Buddhist Literature of Ancient Gandhara: An Introduction with Selected Translations

CLASSICS OF INDIAN BUDDHISM

The flourishing of Buddhism in South Asia during the first millennium of the Common Era produced many texts that deserve a place among the classics of world literature. Exploring the full extent of the human condition and the limits of language and reason, these texts have the power to edify and entertain a wide variety of readers. The Classics of Indian Buddhism series aims to publish widely accessible translations of important texts from the Buddhist traditions of South Asia, with special consideration given to works foundational for the Mahāyāna.

Editorial Board

Andy Rotman (chair), Smith College

Paul Harrison, Stanford University

Jens-Uwe Hartmann, University of Munich

Sara McClintock, Emory University

Parimal Patil, Harvard University

Akira Saitō, University of Tokyo

Discover the fascinating story of a long-hidden Buddhist culture at a historic crossroads.

FOLLOWING ALEXANDER THE GREAT’S conquest of the East, a series of empires rose up along the Silk Road. In the region called Gandhāra, in what is now northern Pakistan, new kingdoms flourished and became increasingly important centers for the development of Buddhism, reaching their apex under King Kaniṣka of the Kuṣāṇas in the second century CE. Gandhāra has long been known for its Greek- Indian synthesis in architecture and statuary, but until about twenty years ago, almost nothing was known of its literature. Manuscripts unearthed during the last few decades show that Gandhāra was a vital link in the early development of Buddhism, crucial to both the rise of the Mahāyāna tradition and the transmission of Buddhism to China. The Buddhist Literature of Ancient Gandhāra surveys what we know about Gandhāra and its Buddhism, and it also provides translations of a dozen different short texts, from similes and stories to treatises on time and reality.

In this remarkable book Richard Salomon makes accessible to general readers and specialists alike the fruits of more than two decades of study. This is a stunning achievement, giving the reader a glimpse of a lost world of Buddhist thought and practice. It belongs on the bookshelf of every reader interested in the history of Buddhism.

— Jan Nattier, author of A Guide to the Earliest Chinese Buddhist Translations

This book opens a window on an early phase of Buddhist literary history that only a century ago seemed lost forever. It is a testament to the indefatigable efforts of a small band of scholars, led by Richard Salomon himself, who have toiled to make this precious material accessible to us.

— Bhikkhu Bodhi, author of In the Buddha’s Words

With a detective’s eye and a storyteller’s pen, Professor Salomon shares a scholar’s delight in these Silk Road texts.

— Hozan Alan Senauke, Berkeley Zen Center

Reveals not only the significance but also the beauty of this literature.

— Ingo Strauch, University of Lausanne

This book is dedicated to the memory of Carol Goldberg Salomon

(July 28, 1948–March 13, 2009)

זיכרונה לברכה

Publisher’s Acknowledgment

The publisher gratefully acknowledges the generous help of the Hershey Family Foundation in sponsoring the production of this book.

Contents

List of Illustrations

Preface

Introduction: The Rediscovery of Gandhāran Buddhist Literature

PART I. CONTEXTS

1.The World of Gandhāran Buddhism

Gandhāra and India’s Northwest Frontier

The Bloody History of Paradise

Early Buddhism and Gandhāra

Aśoka and the Mauryan Empire

The Indo-Greeks

The Age of the Scythian Kingdoms

The Climax of Gandhāran Buddhist Culture under the Kuṣāṇa Empire

The Decline of Buddhism in Gandhāra

The Legacy of Gandhāran Buddhism

2.Buddhist Manuscripts, Buddhist Languages, and Buddhist Canons

Buddhist Texts and Canons

The Languages of Buddhism

Gāndhārī and Kharoṣṭhī

Deciphering Gāndhārī Documents

The Destiny of Gāndhārī and the Triumph of Sanskrit

The Gāndhārī Hypothesis

Further Discoveries of Gandhāran Manuscripts

The Character of the Scrolls

3.The Buddhist Literature of Gandhāra

The Scope of Gandhāran Buddhist Literature

Canonical and Paracanonical Sūtras

Vinaya Texts

Abhidharma and Scholastic Literature

Edifying Narratives

Mahāyāna Texts

Miscellaneous Texts

What Did Gandhāran Buddhists Read?

Oral and Written Texts and Canons

Was There a Gāndhārī Canon?

The Problem of School Affiliation

PART II. TEXTS

Sūtras in Prose

1.Three Numerically Grouped Sūtras

a.The Buddha and the Brahman Dhoṇa

b.The Words of the Buddha

c.The Four Efforts

2.Five Thematically Grouped Sūtras

a.The Four Concentrations

b.Not Yours

c.Living Full of Disenchantment

d.The Adze Handle

e.The Parable of the Log

Poetic Texts

3.The Rhinoceros Sūtra

Translation

4.A Chapter from the Dharmapada

The Monk

Legends and Previous-Life Stories

5.Songs of Lake Anavatapta

a.Mahākāśyapa

b.Nanda

c.Śroṇa Koṭiviṃśa

g.Yaśas

e.Piṇḍola Bharadvāja

f.Vāgīśa

g.Nandika

h.Kusuma

6.Six Stories of Previous Lives and Other Legends

a.The Story of a Rich Man

b.The Previous Life of the Bodhisattva as a Merchant

c.The Previous Life of the Bodhisattva as Prince Sudaṣṇa

d.The Previous Life of Ājñāta Kauṇḍinya as a Potter

e.The Previous Life of Ānanda as a Prince

f.The Monk and the Saka

7.Avadāna Legends

a.The Contest between the Black and White Magicians

b.The Story of Zadamitra

8.The Many Buddhas Sūtra

Translation

Scholarly Commentaries and Debates

9.A Commentary on the Sūtra of Chanting Together

a.The Five Faculties

b.The Six Roots of Argument

10.A Commentary on Canonical Verses

a.Trade What Ages

b.Endowed with Proper Conduct

c.An Angry Man

11.An Abhidharma Treatise on Time and Existence

Translation

The Emerging Mahāyāna

12.The Perfection of Wisdom Sūtra

a.The Practice of the Knowledge of All Forms

b.The Teaching on Merit

Conclusions

State of the Art and Future Prospects

What Have We Learned?

Appendices

1.Specimen of a Verse in Various Buddhist Languages

2.Specimen of a Gandhāran Buddhist Inscription: The Reliquary of Śatruleka

Notes

Glossary

Bibliography

Image Credits

Index

About the Author

List of Illustrations

Figures

1.The clay pot (British Library pot D) that contained the British Library Kharoṣṭhī scrolls

2.The British Library scrolls in their original position inside the clay pot

3.British Library fragments 16–19 before unrolling and conservation

4.British Library conservators unrolling the scrolls

5.A Gandhāran birchbark manuscript unrolled. Senior scroll 19, recto, The Parable of the Log (translation 2e)

6.The Manikiala stūpa

7.A coin of Menander

8.The Shinkot reliquary

9.The silver reliquary of Prince Indravarma

10.The Tīrath footprint inscription

11.Statue of Kaniṣka from Mathurā

12.Donor figures with Central Asian dress

13.The Trojan horse in a Gandhāran sculpture

14.Herakles/Vajrapāṇi from Haḍḍa, Afghanistan

15.Roman glass from the Begram treasure: the rape of Europa

16.A coin of King Kaniṣka

17.Rock carvings and inscriptions at Hodar on the upper Indus River

18.One of 729 inscribed slabs of the Pali Tipiṭaka at Kuthodaw Pagoda, Mandalay, Burma

19.Fragment of a palm-leaf folio from Bamiyan: the Great Nirvāṇa Sūtra

20.Senior scroll 24 as found, rolled up and folded in half lengthwise

21.A portion of British Library fragment 1, the Songs of Lake Anavatapta, showing the blank area between segments that were originally glued together

22.The pot in which the Senior scrolls were found, and its lid

23.Senior scroll 8: the index scroll

24.Upper right corner of Senior scroll 5, with the concluding phrase of the first sūtra added in the margin

25.Senior scroll 5 before unrolling

26.Senior scroll 5 after unrolling and conservation

27.Kizil cave painting of the Sūtra of the Log

28.An Indian rhinoceros

29.The Rhinoceros Sūtra scroll: the debris box

30.A portion of the Khotan Dharmapada scroll

31.The Songs of Lake Anavatapta: Portion of British Library scroll 1 before conservation

32.The Songs of Lake Anavatapta: British Library scroll 1 as originally found in the pot

33.The Buddha and the sleeping women in a Gandhāran sculpture

34.The Viśvantara-jātaka on a stair riser from the Jamālgaṛhī stūpa

35.A Gandhāran relief of the Dīpaṅkara-jātaka

36.Base of a caitya (shrine) from eastern India with rows of Buddha images

37.Section of British Library scroll 9 showing the text translated in sample a

38.The inscribed reliquary of Śatruleka

Maps

Map 1. Greater Gandhāra

Map 2. The World of Early Buddhism

Preface

THIS BOOK IS a distillation of the results of twenty years of concentrated work by my many collaborators and me on the Gandhāran Buddhist manuscripts that were discovered during that period. These manuscripts and fragments, which now number in the hundreds, date from the early centuries before and after the beginning of the Common Era. Written in the Gāndhārī language in Kharoṣṭhī script, they have brought to light the previously unknown literature of Gandhāra, a major center of early Buddhism in the northwestern frontier of the Indian subcontinent. Gandhāra had previously been familiar mainly from archaeological remains, especially its world-renowned tradition of sculpture.

Until now, our findings have been published mostly in scholarly books and articles designed for an audience of specialists in academic Buddhist studies and Indian linguistics. A preliminary book, Ancient Buddhist Scrolls from Gandhāra: The British Library Kharoṣṭhī Fragments (Salomon 1999), introduced the first group of Gandhāran manuscripts to be discovered in a semi-technical format, while scholarly editions and translations of individual manuscripts from this and other collections have been published by the University of Washington Press in the volumes of the Gandhāran Buddhist Texts (GBT) series. These books are not intended for general readers. They are meant to enable those who work with Buddhist literature in more widely known languages such as Sanskrit and Pali to understand the Gāndhārī texts.

But almost from the very beginning of this project, I have also been aware of our obligation to avoid burying these discoveries in technical publications for specialists and to meet what I consider a scholar’s duty to present new knowledge to interested readers in an accessible format. This book is an attempt to meet that obligation by making the newly found Buddhist literature of ancient Gandhāra accessible to a wider audience, be they students of Buddhism or readers with a general interest in ancient religions, languages, or literature.

In addition to translating and explaining the contents of the manuscripts themselves, I have tried to give the reader some sense of the scholar’s agonies and ecstasies involved in studying fragmentary, decrepit manuscripts that have lain unread for nearly two millennia. Each of these translations is a distillation of countless hours of labor by one or, more often, several scholars. In some places, I have presented a degree of technical explanation of the methods and techniques of decipherment to give the reader a sense of not only what has been done but also how and why.

The texts presented here were selected on a variety of criteria. First, an attempt has been made to sample each of the main genres of Buddhist texts now known in the Gāndhārī language. These are grouped together under five main rubrics: prose sūtras, poetry, legends and stories about previous lives, scholarly treatises, and Mahāyāna literature. Within these categories, the specific texts were chosen on the basis of factors such as the amount of material preserved in the fragments, legibility and accessibility, and overall interest and significance. Most of the texts have either been previously published in scholarly editions or at least studied in detail by my colleagues and myself, though not yet fully published. Some of the texts (nos. 1–3, 5–6, 8, 12) are presented in full; these are generally the shorter and/or better-preserved specimens. The rest are representative selections from longer texts, most of them not previously published. From these, I have chosen passages that are relatively comprehensible, interesting, and representative of the text as a whole.

Readers will notice, and perhaps be surprised, that in many cases the introductory material and commentary is longer — sometimes much longer — than the translated text itself. This is a product of the special character of this literature. Nearly all of the texts are incomplete, and in many cases only a small fraction of the entire work survives. I have therefore tried to make up the deficiencies in the surviving material by explaining in some detail their meaning, their context within Buddhism in Gandhāra and beyond, and especially their relationships with parallel or similar texts in other Buddhist languages and traditions that help to clarify their meaning and importance. I hope that in doing so I will be contributing to, rather than distracting from, the reader’s understanding and appreciation of the texts themselves.

Although nearly half of the texts presented here have already been translated in GBT volumes or elsewhere, the translations have been completely revised here and transformed into what I hope will be a more natural and readable style. The previous translations were designed primarily to help scholars and specialists compare them with the original texts and the parallels in other languages, and therefore they are strictly literal, with no pretension to literary qualities or readability. In the new translations I have, however, still stayed as close as possible to the structure if not the individual words of the originals, in the hope of giving the reader a sense of their style and rhythm. Thus, for example, I have translated in full all of the repetitions, sometimes quite extensive, that are so characteristic of Buddhist literature; see, for example, the introduction to translation 12.

Two of the translations presented here (nos. 3 and 5) are based on my own previously published work, and two others (8 and 12) have not been translated before. The rest of the translations are revised, in whole or in part, from the work, published or unpublished, of their original editors, namely Mark Allon (1), Stefan Baums (10), Collett Cox (11), Andrew Glass (2a), Meihuang Lee (2b), and Timothy Lenz (4, 6, and 7), on whose labors I have heavily relied and to whom I am heavily indebted.

In a deeper sense, though, all of the work in this and in the prior publications has been essentially collaborative in nature, representing, for the most part, the combined efforts of members of the research group of the University of Washington Early Buddhist Manuscripts Project (EBMP, formerly called British Library / University of Washington Early Buddhist Manuscripts Project), which was constituted in 1996 to sponsor and coordinate the publication of the then newly discovered Gandhāran manuscripts. The founding members of the group were Professor Collett Cox, the then-graduate students Timothy Lenz and Jason Neelis, and myself, who were joined in the following years by Mark Allon, Stefan Baums, Andrew Glass, Meihuang Lee (Tien-chang Shih), Joseph Marino III, Michael Skinner, and Fei Zhao, among many others. From the very inception of the project and up to the present, members and visitors have attended weekly meetings every Friday afternoon of what has come to be known as the Kharoṣṭhī Klub, in which the texts are read and interpreted in a collaborative effort. Thus the principal author of the definitive edition of each manuscript serves as the leader, compiler, and final authority, but the results reflect, to a considerable extent, the work of all of the members. The work is slow and painstaking; it is not unusual for a single difficult or incomplete word to be discussed for several hours as we explore all possible options for its interpretation, which usually involves searching through a multitude of Buddhist texts in a variety of languages, with each participant contributing on the basis of his or her area of expertise. Sometimes a long afternoon’s work yields only the solution to a single problematic word, but even this much can be a satisfying experience for all concerned.

I therefore must express my debt and appreciation not only to the primary authors mentioned above but to all the many others — far too many to name — who have participated in our discussions over the last twenty years. Here I can mention only a few of the prominent scholars, most of whom have visited Kharoṣṭhī Klub at least once, who have made particularly significant contributions to EBMP studies: Daniel Boucher, Jens Braarvig, Harry Falk, Charles Hallisey, Paul Harrison, Jens-Uwe Hartmann, Oskar von Hinüber, Chanida Jantrasrisalai, Seishi Karashima, Kazunobu Matsuda, Jan Nattier, K. R. Norman, Gregory Schopen, Jonathan Silk, Peter Skilling, Ingo Strauch, and Klaus Wille. It has been a source of deep satisfaction and pleasure for my colleagues and me to have had the opportunity to work closely with so many great scholars of the past, present, and future. The participants in the Hwei-tai Seminar at Stanford University in October 2015, including Luis Gomez, Jan Nattier, and especially Paul Harrison, provided many helpful comments on the interpretation of texts, particularly for the Perfection of Wisdom Sūtra presented in translation 12 below.

I also wish to thank colleagues and others who have read various sections of this book and offered many helpful comments and suggestions: Collett Cox, Robin Dushman, Timothy Lenz, Alan Senauke, and especially Jason Neelis. I must also express my gratitude to the many institutions and individuals who have sponsored, facilitated, or otherwise promoted the work of the EBMP for so many years. Principal among these are the British Library, Centre for Advanced Study at the Norwegian Academy of Science and Letters, Dhammachai International Research Institute, Henry Luce Foundation, International Research Institute for Advanced Buddhology at Soka University, National Endowment for the Humanities, and the University of Washington (including the University of Washington Press). Among many individual supporters, Cris Cyders and Melinda Upton Cyders, along with another generous donor who prefers to remain anonymous, have been reliable friends of the EBMP over many years. We are also deeply grateful to the private owners and public curators who have generously and freely made the manuscripts available to us for study and publication, and thereby made this entire enterprise possible: Martin Schøyen, Robert Senior, the British Library, the Bibliothèque nationale de France, and the Library of Congress. Finally, I wish to thank my friends Tom Lowenstein and Bridget MacCarthy, who offered hospitality and good company during my many early visits to London in connection with this project.

Last but not least, I thank the two editors who helped me with this book. The first is David Kittelstrom, who made innumerable changes small and large that will make this book much better than it would have been without him. The second is my father, George Salomon, who though he died many years ago remains the ever-vigilant editor inside my head, reminding me always to think of the reader and to make every sentence clear, concise, and cogent.

Technical Notes

I have tried as far as possible to minimize untranslated words, but I have retained terms such as arhat, tathāgata, dharma, nirvāṇa, brahman, saṃsāra, saṅgha, and stūpa on the assumption that these will be familiar to readers with a modest background in Buddhism. I have also left unexplained some of the most basic Buddhist principles and concepts, such as the four noble truths or the eightfold path, on the assumption that the reader will be familiar with them or able to easily track them down.

In general, I have striven for consistency, translating each word the same way each time it occurs. But I have avoided a rigid mechanical application of this principle in connection with certain common words with wide ranges of meaning or with complex technical terms such as saṃskāra, which cannot be reduced to a single equivalent in all contexts. In the case of the untranslatable dharma, I have differentiated Dharma in the sense of the Buddhist doctrine from dharma as phenomenon, quality, element, and so on. Some of these decisions and alternatives are clarified in the glossary. Choosing such primary translations for important words often involves difficult and sometimes more or less arbitrary decisions; thus, for example, I have decided to translate bhagavān, the usual honorific title for the Buddha, as lord, but this is only one of several possible choices, including exalted one or master, no one of which is obviously superior. In choosing translations for technical vocabulary, I have often followed the example of Bhikkhu Bodhi’s translations of the Pali Saṃyutta- and Aṅguttara-nikāyas.¹

Because of the several languages involved in this material, it is difficult to maintain perfect consistency in citing Indian names and untranslated Buddhist technical terms that have different forms in Gāndhārī, Sanskrit, and Pali. In general, I have used the Sanskrit form as the default, except where a Gāndhārī or Pali term is specifically being presented or discussed. Thus, for example, I use sūtra with reference to texts in all languages, even in some cases where it might seem odd to refer to a Pali sutta this way. In the case of important terms, I have generally supplied both the Sanskrit and Pali forms, noted as Skt and P respectively, since one or the other of them may be more familiar to readers, depending on their backgrounds. Where no alternate form is offered, as for words like buddha or vinaya, this means that the form is the same in Sanskrit and Pali.

Presenting comprehensible translations of these sometimes very fragmentary texts often requires that missing words or phrases be filled in, either internally on the basis of repeated patterns in the manuscript itself or from external sources such as parallel or related texts in other languages. All such supplementary elements in the translations are indicated in square brackets.

Source Texts

1.Three Numerically Grouped Sūtras: British Library fragments 12+14

2.Five Thematically Grouped Sūtras: Senior scrolls 5 (translations a–d) and 19 (translation e)

3.The Rhinoceros Sūtra: British Library fragment 5B

4.A Chapter from the Gāndhārī Dharmapada: The Monk: Khotan Dharmapada + British Library fragments 16+25

5.Songs of Lake Anavatapta: British Library fragment 1

6.Stories of Previous Lives and Other Legends: British Library fragment 16+25

7.Avadāna Legends: British Library fragment 1

8.The Many-Buddhas Sūtra: Library of Congress scroll

9.A Commentary on the Sūtra of Chanting Together: British Library fragment 15

10.A Commentary on Canonical Verses: British Library fragments 9 and 13

11.An Abhidharma Treatise on Time and Existence: British Library fragment 28

12.The Perfection of Wisdom Sūtra: Split collection, fragment 5

Introduction

THE REDISCOVERY OF GANDHĀRAN BUDDHIST LITERATURE

THE G ANDHĀRAN B UDDHIST scrolls first came to my attention in 1994, when I received a set of blurry black-and-white photographs of some old manuscripts that had recently been acquired by the British Library. Having worked for many years on Buddhist dedicatory inscriptions in stone or on metal plaques from the ancient region of Gandhāra in modern northwestern Pakistan, it was immediately clear to me that these were genuine Buddhist manuscripts written in the Kharoṣṭhī script and Gāndhārī language of that region, dating from about the first or second centuries of the Common Era. This made them the oldest surviving specimens of original Buddhist texts in the world, as well as the oldest Indian manuscripts of any type, and the discovery was widely reported in media worldwide. Subsequent studies have confirmed that these and the other similar materials that were discovered in the following years date from between the first century BCE and the third century CE.

In July 1995 I took the first of many trips to London to see the collection itself, which turned out to consist of twenty-nine fragments of scrolls made of extremely brittle birchbark.² I spent my first few days in London surveying the material in a combined fog of excitement, puzzlement, and jet lag, and the more I pored over the scrolls, the larger loomed the intimidating dimensions of the work that would be involved in reading and translating them. The prospect was daunting not only in terms of the large amount of material but also its generally poor condition, for all of the scrolls were fragmentary in some degree, and some of them were little more than loose scraps. At some time in antiquity they had been rolled up into tight, cigar-shaped packages, packed into a clay water jug, and buried underground. The jug itself provided an important clue as to their sectarian affiliation in the form of an inked inscription recording it as a gift To the universal community, in the possession of the Dharmaguptakas, referring to one of the traditional eighteen nikāyas, or schools, of Indian Buddhism.³

Figure 1. The clay pot (British Library pot D) that contained the British Library Kharoṣṭhī scrolls.

The circumstances of the burial and subsequent rediscovery of the scrolls are unfortunately unknown, as they had already passed through the antiquities market in Pakistan by the time they became known to the scholarly world, and the only information available about their original provenance consisted of unreliable second- or third-hand rumors. Nevertheless, all indications are that they were found in a Buddhist stūpa or monastic complex in the area of Haḍḍa, near modern Jalalabad in eastern Afghanistan, which had been one of the great centers of Buddhist intellectual life in the early centuries of the Common Era. The scrolls had evidently been interred in a ritual burial similar to that accorded to the bodily remains of deceased monks. For from a Buddhist point of view, monks and manuscripts are analogous in that they are both keepers of the Dharma; that is to say, they serve to protect and transmit the precious words of the Buddha. The practice of ritually burying manuscripts may also have been motivated by a desire to insure the survival of the Dharma in future centuries, since the Buddha is said to have predicted that his teachings would be forgotten in the centuries following his earthly decease and entry into nirvāṇa.⁴ At least some of the scrolls were apparently already old and decrepit when they were interred, and several of them had secondary notations such as All has been written today, apparently indicating that they had been recopied onto new scrolls and were ready for interment.

Figure 2. The British Library scrolls in their original position inside the clay pot.

But after nearly two thousand years underground, they had suffered a great deal of damage. Birchbark, when new and fresh, is supple, tough, and attractive. But as it dries out, it becomes extremely brittle, so that many of the scrolls, especially those that had lain at the bottom of the pot, had degenerated into flat stacks of horizontal strips. The osmosis of groundwater through the walls of the pot also caused the portions of the scrolls that had been in contact with the sides to decompose, so that in many cases their edges, and often larger sections, had disintegrated completely. By the time I first saw them, the scrolls had already been painstakingly unrolled by the conservation staff of the British Library’s Oriental and India Office division and permanently mounted between glass sheets in fifty-six separate frames.

Figure 3. British Library fragments 16–19 before unrolling and conservation.

Despite my awe at seeing this collection of dozens of Buddhist texts in Gāndhārī, I was not entirely surprised that it had come to light. For, almost exactly one hundred years earlier, a single specimen of a similar Gandhāran manuscript had been discovered, under unclear circumstances, in the region of the Central Asian city of Khotan in the modern Xinjiang Uyghur Autonomous Region of China. The Khotan scroll contained a previously unknown version in Gāndhārī language of the popular Buddhist verse anthology entitled Dharmapada (Words of the Dharma), one chapter of which is presented as translation 4 in this book. For the next century, this manuscript constituted virtually the only known specimen, except for a few other tiny scraps, of Buddhist literature in the ancient language of the Gandhāra region. But on the basis of this unique manuscript it had been hypothesized that there must have once existed a more extensive Buddhist literature, perhaps even an entire canon, in Gāndhārī. Now, a century later, this hypothesis was about to be confirmed.

Figure 4. British Library conservators unrolling the scrolls.

The first task was to identify the texts and to determine their relationship to other previously known Buddhist literature. After a few days of random grazing through the texts, the first breakthrough came after I noticed that in one of them an unfamiliar word, kharga, recurred in each line. At the time, I was camping out in the living room-cum-library of the London home of an old friend, the independent Buddhist scholar Tom Lowenstein. While randomly scanning his bookshelves one evening, my eye fell on a book by the great British Pali scholar K. R. Norman entitled The Rhinoceros Horn and Other Early Buddhist Poems, which consists of a translation of the Pali anthology of verse sūtras called Suttanipāta (Collection of Sūtras). I then immediately realized that the repeated word kharga must be the Gāndhārī equivalent of Pali khagga, rhinoceros, and that the manuscript consisted of a Gāndhārī version of the well-known Rhinoceros Sūtra (Khaggavisāṇa-sutta), which is one of the poems incorporated into the Suttanipāta. Since each verse of this poem concludes with the refrain Wander alone like the rhinoceros, the repetition in each line of the corresponding word in its Gāndhārī form removed any doubt about its contents. This poem would soon become the focus of my first major project of editing, interpreting, and translating a Gāndhārī manuscript. The results of that project, in the form of a scholarly edition and study,⁵ are summarized here in translation 3.

Figure 5. A Gandhāran birchbark manuscript unrolled. Senior scroll 19, recto, The Parable of the Log (translation 2e).

A second breakthrough came in connection with another scroll in which I noticed the repeated phrase anodate mahasare. Gradually it dawned on me that must correspond to a Sanskrit phrase, anavatapte mahāhrade, at the great Lake Anavatapta, which is the refrain of another well-known poem, the Anavataptagāthā or Songs of Lake Anavatapta. This text describes how the Buddha and his disciples revealed, while seated on giant lotuses in the sacred Lake Anavatapta in the high Himalayas, the karmic factors from their past lives that led them to their present condition. A little further checking quickly confirmed that this was an early Gāndhārī version of this popular poem previously known in Sanskrit, Tibetan, and Chinese versions. This was to be the topic of my second editing project, published in 2008, which is presented in this book as translation 5.

Gradually several other texts were identified, at least as to their genre if not their precise contents. They proved to embrace a wide variety of contents, styles, and subjects: poems, sūtras and commentaries on sūtras, compilations of legends about the lives of famous figures in Buddhist history, commentaries and scholastic treatises on the fine points of Buddhist doctrines, hymns in praise of the Buddha, as well as quite a few unidentified and unclassifiable pieces. It also became evident that the manuscripts were the work of some twenty different scribes, many of whose works could be easily recognized from their distinctive hands. We had, in short, what seemed to be a random selection or culling of the contents of a monastic library, or perhaps of a personal manuscript collection, probably dating from the first century CE.

As work went on in the course of this and many subsequent trips to London, it became abundantly clear to me that the complete study of the British Library scrolls would occupy far more than one lifetime of one scholar. Deciphering fragmentary manuscripts in an incompletely known language requires special expertise in each of the various genres involved, and it was obvious that this was a job for a team, not an individual. For this, I turned first to my colleague and collaborator of many years, Professor Collett Cox, an expert in scholastic works of the abhidharma class, which happened to be abundantly attested among the British Library scrolls, and which were also one of the areas of Buddhist literature that I felt least qualified to work with. The two of us then teamed up, and with the generous support of officials of the British Library and of our home institution, the University of Washington, established in September 1996 the British Library / University of Washington Early Buddhist Manuscripts Project (EBMP), dedicated to the study and publication of the British Library Kharoṣṭhī scrolls. Early on in the project an agreement was reached with the University of Washington Press to publish in a dedicated series entitled Gandhāran Buddhist Texts (GBT) the results of research by EBMP members and associated scholars.⁶

The majority of the manuscripts presented in this book (translations 1, 3, 5–7, and 9–11) belong to the British Library collection, mainly because that group, being the first to be discovered, is the one that has been most intensively studied to date. But in the years and decades following the discovery of the British Library scrolls, several more major groups of similar and equally important manuscripts, as well as some isolated individual texts, were discovered. By now over two hundred significant bodies of text are available for study. The majority have been made available for scholarly research and are being studied and published under the auspices of the University of Washington Early Buddhist Manuscripts Project and the parallel project Buddhist Manuscripts from Gandhāra at the Bavarian Academy of Sciences and Humanities in Munich. The complete detailed study of this material is a huge job that will take many years to complete, but in the meantime Gāndhārī literature has already been established as a separate subfield of Buddhist studies, and from it a steady flow of scholarly publications can be expected to continue for a long time to come.⁷ These new materials are described in chapters 2 and 3, but first we turn to the historical and cultural background of these manuscripts and the world that produced them.

PART I

Contexts

Map 1. Greater Gandhāra.

1. The World of Gandhāran Buddhism

Gandhāra and India’s Northwest Frontier

GANDHĀRA is the ancient name, attested since the time of early Vedic texts dating back at least three thousand years, for the Peshawar Valley and adjoining regions along the Kabul River, stretching for about one hundred miles between the Suleiman Mountains on the edge of the Iranian plateau to the west and the Indus River on the east. In modern terms, Gandhāra corresponds to the area around Peshawar, capital of the Khyber Pakhtunkhwa Province (formerly North-West Frontier Province) of Pakistan.

But the term Gandhāra — or more accurately, Greater Gandhāra — is also applied more broadly to surrounding areas. This territory includes the Swat Valley to the north, the western Punjab including the ancient metropolis of Taxila to the east, eastern Afghanistan to the west, and in the north Bactria (modern northern Afghanistan and southern Uzbekistan), and even parts of the region around the Tarim Basin in Central Asia in the modern Xinjiang Uyghur Autonomous Region of China. All of these regions came under the cultural influence of Gandhāra during its glory days in the early centuries of the Common Era and thereby adopted the Gāndhārī language as a literary and administrative medium and Gandhāran Buddhism as their dominant religion. Thus Greater Gandhāra can be understood as a primarily linguistic rather than a political term, that is, as comprising the regions where Gāndhārī was the indigenous or adopted language.

With the spread of Gandhāran cultural and political power into Central Asia, particularly under the Kuṣāṇa emperors in the first and second centuries CE, Gandhāra came to be directly linked into the commerce of the silk roads, tapping into the lucrative trade in luxury goods between China and the Western world. This source of wealth was no doubt one of the major factors in the power and prosperity of the Kuṣāṇas. Besides the economic benefits that the silk road traffic brought to Gandhāra, it also provided cultural and artistic stimuli leading to the development of an eclectic Buddhist culture incorporating Central Asian and Hellenistic ideas and imagery, while also opening the way for the exportation of Buddhism into Central Asia and China.

The Bloody History of Paradise

In 1978, a Japanese rock group called Godiego recorded in English a song called Gandhara as the theme song for the television drama Saiyūki, or Monkey. These lyrics read in part:

A long time ago when men were all babes,

there was a land of the free.

Fantasy and dreams

were its untouched wealth,

and goodness and love were real.

Each man desires to reach Gandhara,

his very own utopia.

In the striving, in the seeking soul,

man can see Gandhara.

In Gandhara, Gandhara,

they say it was in India.

Gandhara, Gandhara,

the place of light, Gandhara.

Here we see the image of Gandhāra as it was, and still is, imagined by East Asian Buddhists: a magical holy land of peace and harmony. The television show Saiyūki was based on a popular Chinese novel of the sixteenth century, an imaginative account of the famous Buddhist monk Xuanzang’s pilgrimage to Gandhāra and other parts of India nearly a millennium earlier. This idealized presentation of Gandhāra is a distant reflection of the glory days of Gandhāran Buddhism in the early years of the Common Era. Even by the time of Xuanzang’s epic journey to India in the early seventh century, Buddhism had largely declined in Gandhāra. But he was still able to see its legacy in the form of stūpas and other monuments, many of which still dot the landscape today, especially in the Swat Valley, and to collect the legends of its splendors and wonders in the centuries before his time.

Figure 6. The Manikiala stūpa.

The situation in this region today could hardly be more different. It is currently the epicenter of an ongoing bitter struggle between radical Islamists on one side and the ruling powers of Pakistan and Afghanistan on the other, each side fueled by foreign supporters. The willful destruction of the colossal Buddhas of Bamiyan, Afghanistan, by the Taliban in 2001 and the assassination of Osama bin Laden by American military forces in Abbotabad, Pakistan, in 2011, both within the territory of Greater Gandhāra, are only two of the most widely publicized battles in this war between the forces of fundamentalism and modernism. Viewed from a broad historic perspective, these conflicts are a sequel to the three Anglo-Afghan wars of the nineteenth and early twentieth centuries, and to the Great Game, the struggle throughout the nineteenth century between the expanding British and Russian empires for control of Afghanistan.

But even then, strife was nothing new to this region. Throughout history, from the earliest recorded times, Gandhāra has been the scene of frequent wars and invasions from the west into the Indian subcontinent. This turbulent history is the inevitable outcome of its geographical and cultural setting. Gandhāra is an archetypal frontier region situated on a fault line between major geo-cultural zones. Lying on the seam between the Iranian world to the west, the Indian world to the east, and Central Asia to the north, it is the place where, again and again throughout history, these and sometimes other worlds have collided. The passes linking the Iranian plateau and the plains of the Punjab and northern India, including the fabled Khyber Pass, which leads directly into Gandhāra proper, have served for thousands of years as a geographical funnel into India, whose fertility, vast population, and fabled luxuries have enticed conquerors and settlers since the very dawn of history. It was through this funnel that, some four millennia ago, speakers of Indo-European languages began entering northern India, where they would establish the Sanskritic culture that has predominated in the Indian cultural world ever since.

In view of its role as a frontier region and zone of transit between several cultural regions, it is not surprising that Gandhāra has always had a distinct and complex cultural identity. On the one hand, Gandhāra has usually been culturally more part of India than of the Iranian world, despite the constant influence of and frequent political domination from the west, whether Hellenistic, Iranian, Afghan, or Central Asian. In this regard, the rugged mountains that define the western border of Gandhāra proper have also functioned as a cultural boundary, if a porous one. But on the other hand, even within the Indian cultural zone Gandhāra has always stood apart as a land on the fringe, with its own distinct ways and identity, and relationships between Gandhāra and the Indian heartland often seem to involve a certain ambivalence. Even in the early Vedic culture some three millennia ago, which was centered in the neighboring Punjab region to the east, Gandhāra was viewed as a foreign land at the outer limits of the known world, strange and vaguely threatening. The earliest Vedic text, the Ṛg Veda, barely mentions Gandhāra, referring only once to wool from Gandhāra, and also, more or less in passing, to the Kabul (Kubhā) and Swat (Suvāstu) Rivers. The somewhat later Atharva Veda, datable to around the early first millennium BCE, mentions Gandhāra only in a charm intended to dispel fevers to the far distant lands to the west and east: We send the fever to the lands of Gandhāri, Mujavat, Anga, and Magadha.⁸ In a later stage of Vedic literature, the Chāndogya Upaniṣad contains a parable of a man from Gandhāra who must find the way back to his distant homeland.⁹

But as the cultural purview of the heartland Indian culture in the Gangetic plain expanded in subsequent centuries, Gandhāra began to be brought within the pale. Early Buddhist literature contains several lists of the sixteen