ev Zv ⇤ SP 5:27d -PEVk\ t ˛dAt&y\ SP 8:19c -PAEVt\ c EvhFn\ c DP 4:3a -PAEVtA nAfy ⇤dT◆ DP 4:4a -ym ⇤KlAEDd ⇤vtA, DP 7:5b ˙À ⇤ d˝=s ⇤Et m⌃/ ⇤Z DP 8:9a ˙£A ⇤ dvo mh ⇤òr, SP 1:6b ˙£A y, svBtAnA\ SP 8:11a ˙cAg }Ad^ íAdfAΩlAt^ DP 7:18b ˙cAg { }kAdfA zd˝A, DP 7:18a -vd ⇤⇤ h ⌃yAsmÅmm^ DP 11:1d -vnAmm⌃/ {vOqX⌃t {, DP 7:20c -vyMBvcn\ yTA DP 5:13d -vr⌃yAsmnÅmm^ AVi 18b -vE-t no mFEt m⌃/ ⇤Z DP 12:10a -vE-t c ⇤Z bEàmAn^ SP 7:42d -vE-tkBEqtm^ SP 5:16d -v-TAnvEjtA, k⌫XA, DP 6:15a -v-v-TAn ⇤ Env ⇤fy ⇤t^ DP 6:15d -vAhA ˛AZ ⇤ t tAlk ⇤ AVi 34d -vAhAkAr\ ˛Et§y ⇤t^ DP 8:21b -vAhAkArvq÷ArO SP 2:14a; DP 2:12a -vAhAkAr ⇤ E-TtA ⇤ dvA, DP 7:18c -v ⇤ -v ⇤ -TAn ⇤ E-Tt\ mn ⇤ DP 7:20b hkAr\ vAmto ⌃ys ⇤t^ AVi 28b hErtA\ coÅr ⇤ dçAt^ DP 6:30a hErtAmÅr ⇤ c {v SP 5:23c hErtAl\ t dApy ⇤t^ AVi 118d, SP 8:22d hErtAl\ mn,EflAm^ AVi 116a hl\ DnEq s\ytm^ DP 4:12d hEvy‚ {-tT {v c SP 2:13b; DP 2:11b h-t\ h-tA⌃tr ⇤ kyAt^ DP 6:18c h-tmA/˛mAZ ⇤n DP 9:2c hE-tnA fkV ⇤nAEp DP 9:4a h-t ⇤ t dE"Z ⇤ _Ω§ ⇤ AVi 22a h-t ⇤ pnvsO cAEp SP 5:5a; DP 6:6a Ehmk⌃dEnB\ c {v SP 7:13a EhmvE(kESbq\ hE⌃t AVi 82c Ehr⌫my ⇤n pA/ ⇤Z DP 9:31a

214 Ehr⌫ygBm⌃/ ⇤Z DP 9:9a Ehr⌫ygBm-/ ⇤q AVi 47a Ehr⌫ygB ⇤Et m⌃/ ⇤Z AVi 89c hFnAΩ\ k⌫Xgolkm^ DP 10:9b hFnAΩA mrZ\ Bv ⇤t^ DP 4:4b h(vA£ftm ⇤k\ t DP 9:12a h\kAr ⇤Z t pjy ⇤t^ SP 7:19d h\kAr ⇤Z mhA(mnA DP 11:27d „EQCr-t EfKA c {v DP 11:20c „(vA km smArB ⇤t^ SP 5:6d „dy\ t-y smAHyAt\ AVi 103c „dy ⇤ íO bAh gù ev Eh AVi 71b „dy ⇤ Ev⌃ys ⇤Eí Z\ AVi 49a; DP 11:13c „dy ⇤ ôFrEDE§tA AVi 65d; DP 5:8d „dy ⇤nsmAcr ⇤t^ AVi 100d ⇤ hmk⌃dEnBAkAr\ DP 11:10a ⇤ hmgBsmE⌃vtAn^ SP 5:16b; DP 6:22b

Early Tantric Vais.n.avism ⇤ hmvZ◆ t Ec⌃ty ⇤t^ DP 11:8d ⇤ hmvZ◆ EvEc⌃ty ⇤t^ SP 7:11d hot&y\ cAnpvf, AVi 101d hot&yAEn yTAámm^ SP 7:48b hom\ fAE⌃t\ +++++ SP 8:35b ++EvE⌃d ? smAyƒ\ AVi 116c +gSPO íO t kSpy ⇤t^ AVi 43b ++msAm nAsAg ⇤ } AVi 31c +++d ⇤vt {, m⌃/ {, SP 8:24c ++++ - aAv!pAy SP 8:47a +++++aA\ yƒ\ SP 8:54a +++++KA\ sdA AVi 39b +++++mA£km^ SP 8:57b /// m⌃/ ⇤Z AVi 91c /// mcnm^ SP 1:3d /// Em(yAh, SP 8:20c

Index of Tantric Mantras and other Prose Lines

a`ny ⇤ t-y gAy/F t-y m⌃/,. SP 8:3 aE`nmDA iEt m⌃/ ⇤Z d`Dodk ⇤n ‹Apy ⇤t^ AVi 84.ii aE`nd ⇤vtA vAto ⇤ dvtA iEt m⌃/ ⇤Z ⇤ dv\ c {v t pjy ⇤t^ AVi 88.ii aA`n ⇤⇧yA\ EdfABAg ⇤ aAcAyo _E`nkAy◆ ˛ArB ⇤t^ AVi 99.i aA no Bd˝A iEt m⌃/ ⇤Z ˛AsAd-TAn\ ˛v ⇤fy ⇤t^ AVi 96.i im\ m ⇤ gΩ ⇤ ymn ⇤ ’ Et m⌃/ ⇤Z nçodk ⇤n ‹Apy ⇤t^ AVi 85.ii Uz vAm ⇤Et EvB }AX^ b⇥h(sAm\ dE"Z ⇤ jAn. l>jA Eíj vAmjAnk ⇤ AVi 42 : : nArAyZAy nm,. AVi 29 : a\ ˛çflAy nm,. AVi 29 : a, aEnzàAy nm,. AVi 29 : akAl ⇤y ⇤§\ n aA Br RV 6.46.5 AVi 44 i⌃d˝EmãAETno b⇥ht^ ( y⌫vkm^ ) SV 1.198 DP 13:1 i⌃d˝Ay gAv aAEfrm^ RV 8.69.6 DP 8:14 i⌃d˝Ay sAm gAyt RV 8.98.1 AVi 39 i⌃d˝AyAEh Ec/BAno VS 20.87 DP 13:7 i⌃d˝AsomA tpt\ r" uNjtm^ RV 7.104.1 DP 8:8 im\ n sommE⌃tt, RV I.179.5 AVi 85; DP 9:15 im\ m ⇤ gΩ ⇤ ymn ⇤ RV 1.75.5 AVi 41 imA\ vA\ RV 7.36.2 SP 5:14; DP 6:20 iq ⇤ (voj ⇤ (vA VS 1.1 DP 13:8 ih rEt VS 8.51 SP 5:35, 8:8 IfAvA-yEmd\ svm^ VS 40.1 SP 7:1 ut vAt EptAEs n, RV 10.186.2 DP 8:10 uÅAnpZ ⇤ sBg ⇤ RV 10.145.2 AVi 92; DP 9:38 uEŧ b }úZ-pt ⇤ VS 34.56 DP 8:17 ud Et§ -v@vr VS 11.41 DP 13:2 uíy\ tms, pEr RV 1.50.10 AVi 31 u‡yAEm KS 11.8 DP 8:12 uBy\ t ⇤ n "Fyt ⇤ vy-ym^ RV 2.9.5 ¯ G¯ AVi 41 Uzíy SV A. a. 266-267 AVi 34; DP 8:5 c\ vAc\ ˛pç ⇤ VS 36.1 DP 13:2 etA aqE⌃t RV 4.58.5 AVi 39 eto E⌃v⌃d˝ -tvAm fàm^ ( fàv(y, ) (SV 1.350 SP 7:45 eq Ev˛ {rEB£t, RV 9.3.6 DP 12:11 eq -tomo mh ug }Ay vAh ⇤ RV 7.24.5 AVi 97, 119 eqo h ⇤ dv ˛Edf VS 32.4 SP 7:35 e ⇤⌃d˝A`nm^ RV 8.40 AVi 38, 76 k\kto n k\kt RV 1.191.1

Vedic Mantra Index AVi 45 kd˝d˝Ay ˛c ⇤ts ⇤ RV 1.43.1-9 AVi 38 kEná⌃tm^ = kEná\Et hErrA RV 9.95.1 = SV 1.530 AVi 78 kyA nEñ/ VS 36.4/27.39 AVi 76, 93 kA⌫XA(kA⌫XAt^ VS 13.20 DP 13:6 kAlF krAlF c mnojvA MU 1.2.4 DP 12:3 kq\Bk-tdb }vFt^ RV 1.191.16 DP 8:20 k⇥ Zo-yAKr ⇤§o VS 2.1 DP 12:5 g⌃tA no y‚m^ RV 5.87.9 AVi 79; DP 9:19 g⌃DíArA\ drADqA◆ RV khila. 5.87.9 DP 13:5 gAyE⌃t (vA RV 1.10.1 AVi 35 gAEgb }AúZm^ ?? DP 13:5 gAvEñäA sm⌃yv, RV 8.20.21 DP 12:4 g⇥ZAnA jmdE`nnA RV 3.62.18 DP 13:9 Gm ⇤v mD jWr ⇤ sn ⇤z RV 10.106.8 DP 9:24 G⇥tvtF BvnAnAmEBEôyA VS 34.45 AVi 94; DP 8:6 c(vAEr f⇥ΩA-/yo a-y pAdA, VS 17.91 DP 8:15 cr⌃v { mD AB 7.15.5 DP 9:9 Ec/\ ⇤ dvAnAmdgAdnFk\ VS 7.42 AVi 86 jnrƒA pErDAmAn, ?? AVi 37 >y ⇤§sAm SV ?? AVi 34, 99 td ⇤vAE`n-tdAEd(y-tíAy-td c⌃d˝mA, VS 32.1 AVi 41 tío gAy ( zd˝sAm ) SV 1.115 DP 13:10 tnrEs KS 2.1 = AB 8.27.4 AVi 37 tmFfAn\ VS 25.18 DP 13:6 tm £vAm y\ Egrm^ RV 8.95.6 AVi 45 tr(s m⌃dF SV 1.500 DP 8:14 tBÅA aEΩr-tm VS 12.116 DP 13:7 tEvg }AvA RV 8.17.8/5.2.12 AVi 80; DP 9:20 t ⇤jo _Es fámm⇥tmAy pA, VS 22.1 AVi 39, SP 8:2 /AtArEm⌃d˝m^ RV 6.47.11/VS 20.50 DP 12:10 /FEZ jAnA pEr Bq⌃(y-y RV 1.95.3 AVi 39, SP 7:47 ÏyMbk\ yjAmh ⇤ sgE⌃D\ VS 3.60 AVi 79; DP 9:20 dEDáANZo akAErqm^ VS 23.32

221

222

Early Tantric Vais.n.avism

AVi 77; DP 9:37 dFGAy(vAy b⇥ht ⇤ AV 2.4.1 ¯ G¯ AVi 43 ⇤ dvv }t ⇤ SV A. a. 212–14 SP 8:26 ⇤ dv sEvt, ˛sv y‚\ ˛sv y‚pEt\ VS 9.1 AVi 80; DP 9:20 ⇤ dv-y (vA VS 1.10 DP 9:36 ⇤ dvA gAtEvdo gAt\ EvÇvA gAtEmt VS 2.21 DP 9:24 d˝pdAEdv mmcAn, VS 20.20 AVi 96; DP 8:16 D⌃vnA gA, VS 29.39 DP 12:5 DAm⌃t ⇤ Evòm^ RV 4.58.11 AVi 89; DP 9:34 DrEs Dv Dv⌃t\ VS 1.8 AVi 39, SP 7:2 nm, fMBvAy ( fAMBvm^ ) VS 16.41 DP 8:23 nm-t ⇤ zd˝ m⌃yv ⇤ VS 16.1 SP 8:10 nmo _-t sp ⇤jAg }to drm{ pEt VS 34.1 SP 7:44; DP 9:17 y‚Ay‚A vo VS 27.42 or y‚ y‚\ VS 8.22 DP 8:22 y‚ ⇤n vDt jAtv ⇤dsm^ RV 2.2.1 DP 12:7 yÅ ⇤ rAj\ QC ⇥ t\ hEv, RV 9.114.4 SP 7:46 yT ⇤mA\ vAc\ kSyAZFm^ VS 26.2 DP 8:18 yEd⌃d˝ ˛AgpAgdk^ RV 8.4.1 SP 8:4 ymAy (vA mKAy (vA VS 37.11 AVi 84; DP 9:22 yA, PElnFyA aPlA, RV 10.97.15 SP 8:44 yA t ⇤ zd˝ EfvA tn, VS 16.2 AVi 83; DP 9:23, 9:33 yA aoqDF, pvA jAtA, VS 12.75 AVi 88 y“t ⇤ mn ut y“t ⇤ VS 5.14/11.4/37.2 DP 9:35 y“E⌃t b }msrm^ VS 23.5 DP 12:11 y“ATA\ rAsB\ rT ⇤ RV 8.85.7 AVi 87 yvA svAsA, pErvFt aAgAt^ RV 3.8.4 AVi 32 y ⇤n ⇤d\ Bt\ ? VS 34.4 SP 5:25; DP 6:32 yog ⇤ yog ⇤ tv-tr\ VS 11.14 DP 12:6 yo n, -vo arZo yñ En˜o RV 6.75.19 AVi 46; DP 9:5 rT ⇤ Et§‡yEt vAEjn, RV 6.75.6 AVi 35 zd˝A@yAy, VS 16 AVi 37 vzZsAm SV ?? DP 8:8 vEs§A Eh Emy ⇤@y RV 1.26.1 SP 7:34 vAEckm^ RV 10.125 AVi 41 vAt aA vAt B ⇤qj\ SV 5.7.184.1 ¯ G¯ AVi 38 EvdA mGvn^ ( mhAnAflF ) SV A. a. Pari. 1.1 DP 8:22 Ev˛\ Ev˛Aso _vs ⇤ RV 8.11.6 AVi 40, 42 EvB }AX^ b⇥ht^ ( EvkZsAm = b }úsAm ) VS 33.30

224

Early Tantric Vais.n.avism

SP 7:3; DP 9:40 Evòtñ"zt EvòtomK, RV 10.81.3/VS 17.19 DP 9:26 Ev Zsƒm^ RV 1.154.1ff, but there are only 6 stanzas. DP 9:25 Ev Zo rrAVmEs VS 5.21 DP 8:11 Ev Eh sotors⇥"t RV 10.86.1 AVi 85; DP 9:32 v ⇤dAhm ⇤t\ pzq\ mhA⌃t\ VS 31.18 DP 8:3 v ⇤Edqd ⇤ E˛yDAmAy sçt ⇤ RV 1.140.1 SP 5:24; DP 6:31 v ⇤çA v ⇤Ed, smA=yt ⇤ VS 19.17 SP 7:35 v⇥qAkEp RV 10.86 AVi 75 v }Fhyñ VS 18.12 SP 7:43 f\sA mhAEm⌃d˝\ yE-mn^ EvòA, RV 3.49.1 AVi 77, SP 5:15; DP 6:21, 8:5 f‡o ⇤ dvFrEB£y ⇤ RV 10.9.4/VS 36.12/AVP 1.1.1 AVi 35 fEáym^ TA 4.1–42 AVi 87; DP 9:34 ôFñ t ⇤ l#mFñ VS 31.22 AVi 30, 7:34 ôFsƒm^ RV khila. 2.6 AVi 35 öokA@yAy, No definite clue but VS 20 is a candidate DP 13:8 s\á⌃d˝n ⇤nAEnEmq ⇤Z RV 10.103.2/VS 17.34 DP 12:8 smAnF v, RV 10.191.4 AVi 90; DP 9:35 sEmào a“⌃k⇥drm^ VS 29.1 DP 9:15, 9:16 smd˝\ gQC -vAhA VS 6.21 SP 7:2; DP 9:38, 13:1 sh˙fFqA pzq, RV 10.90.1/VS 31.1 SP 7:42 sptp, ptB⇥Et KS.34.16 DP 8:9 s/AmAZ\ p⇥ETvF\ çAmn ⇤hsm^ RV 10.63.10 AVi 31 sysAm SV 5.7.177.1 DP 8:11 somo ddã⌃DvAy RV 10.85.41 AVi 36 -kMBsƒm^ AV 10.7.1-44 DP 8:9 ˙À ⇤ d˝=s-y Dmt, RV 9.73.1 SP 7:42 -vE-t n i⌃d˝o v⇥àôvA, RV.1.89.6a/VS.25.19 DP 12:10 -vE-t no EmmFtAEònA Bg, RV 5.51.11 AVi 34, SP 7:44 -vAhA ˛AZ ⇤