Classic and Contemporary Buddhist Works

Divine Stories - Introduction

Divyāvadāna Part 1

INTRODUCTION

Ordé’s words were the truth. You could see every image, feel every sensation he described. His metaphors (what we thought were metaphors) took on a palpable reality that hung in our nostrils, stuck in the back of our throats.
Halfway through any sermon I would notice that I was no longer listening to his words but instead experiencing the phenomena he described.

—Walter Mosley

The Divyāvadāna (“Divine Stories”) is a large compendium of Indian Buddhist narratives written in Sanskrit from the early centuries of the Common Era whose stories have since spread throughout Asia, as both narrative and narrative art, leaving an indelible mark on Buddhist thought and practice. The stories in the collection were frequently used in the education of both monastics and laity in premodern Asia, exerting a powerful influence as moral exempla and legal precedent, and considered by many to be the word of the Buddha himself. These stories were likewise canonical in their influence on Buddhist art, and representations of them can be found across Asia, from Kizil in China to Sanchi in India to Borobudur in Indonesia. For scores of generations, these stories have been repeatedly recited, reworked, painted, and sculpted. It is not hyperbole to say that these are some of the most influential stories in the history of Buddhism.

The text contains thirty-six avadānas, or stories, along with two sūtras, which chronicle the spiritual development of Buddhist devotees with special attention given to their karmic legacies. There are stories of kings and beggars, monks and prostitutes, gods and hell beings, how they came to their present circumstances, the futures they have created for themselves, and the pivotal role the Buddha and his teachings can play in their betterment.

Generally the avadānas presented here contain three elements: a story in the present tense in which characters discover the benefits of Buddhist practice and meet the Buddha; a story of the past detailing the deeds done by those characters in a previous lifetime that have now come to karmic fruition; and a juncture at which time the Buddha— who is quite literally an “omniscient narrator”—identifies the characters in the story of the past with those in the story of the present. Although some avadānas diverge from this tripartite structure, all of them tend to exemplify the inexorability of karma. As the Buddha often explains at the end of avadānas in the Divyāvadāna,

And so, monks, the result of absolutely evil actions is absolutely evil, the result of absolutely pure actions is absolutely pure, and the result of mixed actions is mixed. Therefore, monks, because of this, you should reject absolutely evil actions and mixed ones as well, and strive to perform only absolutely pure actions. It is this, monks, that you should learn to do.

Yet these avadānas are much more than formulaic accounts of good and bad deeds and their repercussions. They also contain and embody rules and practices integral to a Buddhist identity; in fact, they are amalgams of rules, etiological accounts, and foretellings that function as a complex and interlinking moral code. This is not a moral code, however, that can easily be distilled into pithy maxims, such as the Buddha’s observation above about the laws of karma. The moral universe embodied in these stories far exceeds such confines. Its complexity— the dexterity with which certain ideas are brought to life, then developed, nuanced, and imposed, all within a densely textured narrative— prevents such a distillation. These stories may be didactic in their intent, but along the way to their ultimate lessons they create diverse moral worlds, showing different ways of thinking and being, and portray characters interacting and commenting on their engagements with these worlds. The result is an argument—not through philosophical analysis or through poetry, but through really good stories. These are entertaining pieces of literature, with plenty of miracles and adventures across the cosmos, but they are also stories to live by, stories that demonstrate a variety of ways of living and the consequences of such behavior.

Not surprisingly these avadānas have circulated widely since their creation. Many of these stories are included in the monastic code (vinaya) of the branch of Buddhists known as the Mūlasarvāstivādins (“The Original Sarvāstivādins”), who flourished in the first half of the first millennium in northwest India. This legal code, which stipulates rules for personal behavior, private property, and social relations, helped regulate monastic and lay conduct in many parts of India for nearly a millennium. This text was then translated into Tibetan in the ninth century, and to this day functions as the only monastic legal code for Tibetan Buddhists, regardless of sectarian affiliation. The text was also translated into Chinese and Japanese, influencing economic policy and commercial relations in China between the fifth and tenth centuries (Gernet 1995) and guiding the revival of Buddhist monasticism in Tokugawa Japan, beginning in the seventeenth century and continuing to the present day (Clarke 2006).

Other avadānas in the text have been equally influential, though preserved in a more circuitous fashion. Versions of the Pūrṇaavadāna (“The Story of Pūrṇa”) exist in Pāli, Sanskrit, Tibetan, and Chinese, and images of the story were also painted in the caves at Ajanta outside of Bombay in the fifth century and in those at Kizil in China in the sixth century. Likewise, the Śārdūlakarṇa-avadāna (“The Story of Śārdūlakarṇa”), which will appear in Divine Stories, Part II, was translated into Chinese four times between the second and fourth centuries, and then translated into Tibetan in the ninth century (Mukhopadhyaya 1954: xii–xiii). The first part of this story was translated into French by Eugène Burnouf in 1844, and this in turn inspired Richard Wagner, who in 1856 sketched out an opera, Die Sieger (“The Victors”), based upon it. Though he abandoned this work, doctrinal elements from it can be found in his great musical drama Parsifal. In 1882, Rajendralal Mitra (1981: 223–27) offered a summary of the work in English in his descriptive catalogue of Sanskrit manuscripts in Nepal, and some fifty years later his friend, the Nobel laureate Rabindranath Tagore, wrote a play in Bengali called Caṇḍālikā retelling the story. Tagore later transformed the play into a ballet, for which he wrote the music, and it is still performed quite frequently. One performance took place in 1987 at Smith College where I teach and is still preserved on a video recording in our library.

Though most of the stories in the Divyāvadāna serve a legal function—to establish rules of ethical behavior, such as the prohibition against drinking, and to explain the etiology and importance of such rules (e.g., Divy 167–93)—the text defies simple genre classification, for the stories in the collection are eclectic. There are passages that focus on monastic regulations, practical wisdom, moral prescriptions, philosophical truth, metaphysical hypotheses, and even astrological calculations, and many such passages can be found in a single story. This hybridity of style may help account for the text’s enormous popularity across place and time, among both monastics and laity, and in painting, sculpture, and theater.

Despite this hybridity, the Divyāvadāna offers enormous insight into Buddhist history, both subaltern and royal. Many stories in the text depict a practice of faith (prasāda) that allows the disenfranchised to accrue enormous reserves of merit, establishing them on the spiritual path and enabling them to leapfrog those who have been more fortunate, if not more virtuous, than they themselves have been (Rotman 2008, chaps. 3–6). This practice allows one with little material wealth, little knowledge, even little interest in Buddhism to embark on the Buddhist spiritual path with the promise of great results. One need only come into visual contact with an object that is an “agent of faith” (prāsādika), such as a buddha, an image of a buddha, an arhat, or a stūpa, and faith will invariably arise.

Charged with this form of faith, the downtrodden can make offerings of very little worth or utility and earn huge amounts of merit. The rich, conversely, are excluded from the practice, handicapped by their wealth and success. In the Nagarāvalambikā-avadāna (“The Story of a Woman Dependent on a City for Alms”), for example, a leprous beggar woman sees the venerable Mahākāśyapa, who “instills faith in her through his body and his mind,” and then the woman offers him some rice gruel along with her rotten finger, which happened to fall in. From that single deed, however negligible the use-value of the gift, she earns enough merit to be reborn among the gods in Tuṣita heaven. This practice of faith, unattested in philosophical tracts and inscriptions, seems to have offered wonderful promise for the disenfranchised.

As for royalty, the second half of the text (Divine Stories, Part II) contains a story cycle that chronicles the life of King Aśoka (stories 26–29), the great ruler who controlled an empire in the third century b.c.e. that stretched across India and westward to present-day Afghanistan and Iran. This biographical account offers an important counter-history of Aśoka, one that complements yet complicates the Aśoka who can be gleaned from his famous edicts (Nikam and McKeon 1959). In this account, Aśoka becomes devoted to making donations to the monastic community, and though he achieves his goal of becoming the greatest giver in Buddhist history, he is “deceived by his own actions” and dies imprisoned and in penury, his sovereignty lost and all his orders countermanded. This troubling story figured prominently in the Buddhist (and Indian) imaginary for millennia, testifying to the great difficulties of being a virtuous king.

The stories in the Divyāvadāna also offer great insight into the arthistorical record. Representations from the text can be found throughout India, from Sanchi to Bharhut and Mathura, and there are particularly famous stories, such as the Prātihārya-sūtra (“The Miracle Sūtra”), which features a miracle competition at Śrāvasti, that are popular at Buddhist sites the world over. In this way, the text functions as a wonderful tool for deciphering and interpreting much Buddhist painting and sculpture. The text’s detailed descriptions of constructing stūpas (Divy 244ff.), decorating shrines (Divy 78ff.), and making Buddha images (Divy 547ff.) are all likewise enormously beneficial to art historians.

Although the stories in the Divyāvadāna offer significant insight into the history of early Buddhism, the history of the stories themselves is less clear. While the consensus among scholars has been that the Divyāvadāna was produced in northwest India between 200–350 c.e. by the Mūlasarvāstivādins, the dating of the text is too complicated to make such a straightforward pronouncement. We don’t really know when, where, why, or by whom the text was produced. So how does one put these stories into perspective? While these stories are compelling as literature and moral exempla, how does one make sense of them as part of Buddhism’s historical record?

The Historical Value of the Divyāvadāna

These legends [in the Divyāvadāna] scarcely contain anything of much historical value.
—Moriz Winternitz

Many of the avadānas in the Divyāvadāna seem to be intentionally naturalized and dehistoricized, repeating stock phrases in lieu of historical descriptions of people, places, actions, and events: householders are “rich, wealthy, and prosperous, with vast and extensive holdings…”; kingdoms are “thriving, prosperous, and safe, with plenty of food and crowds of people…”; young boys are “raised by eight nurses who nourish him with milk, yogurt, fresh butter, clarified butter, butter scum, and other special provisions that are very pure…”; and the list goes on. Since the dharma always holds true, regardless of time or place, the reliance on such tropes in these avadānas creates an aura of timeless truth—or perhaps a world of make-believe. Winternitz’s observation in 1913 about the lack of historical value in the stories of the Divyāvadāna is not completely untrue, but the great Indologist’s insight needs to be put in perspective. Scholarship on the Divyāvadāna has often involved attempts to extract historical data directly from its stories, with only somewhat successful results. This scholarship is unfortunately marked by the positivism of its age—a tendency to read texts as unproblematically representing historical events. Even the most erudite scholars have occasionally confused narrative incident with historical fact. In discussing an avadāna from the vinaya of the Mūlasarvāstivādins in which a doctor cures a woman of her venereal disease by inserting a piece of meat into her vagina to entice and capture disease-causing worms, one scholar remarks, “The story speaks for itself regarding the beliefs about venereal diseases and the cures thereof. It reveals the morals of rich, young widows of respectable families, and certainly provides a unique insight into the scruples of a young physician in his relationship with his patients” ( Jaini 1989: 220). This practice of attempting to pick out the historical elements from the nonhistorical elements in Buddhist literature—what Louis de La Vallée Poussin called the “subtraction method”—has not, in my opinion, provided an effective methodology for scholarship on the Divyāvadāna. It has instead yielded dubious results, proving in part Winternitz’s observation.

Yet Winternitz’s assessment is limited in its purview. There are more ways to engage with these avadānas than merely trying to extract what Winternitz (1993: 277n4) refers to as “a historical nucleus.” History can be narrated, but it can also be embedded within narrative, even among avadānas that share the same structure and stock passages, and that claim to reveal the past and predict the future. While on the surface the avadānas in the Divyāvadāna display an unambiguous Buddhist moral discourse—good and bad actions always have, respectively, good and bad results for laymen, monks, and buddhas in the past, present, and future—the reasoning and representation in these stories exposes an intricate and evolving Buddhist world beneath this apparently smooth surface. It is this inscribed representation of Buddhist consciousness—this complex world with its thoughts, desires, practices, and anxieties—which is the historical prize. And the dichotomy between historical and nonhistorical elements is not a principal concern when historicizing consciousness, because everything contained within it—the miraculous and the mundane—is already historically located in time and place. Even this revised methodological pursuit, however, is thwarted by the complex history of the Divyāvadāna. The Divyāvadāna is a compendium of stories most likely produced by multiple authors at different times, whose dates and sites of production are uncertain, whose intended audience is unclear, whose expected use is unknown, and whose intertextual relations are unresolved. A review of the manuscript history of the Divyāvadāna demonstrates these difficulties quite clearly.

Manuscript History

It is one thing to analyse footprints, stars, faeces (animal or human), catarrhs, corneas, pulses, snow-covered fields or dropped cigarette ash; and another to analyse writing or painting or speech.
—Carlo Ginzburg

In producing the first Western edition of the Divyāvadāna in 1886, E. B. Cowell and R. A. Neil used seven manuscripts of the text. According to Cowell and Neil’s (1886: vi) account, they are:

  1. Add. 865 in the [Cambridge] Univ. Library; 258 leaves, 14–15 lines, dated 1873. Fairly written in the ordinary Nepalese character, but not very correct.
  2. Our own MS., 283 leaves, 12–13 lines; very incorrect.
  3. Our own MS., 274 leaves, 14–15 lines; correct.
  4. The MS. given in 1837 by Mr. Hodgson to the Asiatic Society at Paris; 337 leaves, 9 lines. This is a very correct copy, and having been made for Mr. Hodgson more than 50 years ago, it in some places preserves the old text which has since become illegible in the original. Unlike the others, it is written in ordinary Nagari characters…
  5. [-]. The authorities of the Imperial Library at St. Petersburg kindly lent us for a short time their MS. (P.—272 leaves), which is a similar copy to ABC and contains the same omissions in the 34th avadāna. We were also similarly favored with the loan from the Bibliothèque Nationale of [Eugène] Burnouf ’s own MS…but as this is only like our other MSS. we made no use of it beyond collating it for the first few pages. In Appendix C [Divy 663–70], we have given some account of another MS…in the same Library, which was also kindly lent to us for a time.

Unlike many critical editions of Sanskrit texts, however, their edition is not a piecemeal reconstruction of some would-be original; instead, it is a slightly edited version of the single best manuscript. As editors, they corrected spelling mistakes and offered alternate readings for unclear words and phrases, but as their footnotes make clear, the manuscripts with which they worked were nearly uniform. As Cowell and Neil (1886: vi–vii) observe,

All these MSS., except F, are thus only modern copies, made with more or less care from one original…Our MS. authorities therefore go back immediately to only one source, and our various readings are simply the result of the greater or less care of the respective transcribers.

Cowell and Neil then conclude that this one source is the Divyāvadāna manuscript possessed by Pandit Indrānand of Patan, Nepal.

Though Cowell and Neil never saw this original manuscript, the noted Buddhologist Cecil Bendall examined it, and he determined that the text was produced in the seventeenth century (Cowell and Neil 1886: vii). Even if Cowell and Neil had been mistaken about the genealogy of these various Divyāvadāna manuscripts, judging from their remarks and those of Bendall, it does seem that none of these Divyāvadāna manuscripts was written before the seventeenth century. Considering that there is also no mention of the Divyāvadāna by name in any extant Buddhist literature prior to the seventeenth century, the possibility exists that the Divyāvadāna, as the particular compilation of stories reproduced in these manuscripts, is not a thirdor fourthcentury artifact, but a seventeenth-century one.

An analysis of manuscript F, the only text not from this “single” manuscript tradition, shows that it is not simply a variant of the Divyāvadāna but a different text entirely. Cowell and Neil (1886: 663) write that the text is “evidently a modern transcript, very inaccurately written,” which partially agrees with the other Divyāvadāna manuscripts, but which is “plainly a distinct compilation.” According to the manuscript extracts that Cowell and Neil provide in their Appendix C, the text refers to itself throughout as the Divyāvadānamālā (“Garland of Divine Stories”). Four years previously, in 1882, Rajendralala Mitra (1981: 304–16) had provided an extended summary of another manuscript of the Divyāvadānamālā, though the two manuscripts preserve different stories. Most likely, these are examples of a medieval avadānamālā or “garland of avadānas”—one of the many anonymous retellings of earlier avadānas, mostly metrical in form with Mahāyāna characteristics, from some time between the fifth and eleventh centuries.

Compounding this problem of the singularity and non-antiquity of Divyāvadāna manuscripts is the possibility that the extant Divyāvadāna manuscripts are incomplete—that the title “Divyāvadāna” previously referred to a collection of materials other than the one that exists in these manuscripts. The coda “this is found in the glorious Divyāvadāna” (iti śrīdivyāvadāne) is also found once in manuscript F, the “distinct compilation” that generally refers to itself as the Divyāvadānamālā, and in the colophon to the manuscript of the Vīrakuśāvadāna (“The Story of Brave Kuśa”) preserved in the Cambridge University Library (Add. 1538). The colophon of that text reads, “So ends e Story of Kuśa and e Glorification of the Fast on the Eighth Day of the Waxing Moon, which were selected from the glorious Divyāvadāna.” Neither of these texts, however, occurs in any extant Divyāvadāna manuscripts.

In his introduction to the Devanāgarī edition of the Divyāvadāna in 1959, P. L. Vaidya explains that these references to the Divyāvadāna occur because at one time the Divyāvadāna was a larger text that incorporated these and perhaps other avadānas, while the present collection of avadānas in Divyāvadāna manuscripts is abridged. To support his hypothesis, Vaidya (1959a: ix) discusses the case of a Newari writer who translated nine of the thirty-two avadānas in the Vicitrakarṇikāvadānamālā (“The Garland of Stories of Vicitrakarṇikā”) into a separate volume. Vaidya conjectures that perhaps the Divyāvadāna underwent a similar phenomenon—a few choice avadānas were selected from the larger collection and codified as a separate text— but that in this case the larger collection of avadānas that existed under the name Divyāvadāna was lost and only the abridged collection survived.

But Vaidya could be mistaken. Perhaps these references to the Divyāvadāna from other manuscripts simply tell us, as Cowell and Neil (1886: viii) note, that “the name was current in Nepal.” The author or scribe of the Kuśāvadāna may have employed the name of the Divyāvadāna because the text was well known or well regarded at the time, and he wanted his text to share in that acclaim. If this were the case, then the extant Divyāvadāna manuscripts would be complete, though representing only one of the manuscript traditions by that name. Also possible is that they are complete but somehow not really the Divyāvadāna. Although the coda “this is found in the glorious Divyāvadāna” occurs at the end of each avadāna and at the end of the work as a whole in the two older manuscripts, D and E, there are no references to the Divyāvadāna anywhere in the more recent manuscripts, A, B, and C. Did the scribes of these manuscripts not know what they were copying? Nevertheless, the possibility of Vaidya’s claim being correct frustrates any facile hypotheses about the completeness and ordering of the text.

There is also the possibility that the extant Divyāvadāna manuscripts contain accretions to what once was a core original. As Vaidya (1959: x) explains,

The literary qualities of these avadānas vary considerably, and contain elements of old tales in Purāṇa style, tales from the sacred literature, tales modelled on classical style with considerable dramatic element as in no. 26 [the Pāṃśupradānaavadāna (‘The Story of a Gift of Dirt’)], tales in the semiclassical style as in no. 22 [the Candraprabhabodhisattvacarya-avadāna (‘The Story of the Deeds of the Bodhisattva Candraprabha’)], and tales in purely classical style as in no. 38 [the Maitrakanyaka-avadāna (The Story of Maitrakanyaka)].

The Maitrakanyaka-avadāna in particular appears to be a later composition that, as Michael Hahn (1992: 5) observes, then “found its way into the Divyāvadāna, where it does not belong at all.” Likewise, the Prātihārya-sūtra (“The Miracle Sūtra”) and the Dānādhikaraṇamahāyānasūtra (“The Mahāyāna Sūtra Dealing with the Topic of Giving”) are included in the Divyāvadāna even though, as is clear from their names, neither are avadānas. In addition, the latter is the only entry that affiliates itself by name with the Mahāyāna. While the Prātihāryasūtra is at least narrative in form, the Dānādhikaraṇa-mahāyānasūtra is instead an enumeration of proper gifts and their results—a multiple anomaly to the collection.

Hence, not all the stories in the Divyāvadāna necessarily arose at the same time or in the same place or from the same hand. In addition to the possibility that the text contains accretions in the form of extraneous chapters, it is quite possible that the Divyāvadāna is “not an original book, but compilations from various sources” (Nariman 1923: 297). It is also possible that the text is an accumulation of narrative fragments from centuries of Indian discourse (Prakash 1970: 285), a collection of pre-Buddhist stories reworked and revised for many generations (Sarkar 1990: 163), or even a compendium of inspired derivations from an earlier canonical tradition (Lamotte 1988: 591). Cowell and Neil (1886: vii, n1) make the point explicitly: “The stories evidently belong to various authors.”

Yet, even if, as G. K. Nariman (1923: 293) observes, “the component parts of the…[Divyāvadāna] are of unequal age,” this doesn’t necessarily mean that the Divyāvadāna had an original core that was vastly augmented. If Aśvaghoṣa could write his Buddhacarita in classical Sanskrit in the first or second century c.e., similarly classical compositions cannot be immediately dismissed as later accretions. In fact, as I will discuss shortly, there is evidence for the early existence of the Candraprabhabodhisattvacaryā-avadāna. Also possible is that the Divyāvadāna was compiled using materials of differing antiquities, as if the stories it contains were Buddhist heirlooms from different eras put together by a diligent curator. Perhaps, then, it may be better to think of the Divyāvadāna as the work of an editor or compiler, not of an author. The historicity and unity of the avadānas in the Divyāvadāna is further problematized when one examines the occurrence of these avadānas in the Tibetan tradition. Twenty-one of the thirty-eight stories in the Divyāvadāna were translated from Sanskrit and are preserved within the Tibetan canon, in the vinaya section of the Kangyur.

These avadānas, however, only occur separately as individual texts; they aren’t grouped together, and there is no mention anywhere of a text called the Divyāvadāna.

As a final cautionary tale for those trying to understand the Divyāvadāna historically, I will offer an account of one more manuscript. In the process of creating this present volume, I was fortunate to make use of a manuscript from the National Archives Nepal, labeled 5819, A120/5–121/1, which contained 303 leaves, fourteen lines to a side. The copy I examined was preserved in microfilm at the Central Institute of Higher Tibetan Studies in Sarnath, and though I’m not able to attest to the manuscript’s age, it is written reasonably clearly and accurately in Devanāgarī script. I refer to it throughout this volume as manuscript H.

First there is the question of the text that this manuscript preserves. While the title of the text doesn’t occur at the end of any individual stories, at the very end of the manuscript the text identifies itself as the Divyāvadānamālā. It does, however, contain some stories that differ from either of the texts by the same name described by Mitra or Cowell and Neil. This manuscript of the Divyāvadānamālā and manuscripts A, B, and C of the Divyāvadāna contain many nearly identical stories with nearly identical colophons, neither mentioning the name of the collection to which they belong. What differentiates these collections are the other stories that they contain and the final name appended to the manuscript. It is as though the Divyāvadānamālā is just the Divyāvadāna with bonus stories.

One notable difference between manuscripts A, B, and C and this Divyāvadānamālā manuscript is the treatment of the colophons at the end of each story. Though they often bear similar inscriptions, those in the Divyāvadānamālā manuscript are always crossed out. The one exception is the final colophon, which also includes the name of the manuscript. Since little else in the manuscript is crossed out, why cross out the names of the stories? Did the scribe who copied this manuscript have some question, hesitation, or denial about the names of the stories? And why leave the name of the manuscript intact? Was this a case of repackaging old stories with a new name? Was this an effort to create a new and improved collection?

While manuscript H does contain omissions that suggest a later provenance, it also preserves certain unique and helpful readings, some of which may even predate those preserved in the manuscripts of the Divyāvadāna. Then again, maybe this text really is a Divyāvadāna—as opposed to the Divyāvadāna—regardless of what it says in the final colophon. Yutaka Iwamoto, for example, observes that, to quote Joel Tatelman (2000: 13; cf. Iwamoto 1978: 143–48), “there are only seven stories which occur in every manuscript [of the Divyāvadāna] and that, of these, only two, the Koṭikarṇa-avadāna and the Pūrṇa-avadāna, always occur in the same place, as the first and second stories respectively. In fact, Iwamoto defines Divyāvadāna as a collection of Sanskrit avadānas, the first two stories of which are Koṭikarṇa-avadāna and Pūrṇa-avadāna.” By this criterion, the manuscripts of the Divyāvadānamālā examined by Cowell and Neil, Mitra, and myself are all versions of the Divyāvadāna, rendering that designation less a title than a marker of genre. Perhaps the Divyāvadāna was a brand name that marked its contents as valuable, a veritable bible of stories, but only delineated some of the stories that it contained.

A Story of the Stories

The entire map of the lost will be candled.
—Agha Shahid Ali

Setting aside these doubts about the unity and historicity of the Divyāvadāna, other manuscript evidence exists that demonstrates that some of the avadānas in the form in which they exist in the Divyāvadāna date back to the Kuṣāṇa or Gupta periods. The Śārdūlakarṇa-avadāna was first translated into Chinese sometime circa 148–70. Furthermore, fragments from the Svāgata-avadāna (“The Story of Svāgata”) (Divy 183.21–185.7) and the Saṅgharakṣita-avadāna (“The Story of Saṅgharakṣita”) (Divy 336.22–339.5), which were found in Gilgit, in what is now northern Pakistan, in 1931, have been dated to approximately the sixth century (Lévi 1932: 16–20; cf. Bapat 1949). Also among the manuscript finds in Gilgit is a fragmentary avadāna collection that contains excerpts from six avadānas found in the Divyāvadāna. As JensUwe Hartmann (1980: 251) notes, “The homogeneous script, the identical number of lines on all the folios, and—possibly—the corresponding size of the leaves all suggest that the different texts formed part of one collection.” Fortunately, one manuscript folio contains both the end of the Sahasodgata-avadāna (“The Story of Sahasodgata”) and the beginning of the Candraprabhabodhisattvacaryā-avadāna (Vira and Chandra 1995: f. 1487), linking these avadānas in the same order as they occur in the Divyāvadāna (nos. 21–22), though their date is uncertain. Less fortunately, however, as Hartmann (1980: 251) observes, “the stories, in so far as they are complete, give neither titles nor colophons, and there is no hint either as to the title of the collection, if any, or to the numbers of the preserved avadānas.”

In recent years, there have been additional large finds of manuscripts from Pakistan and Afghanistan that contain avadānas. The British Library Collection contains many collections of stories, dating from the first or second century c.e., “most of which,” as Richard Salomon (1999:35) notes, “are explicitly labeled in the manuscripts as ‘avadānas.’” Representatives from the Divyāvadāna are unfortunately not among them. The Schøyen Collection, housed in Norway, which is even larger than the British Library Collection, also contains avadāna materials. In this collection, there are various fragments, perhaps from the sixth century, of all four avadānas in the Aśoka cycle (Wille 2000), as well as individual fragments from the Dānādhikaraṇa-mahāyānasūtra and the Jyotiṣka-avadāna (“The Story of Jyotiṣka”) (Baums 2002). Hence, although there is no complete Divyāvadāna manuscript from before the seventeenth century, there are significant indications that many of the stories in the Divyāvadāna have been circulating independently from the early centuries of the Common Era.

There is also significant linguistic and textual evidence that connects the stories in the Divyāvadāna with the Mūlasarvāstivādins. Most notably, more than half of the stories in the Divyāvadāna also occur in a similar form in the Mūlasarvāstivāda-vinaya (Huber 1906; Lévi 1907), an immense collection of monastic law and moral tales that is preserved partially in Sanskrit (Dutt 1984), more fully in Tibetan and Chinese translations, and was compiled perhaps as early as the first or second century c.e. Since Edouard Huber (1906) first made this observation, Sylvain Lévi (1907), Heinrich Lüders (1926), and D. R. Shackleton Bailey (1950) have all concluded with him that these stories in the Divyāvadāna were deliberate abridgments of their counterparts in the Mūlasarvāstivāda-vinaya. Though Jean Przyluski (1929) came to the opposite conclusion, suggesting as well that both might come from an earlier and no longer extant source, Satoshi Hiraoka (1998) has argued quite convincingly in the tradition of Huber that the stories in the Divyāvadāna that also appear in the Mūlasarvāstivāda-vinaya are reworked versions of the latter. With just a few exceptions, the stories in the Divyāvadāna even follow the same sequence as their counterparts in the Mūlasarvāstivāda-vinaya (Panglung 1981: xiv–xvii). In short, someone (at some time) abridged some stories from the Mūlasarvāstivāda-vinaya and then compiled them together with stories from other vinayas, a few sūtras, and some other favorite narratives to create a “greatest hits” compilation known as the Divyāvadāna. Since many of the stories in the Divyāvadāna have their origin in the Mūlasarvāstivāda-vinaya, there are, not surprisingly, references to Mūlasarvāstivādin texts, strongholds, and perhaps doctrines, though such proprietary claims are put into doubt because of the unclear relationship between the Mūlasarvāstivādins and the Buddhist sect known as the Sarvāstivādins. In the Śikṣāsamuccaya (“Compendium of Training”), for example, an extract from the Cakravartivyākṛta-avadāna (“The Story of One Foretold to Be a Wheel-Turning King”), which occurs in the Divyāvadāna, is introduced as a Sarvāstivādin text: “This is recited by the noble Sarvāstivādins.” This could signify that this story was shared by the Sarvāstivādins and the Mūlasarvāstivādins, that the Sarvāstivādins wanted to claim it for themselves, or that the author of this passage considered the Mūlasarvāstivādins to be somehow the same as the Sarvāstivādins and not a separate sect. Much has been written about the relationship between the Mūlasarvāstivādins and the Sarvāstivādins—who was in Mathurā, who was in Kashmir, who came first, etc. I will nevertheless recuse myself, following Lambert Schmithausen (1987: 379), from making any definitive statements about the relationship between these two schools. As he remarks,

I cannot enter into the controversial question whether the Mūlasarvāstivādins were, originally, a Vinaya school of the Mathurā area, completely independent of the Sarvāstivādins (as advocated by Frauwallner on the basis of a comprehensive and thorough investigation of the Vinayas) or only a comparatively late offshoot of the Sarvāstivāda school with a Vinaya that is nothing but an enlarged and remodelled version of the Vinaya of the Sarvāstivādins (as Lamotte and Iwamoto seem to think, in opposition to Bareau and Gnoli who assert that the Vinaya of the Mūlasarvāstivādins looks more archaic than that of the Sarvāstivādins).

And so, considering how little is known about the Mūlasarvāstivādins outside of the contents of their voluminous monastic literature, and since the possibility still exists that the Divyāvadāna was constructed not by the Mūlasarvāstivādins but by some other Sarvāstivādin sect or, more doubtfully, some other Buddhist sect at some later date, it is more judicious to treat the Divyāvadāna not as a text created exclusively within a Mūlasarvāstivādin framework, but as one created more generally within the context of early Indian Buddhist monastic culture—probably during the period of Sarvāstivādin Buddhism in Northwest India during the first half of the first millennium. This context for interpreting the Divyāvadāna is, unfortunately, rather vague. But there are other contexts for making sense of the text. While it may not be possible to locate the origin of the avadānas in the Divyāvadāna or even the compilation of the text as a whole in a historical place or moment—for most avadānas resist such efforts—it may be possible to interpret these stories within the context of their own telling. Accounts of stories being told and heard in the Divyāvadāna may offer insight into the sociology and mechanics of how avadānas were used.

An Avadāna of Avadānas

“Everything comes from somewhere,” Haroun reasoned, “so these stories can’t simply come out of thin air…”
—Salman Rushdie

Although there is a degree of guesswork in determining the origin and use of the avadānas in the Divyāvadāna, we do know something about the status of avadānas among the Mūlasarvāstivādins. They came to identify certain stories as avadānas, constituents of a new division of Buddhist literature (Thomas 1933), and they also created independent avadāna texts, such as the Sthavirāvadāna (“The Story of the Elders”) and the Avadānaśataka (“The One Hundred Stories”). But why did the Mūlasarvāstivādins suddenly elevate the avadāna, a mode of composition long in circulation and yet not discussed in any traditional commentaries, to a new and canonical genre?

A number of possible explanations for the uses of these avadānas have been offered. They may have been used (1) to popularize Buddhism—“Aśoka’s preference for the life of an ideal upāsaka [lay disciple] as against that of a monk may have stimulated the Buddhist monks to devise ways and means to popularize their religion, and as the result of the efforts of the monks in this direction, we have the large number of the jātakas and avadānas” (Dutt 1930: 20); (2) to inspire the laity— “As it is evident from the subject-matter of most of these stories, the avadāna purports to kindle faith and devotion in the ordinary believer by laying before him the fruits of good acts…and the bad consequences of evil acts” (Perera 1966: 397); (3) to educate the common people— “But the common people could not be made familiar with the glorious deeds of the world-famed heroes (bodhisattvas); they could not understand [that] what the great hero had done they too could do. So there was a need for something humbler, i.e., the glorious deeds of some ordinary humans with whom the common folk could make identity” (Sharma 1985: 19); (4) to educate young monks—“These avadānas…[were] put together for the ease and convenience of instruction of young monks” (Vaidya 1959a: xii); or (5) to offer preliminary teachings—“One begins to teach dharma by telling avadāna stories” (Tatelman 2000: 12).

While these hypotheses may very well be true, recent finds of avadāna manuscripts provide additional data that suggests a more specific sociology of practice. As Richard Salomon (1999: 36) observes, the terse form of the avadānas in the British Library Collection, with their instructions to expand upon various abridgments in their own specific idiom, “give the impression that the texts are merely skeletons or outlines, which were evidently meant to be filled in and expanded by the reader or reciter.” This is also the case with both the Divyāvadāna and Avadānaśataka. Both contain numerous stereotypical passages that are often abridged with the expression “and so on as before” (pūrvavat yāvat). Yet, as Salomon (1999: 35) also notes with regard to the British Library Collection, “an unusual feature of the texts of this genre is that nearly all of them are written in the same distinctive large hand.” Salomon (1999: 36; cf. 1999: 54) concludes from this that the scribe “evidently was a specialist in this genre,” lending support to the hypothesis of John Strong (1985: 869) that there was “a general class of specialists concerned with avadāna literature.” If this is the case, then the avadānas in the British Library Collection were for the use of avadānists, for only they could properly follow the terse instructions found in the manuscripts, and they were also copied and preserved by avadānists. Furthermore, they were also, it seems, composed by avadānists. Unlike their counterparts in the Divyāvadāna, which are primarily simplified narratives from the vinaya, “the specific content of these stories does not correspond to any previously known material” (Salomon 1999: 136).

This hypothesis that these manuscripts were composed, preserved, and used by avadānists is further supported by an examination of the genres represented therein. Among the fragments of the British Library Collection of Gandhāran manuscripts, there is “a total absence…of vinaya texts of any kind” (Salomon 1999: 163), and the same is true among Central Asian Sanskrit Buddhist manuscripts before the fifth century c.e. (Sander 1991: 142; cited in Salomon 1999: 163). This raises the possibility that, as Salomon (1999: 163) notes,

At this relatively early stage of the written preservation and transmission of Buddhist texts, not all texts or genres of texts had yet been set down in writing…It appears that at this point in the development of the written tradition of Buddhist texts, writing was viewed primarily as a practical matter; texts were set down in written form only when this seemed necessary or useful, as for instance when, for one reason or another, a text was not firmly set in memory or was perceived to be in danger of being forgotten.

Considering the number of specialized abbreviations in the avadānas of the British Library Collection, it seems unlikely that these avadānas were written down because they were “perceived to be in danger of being forgotten”; instead, according to Salomon (1999: 165),“the Gāndhārī avadānas seem to be more in the nature of notes of memory aids than of formal written texts. In other words, they fall somewhere between the strict division of written versus oral texts, serving, evidently, as written supplements to oral deliveries.” In short, these avadānas had a different mode of preservation from vinayas, indicating quite possibly a different status or function, and this difference may very well have involved oral recitation.

Yet the avadānas in the Divyāvadāna are not new compositions like those in the British Library Collection, and the reason for their compilation poses a different mystery. Following the traditions recorded by Kumārajīva, Seng-hu, and Hui-chiao during the fourth and fifth centuries c.e., Lamotte (1988: 174) surmises that various avadānas were excerpted from the Mūlasarvāstivādin vinaya because the latter had become unwieldy. As Strong (1985: 876) explains,

Apparently, the vinayadharas [vinaya masters] of Upagupta’s time could simply no longer handle the load; they opted to cut out the avadānas from their repertoire, but, I would argue, they did so not because they were thought to be nonessential or of dubious status canonically, but because there were other oral specialists around—the avadānists—who were already taking charge of them.

Following Salomon’s notions about the orality of avadānas and Strong’s hypothesis about the avadāna specialist, I would like to suggest another possibility for the creation of the Divyāvadāna and the use of its stories. Here the text itself is our best guide, for it describes the telling of stories in detail—who tells them and to whom, and when they are recounted and where. Although I don’t claim that these accounts are historically referential, they do offer traces of past beliefs from which, it seems, a number of storytelling settings and audiences can be identified.

Stories about Telling Stories

This was how she would learn about the world, in sentences at meals; other people’s distillations amid her own vague pain, dumb with itself. This, for her, would be knowledge—a shifting to hear, an emptying of her arms, other people’s experiences walking through the bare rooms of her brain, looking for a place to sit.
—Lorrie Moore

In the stories in the Divyāvadāna, the Buddha, and occasionally one of his disciples: (1) tells a story to the monastic community to resolve their doubts and questions; (2) tells a story to a lay follower or audience who happens to approach him; or (3) tells a story to a lay audience in a lay disciple’s home after having been invited there for a meal and consuming it. In the first instance, the stories that are told are often accounts of someone’s deeds, both previous and future, that are integral to his or her karmic history. In the second instance, the stories that are told are often referred to as “discourses on the dharma.” And in the third instance these stories are referred to as either “discourses on the dharma” or “dharma stories.” One suspects that Buddhist monks had a penchant for expounding, telling stories rather than giving moral directives or performing rituals. This is even parodied in the Saṅgharakṣita-avadāna, with Buddhist monks being shunned and chastised for being “great talkers.” Now, this last scenario—a monk is invited to a layperson’s home for a meal, consumes the food offered to him, and then offers some instruction in the dharma—is well represented in the Divyāvadāna, and examining the rules for lay and monastic behavior that are preserved in other Buddhist sources, one finds regulations that, when enacted, are in accordance with it. In the Aṅguttara-Nikāya, for example, among the duties prescribed for a Buddhist householder are inviting monastics to one’s home for a meal (ii, 65) and listening to dharma teachings (iv, 209). Considering that householders composed a large percentage of lay disciples (Dutt 1940: 166) and were the group most responsible for providing the monastic community with its material needs (Chakravarti 1987: 65–93), it is not surprising that this relationship between householders and the monastic community would be well regulated and well represented.

Intrinsic to this relationship is an exchange—laypeople offer food to monastics, and monastics in turn offer those laypeople merit. This “food for merit” exchange is made explicit in the Sahasodgata-avadāna. A solitary buddha who is traveling through the countryside arrives at a park on the outskirts of a town but then decides to go elsewhere. A householder sees him and says,

Noble one, why are you turning back? You are in want of food, and I, of merit. Take up residence here in this park and I’ll support you with alms without interruption.

But on such occasions, with food given and merit assigned, if dharma teachings were to be given, which ones would they be? It is often said that the Buddha, after finishing a meal that had been offered to him, “instructs, incites, inspires, and delights” the listener with dharma stories. This indicates that these stories were instructive and entertaining, but suggests little else with regard to their specific content.

Two passages in the Dharmaruci-avadāna (“The Story of Dharmaruci”) offer some additional insight. Early in the story, it is said that as a baby, Dharmaruci was insatiably hungry and thirsty and could only be appeased when listening to the stories told by monastics:

From time to time monks and nuns would enter that house for alms and tell a roundabout story. The boy would listen to that roundabout story, and at that time he wouldn’t cry. He’d listen attentively and silently to their stories about listening to the dharma. When the monks and nuns would depart, he would again experience the suffering of thirst and begin to cry.

In this case, the monastics tell a “roundabout story” (parikathā), an apparent variant of the “dharma story” (dharmīkathā). Though its exact meaning is unclear, it may mean “involved story,” “intricate story,” or even “story cycle.” More clear, however, is the notion of listening to stories about listening to the dharma. This involves the mimetic act of listeners listening to stories of others doing what they themselves are already doing.

Later in the story, in response to some monks’ queries about Dharmaruci’s insatiable hunger, the Buddha asks them, “Do you want to hear a dharma story about the former karmic bonds of Dharmaruci?” They assent, and the dharma story that the Buddha tells is a chapter in Dharmaruci’s dharma history when he was a powerful fighter capable of battling a thousand men. These “karmic bonds” (karmaploti) are, in fact, the “connective thread” (ploti) that ties together one’s karmic history; hence the dharma story in this instance is a dharma case study.

But what were these roundabout stories, the dharma biographies, which were presumably instructive and entertaining, that featured characters benefiting from listening to the dharma?

The most likely answer is—these avadānas themselves. The structure of avadānas, with their stories of the present, then of the past and future, all unfolding in an interactive process between teller and listener is far more roundabout than jātaka tales or stories in the sūtra literature. Furthermore, the instructive value of these stories is thematized within the stories themselves, most notably in what I referred to previously as the first scenario. On those occasions, the past and future stories that the Buddha tells are prompted by a query from some inquisitive monks who are unsure how a particular event came about or how a particular prediction will come true. As it is frequently said, “Those monks in doubt asked the Lord Buddha, the remover of all doubts…” And, one presumes, since the Buddha is here introduced as “the remover of all doubts,” his stories have precisely that effect. As Ānanda says to the Buddha on multiple occasions,

O you who are resolute, an ascetic, and an excellent victor, you know at once with your mind the desires of your listeners. Destroy their doubts that have arisen, O best of sages, with words excellent, enduring, and virtuous.

Instructive as well is the way that many avadānas exemplify the utility of making offerings to monastics, with the donor being promised great rewards in the future. Listening to such stories would naturally reinforce the “food for merit” exchange, a give-and-take that is as essential for the physical survival of monastics as it is for the karmic development of the laity.

Since most avadānas also feature the “doubt removal” scenario, listening is necessarily thematized, for the very structure of these avadānas involves monks listening to stories that the Buddha tells. The above extract from the Dharmaruci-avadāna testifies to this experience— Dharmaruci listens to a story about others listening to the dharma. And since the Buddha can remove the doubts of his monks, shouldn’t his avadānas, or even his stories within avadānas, do the same for the listener? Would one be in error if they didn’t? Hence, shouldn’t any doubts about the utility of the “food for merit” exchange between monastics and the laity be removed by listening to the numerous stories that exemplify the utility of making offerings to monastics, with the donor being promised great rewards in the future? Perhaps, then, the reason that the content of the dharma stories told in the “postmealtime” scenario is never revealed is that these stories are the avadānas themselves. In short, the dharma stories alluded to are already being heard.

In most Buddhist vinaya literature, the description of a monastic rule is accompanied by a story recounting the event that necessitated the enactment of that rule, but in the Mūlasarvāstivāda-vinaya, far more stories are included, creating an even stronger connection between the establishment of rules and the telling of stories. In the Divyāvadāna, this connection between rules and stories is clarified by placing both within a larger frame-story that narrativizes the entire process. A common scenario finds the Buddha telling a story to the monastic community in order to explain a particular phenomenon and to exemplify proper behavior, but all of this is placed within a larger story depicting the process of educating the monastic community. This larger story allows the audience to view the process of rule-giving and storytelling within a more elaborate context. For example, unlike the version of the Svāgata story in the Pāli vinaya (Vinayapiṭaka ii, 108–

10) which recounts the drunken exploits of Svāgata and the rule discerned from this story (i.e., drinking liquor is an offense requiring expiation), the Svāgata-avadāna includes the story of the Buddha telling that larger Svāgata story—the tale of his drunken exploits and the rule against drinking—as well as a sequence in which the monks question those actions of Svāgata. Then, in response, the Buddha tells a story about Svāgata’s past life as a householder when he both abused and served a solitary buddha and the results of these actions.

In the Pāli, vinaya rules appear to be determined from stories: a story is told, a judgment is offered, a rule is established. In the Mūlasarvāstivāda-vinaya and in the Divyāvadāna, rules are also determined from stories, but then secondary stories are used to explain the phenomena and karmic connections within those primary stories. These layered stories allow one to view the process of how rules are taught, through stories of origin and stories of explanation, and to see how the intertwining of stories and rules can allow stories to embody rules and, perhaps, even supplant them. In the Divyāvadāna, there are examples of rules being determined from stories, as in the Svāgata story in the Pāli vinaya, and even more examples of rules and stories being intertwined, with the latter functioning as something between a complementary moral code and a contentious one, what A. K. Ramanujan (1999: 446) referred to as a “counter-system.” But more often stories themselves seem to replace rules, as though the latter was the preferred form of moral guidance, in style if not in content. These stories offer moral exempla, possibilities for ethical action that, as Ramanujan (1999: 456) explains of oral literature, provide “forms, presumptions of meaning, that are filled out by later living.” One wonders if this is what the reciters of avadānas had in mind when, later in the Dharmaruciavadāna, the monks respond to the Buddha’s offer to tell a story about Dharmaruci’s former karmic bonds. As they say, “Let the Blessed One tell the monks a dharma story about Dharmaruci. Hearing such a story from the Blessed One, the monks will keep it in mind.” Now if a Buddhist monk in premodern India were to tell a didactic story to educate a group of householders, what kind of story would he tell? What kind of story would he have known? A Mūlasarvāstivādin monk, for example, would presumably have known stories from his vinaya. Considering the strong connection between rules and stories in his vinaya and the critical position of the vinaya to Buddhist identity— disputes over vinaya rules were the main cause of schisms among early Buddhist sects, and differences over vinaya rules the fundamental distinction between these groups (Lamotte 1988: 290–92)—a Mūlasarvāstivādin monk would most likely have learned vinaya stories while learning the rules that defined his order. Certainly one of the fundamental distinctions between the Mūlasarvāstivādins and the Sarvāstivādins was the content of their respective vinayas. Considering as well that vinaya stories contain the genealogy of problems and their solutions (do not commit such-and-such an act, which such-and-such a person previously performed), a narrativization of this process would have provided those uneducated in Buddhist monasticism with a mimicked account of the educational process itself.

All this is not to say definitively that the Divyāvadāna was the creation of Mūlasarvāstivādin monastics who needed dharma tales to tell, particularly on those occasions when they were invited to someone’s house for a meal, but this possibility is given significant support. This hypothesis would also explain the strong genealogical connection between the avadānas in the Divyāvadāna and their counterparts in the Mūlasarvāstivāda-vinaya, as well as the erasure and marginalization of some of the legal content in the Divyāvadāna versions of these stories. For example, while the Meṇḍhaka story in the Mūlasarvāstivāda-vinaya contains a technical discussion regarding the propriety and mechanics of accepting certain kinds of offerings (see appendix 1), this section is elided completely in the Divyāvadāna. A similar elision also occurs in the Svāgata-avadāna. As Satoshi Hiraoka (1998: 423–24) observes,

It seems virtually certain that…the compiler of the Divy. skillfully omitted the sections of the establishment of the Vinaya rule and the commentary on it, directly connecting the part quoted above with the story of Svāgata’s past. Thus he produced a story that looks natural and preserves the typical style of an avadāna.

Taking all of this information into account, perhaps the development of these avadānas can be explained as follows: With the routinization and increase in lay-monastic interactions, or at least with the desire for such ends, there came the need for good, easily accessible stories to be told on those numerous occasions when monks were enjoined to discourse on the dharma. Since these monks were familiar with the stories found in their vinaya, they reworked these very narratives—transforming them from accounts of the origins of monastic rules to accounts of the workings of karma—to meet the didactic needs of preaching to novice monks and lay disciples. One can speculate that since these stories were not easily accessible for monks to learn and consult because of the tremendous size as well as lack of systemization and revision of the Mūlasarvāstivāda-vinaya (Schopen 1994b: 69), they were eventually anthologized into more accessible volumes. It is also possible that these stories were frequently recited, and as the need to recount them grew, they were codified into their own collections to canonize those stories that were considered appropriate or efficacious to tell. These stories became known as avadānas and, judging by the name of the present collection, the Divyāvadāna—“Divine Stories”— this collection was one of great importance, or at least had pretensions to be so. Regardless of whether this text was really important, or merely self-important, it offers the reader an excellent and entertaining way of engaging with early Buddhist moral thought.

 

How to cite this document:
© Andy Rotman Divine Stories (Wisdom Publications, 2008)

Creative Commons License
This selection from Divine Stories by Andy Rotman is licensed under a Creative Commons Attribution-NonCommercial-NoDerivs 3.0 Unported License.
Based on a work at http://www.wisdompubs.org/book/divine-stories.
Permissions beyond the scope of this license may be available at http://www.wisdompubs.org/terms-use.