The Realm of Aksobhya: A Missing Piece in the History of Pure Land Buddhism

by JAN NATTIER

The Realm of Aksobhya: A Missing Piece in the History of Pure Land Buddhism

The form of Buddhism known as "Pure Land" (Ch. ching-t'u, Jpn. jodo Vf-i) has long held an uneasy status among English-speaking scholars of Buddhism. Exposed early in their training to the first substantial body of Buddhist texts to be made available in English - the canonical texts of the Theravada school, translated from Pali into English beginning in the late 19th century - British and American scholars have often found it difficult to find any connection between the seemingly austere and contemplative

teachings of the so-called historical Buddha and the lush celestial imagery and faith-oriented language of scriptures like the longer and shorter SukhavatTvyuha.1 Writing in the late 1920s, British historian Sir Charles ELIOT went so far as to ask whether the Pure Land Buddhism of Japan should be referred to as "Buddhism" at all: It has grown out of Buddhism, no doubt: all the stages except the very earliest are perfectly clear, but has not the process of development resulted in such a complete transformation that one can no longer apply the same name to the teaching of Gotama and the teaching of Shinran?2

• This paper was originally presented at a conference held at the Chung-Hwa Institute in Taipei, Taiwan in July 1997. I am grateful to the conference organizers, especially the Ven. Sheng-yen, for the rich intellectual experience afforded by that gathering, as well as for their permission to publish this paper here. I would also like to thank Paul Harrison and Daniel Boucher for their helpful comments on an earlier draft. Any errors that remain are my own. 1. The difference is more apparent than real. Scholars of the

Pali texts are at long last beginning to break free of the European rationalism that informed (or better, constrained) the interpretation of these texts during the late 19th and early 20th century, and are now realizing that the Pali texts contain abundant references to the use of paranormal powers by enlightened beings, the existence (and importance) of heavenly realms, the activities of ghosts and spirits of a variety of species, and so on. 2. At his death in 1931 ELIOT left a nearly completed manuscript which was published posthumously (together with a chapter on Nichiren contributed by G. B. SANSOM) as Japanese Buddhism (London: Edward Arnold & Co. 1935). The comments cited above appear on pp. 389-390 of the reprint edition (London:

ELIOT was hardly the last to express incredulity that two such seemingly divergent forms of Buddhism could be related. Indeed I would venture the guess that most scholars who have had the experience of teaching an "Introduction to Buddhism" course in an American or European university have heard these same sentiments expressed by our own undergraduate students. Though the thrust of ELIOT's remarks was to call into question the legitimacy of "Pure Land" Buddhism (or at least of the Japanese interpretation of that form of Buddhism set forth by Shinran), in so doing he also pointed - if perhaps inadvertently - to a way of overcoming this sense of unease. What has made Pure Land Buddhism so difficult for Westerners to appreciate, I believe, is that the "problem" has been posed in precisely the terms expressed by Eliot: that of the apparent discontinuity between the teachings of Gotama found in the early Pali sources, on the one hand, and the teachings associated with scriptures devoted to Amitabha3 on the other. But the latter are not, historically speaking, directly related to the former. On the contrary, prior to the emergence of the belief in Amitabha several intermediate

developments had taken place, and without a clear understanding of these prior stages the cultivation of devotion to Amitabha does indeed appear anomalous. As I hope to demonstrate, however, there is a significant body of evidence that has been largely overlooked in the study of Pure Land Buddhism: evidence concerning the Buddha Aksobhya (Ch. ^Ijj; Tib. Mi'khrugs-pa) who presides over an "eastern paradise" known as Abhirati (Ch. l^jrbfiJi, ^U; Tib. Mnon-par dga'-ba) that resembles in many respects the paradise-like world of Amitabha. Despite the abundant similarities (as well as illuminating divergences) between these two figures, it remains a fact that in both Asian and Western scholarly

circles Amitabha has been studied in isolation, while Aksobhya has hardly been

Routledge & Kegan Paul 1959). The idea of opening this discussion with ELIOT's comments on Pure Land Buddhism was inspired by the use of the same quotation in FUJITA Kotatsu, "Pure Land Buddhism in India," translated by Taitetsu Unno, in James Foard, Michael Solomon, and Richard K. Payne, eds., The Pure Land Tradition: History and Development (Berkeley, CA: Berkeley Buddhist Studies Series 1996), pp. 1-42 (p. 3). 3. As a matter of convenience I will use the name Amitabha ("Unlimited Light") throughout this paper to refer to the Buddha known both by this name (Ch. Mm-yu) and by the name Amitayus ("Unlimited Life," Ch. ^m#) in Sanskrit sources. The Chinese abbreviation A-mi-t'o can, of course, refer to either.

studied at all.4 This paper is intended as a small contribution toward remedying this situation. By examining in detail the information about Aksobhya and his world contained in Indian sources (above all in the Aksobhyavyuha, a sutra devoted entirely to the career and the "pure land" of this figure) I hope to show that this material provides valuable clues concerning the "very earliest stage" in the development of Pure Land ideas to which Eliot referred. Far from being simply a "different lineage" of Indian thought concerning other Buddha-worlds than the one based on Amitabha, as FUJITA Kotatsu has suggested,5 it may well have been an earlier one, which rather than running parallel to beliefs about Amitabha was actually assumed and elaborated upon by Amitabha's devotees.

"Pure Land" Thought in Early Mahayana Literature The term "Pure Land" (Ch. ching-fu I Jpn. jodo is not, of course, an Indian term. It has no known Sanskrit antecedent, and it is now widely agreed that this expression was first coined in China.6

4. The only studies of Aksobhya that have appeared in Western languages to date are Jean DANTINNE: La Splendeur de llnebranlable, Tome 1, Chapitres I-III: Les Auditeurs (Sravaka) (Louvain-la-Neuve, Institut Orientaliste 1983; no subsequent volumes have been issued) and an unpublished doctoral dissertation by KWAN Tai-wo, "A Study of the Teaching Regarding the Pure Land of Aksobhya Buddha in Early Mahayana" (UCLA: Dept. of East Asian Languages and Cultures 1985). A partial translation of Bodhiruci's version of the Aksobhyavyuha (T No. 310[6]) is included in Garma C. C. Chang, ed., A Treasury of Mahayana Sutras (University Park, PA: Pennsylvania State University Press 1983), pp. 315-338. See also the brief but useful entry "Ashuku" in Paul DEMIEVILLE et al., eds., Hobogirin, vol. II (Tokyo: Maison Franco-Japonaise 1930), pp. 39a-40b. Far more surprising

than the relative paucity of studies of Aksobhya in the West is the notable lack of attention to Aksobhya in Japan, where Pure Land Buddhism is a topic of significant academic concern. The statement of FUJTTA Kotatsu (op. cit., p. 9) that "Pure Land ideas are non-existent in such Mahayana sutras as the Astasahasrika Prajnaparamita and Aksobhyavyuha" - both of which provide detailed information on Aksobhya and his world - is indicative of the deep chasm that has separated the study of Amitabha Buddha from the study of his fellow celestial Buddhas in Japanese scholarly circles. 5. FUJITA, op. cit., p. 9. 6. FUJITA, op. cit., p. 20, and more recently in "The Origin of the Pure Land" by the same author, in Eastern Buddhist, vol. 29, no. 1 (1996): 33. The peculiar fact that the word vyuha is sometimes translated as ching if- "pure" by early Chinese translators, most notably Chih Ch'ien and Dharmaraksa ^SH, raises the

Expressions such as "Pure Land teachings" (Ch. ching-t'u chiao ffliMQ or "Pure Land thought" (Jpn. jodo shiso however, have come to be widely used, and in Taiwan and Japan they are generally understood as referring solely to ideas and practices associated with the Buddha Amitabha.7 I would like to propose here, however, that we extend the category of "Pure Land Buddhism" to include scriptures devoted to Aksobhya and other celestial Buddhas as well, at least for purposes of the discussion at hand. In so doing I am deliberately diverging from established tradition in order to engage in a particular type of comparative study. Specifically, I would like to define "Pure Land Buddhism" as the set of all ideas and practices related to Buddhas who are presently living in worldsystems other than our own, a category which would include not only Amitabha but also Aksobhya and the countless Buddha figures described in Mahayana texts as presiding over world-systems in all of the ten directions.8 Defined in this way, Pure Land Buddhism consists of all Buddhist teachings that look forward to the possibility of rebirth in another

possibility that the term ching-t'u #± resulted from a confusion between a Prakrit form of vyuha (*viyuha) and visuddha "pure," a confusion that could occur most easily in the Kharosthi script, where the characters for yu p and su J] are virtually identical (see the chart given in Plate XIV at the end of Part II of A.M. BOYER et al, KharosthT Inscriptions [[[Oxford]]: Clarendon Press 1927]). See for example the translation of the VimalakTrtinirdesa by Chih Ch'ien (T No. 474), where the character ching is used in translating the proper names Prabhavyuha (T 14.519b6 and 14.524a21), Mahavyuha (14.519b7), Padmavyuha (14.519M6), and Ratnavyuha (14.529a7), and the title of the translation of the ManjusrTbuddhaksetragunavyuha by Dharmaraksa (T No. 318), where the characters yen-ching Jt7# are apparently used to translate guna-vyuha. If this line of reasoning is

correct, the expression ching-t'u might well have originated as a rendering of ksetra-vyuha ("field-array"), itself a very common expression in Mahayana sutras. 7. This is not a recent development. According to FUJITA ("The Origin of the Pure Land," p. 36) this usage had already become current in T'ang-period China. 8. I will exclude from this category only those Buddha-figures who have appeared, or will appear, in our own world-system, viz., Sakyamuni, Maitreya, and their numerous predecessors (including Krakucchanda, Kanakamuni, Kasyapa, and so on). Maitreya might in fact be described as a borderline figure: though he will appear in our own world-system, which is generally described as characterized by undesirable qualities (in contrast to other more glorious Buddha-worlds), he will do so in a distant future age when our world has reached the peak of its potential.

world-system (lokadhatu) or Buddha-field (buddha-ksetra), where a Buddha is presently teaching the Dharma. Such worlds are commonly described as far more glorious than our own, but this is not their defining feature; indeed many of these glories are shared with the devarealms (e.g., the Trayastrimsa and Paranirmitavasavartin heavens) and even with parts of our own world (e.g., the northern continent of Uttarakuru, or the continent of Jambudvlpa itself during Mai trey a's future time).9 The essential feature of a Pure Land is thus not its physical attributes, lovely as they may be, but the opportunity to live in the presence of a Buddha. By defining "Pure Land Buddhism" in this more inclusive sense, we will be able to ask a number of important questions of this material as a whole. What brings such delightful world-systems into being, and what must one do

to be born there? How are the (previous) practices of the presiding Buddha in such a land, on the one hand, and the (current) practices of the community of his devotees, on the other, related to traditional Mainstream10 and Mahayana ideas of the paths leading to Arhatship and to Buddhahood? To what extent are such Pure Lands analogous to the heavens in which many early Buddhists hoped to be reborn, and in what respects do they differ (in both form and function) from such worlds? In short: in what ways are ideas about Pure Lands and the possibility of rebirth there continuous with earlier Mainstream and Mahayana notions, and in what ways do they represent genuine innovations? By viewing a wide range of such materials in conjunction we will not only be able to gain a clearer sense of the process of development of Pure Land thought and practice in India, but also to get a better sense of

9. There was clearly a widely accepted body of ideas in India about what an ideal world should be, for the same attributes - soft earth, golden color, pleasant breezes, fragrant scents, easily accessible food and clothing, abundant pools, flowers and fruit, a large population, and so on - recur in a wide range of literature. The appearance of these "utopian tropes" is thus not sufficient in itself to support the argument that one such text is directly related to another. On the most unexpected of such tropes - the notion that an ideal world is entirely flat - see below, note 23. 10. I am using the term "Mainstream" in the sense suggested by Paul HARRISON (in The Samadhi of Direct Encounter with the Buddhas of the Present [[[Tokyo]]: The International Institute for Buddhist Studies 1990], p. xviii, n. 8) to refer to those Indian Buddhists who continued to pursue the traditional path to Arhatship rather than adopting the newer option of the bodhisattva vocation.

what is unique to traditions concerning Amitabha and what features the Amitabha scriptures share with the larger Pure Land tradition. The result of such a study should thus be of use not only to those interested in Indian Mahayana Buddhism in general, but also to those concerned specifically with the development of faith in Amitabha. In a paper of this length it will not, of course, be possible to deal with the full range of Indian Pure Land literature.11 Instead, I will concentrate on what are arguably two of the earliest extant Pure Land texts: the Aksobhyavyuha,12 on the one hand, and the larger Sukhavativyuha on

11. By "Indian Pure Land literature" I mean not only those few texts that have survived in Indic-language versions (such as the shorter and longer Sukhavativyuha sutras) but also scriptures that were surely composed in India but have been preserved only in Chinese and/or Tibetan translations. 12. The two extant Chinese translations are Taisho no. 313, Sf&lffiHS (translated by Lokaksema no later than 186 CE) and T no. 310(6), ^fWlklMlk (translated by Bodhiruci in 706-713 CE). The sole Tibetan translation, 'Phags-pa de-bfin gsegs-pa Mi- 'khrugs-pa 'i bkod-pa ies bya-ba theg-pa chen-po 'i mdo (Stog Palace no. 11 [6], Peking/Otani no. 760[6], Derge/Tohoku no. 50, etc.) by Jinamitra, Surendrabodhi, and Ye-ses-sde, is undated, but was probably produced in the late 8th or early 9th century. No Sanskrit or Prakrit version of this sutra has survived; a

transliterated title Aksobhya-tathdgatasya-vyuha is given in the Tibetan version, but this is in all probability (like many other such titles) only a reconstruction. 13. Following the chronology set forth by FUJITA Kotatsu IS EH ;£ St in his Genshi jodo shiso no kenkyu ]g#p?#±S$l£>ffi£E (Tokyo: Iwanami Shoten 1970), most scholars now hold that the earliest extant Chinese translation of the larger Sukhavativyuha is T No. 362, attributed to Chih Ch'ien and assigned to the period 223-253 C.E., followed by T No. 361 (attributed to Lokaksema, but considered by FUJITA to be the work of Po-yen, c. 258 CE), No. 360 (attributed to Sahghavarman, but attributed by FUJITA to Buddhabhadra and Pao-yiin, c. 421 C.E.), The attribution of T Nos. 310[5] (translated by Bodhiruci in 706-713) and 363 (translated by Fa-hsien in 991 CE.) is not controversial. More recently, however, in a paper presented at the IABS meeting in Lausanne in August 1999 Paul HARRISON offered detailed evidence pointing to the likelihood that T No. 362 may be the work

(or a revision of the work) of Lokaksema, and that T No. 361 should be assigned to Chih Ch'ien (i.e., that the attributions were switched by early cataloguers). It is hoped that this important study will soon be available in print. The relative ages of the shorter and longer Sukhavativyuha sutras (as they are known to Western scholars) is still a matter of debate. I am somewhat inclined to consider the longer version to be the earlier one, though arguments can be made in either direction. In any event it should be emphasized that we are not dealing here with the condensation or expansion of a single text, but with two quite different sutras on the same topic. Whatever the date of the composition of the

the other. Since the Aksobhyavyuha is far less familiar to most scholars of Buddhism I will first describe in some detail what is said in this sutra about Aksobhya's world and the possibility of rebirth there. Having done so, I will then turn to a comparison of this text with the Sukhavativyuha, showing that a thorough study of Aksobhya can indeed illuminate our understanding of Amitabha and his world.

The Emergence of Pure Land Thought in India Any study of early Mahayana Buddhism is subject to one overarching constraint: the absence of any written sources that could document directly the nascent phase of these new ideas and practices. To put it another way, the initial stages of the development of ideas about the practice of the bodhisattva path took place off-camera, and only after the basic ideas associated with this practice had undergone considerable development were the earliest texts that we now refer to as Mahayana sutras composed. Rather than showing us the incipient phase of Mahayana thinking, these scriptures already represent a somewhat later phase of development, in which the viability of the bodhisattva path (at least for some members of the Buddhist community) is already taken for granted.14 Much the same problem attends the study of

the subset of early Mahayana thought with which we are concerned here: the emergence of ideas about other Buddhas and other worlds. What I will attempt to do in this section, therefore, is not to establish precisely when and in what form Pure Land ideas first appeared in India (for our sources do not allow us the luxury of such specificity), but simply to review briefly what the earliest Mahayana sutras translated into Chinese can tell us about ideas and practices associated with these Buddha-figures. Such texts (for which a precise or at least an approximate translation date is generally known) cannot of course provide us with an absolute chronology of developments in India, but they at least offer us a terminus ante quern, a date by which the ideas and practices they contain must have been known at their Indian source.

shorter Sukhavativyuha in India, no Chinese translation is attested prior to that of Kumaraijva, completed in 402 CE (T No. 366). The only other Chinese version is that of Hsiian-tsang (T No. 367), translated in 650 CE. 14. I have discussed this issue in greater detail in a forthcoming study of the Ugrapariprccha-sutra, provisionally titled A Few Good Men: The Bodhisattva Path According to the Inquiry of Ugra.

Pure Land assumptions already play a central role in a number of Buddhist scriptures translated into Chinese prior to 200 CE. There are numerous references to the "Buddhas of the ten directions," i.e., to Buddhas who are presently living and teaching the Dharma in other worlds, itself an innovative concept vis-a-vis earlier Buddhist ideas of a single universe with long intervals between the appearance of Buddha figures. But among these many Buddhas two in particular - Aksobhya and Amitabha - receive by far the most attention. Among the small number of Buddhist scriptures whose appearance in Chinese prior to 200 CE can be confirmed,15 Aksobhya is the subject of one entire (and quite lengthy) sutra, the Aksobhyavyuha MT$$>MM (Taisho No. 313, translated in 186 CE or before) and receives substantial attention in another, the Astasdhasrika-prajnaparamita-sutra M?T$&^=fj§^ (T No. 224, translated in 179 CE). Neither of these sutras ever mentions Amitabha, but he is referred to several times in the

Pratyutpannabuddhasammukhdvasthitasamddhisutra IS^HB^M (T No. 418, translated in 179 CE) as one of the numerous Buddhas of the ten directions who may be visualized in meditation. Both Buddhas, therefore, must have been well known in India prior to the translation of these texts, all of which are the work of the pioneering Yueh-chih translator, Lokaksema %MMWkBy the middle of the third century still other scriptures dealing with these figures had been translated into Chinese, of which we may mention in particular two attributed to Chih Ch'ien the larger Sukhavativyuha MKH^HMflWM^AMM (T No. 362, assigned to the period 223-253 CE), which of course is devoted entirely to Amitabha,16 and the Vimalakirtinirdesa MMWtM (T NO. 474, translated in 223-228 CE), in which the name of Amitabha is mentioned once in

15. For an authoritative discussion of which translated sutras can legitimately be assigned to this early date see Erik ZURCHER, 'A New Look at the Earliest Chinese Buddhist Texts," in Koichi Shinohara and Gregory Schopen, eds., From Benares to Beijing: Essays on Buddhism and Chinese Religion (Oakville, Ontario: Mosaic Press 1991), pp. 277-304. On the translations of Lokaksema see also Paul HARRISON, "The Earliest Chinese translations of Mahayana Buddhist Sutras: Some Notes on the Works of Lokaksema," Buddhist Studies Review, vol. 10, no. 2 (1993), pp. 135-177. 16. If the revised attributions proposed by HARRISON are accepted - as I am strongly inclined to do - T No. 361, and not T No. 362, would now be considered the work of Chih Ch'ien. In either event the translation of a version of the Sukadvativyuha by Chih Ch'ien during the early to mid-third century CE is assured.

passing (in a list of various Buddhas), while Aksobhya and his world are discussed in considerable detail. While still other Buddhas and their worlds would eventually appear in translated scriptures - most notably perhaps the (not yet actualized) world of MafijusiT in the MahjusrTbuddhaksetragunavyuhasutra 3t^Sf ffjf$±iir§M (T No. 318), first translated into Chinese by Dharmaraksa j^SfeSl in 290 CE, and the world of Bhaisajyaguru, described in chapter 12 of the [[Kuan-ting] ching]] WHRM. translated by Po Srimitra in the first half of the fourth century - the evidence provided by the earliest Chinese translations points clearly in the direction of viewing Aksobhya and Amitabha as the first and most important Buddhas of their type. The fact that Aksobhya appears to be better represented than Amitabha in scriptures translated prior to the beginning of the third century

cannot of course be treated as decisive evidence for the situation in India.17 There is no reason to think that the scriptures that happened to have arrived in China by that date were at all representative of the body of Buddhist literature then circulating in India, nor for that matter that translators such as Lokaksema actually succeeded in rendering into Chinese all the Indian texts that were available to them. All we can say for sure, based on the Chinese evidence, is that these early translations demonstrate with certainty that the cult of Aksobhya (to a significant degree) and the cult of Amitabha (perhaps to a lesser degree) were already well established in India by this time. In scriptures translated from the late third century onward, however, the relationship between Aksobhya and Amitabha is reversed, for no new scripture devoted wholly to Aksobhya is ever translated (though the Aksobhyavyuha is retranslated once), while works extolling Amitabha (including several re-translations of the larger Sukhavativyuha, two translations of the

17. Indeed there is considerable reason to suppose the opposite. To take just one example: the scriptures translated into Chinese during the 2nd-4th centuries CE. are overwhelmingly Mahayanist in content, while as late as the mid-7th century CE. (when Hsiian-tsang recorded his famous statistics on the number of members of various Buddhist schools) Mahayanists still represented less than half of the Buddhist population in India (for a convenient tabulation of Hsiiantsang's figures see Etienne LAMOTTE, Histoire du bouddhisme indien [Louvain: Institut Orientaliste 1958], pp. 597-600). It seems quite possible that partisans of the Mahayana path appeared in China as missionaries and translators in disproportionate numbers precisely because they were a minority - and in some cases a despised one - in their own homeland. shorter Sukhavativyuha, and one version of the apocryphal Kuan Wuliang-shou ching MMmWM) appear with considerable frequency. There is some evidence to suggest that this

increasing attention to Amitabha is an accurate reflection of the situation in India, for it is during this same period of time that we see a proliferation of references to Amitabha in Indian Mahayana texts.18 A particularly intriguing tidbit of evidence is found in the Vimalakirtinirdesa, where a change in the sequence of names in a list of Buddhas suggests that the cult of Aksobhya was gradually being eclipsed by that of Amitabha.19 We will return to the difficult question of the chronological relationship between these two Pure Land figures below. First, however, we must examine the contents of the Aksobhyavyuha in detail. The Nature of Aksobhya's World Though the existence of Aksobhya's eastern paradise is taken for granted in several early Mahayana sutras, it is in the Aksobhyavyuha that this world is discussed in the greatest detail. I will rely primarily on this scripture, therefore, in the discussion given below.20 It should also be pointed out that the considerably shorter discussions of Aksobhya and his realm contained in texts like the Astasdhasrika and the Vimalakirti conform to the description given in the Aksobhyavyuha in most of the relevant details.21 Thus there is every reason to believe that a coherent

18. For a discussion of a number of such references and their implications for the history of Pure Land Buddhism see Gregory SCHOPEN, "SukhavatI as a Generalized Religious Goal in Sanskrit Mahayana Sutra Literature," Indo-Iranian Journal vol. 19 (1977): 177-210. 19. Aksobhya appears first after Sakyamuni in the list of Buddhas given in Chih Ch'ien's translation of the Vimalakirtinirdesa (T No. 474, 14.529a7-8), while by the time of KumarajTva's translation in the early 5th century Amitabha (who appears in sixth place after Sakyamuni in Chih Ch'ien's version) has now been moved into first place (T No. 475, 14.548M4-16). Amitabha remains in first place in Hsiian-tsang's mid-7th century translation (T No. 476, 14.574b8-ll), and the same order is found in the Tibetan version (Peking/Otani No. 843, vol. 34, 90.2.8). 20. All citations from the Aksobhyavyuha, unless otherwise indicated, will be taken from the earliest available version of the text, translated by Lokaksema in the late 2nd century CE (see above, n. 12). 21. The sole exception is that the Vimalakirti refers to the presence of mountains in Abhirati.

body of thought concerning the celestial realm of the Buddha Aksobhya was already circulating in India, at least in certain Mahayana circles. The first point to note is that Aksobhya's realm is not, in the technical sense, a heaven: on the contrary, it comprises an entire world-system (lokadhatu) endowed with heavens of its own. Indeed the sutra makes much of the fact that in Aksobhya's world the human realm and the Trayastrimsa Heaven are connected by a staircase, and that the gods frequently descend to the human realm, drawn by the presence of Aksobhya there.22 Abhirati is thus a multi-layered universe much like our own Saha world, but with two important exceptions: it lacks the lower three realms, or durgath (hell-beings, animals, and ghosts), and it lacks Mt. Sumeru and the other mountain ranges that are so central to Indian (including Buddhist) cosmology.23 In other respects, however, Aksobhya's land is clearly modeled on that of Sakyamuni, so much so that the human realm within it is even referred to repeatedly as "that Jambudvipa." It is thus not a heaven in the traditional Buddhist sense that is, a realm located in the upper reaches of the Desire Realm or in the realms of Form or Formlessness - but an entire (if slightly truncated) world-system, shorn only of what the Aksobhyavyuha's authors apparently considered to be our own world's most unattractive features. In a number of respects Aksobhya's world appears simply as a much improved version of our own. Here we find no reference to the "apparitional birth" (hua-sheng {b^fe) by which living beings are born into the various heaven-realms (or, for that matter, into SukhavatI);

22. T 11.757a-b. 23. The absence of mountains is a regular feature of ideal lands in Indian Buddhist literature, including Amitabha's world, the future worlds of the various sravakas predicted to Buddhahood in the Lotus Sutra, and even our own Jambudvipa during the time of the future Buddha Maitreya. Given the centrality of mountains (and indeed, their positive valence) in Indian cosmographic thought, it seems surprising that they should be entirely absent from these Utopian realms. One possible explanation is that the "flat-earth" scenario did not emerge out of Indian Utopian speculation, but was borrowed from another source. In fact this motif appears with some regularity in Iranian apocalyptic literature, where - in contrast to the Indian texts, where the motif of flatness is isolated and plays no productive role - it is explicitly associated with a leveling of social status, and thus with the promise (or threat, depending on the text in question) of an egalitarian society. For a discussion of this motif and the Iranian

sources in which it occurs see Bruce LINCOLN, "The Earth Becomes Flat' - A Study of Apocalyptic Imagery," Comparative Studies in Society and History vol. 25 (1983): 136-153. rather, men and women are born in the normal manner, but without any impurity or suffering on the mother's part. The version preserved in Tibetan translation supplies additional details not found in the Chinese, hastening to add that in Aksobhya's world birth does not result from ordinary sexual intercourse. On the contrary, whenever a man looks at a woman with desire (for in this world desire has not been completely eliminated) his lust is immediately cooled, and he enters into a state of samadhi; as for the woman, she immediately conceives a child.24 All this takes place, in other words, without any physical contact between the "parents" whatsoever. Just as the manner of conception and birth is simply a more rarefied version of processes that take place here in this Jambudvipa, so are the other physical aspects of Aksobhya's realm best described as upgraded versions of our own. His land is free of sickness, people are never ugly* and (on a doctrinal note) there are no "heretical religions" there. Jewelry and clothing grow on trees, and once picked these garments always remain fresh and clean, imbued with the scent of heavenly flowers (thus averting the drudgery of laundry). Nor

does food need to be planted, harvested, or cooked: like the gods of the Trayastrimsa Heaven, as soon as the inhabitants of Abhirati think of food and drink, they immediately attain whatever they desire. In Aksobhya's world people do not have to exert any effort to earn a living, and buying and selling are unknown. Thus those fortunate enough to be born in Abhirati are free to relax and enjoy a paradise-like climate free from the extremes of heat or cold, where a gentle, scented breeze blows in accord with people's wishes.25 I will not enumerate here all of the myriad features of Abhirati, virtually all of which (an abundance of jewels, lotuses, ponds, celestial music, and so on) will be familiar to those who have studied other Pure Land texts. Before moving on to describe other aspects of life in this land, however, we should pause to take note of the use to which these enticing features are put within the text. Contrary to what we might expect, the sutra does not use these attractive qualities - not, at least, in this portion of the text26 - to encourage rank-and-file Buddhists to look

24. Tib. 74.7-75.3. Unless otherwise indicated, all references to the Tibetan text are to the version contained in vol. 36 of the Stog Palace edition. 25. 11.755c-756b. 26. Cf. however what I have described below as a "coda" to the sutra, in which devotees are urged to seek rebirth there by remembering and reciting the sutra itself. forward to rebirth in Abhirati. Instead the delightful features of that land are marshaled to elicit a very different response: bodhisattvas are urged to study and emulate Aksobhya's conduct so that they will eventually obtain such a world for themselves.27

The Practice of Buddhism in Aksobhya's World Not all of the Aksobhyavyuha is devoted to enumerating the physical attributes of the Abhirati realm. Considerable attention is also given to how those fortunate enough to be reborn there will carry out Buddhist practices once they arrive. In particular, the text describes at length how much easier it is to attain Arhatship in Aksobhya's world than in our own. Innumerable listeners attain Arhatship each time Aksobhya preaches the Dharma, and those who require four such lectures to progress step by step from stream-enterer to Arhatship are considered the "slow learners" of the group. No one, apparently, will require rebirth elsewhere before attaining final liberation; thus birth in Aksobhya's land is tantamount to the last birth of the non-returner. The description of life in Abhirati closely resembles that of an

idealized monastic community. Aksobhya's disciples do not need to beg for food, nor do they have to cut and sew their own monastic robes; robes and bowls simply appear before them as needed, and at mealtime their bowls automatically fill up with food. Nor is there any need for the monks and nuns to wash their dishes, for at the end of the meal their bowls simply disappear.28 Since no one in Abhirati would even think of doing an evil deed, Aksobhya does not confer the prohibitive precepts on his congregation, but preaches only about the positive aspects of the Dharma,29 and the entire congregation listens attentively as he does.30 When one of Aksobhya's disciples at last enters into extinction the ground quakes in recognition, and many of them exhibit various marvels (such as spontaneous disappearance, self-cremation in the sky, or the emanation of a rainbow body) as they pass into final nirvana.31 Not all the inhabitants of Aksobhya's realm, however, are engaged in pursuing the path to Arhatship. Comparably gargantuan numbers of 27. This theme is repeated throughout the sutra, but see especially 11.756b22-24. 28. 11.757M6-22. 29. 11.757b22-28,757c4-10. 30. 11.757c21. 31. 11.757c26-758a6. bodhisattvas are also present, and just as was the case with Aksobhya's iravafca-assembly, the bodhisattvas reborn in his land have a much easier time pursuing their chosen

path than do their counterparts in our world. Whenever Aksobhya preaches, for example, those bodhisattvas will be able to remember and recite all that they hear.32 Though Mara is present in Aksobhya's world - or rather, though Abhirati has its own Mara figure - he will not attempt to obstruct the bodhisattvas' progress, and without such interference they will all be able to attain the state of nonretrogression from Buddhahood.33 Despite all the wonders of Aksobhya's realm, the bodhisattvas there will not be content simply to enjoy life in Abhirati itself, for the sutra emphasizes the fact that they will be able to travel freely to other Buddha-worlds. Upon their arrival they will make offerings to the Tathagata who resides there, listen to his Dharma-teachings, and clarify their understanding by asking relevant questions before returning to Aksobhya's realm.34 Elaborating upon the straightforward description of this process found in Lokaksema's version, the editor of a later recension displays a rather cosmopolitan sense of the difficulties that sometimes attend travelers to other cultures:

... if those bodhisattvas wish to go to another Buddha4and, they will no sooner think of that land than they will arrive there, wearing the native costume, speaking the local language with no accent, and acting in harmony with the customs of that land.35

It is noteworthy, incidentally, that only bodhisattvas (and not sravakas) are described as engaging in this inter-galactic travel, an issue to which we will return below. The benefits experienced by bodhisattvas in that land continue to accrue even after death, for the sutra tells us that they will be able to see all of the nine hundred ninety-six Buddhas who are yet to come in this Good Aeon (bhadrakalpa),36 and will be reborn in one Buddha-field after another until enlightenment is attained.37 In this connection the 32. 11.758bl-2,758cl-2,etc. 33. 11.758c24ff.,759bl6ff.,etc. 34. 11.758b9-13. 35. This passage is found only in Bodhiruci's version; see 11.107a27-28. The English translation is taken from CHANG, Treasury, p. 327 (cited above, n. 4). 36. 11.758M3-15. 37. 11.760al6-18,760b7-8,etc.

theme of being able to remember the Buddha's teachings is sounded once again, for it is said that even across the cycle of death-and-rebirth, these bodhisattvas will never forget the sutras they have heard.38

The Conclusion of Aksobhya's Career Glorious as it is, Aksobhya's lifespan as a Buddha will not last forever, and the Aksobhyavyuha devotes considerable attention to the circumstances that will attend his demise. On the last day of his life Aksobhya will send out magically produced versions of himself which will appear throughout all the worlds, preaching the Dharma and causing sentient beings to attain Arhatship.39 He will also issue a prophecy (vyakarana) to his successor, the bodhisattva Gandhahastin, predicting his future attainment of Buddhahood.40 Upon his entry into extinction various auspicious portents will occur (the shaking of the earth, the sound of a great roar, etc.).41 Finally Aksobhya will bring forth fire from his own body,42 thus performing the same kind of self-cremation previously displayed by many of the Arhats in his realm. After his death Aksobhya's Dharma will endure for several hundred thousand kalpas,43 after which it will gradually fade away as the inhabitants of his world lose interest in the Buddhist teachings.44

The Making of a Buddha-Field: Aksobhya's Path to Buddhahood Like the longer Sukhavativyuha, the Aksobhyavyuha begins with an encounter between a Buddha and a monk. Inspired by the preaching of a Buddha known as "Great Eyes" (Ch. ta-mu kuang-mu If |=[; Tib. Spyan chen-po),45 the initially nameless monk who is to become 38. 11.758c5-7. 39. 11.760b26-28. 40. 11.760b28-c2. 41. 11.760c5-8. 42. 11.761al3. 43. 11.761bl4-15. 44. 11.761b20-24. Even in its later recensions the Aksobhyavyuha does not refer to any of the various periodization systems that some Indian writers used to divide the duration of the Buddhist religion into periods of saddharma and saddharmapratirilpaka. 45. The underlying Sanskrit name is uncertain, but Visalanetra - a word which occurs as an epithet of the Buddha in the larger Sukhavativyuha, §9(6), and also at Mvy. 678 as the name of a bodhisattva - would be one possible candidate.

Aksobhya states his intention to undertake the training of the bodhisattva.46 Rather than simply validating his disciple's ambition, however, the Buddha offers a realistic caution. Just as a contemporary professor of Buddhist Studies might try to discourage an eager student from setting out on the long and difficult path that would culminate in the student's becoming a professor herself, so Great Eyes attempts to dissuade his eager disciple. The training of the bodhisattva is very difficult, he points out, implying that the monk should consider the matter carefully before embarking on such a daunting course. Specifically, the Buddha states that the bodhisattva path is extraordinarily demanding because a bodhisattva must bear no malice toward any living being 47 The monk is not easily dissuaded, however, and he immediately pronounces a series of resolutions, beginning with the promise not to bring forth anger, malice or ire toward any living being from then on until his attainment of Buddhahood. Impressed by this long list of vows,

another unnamed monk in the audience suggests that it would be good if this bodhisattva-monk were to be called Aksobhya ("the unperturbed"). Others present in the assembly also take up this usage, and "Aksobhya" becomes the monk's name for the remainder of his lifetimes. Aksobhya then goes on to make an extensive series of vows detailing the specifics of his intended practices. And the majority of these reflect the acts of strenuous self-denial that the writers of a number of early Mahayana sutras (e.g., the Ugrapariprccha, the Kasyapaparivarta, and the Ratnardsi) saw as essential prerequisites for attaining Buddhahood: not simply observing all ten of the "good deeds" (kusala-karmapatha), though these are enumerated in full, but leaving home in every lifetime to become a monk, practicing the twelve (sometimes thirteen) ascetic practices known as dhiitagunas,4* maintaining utter detachment from one's audience while preaching the Dharma, and so on. The validity of Aksobhya's vows - that is, the fact that he will actually succeed in carrying them out - is subsequently confirmed by an

46. 11.752al-2. 47. 11.752a3-5. 48. Only four of the most important of these - wearing a patchwork robe, limiting oneself to the three robes of the sramana, dwelling at the foot of a tree, and constantly standing, sitting, or walking (i.e., never lying down) - are singled out for attention in Lokahsema's version (11.752bl 1-12 and 23-25); in Bodhiruci's translation and in the Tibetan, however, the list of dhutagunas is spelled out in full.

"Act of Truth" voiced by another monk in the audience.49 If it is true that Aksobhya will succeed in fulfilling his vows, the monk proposes, "may he press the earth with the finger of his right hand and cause a great quake."50 The earth indeed shakes six ways in confirmation, and Great Eyes then confers on Aksobhya the formal prediction (vyakarana) of his future Buddhahood, a prophecy which - as Sakyamuni remarks parenthetically to Ananda - is just like the one Sakyamuni himself received from the Buddha DIpamkara.51 The Bodhisattva Path According to the Aksobhyavyuha But that is not the end of the story. The sutra goes on to describe how Aksobhya successfully carried out his vows, never shrinking from giving away parts of his body and going from Buddha-field to Buddha-field in one life after another, always making offerings to the Buddhas there and practicing brahmacarya in their presence.53 The story of Aksobhya makes explicit, in other words, the kinds of activities that early

49. The "act of truth" (for which several different Indian terms - including Pali saccakiriyd and Skt. satyavacana, satyavddya, satyavdkya, etc. - are used) is often conflated with the bodhisattva's vows (pratijnd, pranidhdna) in Japanese and Western scholarship, but it is in fact a speech-act of a very different type. The act of truth is not a promise or pledge to do something; on the contrary, it is based on a simple declaration of a fact (though this may be a fact that is to occur in the future). What makes this an "act" rather than a mere pronouncement is the request made by the speaker for a confirmation of his or her statement by the forces of nature: "If what I have said is true, may

this river run backwards," for example, or "If Aksobhya will succeed in carrying out his vows, may the earth quake in response." Such an act of truth may occur, as here, when a vow or series of vows has just been made, but it may also occur independently, as when the prostitute BindumatI causes the Ganges to flow backwards in confirmation of her statement that she has treated all of her customers equally regardless of their social status (Milindapahha 4.1.42), or when King Sivi's eye is restored by his statement that "whatsoever sort or kind of beggar comes to me is dear to my heart" (Jataka no. 499). What is common to all accounts of the performance of an act of truth is that a proposition is first set forth and then confirmed (or disconfirmed, as the case may be) by a miraculous response from the environment. For a thorough discussion of the act of truth as presented in both Buddhist and Hindu sources see W. Norman BROWN, "Duty as Truth in Ancient India," Proceedings of the American Philosophical Society, vol. 116, no. 3 (1972), pp. 252-268. 50. 11.753al8-19. This is of course the bhumi-sparsa-mudrd. 51. 11.753M0-15. 52. 11.754b26-27. 53. 11.754c2-5.

Mahayana Buddhists believed were required in order to amass the vast amounts of merit needed to procure all the qualities of a Buddha.54 Using a script supplied at least in part by the jdtaka tales, these pioneering bodhisattvas had to look forward to thousands of lifetimes of selfsacrifice before Buddhahood could be attained. And this self-sacrifice did not consist simply of renouncing one's belongings or being kind and compassionate to others; on the contrary, it required dramatic acts of "giving up the body" (dtmabhava-paritydga), narrated in such tales as the Khdntivddi-jataka (no. 313), where the bodhisattva as a renunciant sage (rsi, Pali isi) is cut to pieces at the order of an

evil king, or the VydghrT-jdtaka (found in, among other sources, chapter 19 of the Sanskrit Suvarnaprabhasasutra) in which the bodhisattva as a young prince sacrifices his body to feed a hungry tigress and her cubs. It is no wonder, then, that the bodhisattva is referred to in a number of early Mahayana sutras as a "doer of what is hard" (duskara-kdraka).55 But there were other means through which merit could be attained as well. From the very beginning of Buddhist history, the act of giving (ddna) to members of the Buddhist monastic community was seen as the merit-making activity par excellence. Though the notion of the recipient as a "field of merit" (punya-ksetra) was subject to a variety of interpre

54. Opinions on the length of time required to complete all the prerequisites for Buddhahood varied according to school, but one common figure was three asamkhyeya kalpas and one hundred mahakalpas. See for example the Ta chih-tu lun -X^WW, (T No. 1509), 25.86c-87c and the French translation in Etienne LAMOTTE, Le traite de la grande vertu de sages se, vol. 1 (Louvain-la-Neuve: Institut Orientaliste 1981), pp. 246-255. According to the Ta chih-tu lun each of the thirty-two marks of a Buddha is "adorned with one hundred merits." Each one of these merits, in turn, is the equivalent of the amount of merit necessary to become a cakravartin (according to some sources cited by the Traite), Indra (according to others), or Mara, the chief of the Paranirmitavasavartin gods (according to yet another opinion). Some authorities gave even more immense equivalents, e.g. "the

collective merit of all beings at the end of the kalpa," which is the amount of merit that results in the re-formation of a trisahasramahasdhasra world-system after a period of dissolution (25.87b; Traite I, 250-251). 55. The prominence given to the Lotus Sutra (with its promises that even a child who offers to the Buddha a stupa made of sand will eventually become a Buddha himself) in both Western and East Asian interpretations of Mahayana Buddhism is probably one of the factors (together with a certain squeamishness about bloody acts of self-sacrifice) that have made it difficult for modern readers to appreciate the extent to which the bodhisattva career was viewed by most Indian Buddhists as an excruciatingly difficult path.

tations, in at least some circles it was maintained that it was not merely the gift itself, or even the intention with which it was given, but the virtuousness of the recipient that determined the amount of merit the donor received. What better object of giving could there be, therefore, than a fully enlightened Buddha? Thus for a bodhisattva-in-training the possibility of meeting with an endless series of Buddhas - in the course of "traveling from Buddhafield to Buddha-field," as so many early Mahayana sutras put it - is not merely optional but required, for there is simply no other way to attain the vast quantities of merit required in order to become a Buddha oneself. "Serving billions of Buddhas" thus becomes a kind of internship, as it were, during which the bodhisattva can learn how to be a Buddha by serving one, while simultaneously amassing vast stores of merit by making offerings to the Best of Men. But there is yet another prerequisite to becoming a Buddha according to these early texts. Not only the equipment of merit (punya-

sambhara) but also the equipment of knowledge (jhana-sambhdra) is required. Specifically, the bodhisattva must attain the same degree of knowledge that the Buddha had (now frequently referred to as sarvajfia or "omniscience"), a knowledge that seems to be associated, for many early Mahayana thinkers, with the teachings contained in Buddhist Dharmatexts. Hence the emphasis on traveling to other Buddha-fields - even while inhabiting a place as glorious as Aksobhya's realm - to hear additional discourses on the Dharma, as well as the importance placed on being able to actually remember those discourses, not only during one's present lifetime but even after death. Finally, it should be noted that while myriads of beings succeed in attaining Arhatship in Aksobhya's land, not a single bodhisattva (other than, of course, Aksobhya himself) is described as attaining Buddhahood there. Indeed such cannot be the case, for (as the Tibetan version makes explicit) "since only one bodhisattva, not two, can attain anuttarasamyaksambodhi" in a given time and place, those who want to attain Buddhahood "should do as Aksobhya did."56 Those who wish to become Buddhas, in other words, cannot simply do so in Aksobhya's encouraging presence, but must themselves retrace the steps of his pa

th.57 56. Tib. 195.2-6. 57. 11.753a21-25; cf. the Tibetan version at 31

This brings us to yet another factor that must surely have contributed to the formulation of ideas concerning other Buddha-worlds, a problem that we might describe (in contemporary parlance) as "bodhisattva jobmarket crowding." As more and more people within the Buddhist community opted for Buddhahood rather than Arhatship, the problem of where one might find "employment" as a Buddha emerged in sharp relief. Given the axiomatic assumption that there could be only one Buddha in a given world at a time (for the very definition of a Buddha is one who discovers the path to enlightenment by himself and then teaches it to others in a world where no "Buddhism" exists), one could not of course become a Buddha while still a member of an existing Buddhist community; the attainment of the final goal would have to wait until a later life when one is reborn into a

world with no knowledge of the Buddha's teachings. Those desiring to become Buddhas in our own world-system might, of course, simply get in line behind the bodhisattva Maitreya (recognized by virtually all Buddhists as the next Buddha-tobe), but this would mean a wait of several billion years, as Maitreya's own descent from the Tusita heaven was not expected to occur for some five and a half billion years.58 Alternatively - and much more appealingly - one might seek rebirth in another realm whose qualities are the result

of one's own bodhisattva activities, and where Buddhahood could far more quickly be attained. All of the above suggests a radically new way of understanding the emergence of so-called "Pure Land" ideas in Indian Buddhism. No longer do these paradise-like realms appear as a concession to the needs of an under-achieving laity, much less as evidence for the incorporation of foreign (e.g., Iranian) or non-Buddhist (e.g., Hindu) ideas. On the contrary, the existence of other Buddha-fields now appears as a logical necessity, elicited by the mainstream understanding of the requirements of the bodhisattva path itself. Whatever the other factors in Indian culture at the time that might have contributed to this expanded vision of the cosmos,59 the idea of the bodhisattva path as a viable option for a

58. On some of the ways in which Buddhists have responded to this anticipated delay see Jan NATTIER, "The Meanings of the Maitreya Myth: A Typological Analysis," in Alan Sponberg and Helen Hardacre, eds., Maitreya, the Future Buddha (Cambridge: Cambridge University Press 1988), pp. 23-47. 59. A wide-ranging comparative study of the emergence of ideas of other worldsystems in India would be a valuable contribution not just to Buddhist Studies, but to the study of Indian religious history as a whole. small but significant minority within the Buddhist community virtually required that such a worldview be produced. * * * If the existence of other worlds where bodhisattvas might be reborn in the presence of Buddhas in the near future and ultimately become Buddhas themselves was viewed as a

necessity by those those who had taken bodhisattva vows, such worlds surely offered an attractive possibility for rank-and-file Buddhist devotees as well. Before turning to a comparison of the Aksobhyavyuha and the Sukhavativyuha, therefore, we must first pause to consider what the Aksobhya literature can tell us about the emergence of a generalized hope for rebirth in a "Pure Land." It is clear from the outset that, while rebirth in Abhirati is portrayed as an option for many Buddhists, it is far from easy or automatic.

"Those born into [[[Aksobhya's]]] Buddha-world," the gods remark after hearing his vows, "will not [be people who] have inferior merits."60 Somewhat later in the sutra the difficulty of entry into Abhirati is made even clearer when an unnamed monk in the audience naively expresses his desire to be reborn in Aksobhya's world and is rebuked by Sakyamuni: "A deluded man like you cannot be born in that Buddha-field!" Sakyamuni replies. "One who has a lustful mind cannot be born there. Beings are born there by virtue of meritorious conduct, righteous conduct, pure conduct [[[brahmacarya]], i.e., celibacy] and correct conduct."61 Based on these statements it would seem that birth into Abhirati

operates in much the same way as birth into the various heavens known in earlier Buddhist literature: it is a reward for one's own merit, and considerable quantities are required. This stance is maintained throughout most of the Aksobhyavyuha, with repeated mentions of the importance of dedicating one's merit to rebirth in Abhirati (for which, of course, one must have accumulated the requisite merit). Toward the end of the sutra, however, it suddenly appears that birth in Abhirati might be much easier than the text had previously suggested. Now we are told that rebirth in Aksobhya's presence can rather easily be ensured - not by visualizing him or remembering his name (as those familiar with the Sukhavati literature

60. 11.753b3 (reading with note 6 to the Taisho edition; cf. Bodhiruci's version, 11.103b4). 61. 11.756al 8-22. Bodhiruci and the Tibetan have essentially the same reading. might suppose), but rather by the acceptance, memorization, and diffusion of the text of the Aksobhyavyuha itself.62 What we have here, in other words, is a series of classic "book-cult" exhortations of the type that appear at the conclusion of so many Mahayana sutras.63 * * #

Having examined the contents of the Aksobhyavyuha in some detail, we may now conclude with some comparative observations on the content of the larger Sukhavativyuha.64 These two texts - both of which describe the career of a bodhisattva-monk from his initial vow in the presence of a Buddha to his present-day life in a glorious Buddha-field - are strikingly similar in structure and content, and thus they provide an ideal laboratory for comparative analysis. If we use the Aksobhyavyuha as a mirror for comparison with the larger Sukhavativyuha, what can we see in each of these sutras that might have been invisible before?65 Both sutras begin with the narration by Sakyamuni of a long-ago encounter between a Buddha and a monk, but even at the outset they begin to diverge in interesting ways. In the Aksobhyavyuha the Buddha Great Eyes, as we have seen, tries at first to dissuade the future

62. 11.762c-764a. It is interesting to note that the two methods of attaining rebirth in Abhirati described in the Aksobhyavyuha - directing one's accumulated merit toward that end and revering the text of the sutra itself - are the same methods recommended in a set of verses on how to be reborn in Abhirati included in the Mahayana Mahaparinirvana sutra (T No. 375, 12.734a-b); cf. the English translation in John STRONG, The Experience of Buddhism: Sources and Interpretations (Belmont, CA: Wadsworth 1995), pp. 193-195. 63. A casual perusal of the concluding sections of the Mahayana sutras preserved in the Taisho edition of the canon will turn up dozens, if not hundreds, of such passages. 64.

The larger Sukhavativyuha has been chosen for this purpose mainly because its more extensive content allows for a much more detailed comparison. It is also possible (and in my view rather likely) that it is older than the shorter text. The authors of the shorter Sukhavativyuha, incidentally, were clearly aware of the existence of Aksobhya, who is correctly placed (along with several other Buddhas) in the East (§11). In the Aksobhyavyuha, by contrast, there is no mention of Amitabha. 65. All references to the larger Sukhavativyuha in this section, unless otherwise noted, will be taken from the Sanskrit edition edited by F. MAX MULLER. A thorough comparative study of the treatment of each of the issues discussed here in all extant versions of the sutra - Chinese and Tibetan as well as Sanskrit - is highly desirable, but unfortunately this lies outside the scope of the present paper.

Aksobhya from embarking on the bodhisattva path; in the Sukhavativyuha, by contrast, Lokesvararaja makes no attempt to discourage Dharmakara. On the contrary, he simply replies that the monk should obtain for himself the vast array (vyuha) of qualities that contitute a Buddha-field, and then proceeds to tutor him for a full million years on what constitutes this "array." Already we can discern the outlines of two major differences in perspective between these texts. First, it is clear that for the writers of the Aksobhyavyuha the bodhisattva path is not intended for everyone, and should only be undertaken after serious reflection on what it entails. For the authors of the Sukhavativyuha, by contrast, the bodhisattva path is simply taken for granted, and the appropriateness of Dharmakara's "choice" is never called into question. Second, while the future Aksobhya's initial resolution is simply to train himself in the bodhisattva path, Dharmakara's initial resolutions are already focused on the final goal of Buddhahood.

And that goal is understood in a very specific way: it involves, above all, the design and construction of one's own Buddhaland. These foreshadowings are confirmed by what is found later in the texts. In the Aksobhyavyuha, as we have seen, Abhirati is praised as a place where Arhatship can easily be attained, thus making it clear that Arhatship is still viewed as a viable (and indeed valuable) path. In the Sukhavativyuha, by contrast, there is no discussion whatsoever concerning the ease of attaining Arhatship in

Amitabha's land. On the contrary, though Arhats are mentioned briefly,66 the sutra seems to suggest that the only "live option" for members of its Buddhist audience is the attainment of anuttarasamyaksambodhi. But this notion is bifurcated in a peculiar way: a distinction is made between rank-and-file bodhisattvas, described as ekajatipratibaddha "bound to (only) one more birth," and bodhisattvas who have taken on vows to rescue other beings.67 The classical notion of the bodhisattva as one who has vowed to attain 66. Arhats receive considerably more notice in the the two earliest versions of the Sukhavativyuha (T Nos. 361 and 362) than in the extant Sanskrit text or, for that matter, in the Chinese version attributed to Sanghavarman. A thorough comparison of these two early renditions of the Sukhavativyuha with Lokahsema's translation of the Aksobhyavyuha could provide valuable additional information on the stages by which the Arhats gradually became marginalized to the point of nearinvisibility in later recensions of the Sukhavativyuha. 67. The most vivid instance of this bifurcation is in §33.

Buddhahood for the sake of others thus appears here only in vestigial form, suggesting that the idea of the bodhisattva path as a challenging alternative suited for only the few has now been left far behind. What we see in the Aksobhyavyuha, in other words, is the standard scenario of the "three vehicles," according to which some Buddhists - but not all can and should undertake the difficult practices leading to Buddhahood so that they may in turn assist others in attaining Arhatship, just as Sakyamuni did.68 In the

Sukhavativyuha, by contrast, this notion has receded into the background, and the bodhisattva path (now understood in far less rigorous terms) has now been generalized as appropriate to all. In fact these two developments would be expected to occur in conjunction, for it is quite improbable that the bodhisattva path - conceived of in the intensely demanding terms in which it was initially formulated - could be viewed as appropriate for, much less attractive to, the entire Buddhist community. On the contrary, the bodhisattva path in its incipient stages is best viewed not as a new "school" of Buddhism at all but simply as a higher and more demanding vocation suitable for a few within the larger Buddhist sangha, analogous to the monastic vocation

68. It is important to point out that the notion of "one vehicle" (ekaydna) was never universally accepted by Indian Mahayana Buddhists. The Astasdhdsrikaprajndpdramitd-sutra and its descendants, for example, take for granted the validity of all three vehicles even as they direct their message (mainly) toward those on the bodhisattva path, and the widely used fwt-gotra theory is explicitly based on the understanding that the bodhisattva path is not an appropriate choice for all Buddhists. Even Santideva - the 8th-century Madhyamika scholar-monk who so eloquently describes the requirements of the bodhisattva path in his Bodhicary avatar a and Siksasamuccaya - makes it clear at the beginning of

the latter that his instructions are intended not for all Buddhists, but specifically for those who belong to the Buddha-gofra (BENDALL ed., p. 2, line 9). Much confusion has been created by the widespread practice of interpreting all negative comments found in Mahayana sources about "falling to the level of the srdvakas and pratyekabuddhas" (to use the wording found in the Astasahasrikd) as if they were criticisms directed toward a competing Buddhist school. On the contrary to use an academic analogy once again - they are better understood as the exhortations of a professor to a Ph.D. student not to take a terminal M.A. degree and be done with it, but to strive to complete the much more demanding doctoral degree, at which point the student (having become a professor) can then teach others, leading them to the attainment of the (admittedly lower) B.A. and M.A. degrees.

within the Roman Catholic church.69 In those circles where the bodhisattva path came to be viewed as appropriate for all members of the Buddhist community, however, the requirements for being a bodhisattva had of necessity to be reduced. To put it simply: the bodhisattva path as originally conceived was suitable (and indeed intended) only for a small elite within the larger Buddhist community. Once it had been generalized to apply to the entire population of Buddhist practitioners - old and young, male and female, householder and renunciant - the difficult requirements of the bodhisattva path had either to be deferred to a future lifetime for most members of the Buddhist community or to

be radically reduced in the present. To explore the significance of the second difference in perspective foreshadowed above - the centrality of the traditional bodhisattva path in the Aksobhyavyuha versus the centrality of the construction of a Buddha-field in the Sukhavativyuha - we must turn to what is widely considered to be the heart of the SukhavatI literature: the vows made by the monk Dharmakara. These vows are almost too well known to require discussion, but we may summarize them briefly by saying that in a series of resolutions - twenty-four in the earliest extant versions (T Nos. 361 and 362), forty-eight in the version most widely used in East Asia (T No. 360) - the future Amitabha details the specific qualities of his future Buddha-field. Each vow concludes with what might be described as a "sanction" clause, in which Dharmakara states "[If I do not succeed in bringing this about], may I not attain Buddhahood."70 Aksobhya's vows are also followed by a sanction clause, but the formulation of the penalty is entirely different. In wording widely echoed in other Mahayana sutras,71 the future Aksobhya underscores the

69. In the Buddhist case, of course, the bodhisattva option is initially presented as a more rigorous path within the monastic community, thus making it (at least at the outset) an even more demanding religious option. I am drawing here on several sources, including my own study and translation of the Ugrapariprccha, the content of several other early Mahayana sutras, and the work of a number of other scholars including Paul HARRISON, Paul WILLIAMS, and SHIZUTANI Masao PSJESI. 70. Skt. ma tavad aham anuttaram samyaksambodhim abhisambudhyeyam; Ch. BI^fFW in T No. 362, ^MlEM in T No. 360. 71. See for example the ManjusrTbuddhaksetragunavyuhasutra (T Nos. 318, 310[15], 319; Tibetan Peking/Otani No. 760[15]); the Vimaladattapariprccha (T Nos. 338, 3110[33], 339; Pek. No. 760[33]); and the Sumatidarika-pariprccha (T Nos. 334, 335, 336, 310[30]; Pek. No. 760[30]).

seriousness of his vows by stating "[If I should fail to fulfill these vows] I would be breaking faith with all the Buddhas, the Blessed Ones, who are even now teaching the Dharma in incalculable, innumerable, inconceivable, immeasurable world-systems - those Buddhas, the Blessed Ones."72 The notion of breaking a promise made to the Buddhas may not seem, to the contemporary reader, nearly as serious as the prospect of renouncing the possibility of Buddhahood; but as the discussion in Santideva's Siksasamuccaya makes clear, in the Indian religious context, where the relationship with one's guru was of paramount importance, the notion of breaking a pledge made to one's spiritual teacher was a dreadful prospect indeed.73 More important than the difference in the wording of the sanction clause, however, is the difference in the content of the vows themselves. In Dharmakara's case, as we have seen, the majority of his vows are

72. The version I have cited is from Lokaksema's translation, where the sanction clause, repeated after each of the eight sets of vows, reads ^^bWMiWM^WMm^mm^m^mmmm^^mm^^M^m&^ m.752a-c). Bodhiruci's version is generally shorter, reading in most cases simply "that would be to deceive all the Buddhas" fJ&Sfcffi—#J|§# (e.g., 11.102b4-5). In a few cases, however, his version reads "that would be to turn my back on all the Buddhas" SU^JStf—SJflffr (e.g. 11.102b26-27). The same expression occurs in a longer version of the same clause at 102b2-3 ("that would be to turn my back on the Buddhas, the Tathagatas, who are presently teaching the Dharma in immeasurable, innumerable, unlimited world-systems"). The Tibetans seem to have interpreted the underlying optative verb not as a conditional statement ("I would be...") but as an imperative ("may I ..."), for the Tibetan version

reads bdag-gis ... sans-rgyas bcom-ldan-'das de-dag bslus-par gyur-cig, "May I deceive (sic) the Buddhas, the Blessed Ones" (15.6-16.1 and passim). Happily a Sanskrit version of this statement has been preserved in Santideva's Siksasamuccaya, which makes it clear that the underlying verb form was indeed an optative. Singling out Aksobhya's vow to become a monk in every future lifetime, Santideva quotes him as saying visamvdditd me buddha bhagavanto bhaveyur yadi sarvasydm jatau na pravrajeyam "the Buddhas, the Blessed Ones would be broken faith with by me if in all my lives I do not go forth [as a monk]" (BENDALL ed., 14.15). (I do not know the source of the phrasing tdvad aham buddhdms tathagatdn vahceyam ye 'parimanesv asamkhyeyesv aparamitesu lokadhatusv etarhi dharmam desayanti given without attribution in DANTINNE [La splendeur, p. 81]; presumably it is a reconstruction based on the Tibetan cited in his previous note.) Since a bodhisattva necessarily makes his vows in the presence of the Buddhas of the ten directions (who are, after all, all-seeing and all-knowing), the degree of his offence if he should he break one of his vows would be immense. 73. Siksasamuccaya (BENDALL ed.), pp. 12-13.

devoted to a description of the precise qualities of the Buddha-field he intends to create. Aksobhya's vows, however, are entirely different in focus. Rather than describing the particular features of his future Buddha-world, he sets forth in detail the bodhisattva-practices that he promises to carry out. These practices are in no way innovative, but simply draw on standard Mahayana assumptions about the necessity of spending countless eons in the cycle of rebirth, engaging (via self-sacrifice as well as dana and other practices) in the acquisition of the tremendous quantity of merit required for Buddhahood.74 To be sure, the Aksobhyavyuha devotes considerable attention to the delightful features of the Abhirati realm, but Aksobhya is never described as vowing to bring these particular qualities about.75 On the

74. In the earliest version of the Aksobhyavyuha there are eight sets of such vows (11.752a-c), containing an average of five vows each. Following the pronouncement of these eight sets of vows the sutra states that Great Eyes and various gods and humans "stood as guarantors" for the vows, and the Buddha declares that if bodhisattva-monks put on that kind of armor (i.e., take vows of this kind) they will all attain anuttarasamyaksambodhi. 75. Certain passages within the sutra might be viewed at first glance as exceptions, but they can, I believe, be easily explained. In the section of the sutra that follows the eight sets of vows in Lokaksema's version (and are set off from these main vows by the statement that Great Eyes and various gods, asuras and humans witnessed Aksobhya's vows, thus arousing the suspicion that these additional resolutions are an interpolation) the future Aksobhya makes some additional promises, including the promise not to find fault with the members of the fourfold sangha and not to emit semen even in

a dream (752cl8-753al5). Each of these additional two resolutions has a correlate in the nature of his future world: in the first case, that the members of his future sangha will in fact be faultless; and in the second, that the members of his community will be free from certain types of contamination - specifically, that renunciant bodhisattvas will not suffer from the emission of semen even during dreams, and that the women (literally "mothers") of Abhirati will not suffer from any impurity. My point is that even

though these two resolutions appear to be phrased in such a way as to constitute an exception to the rule, they probably originated simply as by-products of actions Aksobhya himself promised to do. In Bodhiruci's version and in the Tibetan, only one of Aksobhya's vows appears to violate this rule: the vow in which he promises that women will have no impurity (now understood, in these two versions, as being free of faults). It is interesting to note that in both versions (or rather, in the branch of the Indian textual tradition that underlies them) this vow seems to have been altered in transmission by an editor familiar with the sanction clause of the Sukhavativyuha: in both cases, the future Aksobhya states that if women in his land are not free of

contrary, the lovely features of his Buddha-field are portrayed as a side effect of the excellence of what he did in fact vow: to undergo intensive and sometimes grueling bodhisattva practices. Just as rebirth in more comfortable circumstances in this world is regularly described in Buddhist literature as the result of good actions which often have no direct connection to the reward received, so the specific features of Aksobhya's world are described as the by-product of the "excellence of his vows" (pranidhana-visesa).16 This same expression occurs, as is well known, in the Sukhavativyuha, both before and after the list of Dharmakara's vows.77 But there it is used in a quite different sense. Now it is understood not as referring to the "excellence" or "distinguished quality" of the bodhisattva's vows in general, but to the specific (and individual) promises contained in those vows. In the Sukhavativyuha, in other words, the expressionpranidhdnavisesdh (here used in the plural) is understood as referring to the distinctive

- i.e., unique and individual - vows made by Dharmakara that determined what his Buddha-field would be like. It is easy to see how someone familiar with the kind of language used in the Aksobhyavyuha could have interpreted the expression in this sense. It occurs only in the section of the sutra in which Aksobhya's future Buddha-field is being described, and the context in which it is used could well have led a listener to assume that it referred to some unstated list of vows concerning the features of that field. What we may have in the Sukhavativyuha, in other words, is a reinterpretation of the expression pranidhana-visesa as used in the Aksobhyavyuha in a way that brought the notion of making vows concerning the nature of one's future Buddha-field onto center stage, while the more generic vows concerning the requirements of bodhisattva practice receded into the background. * * *

faults, he will renounce his own attainment of Buddhahood (11.103al5ff., Tib 29.3-7). No such statement appears in Lokaksema's version. 76. The wording in Lokaksema's version is ^^MfM^MWil^WMM^MMM ft (11.756b20-21, reading the final character with the Taisho editors' note 10, a reading that is confirmed by the Tibetan, where khyad-par is also surely a translation of Skt. visesa "excellence, distinctive [[[quality]]]"). 77. See §7 (MAX MULLER ed., p. 11, line 7) and §9 (p. 21, line 19 - p. 22, line 1). The Chinese reads WM±WBZM By reading the Aksobhyavyuha and the Sukhavativyuha in conjunction we have been able to identify a number of elements shared by these two texts, as well as a number of important differences. To return now to the question posed at the outset: in what ways are the ideas about Pure Lands contained in these sutras continuous with earlier Mainstream and Mahayana notions, and in what ways do they represent genuine innovations? Both sutras - to begin with elements they share in common - present a vision of a world in which a Buddhist practitioner may aspire to be born which is in many respects like the early Buddhist vision of heaven. Indeed both texts, in searching for analogies to the

glories of Abhirati or Sukhavati, refer explicitly to heavens known already in earlier Buddhist cosmology.78 More specifically, the fact that Abhirati provides an optimal setting for the rapid attainment of Arhatship makes it analogous to the "Pure Abodes" (suddhavasa) of the Theravada (and indeed the general Mainstream) tradition, the upper heavens of the Form Realm in which the non-returner (anagamiri) is reborn, attaining Arhatship there and never returning to our world. Here the Sukhavativyuha diverges, however, for it contains - at least in the Sanskrit and "Sanghavarman" recensions - no discussion of the ease of attaining Arhatship there.79 On the contrary, Sukhavati appears

to provide a way-station only for bodhisattvas, who (apparently with little preparation here on earth) will become irreversible from the enlightenment of a Buddha. If we look at what is necessary to attain rebirth in these worlds, once again the Aksobhyavyuha appears to be closer to earlier tradition. What is necessary to attain rebirth in Abhirati (according to all sections of the text except the closing "book-cult" passage) is very much in line with the entrance requirements for traditional heavens: it is the result of one's own good actions, and a considerable quantity of such merit is required. In the Sukhavativyuha, by contrast, access to Amitabha's world is considerably easier: all that is necessary is remembering Amitabha ten times, and avoiding particularly heinous actions (viz., the five anantarya sins and slandering the Dharma). What is especially noteworthy in the Sukhavativyuha is what replaces individual merit as the means of

78. The Aksobhyavyuha refers several times to the Trayastrimsa heaven, while the Sukhavativyuha takes its analogy from the highest heaven in the Desire Realm, the Paranirmitavasavartin heaven. 79. As mentioned above, however (note 66), Arhats are considerably more prominent in the two earliest Chinese versions. entrance, for rebirth in Sukhavati is portrayed as the result of a relationship between the believer and Amitabha. We have here the seeds - if not yet the full fruition - of what would come to be known as reliance on "other-power" in the Pure Land Buddhism of East Asia. Thus in several respects the Aksobhyavyuha appears to lie slightly closer than

does the Sukhavativyuha to the pre-existing worldview of Mainstream Buddhism. But how do these two sutras align themselves with what we have referred to above as the "trunk-line" Mahayana tradition? That is, in what ways does the Pure Land Buddhism of these two texts appear to be continuous with early Mahayana beliefs and practices, and in what ways do these sutras diverge from those conventions in unique and innovative ways? Once again it is the Aksobhyavyuha that appears to be the more conservative of the two. Here we still find the standard early Mahayana assumption that Buddhahood is only for certain members of the community, while other Buddhists should be content to attain Arhatship.

Indeed it is the role of a Buddha to help them do so quickly, a part that Aksobhya plays with great effectiveness. Moreover, the path to Buddhahood is still conceived of as tremendously difficult, requiring eons of self-sacrifice to attain. Though Aksobhya offers an "accelerated course" (as it were) to those fortunate enough to be born in his realm, it is also reiterated throughout the sutra that those who wish to become Buddhas must imitate the rigorous and extensive bodhisattva training undergone by Aksobhya

himself. Thus while the path to Buddhahood may be shortened by hearing the Dharma and making offerings both in the presence of Aksobhya and in other Buddha-worlds, the basic requirements of the path have not been changed. Nor has the final destiny of the bodhisattva been revised in any way: the ultimate goal is to become a Buddha oneself in a world that has no Buddhism, and there to preside over one's own immeasurable assembly of sravakas. When we turn to the Sukhavativyuha, by contrast, we see that many of these ideas have changed. Buddhahood now appears to be viewed as appropriate for everyone - or at any rate, there is no discussion of the attainment of Arhatship by the denizens of SukhavatI,80 and the term anuttarasamyaksambodhi is now generalized to the extent that it seems

80. Once again these comments are based solely on the text of the sutra contained in the extant Sanskrit version and in T No. 360. to refer to the spiritual destination of all beings.81 Even in the case of Amitabha himself the path appears far less difficult: in the sole passage in which Dharmakara's bodhisattva-conduct is actually described, only such practices as self-restraint, kindness to others, and the practice of the paramitas are named.82 There is no reference to bloody acts of selfsacrifice like those described in the jataka stories, nor to the solitary and rigorous self-cultivation in the forest that characterized the pursuit of enlightenment in Sakyamuni's final life. Above all, though, there is a shift in focus in the Sukhavativyuha toward viewing the creation of a Pure Land as the primary aim of the bodhisattva path. Virtually all of Dharmakara's bodhisattva vows are centered on this project, while in the Aksobhyavyuha the qualities of Abhirati are portrayed not as the

objective but as the by-product of Aksobhya's vows. Indeed we have even seen evidence that the authors of the Sukhavativyuha may have formulated their own understanding of the content of Dharmakara's resolutions with the wording of the Aksobhyavyuha in mind. Whatever the direct relationship between the Aksobhyavyuha and the Sukhavativyuha, however (and we should recall that we do not have the "original" version of either text), one thing is clear. The Aksobhyavyuha - though it is explicitly a "Pure Land" sutra - remains quite close in its content and expectations to the world-view of earlier Mainstream and Mahayana Buddhism. The Sukhavativyuha, by contrast, has introduced a number of important innovations, including the generalization of the bodhisattva path to all practitioners, the ease of the attainment of Buddhahood, and the importance of relying on Amitabha with

faith. All of these suggest that the Sukhavativyuha - at least in the form of the text most widely used in East Asia today - is the product of a community that was operating on the principle of "one vehicle" for all, in which the requirements for Buddhahood had accordingly been radically revised. It seems clear, therefore, that the ideas contained in the Sukhavativyuha represent a further development of those found in the Aksobhyavyuha, and not vice versa. The type of Pure Land Buddhism presented in the Aksobhyavyuha thus constitutes a transitional stage

81. This issue is explored in detail in an unpublished paper by Andrew JUNKER, "A Look at Oddities and Influence in the Shorter and Longer Sukhavativyuha Sutras" (Indiana University 1997). 82. MAX MULLERed., §10.

between the earliest understanding of the bodhisattva path and the distinctive form of Pure Land Buddhism articulated in the larger Sukhavativyuha. As such it provides an important missing piece in the historical puzzle, revealing that "very earliest stage" of Indian Pure Land Buddhism to which Charles Eliot so long ago referred. Perhaps with the addition of this new material we will more accurately be able to trace the steps by which Pure Land Buddhism gradually became an important element in Indian Buddhist thought and practice, and ultimately through its East Asian manifestations - a major player on the world religious scene today.