Reviews Of Diffeent Books

Author(s): J W De Jong
Publisher: J W De Jong
Catalog link: https://jainqq.org/explore/269556/1

Page #1
REVIEWS The Ramayana of Valmiki. Volume I: Balakanda. Introduction and Translation by Robert P. Goldman. Annotation by Robert P. Goldman and Sally J. Sutherland. Princeton University Press, 1984. xx, 429 pp. PS52.50. Volume II: Ayodhyakanda. Introduction and Translation by Sheldon I. Pollock. Princeton University Press, 1987. xviii, 560 pp. PS40.00. The entire Ramayana has been translated into English, French and Italian. Goldman states that there are at least three translations into English. With the recent publication of N. Raghunathan's translation (Madras, 19811982) there are now five translations of the entire poem into English, the others being by R. T. H. Griffith (1870-1875), M. N. Dutt (1891-1894), M. L. Sen (1927) and H. P. Shastri (1952-1959). The Ramayana has been translated into French by Hippolyte Fauche (1854-1858) and by Alfred Roussel (1903), and into Italian by Gaspare Gorresio (1847-1858 and 1870). Goldman makes the following remarks with regard to the existing translations: "Whatever their literary merit, and this varies radically, all of the translations so far made have been based upon a single version or recension of the text, either the vulgate or the Bengal version, and their authors have not been in a position to judge accurately the vital text-critical problems that the epic presents. None of them, moreover, has attempted to put before its audience the results of a close and critical reading of the extensive commentarial literature that has grown up around the poem" (Vol. I, p. 96). This new translation undertaken by a group of American scholars is based upon the critical edition published in Baroda (1960-1975). However, the text of the critical edition has not been followed slavishly as can be seen from the many notes in which decisions taken by the editors of this edition are queried and from the list of emendations and corrections at the end of volume two. Sheldon I. Pollock has contributed an essay on the Ramayana text and the critical edition to Volume I (pp. 82-93) in which he discusses in some detail the relation between the northern and the southern recensions. According to Pollock the northern recension presents some sort of revision but not because, as has been maintained in the past, grammatical irregularities of the southern recension have been eliminated in Indo-Iranian Journal 32: 195-254, 1989. 1989 Kluwer Academic Publishers. Printed in the Netherlands.

Page #2
196 REVIEWS the northern recension. The differences between the two recensions are that the wording of the northern recension frequently differs from that of the southern without appreciably altering the text's grammatical regularity or poetic acceptability. In the second place, the northern recension often tends toward a popularization or glossing of the southern text. Pollock remarks that the second point appears to have gone unnoticed. This is a matter of great importance for the interpretation of difficult passages as is obvious from remarks made by Pollock in the notes to the translation of volume two. Although Pollock and Goldman do not always agree with the decisions taken by the editors of the critical edition, they do not hesitate to acknowledge its merits. Pollock writes: "The critical edition, then, we believe, puts us in possession of the most uniform, intelligible, and archaic recension of the Valmiki Ramanaya, corrected and purified on the basis of the other recensions and versions that are descended from the common oral original." (p. 92). In the introduction to the first volume (pp. 1-59) the general editor, Robert Goldman, examines the various theories as to the date and the historicity of the Ramayana, problems which have been discussed by several generations of scholars. As to the date of the Ramayana Goldman arrives at the conclusion that the Ramayana reflects the political situation in northern India at a time prior to the rise of Buddhism. According to Goldman the oldest parts must have been composed after the beginning of the seventh century B.C. and before the middle of the sixth century B.C. However, the fact that the oldest parts of the Ramayana agree with the historical situation in northern India in the period between 750 and 500 B.C. does not necessarily imply that they have been composed in the same period. In his book. on the Indian epic P. A. Grintser points out that usually there is a gap of several centuries between events described in epics and the composition of the epics themselves.' It is certainly quite well possible that the oldest parts of the Ramayana were composed at a later period. Goldman discusses in some detail the relation between the Ramayana and the Ramopakhyana in the Mahabharata. Goldman shares the opinion of Jacobi, Sukthankar and Raghavan that the Ramopakhyana is based upon Valmiki's poem and he takes pains to refute Vaidya's and van Buitenen's efforts to disprove the priority of the Ramayana. In order to explain the enormous influence of the Ramayana in India and Southeast Asia, Goldman pays particular attention to the characterization of Rama. He stresses Rama's exaggerated self-denial and general lack of emotion in the face of personal tragedy. However, his self-control is not absolute; Goldman cites a passage in book two (47.8-10) in which Rama

Page #3
REVIEWS 197 shows his bitterness towards his father. Goldman discerns a delicate balance in the characterization of Rama. Goldman writes: "On the one hand, he is the most important, if not the most extreme example of traditional India's ideal man, the son who subordinates the goals of his own life to those of his father. On the other hand, he has a sufficient degree of ambivalence and a sufficiently rounded character to enable him to serve as a model for countless generations of Indians, while remaining the compelling hero of a fundamentally tragic epic" (p. 58). It is easy to agree with Goldman on the first point but one hesitates to subscribe to this view that Rama has a sufficiently rounded character. Passages such as the one quoted by Goldman (2.47.810) are rare and far from being typical of the righteous (dharmatman) Rama. One wonders whether it is not rather due to his being the example of filial piety and of married love that he has been able to exercise such a lasting influence on the imagination of the Indians. Goldman remarks that there is no reason to regard Valmiki as having any greater claim to historicity than Vyasa. However, he is not unwilling to attribute the central portion of the Ramayana to a poet named Valmiki. Goldman writes: "In the end, the most that can be said is that there appears to be no real evidence to contradict the proposition that the central portion of the Ramayana has a single author. On the basis of the unanimous tradition, there is no reason for us to doubt that this author's name was Valmiki" (p. 31). Goldman chooses his words very carefully but leaves us uncertain whether or not it is possible to demonstrate with some degree of probability that "the central portion" (whatever that may be) is the work of a single author. Goldman's introduction to the Balakanda (pp. 60-81) examines the composition of this book in greater detail than has been done by previous scholars. He is willing to admit that substantial portions are known to the central books and formed part of the original stratum. As to the remaining portions Goldman tries to explain why they have been added in the course of time. His analysis of the composition of the Balakanda is carried through very carefully and will be very helpful for a better understanding of this book to which not enough attention has been paid in the past. In his essay on the translation and annotation (pp. 96-117) Goldman. compares a passage of the Iliad (5.259-268) with a passage of the Ramayana (6.55.120-125) and shows the different treatment of a battle-scene in both works. He writes: "Where the Greek poet strives to speak to the sensory experience of his audience, the Sanskrit bard aims to stir the emotions of awe and amazement with a dense texture of descriptions and comparisons that move us away from the world of mundane experience to that of the

Page #4
198 REVIEWS supernatural" (p. 103). Goldman points out that most of the epithets used by Valmiki allude to moral, ethical, intellectual, and emotional rather than physical qualities, qualities which are generally vague, often obscure, and always sensorially opaque. This makes the task of the translator much more difficult. Goldman remarks that it is often more difficult to be sure of the meaning of a Ramayana passage than of passages in far more elaborate and difficult genres of Sanskrit. For this reason Goldman and his colleagues decided that it would be necessary to add annotations that would deal with every unclear and difficult passage in the poem. Great use has been made of the commentaries. Particularly welcome is the plan to study the works and relationships of the major commentators in a planned future companion volume. Undoubtedly the careful annotation is one of the great achievements of this new translation of the Ramayana. How detailed the notes are is obvious from the fact that in book one the translation occupies 151 pages, and the notes 126 pages in small print. For book two the figures are respectively 248 and 203. Pollock's introduction to his translation of the Ayodhyakanda deals with several important problems related not only to this book but also to the meaning of the entire epic and its relation to the Mahabharata. Pollock points out that at the root of many Indian epic narratives is a political problem similar to the one found in the Ayodhyakanda. Pollock writes: "An integral theme of Sanskrit epic literature is kingship itself and its attendant problems: the acquisition, maintenance, and execution of royal power, the legitimacy of succession, the predicament of transferring hereditary power within a royal dynasty" (p. 10). He remarks that, whereas in the Mahabharata armed combat was considered to be the only means for the resolution of political and dynastic conflict, we find in the Ramayana the first literary attempt in India to "moralize" the exercise of political power. Pollock remarks that the perspectives and attitudes we find in the Ayodhyakanda were also those of Asoka. In section four of the introduction (A Problem of Narrative and Its Significance) Pollock examines a number of important discrepancies and inconsistencies. According to him the original tale turned on the bride-price (rajyasulka), on the basis of which Kaikeyi asked that her son be crowned, and the one boon granted by the king to Kaikeyi. In revising this story Valmiki introduced the episode of the two boons granted by Dasaratha in gratitude to Kaikeyi for her assistance to him during the battle of the gods and the demons. Pollock distinguishes several reasons for this revision. First, the revision preserves the honesty and integrity of Dasaratha. Second, a considerable intensification in the dramatic action has been achieved by

Page #5
REVIEWS 199 the alteration. Furthermore, by amalgamating into a single act of Kaikeyi's perversity Rama's dethronement and exile, the principal didactic thrust of the book is reinforced, for obedience to the word of his parents becomes Rama's guiding principle. Pollock stresses the importance of fate in the Ayodhyakanda and points out how fatalism has taken on a more specific shape in the recurrent appeal to karma. Destiny plays an important role in the life of Dasaratha who, in his youth, killed, in error, a young ascetic. It is due to this deed that he is destined to end his days grieving for a son. According to Pollock in this episode Valmiki appears to have reworked and inserted a narrative derived from the folk tradition as preserved in the Samajataka. This story weakened considerably the force of the original tragedy of an old king overmastered by his passion for a beautiful young woman and contributed to the tendency of idealizing Dasaratha as venerable king. In several instances (the plot of the Ayodhyakanda, the two boons, the characterisation of Dasaratha and Rama, etc.), Pollock finds traces of a revision of an original, more primitive tale by the composer of the monumental Ramayana to whom he refers as "Valmiki" by way of convenience. Valmiki is inspired by a different and larger concept of dharma than the kshatriya's code [ksatram dharmam], "where unrighteousness and righteousness go hand in hand (adharmam dharmasamhitam), a code that only debased, vicious, covetous and evil men observe" (101.20). The kshatriya's code is dominant in the Mahabharata and this poses the problem of the relation of the two epics. Pollock points out both the congruence and the disagreements between the Ayodhyakanda and the Sabhaparvan and remarks that "the structural congruence of the two books thus seems to point to a . more fundamental relationship between the epics than generic affiliation" (p. 41). According to Pollock the question of priority with respect to preliterate epic or popular texts is in general misleading. He suggests that one has to think of the two epic traditions as coextensive processes that were underway throughout the second half of the first millennium B.C., until the monumental poet of the Ramayana and the redactors of the Mahabharata authoritatively synthesized their respective materials and thereby in effect terminated the creative oral process (cf. pp. 42-43). Notwithstanding the discrepancies and inconsistencies in the narrative of the Ayodhyakanda Pollock considers this book to be the most skillfully structured of the seven books (cf. pp. 45-47). Although he does not state this expressly he appears to be of the opinion that this book must be the work of a major poet, a poet who was at the same time deeply inspired by a concept of dharma similar to that found in the inscriptions of Asoka (cf.

Page #6
200 REVIEWS p. 70, n. 12). Pollock remarks that Valmiki's chronological relationship with the Mauryan king is problematical. However, the fact that as he remarks "so similar a discourse" is shared by both seems to suggest that he must have lived in a period not too far removed from that of Asoka. In his preface Goldman alludes to some reservations which were voiced with regard to a "committee translation". The publication of these first two volumes shows that it is quite possible to maintain the same tone even when different translators are at work. It is unnecessary to quote some of the older translations to demonstrate the stylistic difference between this translation and previous English translations. The translation flows smoothly and is at the same time accurate and readable. Goldman writes that Gorresio's Italian translation and Roussel's French translation are elegant and readable. I have not been able to consult Gorresio's translation but Roussel's translation is certainly far from elegant and one has to agree with Renou's opinion in this regard. The translations of books one and two by Goldman and Pollock are undoubtedly a pleasure to read and will certainly be enjoyed by the general reader who has no special knowledge of ancient India. It is only in very rare cases that one has the feeling that the tone is not entirely just, and that a different rendering would be preferable. Is it, for instance, quite appropriate to say that Rama is "a decent man" (2.2.20)? Both translators have tried to render descriptive epithets by various English equivalents. For instance, in book two manasvin is rendered as 'in better spirits', 'sensible', 'in high spirits' and 'spirited'. 'In high spirits' (27.33; 49.8) is less felicitous than the other equivalents. The quality of the translation is admirable. When reading it together with the text one sometimes wonders why a certain translation has been adopted. Almost always one finds a note which shows how carefully different possibilities have been considered in the light of the various explanations given by the commentators. It is perhaps largely due to the different nature of the books that in the notes to book one the accent is put more on elucidating the meaning of the text whereas in the notes to book two philological problems are discussed with great perspicacity. It is of course not possible always to agree with the translation. One also does not always find a note where one would expect one. In reading these two volumes I made some marginal remarks which I submit mainly to show with how much pleasure I have read the admirable work done by Goldman and Pollock. 1.2.28: "Grieving once more (muhuh) for the kraunca hen". Rather 'grieving again and again'. PW quotes this place for muhuh in the meaning. of muhurmuhuh jeden Augenblick wiederholt'. 1.2.30: "Brahman, it was by my will alone..." One is struck by the fact

Page #7
REVIEWS 201 that Brahma as speaker uses the vocative Brahman. Goldman has no note on it. 1.5.10: "its markets had well-ordered interiors" (suvibhaktantarapanam). For antarapana 'inner shops' see Pollock's note on 2.51.13. 1.5.20: "the sole support of his family, in hiding, or in flight" (aparaparam / sabdavedhyam ca vitatam). Goldman does not explain his rendering of aparaparam. The commentaries have 'without father, etc., and son, etc.'. Goldman's translation of vitata follows the commentaries (palayita) but, here as in so many other cases one wonders whether this is an explanation ad hoc or not. 1.6.2: "He was renowned in the three worlds as a masterful man and a royal seer like one of the great seers" (maharsikalpo rajarsis trisu lokesu visrutah). It seems preferable to translate these two padas separately for trisu lokesu visrutah which occurs several times in the Ramayana is typical for the oral style: "He was a royal seer like one of the great seers and renowned in the three worlds". 1.7.17: "surrounded by ... counsellors, skilled both in counsel and strategy" (mantribhir mantrahitair nivistair vrto). I understand neither the Sanskrit text nor the translation. Several manuscripts have mantrahite. 1.13.15: "Day after day, skilled brahmans, following instructions, performed, in accordance with the ritual texts, all the rites that make up the sacrifice" (divase divase tatra samstare kusula dvijah sarvakarmani cakrus te yathasastram pracoditah). Whatever may be the exact meaning of samstare, it depends on kusala. The expression yajnasamstare occurs several times in the Mahabharata, for instance 1.54.9: rtvigbhir devakalpais ca kucalair yajnasamstare. Probably samstare stands for yajnasamstare. 1.16.18: The translation omits sahasrasah. 1.25.13: The translation omits krodhamurchita. 1.32.2: "Your Majesty, Vayu who lives in all creatures wishes to ruin us by approaching us in an improper manner. He has no regard for what is proper" (vayuh sarvatmako rajan pradharsayitum icchati / asubham margam asthaya na dharmam pratyaveksate). It is better to separate the two halves of the verse: "Your Majesty, Vayu who lives in all creatures wishes to ruin us. Approaching us in an improper manner he has no regard for what is proper". 1.43.11: "Dilipa ... could not, though he too strove to do so, think of a way to bring down the Ganges" (dilipena ... punar na sarkita netum gargam prarthayatanagha). Whether one reads sarkita or sakita which seems much more probable, in any case one must correct gangam to ganga. See also PW's.v. sak-, p. 9a. 1.58.1: "and spoke sweet words directly to him, although his appearance

Page #8
202 REVIEWS was that of a pariah" (abravin madhuram vakyam saksaccandalarupinam). Saksat has to be connected with candalarupinam "although in person his appearance was that of a pariah", cf. PW saksaccandalatam gatah'zu einem leibhaftigen K. geworden'. 1.68.18: "Mighty Janaka, gifted with insight, also passed the night after completing both his sacrifice and the rites for his two daughters, in accordance, with traditional law" (janako 'pi mahatejah kriya dharmena tattvavit / yajnasya ca sutabhyam ca krtva ratrim uvasa ha). According to the Tilaka yajnasya depends on kriya: yajnasyavasistah kriyah. The two ca's are troublesome. 2.8.25: "your son is about to be set upon by Rama" (pracchadyamanam ramena bharatam). It is possible to translate pracchadyamanam more precisely, cf. PW "Jmd verdunkeln, im Wege stehen (West.: insidiari; Schlegel: proculcare)". 2.10.32: "Gradually the king regained his senses" (cirena tu nrpah samjnam pratilabhya). Here cirena certainly means 'after a long time, at last', cf. PW 'nach langer Zeit, spat, nicht gleich' with a reference to Ramayana 5.30.13 (Cf. ed. 5.30.3). The meaning 'gradually' ('langsam') for cirena is not found in PW but only in pw. 2.13.28: "a mountain peak or a motionless cloud" (adrikutacalamegha). Rather 'a motionless cloud on a mountain peak', cf. Tilaka: adrikute. 'drisrniga arudho 'calo yo mahameghas. 2.41.28: "a broad pathway, ... free from ... the dangers of dangerous beasts" (mahamargam abhayam bhayadarsinam). Pollock's translation follows Govindaraja's commentary. Mahesvaratirtha has bhayadarsinam . bhayasarkinam which seems preferable. Is bhayadarsin in the meaning 'dangerous beast' found elsewhere? 2.46.34: "Just now, while you still stood before them, the people felt sick at heart just to contemplate you, in their mind's eye, living far away" (dure 'pi nivasantam tvam manasenagratah sthitam / cintayantyo 'dya nunam tvam niraharah krtah prajah). In a note Pollock remarks: "The translation here accords to some extent with that of Cm. The alternatives (of Cg, Ct, Cr and the Mylapore editors) in one way or another turn the sense around (though you are far away, they think of you as nearby'), but with impossible, or jejune, consequences." The consequences may be jejune but this seems to be the only explanation possible. 2.51.15: "crowded with men of importance" (mahajanasamakulah). Pollock does not explain why he rejects the meaning crowd for mahajana. 2.58.10: "With effort I managed to collect my thoughts and recover the power of speech" (manasah karma cestabhir abhisamstabhya vagbalam).

Page #9
REVIEWS 203 Pollock follows.Cm's rather far-fetched explanation. It is certainly preferable to read karmacestabhir, cf. PW 'durch Kraftanstrengung des Geistes'. See also PW for abhisamstabhya 'kraftigen, aufrichten'. 2.66.32: "for he anticipated yet another calamity" (dvitiyapriyasamsanat). Rather 'because she told another unpleasant news.' 2.72.1: "Now, as the grief-stricken Bharata was making the journey back" (atha yatram samihantam ... bharatam sokasamtaptam). Pollock notes that his translation is in disagreement with the commentators, who understand "preparing for contemplating') the journey. It is difficult to understand Pollock's translation for the conversation between Bharata and Satrughna takes place in Ayodhya after Bharata's return from Rajagtha and before his departure to the forest to see Rama. 2.74.18ab: The translation omits these two padas. 2.93.36: "he reached in vain for Rama's feet and collapsed in tears" (padav aprapya ramasya papata bharato rudan). Rather 'before reaching Rama's feet'. . 2.94.48: "And when a thief, either caught in the act or discovered with the stolen property" (grhitas caiva prstas ca kale drstah sakaranah). In Mahabharata 2.5.94 van Buitenen translated sakarana as 'with his tools'. PW has 'Anzeichen, Beleg, Beweisgrund' which is less probable. In his article on some lexical problems in the Valmiki Ramayana Pollock remarked that we have no complete grammar of the epic dialect, no adequate dictionary (let alone specialized lexica, as of particles), and worst of all no concordances. In the age of the computer it ought not to be too difficult to make a concordance. There is no dictionary of the Ramayana but one can find much lexical material on the Ramayana in the great Petersburg dictionary. It would be useful to list all the references to the Ramayana as, for instance, Scharpe has done for the references to the Kadambari in the small Petersburg dictionary. Both Goldman and Pollock could have given more references to the great Petersburg dictionary. In the annotation of these two volumes and especially in that of the second one there are many important remarks on lexical matters and it is to be hoped that after the completion of the translation of the entire Ramayana all these items will be indexed in a supplementary volume. Misprints are very rare. I have noted the following ones in volume 2: p. 450, note 22: prsthah; p. 478, note 22: rupap; p. 482, note 69: bristle toothbrushes; p. 500, note 19: duhkam; p. 546: Raghaven. In volume one (p. 89, n. 21) there is a reference to Kenney 1974 but this is not listed in the bibliography.

Page #10
204 REVIEWS NOTES I P. A. Grincer, Drevneindijskij epos (Moskva, 1974), pp. 166. 2 'exacte, peu litteraire' (Litterature sanskrite, Paris, 1946, p. 102). 3 'Some lexical problems in the Valmiki Ramayana', Babu Rama Saksena Volume (Lucknow, 1983), p. 275. 4 A. Scharpe, Bana's Kadambari (Leuven, 1937), pp. 398-446. Australian National University J. W. DE JONG Constantina Rhodes Bailly, Shaiva Devotional Songs of Kashmir. A Translation and Study of Utpaladeva's Shivastotravali. State University of New York Press, 1987. X, 196 pp. Paper $12.95, Cloth $39.50. Mrs. Bailly's book is based upon a doctoral dissertation which comprises a translation of a selection of the songs and an "in-depth inquiry into the status of the manuscripts and the preservation of the textual tradition of the Sivastotravali". According to the author an examination of seven manuscripts, two in devanagari and the other five in sarada, showed that there were no major variants in any of these manuscripts. The differences between the manuscripts and the text printed in 1964 in the Chowkhamba Sanskrit Series were "for the most part simply errors in samdhi or the use of synonymous terms that fit into the meter exactly." The text reproduced in this book is based upon the above-mentioned edition and does not indicate any variant readings, although in appendix B the author refers to footnotes to the text (p. 175). The only information about variant readings is found on p. 174 where the author mentions the insertion of synonyms, as svarupa for sarira (4.25), jagatam for lokanam (10.1), and visrto or galito for prassto (4.14) and the interchanging of vibho, prabho and sometimes natha in the vocative. These variant readings are certainly not metrically equivalent. The introduction deals with the spiritual contents of the verses but says very little about the literary form. The author remarks that Utpala employed a variety of complex Sanskrit metres, both syllabic (sloka, sikharini, rathoddhata, and sardulavikridita) and moraic (arya and vaitaliya) but fails to present a list of the metres in the songs. The translation is on the whole faithful and agreeable to read. It gives a good idea of Utpaladeva's delight in his union with Siva. However, technical terms are sometimes not correctly rendered or explained in notes. For example, I.13a amulad vaglata seyam kramavispharasalini is rendered as Indo-Iranian Journal 32: 1989.

Page #11
REVIEWS 205 follows: "May this vine of speech, Rising steadily from the root, Everywhere adorned with blossoms", although Ksemaraja explains that krama refers to pasyanti, etc. Sometimes, the translation omits words found in the original. For instance, VI.2: viyogasare samsare priyena prabhuna tvaya / aviyuktah sadaiva syam jagatapi viyojitah // Mrs. Bailly translates: "Even if I am separated - From the world of samsara - May I not be separated from you, My beloved." The romanised text transliterates the devanagari text without separating words which do not form a compound.' More troublesome are the numerous misprints which cannot all be listed here. For example, the first chapter contains the following misprints: 1.3 labdhatatsampadam read labdhatvatsampadam; 1.6 tadria read tadrsa; I.18 praniyate read praniyate; 1.21 santakallolasitaccha- read sitaccha-. Three syllables are omitted in IV.15: bhavatpurargala- read bhavatpuragopurargala-. In other instances, the text seems to have been romanised by somebody who had difficulties in reading the devanagari script, for example: III.10 syadrsa read syad dasa; V.10 kunyaddhatadapi read kudyad ghatad api; X.1.5 'tyantatarasutkanthito read 'tyantataram utkanthito; XVI.7 udyam read rudyam; XIX.17 janaki'tha read janako'tha. In all these cases the translation is based on the text as printed in the Chowkhamba Sanskrit Series. NOTE ! On the romanisation of Sanskrit texts see F. B. J. Kuiper's remarks, Gopalakelicandrika Amsterdam, 1987), p. 6. Australian National University J. W. DE JONG Alfred Hillebrandt, Kleine Schriften. Herausgegeben von Rahul Peter Das (Glasenapp-Stiftung, Band 28). Stuttgart, Franz Steiner Verlag GmbH, 1987. XXXI, 685 pp. DM 98,-. The publication of Hillebrandt's Kleine Schriften is extremely welcome. Although he died sixty years ago, his work is still being used by scholars and in recent years several of his books have been reprinted and translated. The bibliography which Rahul Peter Das has compiled with great care Indo-Iranian Journal 32: 1989.

Page #12
206 REVIEWS shows that Hillebrandt has been active in many fields: Vedic studies, linguistics, Arthasastra, drama, Kalidasa, Indian materialism and Buddhism. Das has been able to incorporate in this volume a large selection of his articles, omitting mainly those which were written for the general reader or which were already published in Alt-Indien (Breslau, 1899). Of his many reviews twenty have been reprinted, the most important certainly being his long review of Oldenberg's Metrische und textgeschichtliche Prolegomena (GGA 1889, pp. 387-424). Hillebrandt's detailed review of Telang's edition of the Mudraraksasa (ZDMG 39, 1885, pp. 107-132) was followed in 1912 by his edition of the Mudraraksasa and by several articles which are all reprinted in this volume. Also reprinted are his articles on Kautilya whose Arthasastra was discovered by Hillebrandt in 1908. Hillebrandt's parody of Pischel's Vedic studies: Die Gotter des Rgveda, published under the pseudonym of Fritz Bonsens in 1894 is also included in it. We must be very grateful to the editor of this volume for having added a list of corrigenda and indexes which are much more detailed than most indexes in this series: A. Sach- und Namensverzeichnis (pp. 631-645); B. Verzeichnis der Worter, Stamme, Wurzeln, Formen (pp. 646-652); C. Verzeichnis der Textstellen (pp. 653-685). This volume is a fitting tribute to the memory of a great scholar. Australian National University J. W. DE JONG Jaan Puhvel, Comparative mythology. Baltimore and London, The Johns Hopkins University Press, 1987. X, 302 pp. $32.95. Indo-Iranian Journal 32: 1989. The title of Puhvel's book is slightly misleading because this is not a study of the methods and aims of comparative mythology, but a study of IndoEuropean comparative mythology, a topic which for many decades was anathematized by scholars. In an introductory chapter Puhvel sketches the history of the study of myth and characterizes briefly four different approaches in the twentieth century, the ritualistic (Jane Harrison and others), the psychoanalytic, the sociological and the structuralist. He is critical of Levi-Strauss: "Overlaying known data with binaristic gimmickry in the name of greater "understanding" is no substitute for a deeper probing of the records themselves as documents of a specific synchronic culture on the one hand and as outcomes of diachronic evolutionary processes on the other" (p. 19). Puhvel's own point of view is stated in the following words:

Page #13
REVIEWS 207 "Thus the twentieth-century lessons of ritualism, psychoanalysis, sociology, and structural anthropology alike deserve to be heeded by the historical and comparative student of myth and religion, but only to the extent that they offer viable insights into a study that is by definition historical, and more specifically philological, rooted in the minute and sensitive probing and comparison of primary written records" (p. 20). In this connection Puhvel cites the names of Emile Benveniste, Georges Dumezil, Stig Wikander, Jan de Vries and Jacques Duchesne-Guillemin as those who over the past halfcentury have been active in this direction. However, it is above all Georges Dumezil who is Puhvel's guide and mentor. The main part of Puhvel's book is a systematic study of the mythology of the Indo-European peoples: Vedic India, Epic India, Ancient Iran, Epic Iran, Ancient Greece, Ancient Rome, Celtic Myth, Germanic Myth, Baltic and Slavic Myth. In an introductory chapter Puhvel examines the creation myth in the Ancient Near East and argues against Stig Wikander who maintained that the Kingship in Heaven myth is an Indo-European theme independently preserved in Greece, Anatolia, and Iran. According to him "this myth is a textbook case of ancient Near Eastern diffusionism and entails a methodological lesson about proper procedure in comparative mythology." In another introductory chapter Puhvel studies the concepts "Indo-European" and "Indo-Iranian". He points out the importance of agreements between Indic and Iranian on the one hand and Italic and Celtic on the other for the reconstruction of Proto-Indo-European lore and IndoEuropean myth. To India and Rome he adds as a third anchor the Germanic peoples, especially Scandinavia and its colonial cranny of Iceland. Puhvel briefly draws attention to the transposition of myth to heroic saga in India, Iran, Scandinavia and Ireland The third part of Puhvel's book takes up a number of special themes: god and warrior, king and virgin, horse and ruler, fire and water, twin and brother. As can be expected, in these chapters comparative mythology plays a greater role than in the preceding chapters on the Indo-European traditions. According to the preface Puhvel's book is not so much a research treatise as a compendium born of pedagogy. However, the author of the Hittite Etymological Dictionary does not spare the reader many etymologies which will probably be beyond the reach of the wider educated public to which this book is addressed. The specialists will regret that they have not been registered in the detailed index which mainly comprises proper names. Puhvel has abstained from notes but has added recommendations for further reading at the end of chapters. Georges Dumezil and his followers have developed an impressive structure

Page #14
208 REVIEWS of comparative mythology which will continue to provoke both approval and opposition. Puhvel's judicious and careful exposition which contains many personal views is to be welcomed as a very lucid introduction to the study of this fascinating topic. Australian National University J. W. DE JONG Gilgul. Essays on transformation, revolution and permanence in the history of religions. Dedicated to R. J. Zwi Werblowsky (Supplements to Numen L). Edited by S. Shaked, D. Shulman, G. G. Strousma. Leiden, E. J. Brill, 1987. VIII, 326 pp. 108 guilders. ISBN 90 04 08509 2. This collection of twenty-two essays is dedicated to R. J. Zwi Werblowsky, a well-deserved tribute to a scholar who has been active in many fields of religious studies. It is therefore not surprising to see that this volume comprises several articles relating to Indian and Iranian religions. Jan C. Heesterman writes on the essential ambivalence of sacrifice which results from the interplay of self-sacrifice and sacrifice by an outside agent in "Self-sacrifice in Vedic ritual" (pp. 91-106). Wendy Doniger O'Flaherty examines the metaphor of the shepherd in Hinduism and Christianity and shows that Christianity emphasizes the positive aspect of the shepherd while Hinduism emphasizes the negative: "The good and evil shepherd" (pp. 169--191). David Shulman compares the passage where Sugriva reveals to Rama the jewels that Sita had desperately cast to earth while she was being carried off to Lanka by the demon Ravana in Valmiki's Ramayana and in Kampan's Tamil version of the Ramayana and points out the differences between the two texts: "the hero of the Sanskrit epic moves, in his consciousness, from the level of man to that of god .... The Tamil text, with its theistic axiology, stresses the opposite direction - that of God's similarity to man": "The anthropology of the avatar in Kampan's Iramavataram" (pp. 270-287). Saul Shaked traces the changes undergone by the two sets of First Man figures, Yima (Jamshid) with his associates, and Gaya Maretan (Gayomard) with the Bull and the first human couple Mashye and Mashyane: "First man, first king. Notes on Semitic-Iranian syncretism and Iranian mythological transformations" (pp. 238-256). In "Religious visions of the end of the world" (pp. 125-137) Joseph Kitagawa treats briefly of the concepts of Saosyant, the Cakravartin, Maitreya and Amitabha. He writes that Heinrich Zimmer has traced the concept of the Cakravartin not only to the earliest Vedic, but also to the pre-Vedic, pre-Aryan tradition of India, Indo-Iranian Journal 32: 1989.

Page #15
REVIEWS although without pointing out that, to say the least, Zimmer's opinion is not shared by many scholars. Important for Indian studies is Ninian Smart's article "The importance of diasporas" (pp. 288-297), in which he examines the effects of Hindu diaspora on religious developments. 209 Two other articles must be mentioned, although they relate to Chinese and Japanese religion. Anna Seidel treats of the resurrection of the body in Taoism: "Post-mortem immortality - or: the Taoist resurrection of the body" (pp. 223-237). Michael Pye examines the attitude of Shinran Shonin (1173-1262), the founder of Shin Buddhism, towards the concept of thisworldly benefits (genzeriyaku): "This-wordly benefits in Shin Buddhism" (pp. 192-202). The only negative comment which one must make with regard to this volume concerns the many misprints which one would not expect to find in such a publication. Australian National University J. W. DE JONG Eli Franco, Perception, knowledge and disbelief. A Study of Jayarasi's Scepticism (Alt- und Neu-indische Studien, Band 35). Stuttgart, Franz Steiner Verlag Wiesbaden GmbH., 1987. IV, 584 pp. DM 58,-. Jayarasi's Tattvopaplavasimha "The lion which annihilates [all] principles" is the only text of the Lokayata or Carvaka school which has been preserved. The only manuscript known so far dates from the end of the 13th century and was discovered by Pt. Sukhlalji Sanghavi and Pt. Becherdas Doshi in 1926. Pt. Sukhlalji Sanghavi and Rasiklal C. Parikh published the text in 1940 as volume 87 of Gaekwad's Oriental Series. The TUS (Tattvopaplavasimha) comprises 14 chapters of which the first six have been translated by Eli Franco. In his introduction he examines first the relation of Jayarasi to traditional Lokayata which was materialistic and recognised only one means of valid cognition, i.e. perception (pratyaksam ekam pramanam). Jayarasi was not a materialist, but a radical sceptic who rejected even perception as a means of valid cognition. Franco points out that "there is no doubt that Jayarasi regarded himself as a follower of Brhaspati, the guru of the gods and legendary founder of the Lokayata" (p. 4), although he boasted in the concluding verses of the TUS to have discovered new arguments which did not occur to Brhaspati. Sanghavi and Parikh came to the conclusion that Jayarasi had lived in the eighth century A.D. This dating is confirmed by Franco. He remarks Indo-Iranian Journal 32: 1989.

Page #16
210 REVIEWS that Jayarasi seems to have played a role in the development of the vikalpa reasoning in the Jaina school and that this is probably the reason which saved him from oblivion. Franco points out that the difficulty of the TUS resides in the fact that Jayarasi is extremely laconic in referring to the theories he is criticizing. An additional difficulty is that he often uses texts which have not survived. The Nyaya philosophy is represented by the Acaryas, a group of commentators on the Nyayavarttika, the Mimamsa school by the Brhattika and the commentaries thereupon, and the Sankhya by the Sastitantra of Varsaganya and Vindhyavasin's commentary on it. As to the Buddhists his main opponent is Dharmakirti and he refers often to his Pramanaviniscaya. In the compendious notes to the translation (pp. 299-551) Franco identifies and quotes Jayarasi's sources, discusses difficulties of interpretation, etc. with great learning and penetration. Franco has also re-edited the text, with the help of a microfilm of the manuscript. In many places he has arrived at establishing better readings. In his introduction Franco examines the arguments advanced against scepticism by its opponents and tries to demonstrate that "sceptics both in India and in Europe were charged with certain forms of the fallacy of selfrefutation, and that in both traditions this charge is mainly due to a misunderstanding of the sceptic's use of language and of his frame of mind" (p. 37). He remarks that the argument of practical impossibility of scepticism seems never to have been employed in Indian philosophy, whereas in European philosophy it is even more common than the argument of selfrefutation (p. 41). Franco has a high regard for scepticism and places the TUS in a larger context with the following remarks: "It is my firm belief that the study of the sceptics and their influence on the so-called dogmatic philosophers will improve and deepen our understanding of the historical development of Indian thought, and that the evaluation of its role in India, compared to and contrasted with its role in Western philosophy, may turn out to be one of the most gratifying tasks for comparative philosophy" (p. 2). It is to be hoped that Franco's book will be read not only by the students of Indian philosophy but also by all those interested in the study of scepticism. The TUS ends with the words: tad evam upaplutesv eva tattvesv avicaritaramaniyah sarve vyavahara ghatanta iti, of which Franco gives the following translation: "Thus, when the principles are completely annihilated, all every day practice (or: all thinking, speaking and acting) can be delightful in as much as it no [longer has to be] deliberated" (p. 44). Franco does not comment upon the expression avicaritaramaniya apart from remarking that

Page #17
REVIEWS 211 it is used pejoratively in non-Carvaka texts, cf. e.g. Prakaranapancika p. 373.1 (n. 59). In another note he quotes two passages from Jayanta's works in which the same expression is found (n. 23).' It is interesting to note that in Buddhist texts dating from the eighth century avicararamaniya and avicaraikaramaniya are often used to characterize the conventional truth or reality (samvrti).? M. Ichigo renders karika 64 of the Madhyamakalamkarakarika as follows: "One should understand that conventional (truth) is in essence (1) that which is agreeable and acceptable only as long as it is not investigated critically (ma-brtags gcig-punams-dga'-zin = avicaraikaramaniya or avicaraikaramya), (2) that which is characterized by arising and decay, and (3) whatever has causal efficiency."3 Most scholars agree with Giuseppe Morichini that Santaraksita must have lived roughly from 725 to 785.4 Haribhadra lived in the second half of the eighth century during the reign of king Dharmapala of the Pala dynasty. Avicaraikaramya is the expression most commonly used by Haribhadra. Other expressions used by him are avicararamaniya, avicaraikamanohara and avicaraikaramaniya. Similar expressions are used by later authors. Avicaramanohara occurs once in the Bodhicaryavatarapanjika written by Prajnakaramati (10th century ?), cf. p. 357.17 (ed. L. de La Vallee Poussin). Avicaritaramya is found in Jnanasrimitra's Ksanabhangadhyaya p. 6.8: avicaritaramya pratitih samvrtih (Jnanasrimitranibandhavali, ed. A. Thakur, Patna, 1959, p. 6.8). According to Thakur Jnanasrimitra lived in the first half of the eleventh century (cf. Introduction p. 3). The term (Derge ed., p. 260a2: ma-brtags nams-dga') occurs also in the Madhyamakaratnapradipa attributed to Bhavya (c. 490- 570). Lindtner is convinced that this work was written by Bhavya but other scholars do not agree with him. In his Satyadvayavatara Atisa (928-1054) uses the term avicaraikaramaniya. It is not possible to know whether Jayarasi was the first to use this expression or the Buddhists. Both vyavahara and samvrti express similar concepts, but Jayarasi maintains that one must be content with every day practice whereas the Buddhist denied all validity to samvrti from the point of view of the absolute truth or reality (paramartha). It would be worthwhile to further investigate the history and use of avicaritaramaniya and its synonyms. NOTES Franco's index wrongly refers to p. 48.23. 2 EJIMA Yasunori points out that this seems to be a new expression commonly used in this period to define samvrti. He refers to Santaraksita's Madhyamakalamkarakarika 64, and to his Vadanyavrtti, p. 26.10 (ed. R. Sankrityayana, Appendix to JOBRS, vols. XXI-XXII,

Page #18
212 REVIEWS 1935-36) and to the General Index to Haribhadra's Abhisamayalamkaraloka (Chugan shiso no tenkai. Bhavaviveka no kenkyu, Tokyo, 1980, p. 253, n. 46). 3 ICHIGO Masamichi, Madhyamakalamkara, Tokyo, 1985, p. CXLIIX. The same karika has been translated and explained by KAJIYAMA Yuichi, cf. 'Chugan shiso no rekishi to bunken', pp. 55-56 (Koza daijo bukkyo, vol. 7, Tokyo, 1982). 4 Cf. 'Early Vedanta Philosophy', East and West N.S. 11 (1960), p. 35 and NAKAMURA Hajime, Shoki no Vedanta tetsugaku (Tokyo, 1981), pp. 653-657. 5 Cf. David Seyfort Ruegg, La theorie du tathagatagarbha et du gotra (Paris, 1969), p. 124. Chr. Lindtner, 'Atisa's introduction to the two truths, and its sources', JIPh 9 (1981), p. 170, verse 4: "When you do not examine it, it affords pleasure." 7 Cf. Chr. Lindtner, 'On Bhavya's Madhyamakaratnapradipa', Indologica Taurinensia XII (1984), pp. 163-164 and p. 164, n. 3. 8 Chr. Lindtner, JIPh 9 (1981), p. 190, verse 3: ma brtags gcig pu nams dga' ba'i which translates avicaraikaramaniya and not avicararamaniya. Australian National University J. W. DE JONG Patrick Olivelle, Renunciation in Hinduism. A Medieval Debate. Volume One: The Debate and the Advaita Argument, 1986. 156 pp., Volume Two: The Visistadvaita Argument, 1987. 193 pp. Publications of the De Nobili Research Library, vols. 13 and 14. AS 360 and 360 (available only jointly). Commission Agents: E. J. Brill, Leiden - Gerold & Co., Vienna - Motilal Banarsidass, Delhi. The debate studied by Patrick Olivelle concerns the proper linga 'emblem' of a Brahmanical renouncer. The controversy between the Advaitins and . the Visistadvaitins centers on three points: 1. Should renouncers continue to wear the sacrificial cord (yajnopavita) and topknot (sikha)? 2. Should they carry a triple staff (tridanda), a single staff (ekadanda), or no staff at all? 3. Is there a level of renunciation that transcends dharma, and is it permissible for renouncers at that level to wear no emblem at all? Patrick Olivelle sketches the history of the controversy in chapter three of part one: the debate. One of the problems discussed by him is the distinction between the ekadanda and the tridanda. Olivelle shows that in ancient documents there is no clear distinction between tridanda and trivistabdha which both indicate a tripod used to hang a waterpot. In medieval texts tridanda came to mean a walking stick made of three reeds. Olivelle remarks that the controversy between Advaita and Visistadvaita on the renouncer's place within dharma operates at two levels, depending on whether the renouncer is already enlightened or is only aspiring to achieve enlightenment. The two schools were deeply divided on the importance of rites and acts for renouncers and the type of linga carried by the Indo-Iranian Journal 32 (1989).

Page #19
REVIEWS 213 renouncer communities of the two schools was a symbol of a deep doctrinal division. The debate on the liriga of the renouncer was carried on by appealing to the authority of sruti and smrti. Olivelle studies the problems which arose in this connection regarding the distinction between authentic Vedic or smrti texts and spurious texts. He examines the hermeneutical principles applied by the pandits and the criteria for the authenticity of Vedic and smrti texts. Olivelle shows very well the importance of this debate not only for understanding the status of the renouncer according to the two Vedanta schools, but also for the problem of determining the authenticity of Vedic smrti texts. Volume one contains the text and translation of the Advaita texts: Samkara's commentary on Bihadaranyaka Upanisad 3.5.1, Anandanubhava's Nyayaratnadipavali, Madhava's Parasaramadhaviya and the Pancamasramavidhana. The Pancamasramavidhana is edited by Olivelle on the basis of seven manuscripts. Volume two comprises the following Visistadvaita texts: Linganirupana, the third chapter of Yadavaprakasa's Yatidharmasamuccaya, the Satyayaniya Upanisad, one of the Samnyasa Upanisads edited by F. O. Schrader (Madras, 1912) and studied by J. F. Sprockhoff (Samnyasa, Wiesbaden, 1976), the Yatilingasamarthana, the tenth chapter of Varada's Sribhasyaprameyamala, and two chapters of Vedanta Desika's Satadusani, the Yatilingabhedabhangavada and the Alepakamatabhangavada. The Yatilingasamarthana is edited by Olivelle on the basis of eleven manuscripts and one printed edition. A manuscript of the text had been used by Sprockhoff (cf. op. cit., p. 269, n. 18). Almost all the texts are translated for the first time by Olivelle. On p. 23 he states that the Satyayaniya Upanisad is translated here for the first time but in the notes to his translation he refers to Ramanathan's translation: The Samnyasa Upanisad-s (on renunciation). Tr. A. A. Ramanathan. Madras, 1978 (cf. vol. 1, p. 13). The editions used by Olivelle are listed in the bibliography which, however, does not mention the edition of Vedanta Desika's Satadusani used by him. Olivelle's translations are done with great care and only in very rare instances would one like to suggest a different rendering. In volume one on p. 88 (phrase 41) Olivelle has omitted to translate esanatvac, and on p. 102 (phrase 21) asutra, cf. Sprockhoff p. 71: asutrodarapatrah" (er ist) ohne Opferschnur, (er hat) den Bauch als Gefass." On p. 102 Olivelle translates yatkimcin nadyat as "let him not eat anything else" instead of "let him not eat anything whatsoever." Olivelle notes that the reading here is quite

Page #20
214 REVIEWS different from that of the critical edition of the Kathasruti Upanisad. Anandanubhava's reading is found in the Katharudra Upanisad. On p. 123 the text quotes Harita: Naste jalapavitre va tridande vapramadatah/ekam tu vainavam dandam palasam bailvam eva va/grhitva vicaret tavad yaval labhyet tridandakam // Olivelle remarks that most editions read va pramadatah (p. 138, n. 135). He is certainly correct in reading vapramadatah, but I do not think that apramadatah refers to grhitva "he should diligently take" and I would prefer to connect it with naste: "When the water strainer or the triple staff is spoilt without this being due to negligence." In volume two on p. 10 (phrase 43) the text has matran dasa which is probably a misprint for matra dasa, for matra is the technical term for the personal belongings of a renouncer (cf. volume one, p. 27), and not matra. On p. 18 (phrase 69) Olivelle translates atmasammitam by 'one that he likes'. A danda atmasammita is 'a stick of he same size as himself', cf. Dattatreya's murdhni sammitan (p. 11, phrase 76) and murdhasammitan (p. 49, phrase 26). On p. 30 Olivelle translates atha khalu saumyaite parivrajak, yatha pradurbhavanti tatha bhavanti with "Now, my dear, these wanderers become (after death) what they appear to be (here)." Olivelle remarks that he follows Schrader and Sprockhoff in translating this phrase. I wonder whether it is not possible to translate: "... these wanderers are as they appear [i.e. their inner disposition agrees with their behaviour and appearancel." P. 90 'six faults' (phrase 114) is a slip of the pen for 'eight faults'. On p. 120 Olivelle translates: "In the evening (he should go) begging clockwise to households of all varnas without discrimination, avoiding those of a bad repute and of the outcastes" (phrase 32). Avicikitsitam means rather 'without doubt, without hesitation than 'without discrimination'. On p. 99 phrase 55 reads: iha tavat tvatpaksapativijnanesvaradyanudahrtatvat, tvatpaksapratiksepakair yadavacaryabhaskaraprabhrtibhir bahusruter upanyasya dusitatvabhavac ca samdigdhasrutitvadisu kesucid vacanesu apratyayad eva pariharah. Olivelle translates: "... Yadavaprakasa, Bhaskara Acarya, and the like who have refuted your position do not allude to and censure many of these Vedic texts." In a note he remarks that the precise meaning of bahusruteh in this context is unclear. I think that one must read bahusrutaih: "... scholars such as Yadavaprakasa, Bhaskara Acarya and others who have refuted your position have referred to them and have not censured them." P. 116 phrase 403 reads: na ca jatyadyanurupasambandhadivarjanat bhagavadbhaktesv api tesv avajnadiprasanga iti vacyam. Olivelle translates: "It cannot be said that, since we avoid certain types of relationships with them in accordance with their caste and the like, we should also show marks of disrespect to them, even though they are

Page #21
REVIEWS 215 devotees of the Lord." I believe that the meaning is as follows: "It cannot be said that by avoiding certain types of relationships with them in accordance with their caste and the like, we show disrespect, etc. to them, even though they are devotees of the Lord." A few misprints have to be corrected. Volume two, p. 13 (phrase 126) read kalau for kalu; p. 75 (phrase 105) read vyasadisamakhyam for vyadismakhyam; p. 100 (phrase 71) read vaikhanasasutre for vikhanassutre. NOTE Isadivimsottarasatopanisadah, ed. by Narayana Rama Acarya (Bombay, NSP 1948), p. 546, 1.5. Australian National University J. W. DE JONG Mark Tatz, Asarga's Chapter of Ethics With the Commentary of Tsong-KhaPa, The Basic Path to Awakening, The Complete Bodhisattva (Studies in Asian Thought and Religion, Volume 4). Lewiston/Queenston, The Edwin Mellen Press, 1986. VIII, 358 pp. $49.95. Mark Tatz's book comprises translations of the silapatala from Asanga's Bodhisattvabhumi and of Tson-kha-pa's commentary: Byar-chub sems-dpa'i tshul-khrims-kyi rnam-bsad, byar-chub gzun-lam. In his introduction Tatz gives a good explanation of bodhisattva ethics and of the literature on the subject in India and Tibet. Tson-kha-pa's work is very instructive because it is based upon a detailed knowledge of the Indian literature on bodhisattva ethics. Tatz remarks: 'Tsong-kha-pa bases his comments upon the Indian commentaries to the Chapter on Ethics and the TV (Candragomin's Twenty Verses of the Bodhisattva Vow). That of Jinaputra/Samudramegha is most often utilized for explanation of terms. Santaraksita, Atisa, Ratnakarasanti, Santideva and others are introduced in the course of discussing complicated questions' (p. 35). Tison-kha-pa's work occupies 98 folios in the Peking edition and 109 folios in the Tashilhunpo edition. Tatz's translation (pp. 91-263) is followed by no less than 592 notes. The appendices comprise a translation of a fragment of a commentary to Candragomin's Twenty Verses from Tun-huang, a translation of a passage on seminal transgression from the Akasagarbha-sutra and a translation of two sections from the commentary by Jinaputra/Samudramegha: on alcohol and on permission to murder. A detailed index (pp. 343-358) concludes the work. Indo-Iranian Journal 32: 1989.

Page #22
216 REVIEWS Tatz's book is a useful contribution to the study of bodhisattva ethics in India and Tibet. The texts translated by him are of great importance and the translation is very readable. One has the impression that Tatz was more interested in Tson-kha-pa's work than in Asanga's silapatala. The translation is without annotation which, however, is to be found in the notes to the translation of the Basic Path, a rather inconvenient arrangement for those who are mainly interested in the silapatala. Tatz gives no page references to the editions by Wogihara and Dutt. It would also have been very helpful if Tatz had subdivided the text into sections and subsections in accordance with the synopsis which is to be found in Wogihara's edition as has been done by Demieville in his translation of the dhyana chapter. On the whole the translation of the Chapter on Ethics is satisfactory, but Tatz has not solved all problems relating to the text. For instance, on p. 140 of Wogihara's edition one reads: samasatah sarvanarthopasamhitamanapasamudacaraparivarjanaih cittanuvartanata in brief, he complies with the thought [of everyone] but avoids all conduct which is useless and disagreeable'. Tatz translates: 'briefly, he complies with the thought [of everyone] excepting those who are useless and disagreeable' (p. 50). There is no divergence here between the Sanskrit text and the Tibetan version and Tatz fails to point out the reason for his translation. The next sentence reads: bhutais ca gunaih sampraharsanata rahah prakasam vodbhavanam upadaya 'he makes [everyone] rejoice in genuine good qualities by making them known privately or publicly'. Tatz translates: 'He applauds the genuine good qualities of others, or reveals those that are hidden' (p. 50). On p. 187 the Sanskrit text reads: kayasya bhedat tatropapadyate yatrasya samanadhikasila bodhisattvah sabhagah sahadharmikah kalyanamitrabhuta bhavamti. Tatz translates: 'After his physical dissolution, wherever he is reborn those with the same higher ethics are found, bodhisattva colleagues with the same doctrine who function as spiritual advisers' (p. 88). Tatz wrongly translates samanadhikasila 'with the same or higher ethics', cf. Tib. tshul-khrims mtshuns-pa 'am Ihag-pa'i byan-chub-sems-dpa'. Tatz has also misunderstood the Tibetan translation. In note 577 he states that the Tibetan translation reads samana-silah va-adhika-bodhisattvah sabhagah! Tson-kha-pa (Tashilhunpo ed. 106a) has: lus-zig-nas kyan gan-na randan-tshul-khrims mtshuns-pa'am Ihag-pa'i byan-sems skal-pa 'dra-zin... Tatz translates: 'After your physical dissolution, wherever you are reborn, those with the same ethics as yours are found in other words, higher bodhisattva colleagues with the same doctrine... (p. 258). Hsuan-tsang also interpreted samanadhikasila correctly as Tatz could have seen if he had consulted Ui's work on the Bodhisattvabhumi.2 Moreover, Tatz does not explain why he translates sabhaga 'similar' as 'with the same doctrine".

Page #23
REVIEWS 217 Tatz suggests very few corrections to the Sanskrit text. On p. 152 Wogihara reads: sahadharmikasya bodhisattvasya krtapranidhanataya vijnasya but Dutt has krtapranidhanasya. Tatz suggests to read samvarasamadatta-vijnasya because the Tibetan version reads sdom-pa blanspa mkhas-pa (p. 276, note 148). It would at most be possible to read samvarasamadanavijnasya but krtapranidhanasya is confirmed by Hsuantsang's Chinese version. On p. 38 Tatz writes that 'the divergences are not sufficient to make the Chinese crucial for work on the Bodhisattvabhumi, if one has the Sanskrit and Tibetan at hand' (p. 38). However, it is wrong to neglect the Chinese version when one tries to correct the text." On p. 153 Wogihara reads: tathagatapratimam puratah sthapayitva sammukhikrtyaivam syad vacaniyah. Dutt has sampuraskrtyaivam instead of sammukhikrtyaivam. Tatz suggests to read sampujya puraskrtya after Bodhisattvapratimoksasutra and Tibetan sources (p. 279, note 183). The Tibetan translation has legs-par mchod-pa byas-nas which probably translates sampujya. However, there is nothing corresponding to puraskrtya in the Tibetan translation. Obvious mistakes in the Sanskrit text are overlooked by Tatz. For instance, on p. 153 Wogihara reads yasya va yacati saktir which must of course be corrected to yasya va yavati saktir (Tib. yar-na des ci nus-pa). In quoting this passage Tson-kha-pa has da-Itar-gyi rtsol-pa'i ci-nus-pa 'making whatever effort of which he is presently capable' (Tatz's translation, p. 146). In a note Tatz queries the word rtsol: ''Effort' rtsol, but this may be a misreading. Gunaprabha has 'whatever human skill rtsal he may presently possess in generating a serene thought.' (p. 279, note 181). However, Gunaprabha's skyes-bu'i rtsal translates purusakara 'effort'.S In his translation of Tson-kha-pa's commentary Tatz underlines passages cited from the chapter on ethics. Tatz remarks that Tson-kha-pa often paraphrases (p. 264, note 17). On p. 118 he underlines the following passage: 'He generates compassion, envisioning him as subject to defilement (non-mons-pa'i chos-can la dmigs-nas snin-rje bskyed-do). According to note 71 (referring to note 72 in the text; note 72 is missing) 'Subject to' (chos can, *dharmin) in place of Bodhisattvabhumi 'according to the doctrine' (chos can gyi, dharma). However, the Sanskrit text has: dharmamahakarunatam upadaya (chos-bzin-gyis snin-rje chen-po la brten-nas). Tatz translates: 'based upon great compassion according to the doctrine' (p. 52). In a passage from the Siksasamuccaya quoted by Tson-kha-pa one finds the expression bodhisattvasiksapadabhyasaparamasya 'intent on the practice of the precept of the Bodhisattva' (transl. Bendall and Rouse, p. 12). Tatz translates: 'who takes it seriously and is familiar with the bases of bodhisattva training' (p. 101). Tatz's wrong translation is due to a misinterpreta

Page #24
218 REVIEWS tion of the Tibetan text (byan-chub sems-dpa'i bslab-pa'i gnas-la goms-pa lhur-byed-pa). He does not seem to have consulted the Sanskrit text of this passage (p. 11.15-12.3) and refers in a note to Skt. 10.12-15 (p. 264, note 21). The undertaking of the pratimoksa vow belongs to seven classes: saptanaikayikam pratimoksasamvarasamadanam (Wogihara, p. 138). In Tibetan saptanaikayika is rendered by ris bdun-po. In note 37 Tatz remarks: "Classes', Tib ris; but Bbh Skt reads naikayika, glossed by Jinaputra as 'those who follow the teachings'' I have not found this explanation in Jinaputra who explains: ris bdun-po zes bya-ba smos-te / ris-bdun-po la brten-nas ris-bdun-gyis phye-ba'i phyir-ro Seven classes' is said because based upon seven classes (nikaya) it is divided into seven classes'. On p. 150 Wogihara reads: napi tesam (see Dutt) purastat priyavigarhako bhavati napy apriyaprasamsakah 'In the presence of them (i.e. others) he does not censure their friends nor praise their enemies'. Tatz translates: 'He does not censure someone before his friends, nor praise him before his foes' (p. 58). Tatz's wrong translation is due to the fact that he does not translate tesam purastat. In translating Tson-kha-pa's commentary he likewise omits gzan-gyi mdun-du 'in the presence of others' (p. 129). Sometimes it is difficult to understand Tatz's translation. For instance, after having enumerated a number of items Tson-kha-pa states: don-tshan de-dag re-re'i mgor byan-chub-sems-dpa' gzan-gyi sems dar-mthun-par byed-pa ni zes sbyar-ro 'At the beginning of each of these sections one must add 'the bodhisattva acts in compliance with the expectations of others'. . Tatz translates: 'The application is that on top of these items, the bodhisattva complies with others' expectations' (p. 130). Tatz strangely mistranslates de-dan-de 'this and that'. Tson-kha-pa states that he extols those endowed with the qualities of faith, morality, learning, liberality and wisdom by mentioning them one by one (dad-pa dan tshulkhrims dan thos-pa dan gton-ba dan ses-rab kyi yon-tan dar-ldan-pa la dedan-de'i gtam-gyis gzers-bstod-pa'o). Tatz translates: ... by discussing it [the quality and them' (p. 130). Tatz also wrongly translates gton-ba liberality' by 'renunciation'. Tson-kha-pa explains that morality and liberality are causes for temporal happiness (mron-mtho, abhyudaya) and wisdom for ultimate bliss (res-legs, nihsreyasa). Tatz translates mron-mtho and res-legs by 'elevation' and 'sublimity' without indicating the Sanskrit equivalents (p. 130). Tson-kha-pa explains that learning brings about wisdom (de 'dren-pa ni thos-pa'o). Tatz translates: 'Its leader is learnedness' (ibid.). Tibetan 'dren-pa is used to render avahaka (Wogihara p. 138.4). It is a pity that Tatz's work suffers from a series of minor blemishes

Page #25
REVIEWS because, in general, he understands the texts correctly. His work would have gained much by having been carefully vetted before publication. In any case, we must be glad that Tatz has made these two texts more accessible by his translation. The chapter on ethics from the Bodhisattvabhumi does not seem to have attracted much attention from Western scholars. Several prominent Japanese scholars such as Hirakawa Akira, Aramaki Noritoshi and Hadano Hakuyu have published studies of this chapter. Tatz does not mention them although he lists in his bibliography an article in Japanese by Ueyama Daishun. They are all enumerated in the excellent bibliography which Hakamaya Noriaki has added to the photomechanic reprint of the Derge edition of volume seven of the Sems-tsam section of the bsTan-'gyur (Tokyo, 1980). Australian National University 219 NOTES I'Le chapitre de la Bodhisattvabhumi sur la Perfection du Dhyana', Rocznik Orientalistyczny 21 (1957), 109-128 Choix d'etudes bouddhiques (Leiden, 1973), pp. 300-319. 2 Bonkantaisho Bosatsujisakuin (Tokyo, 1961), pp. 202-203. 3 Ui's translation (p. 170) is wrong because he follows Wogihara's text. 4 Hsuan-tsang's translation of the chapter on ethics is to be found in Vol. 30 of the Taisho edition, pp. 510-523 and not pp. 516-521 as indicated by Tatz (p. 329). 5 Cf. Akira Hirakawa, etc., Index to the Abhidharmakosabhasya, Part one (Tokyo, 1973), P. 238. 6 On p. 333 his name is spelled Hadano, Kakukyo by Tatz. Indo-Iranian Journal 32: 1989. J. W. DE JONG David Snellgrove, Indo-Tibetan Buddhism. Indian Buddhists & Their Tibetan Successors. London, Serindia Publications, 1987. xxiv, 640 pp., 88 pl. PS30.00. David Snellgrove retired in 1982 from his chair of Tibetan studies in the university of London. The present publication shows that he has used his retirement to write a comprehensive work on Buddhism in India, Central Asia, Nepal and Tibet up to about 1300 A.D. Snellgrove's book comprises five chapters of unequal length: I. Origins in India (pp. 1-43); II. Later developments in India (pp. 44-116); III. Tantric Buddhism (pp. 117-303); IV. Buddhist communities in India and beyond (pp. 305-380); V. The conversion of Tibet (pp. 381-526). In his preface Snellgrove explains his reasons for dealing only briefly with the Buddhist origins in India: "For the first chapter on Buddhist origins

Page #26
220 REVIEWS in India there is already a very large amount of published work available, mainly in English, but also in French and German. This subject is thus comparatively easily treated, and one's main task in this case is to correct the liberal-minded rationalizing approach, from which the retelling of Sakyamuni's life may suffer distortion." It is certainly true that much more has been written on the Buddhist origins in India than on Tantrism and the conversion of Tibet to Buddhism but whether this subject is therefore comparatively easily treated is another matter. For instance, in recent years several scholars have been trying to determine the nature of the earliest Buddhist doctrine by distinguishing different layers in the sutras, by isolating verse passages and by comparing Buddhist doctrines with Jain doctrines and teachings found in the Upanisads. Snellgrove does not once refer to the writings of scholars such as Schmithausen and Vetter in Europe and Aramaki and Nakamura in Japan. Secondly, one wonders whether it is still necessary in the nineteen-eighties to correct the liberal-minded rationalizing approach which was widespread towards the end of the last century and which is only rarely found in recent times. It was obviously Snellgrove's intention to dwell only briefly on topics which have already been studied by other scholars unless he considered it necessary to correct wrong ideas. It is necessary to keep this in mind when reading Snellgrove's book and not to expect a well-balanced treatment of the various aspects of Buddhism Snellgrove has made a wide use of Sanskrit and Tibetan materials, translating many passages of hitherto untranslated texts. In the case of texts. already translated he presents his own translations. The first two chapters deal mainly with Hinayana and Mahayana. Snellgrove stresses that one cannot treat them as two distinct phases: "They may be regarded, however, as two fluctuating tendencies, usually mingling together, and only kept strictly apart in certain philosophical texts" (p. 32). He also points out that in Mahayana Sakyamuni, the Buddha of the present world-age, remains the centre of Buddhist devotion. According to Snellgrove "The central object of worship was in the first instance the stupa and then later as the anthropomorphic form of Sakyamuni was elaborated, the Buddha-image itself, either superimposed upon the stupa or enthroned alone" (p. 51). In chapter four Snellgrove remarks that in the seventh century Hsuan-tsang is greatly interested in all the historical and legendary sites connected with Sakyamuni's life and that the only celestial Bodhisattvas, whose cult impressed him, are Maitreya and Avalokitesvara. He also points out that there seems to be a time lag of two or three centuries between Mahayana sutras preaching the benefits of devotion to other Buddhas as well as Sakyamuni and their iconographical representation. As to Tantric iconography Snellgrove writes that "there is little surviving iconographic

Page #27
REVIEWS 221 material that can be firmly classed as tantric earlier than the ninth century" (p. 316). Snellgrove makes no reference to inscriptions but it is interesting to note that the oldest mention of the term Mahayana is found in an inscription dated A.D. 507-8. Although the oldest Mahayana texts were probably written already in the first century B.C., it seems that the sentiment of belonging to a specific and different school of Buddhism was felt only after a lapse of several centuries. Snellgrove does not attach much importance to the influence of layfolk on the development of Mahayana: "It is sometimes suggested that the Mahayana represents a more popular form of Buddhism in which the attitudes and aspirations of the layfolk are accorded greater scope, but such a suggestion, while partly true, can also be quite misleading. There would seem to be no doubt that the real protagonists of the Mahayana were monks, and the new scriptures were compiled by monks, some of whom were renowned as masters of philosophy. Also the career of the Bodhisattva, as described in the texts, assumes that the life of a monk and at least a life of celibacy are essential conditions for his progress" (pp. 62-63). It is a pity that Snellgrove does not discuss in greater detail the role of the layfolk in the development of Mahayana. Since the publication of Hirakawa's article on "The Rise of Mahayana Buddhism and Its Relation to the Worship of Stupas' in 1963, the influence of lay-followers especially with regard to the worship of stupas has been a lively subject of discussion in Japan. Probably some scholars have overrated the importance of layfollowers, but one must not forget the fact that the Mahayana sutras are inspired by a very different spirit. It is perhaps not inappropriate to quote the following words by Etienne Lamotte: "Le Mahayana est traverse de part en part par un esprit laic et democratique. Les Mahayanasutra ne s'adressent plus avant tout aux moines qui ont quitte leur maison pour mener une vie errante, mais aux nobles fils et filles de famille demeures dans le siecle." On p. 7 Snellgrove remarks that the story of Sakyamuni's life was regarded as representing the set course that any Buddha should follow in his last life in this world and that it was defined as the Twelve Acts of a Buddha. To my knowledge the Twelve Acts of the Buddha (mdzad-pa bcugnis) are not mentioned in Indian Buddhist texts but have been formulated for the first time in Tibet. In an article published in 1839 Csoma de Koros gave a list of the twelve acts which is different from the one published by Snellgrove. It would have been very welcome if such an excellent specialist in Tibetan Buddhism as Snellgrove would have given us some more information on the origin and history of the different lists of the Twelve Acts. Whereas chapters one, two and four suffer somewhat from the fact that

Page #28
222 REVIEWS Snellgrove tried to deal with many topics without being able to treat them adequately within the limited space given to them, a much more satisfactory picture is to be found in chapters three and five which deal respectively with Tantric Buddhism and the conversion of Buddhism. Snellgrove is one of the leading specialists in Buddhist Tantrism. His edition and translation of the Hevajratantra (London, 1959) is one of the best studies of a Tantric text ever published. His book-length chapter on Tantrism is a major contribution to the study of Tantrism and will be welcomed both by the general public and by specialists. The Tantric literature is very vast and only very few texts have been adequately studied by modern scholars. What Snellgrove has achieved in this chapter is an overview which forms a solid basis for further studies. He makes a distinction "between Mahayana Buddhism including those tantras that are closely relatable on the one hand and the form of tantric Buddhism, which may accurately be described as Vajrayana, on the other" (p. 279). According to Snellgrove the traditional arrangement of four categories of tantras (kriya, carya, yoga and anuttarayoga) is of a rather arbitrary nature (p. 119, n. 5). He devotes one section to various kinds of tantra in which he distinguishes tantras relatable to Mahayana sutras and tantras with non-Buddhist associations (pp. 147-160). To these later groups belong the anuttarayogatantras which introduce many new and un-Buddhist concepts. Throughout this chapter on Tantrism Snellgrove shows the progressive tantric development from Mahayana sutras to Vajrayana tantras. In the course of his treatment he explains the main characteristics of the tantric literature such as magical formulas, mandalas, Buddha-families, initiations and consecrations, etc. Snellgrove takes great pains to determine the meaning of important tantric concepts: siddha, sadhana, vajra, vidya, hrdaya, bija and samaya, etc. Nobody will read this chapter without learning much about Buddhist tantrism. Especially welcome are the many passages from tantric texts of which most have not been translated before. One of the texts most quoted by Snellgrove is the Sarvatathagatatattvasamgraha which Snellgrove has studied for many years. In 1959 he announced the discovery of an early manuscript of it in Nepal. In 1981 Snellgrove published a facsimile edition of this manuscript provided with a long introduction of more than sixty pages. In the same year Isshi Yamada brought out a romanised edition of this same manuscript compared with the Tibetan and Chinese translations. Recently Lokesh Chandra published a devanagari edition which we have not yet been able to consult.8 The STTS has also been edited in Japan by Horiuchi Kanjin. In translating passages from the STTS Snellgrove refers only to Yamada's edition. Even with the

Page #29
REVIEWS 223 help of the editions by Yamada and Horiuchi many problems remain. For instance, on p. 138 Snellgrove translates: "They replied: "Let it be so, we enter into this pledge of yours."". Yamada's edition reads: ta evam ahuh / "evam astv iti / kin tu tava samayam [akovidah]"/. Yamada remarks in a note that the word akovidah is illegible. Both the Tibetan and the Chinese versions translate 'we do not understand'. Horiuchi reads: kim tu [vayam] bhagavad- (or "vatah) ajna-samayam nava (or adhi) gacchamah].10 Snellgrove's translation is not accompanied by philological notes and it is difficult to know how he arrived at his interpretation. It is very much to be hoped that Snellgrove will publish an edition and translation of the STTS as he has done so admirably for the Hevajratantra. Snellgrove's translations are not always free from errors. On p. 126 he translates a passage from the Mahayanasutralamkara (ed. S. Levi, p. 87, line 7): yasmad dharmadhatuvinirmukto dharmo nasti dharmatavyatirekena dharmabhavat: "As there is no element apart from the elemental sphere, so there is no essential truth (or elemental essence, dharmata) apart from the elements." The meaning is just the opposite: "As there is no element apart from the elemental sphere because there is no element apart from the elemental essence." Snellgrove refers to Levi's translation which he misunderstood: "Puisqu'il n'y a pas d'Ideal qui soit emancipe du Plan des Ideaux en effet, pas d'Ideal sans Idealite." The Sekoddesatika mentions a Wisdom-maiden (prajna) from twelve to twenty years old. Snellgrove says that she is from sixteen to twelve years old (p. 263). Chapter V The Conversion of Tibet' (pp. 381-526) describes the background to the introduction of Buddhism, the first diffusion of Buddhism during the eighth and ninth centuries and the second diffusion from about the year 1000 onwards. Snellgrove insists on the fact that Bon was not the original pre-Buddhist religion of Tibet but a heterodox form of Buddhism. He draws attention to the close relationship between Bonpos and rNin-mapas. Snellgrove writes: "They differ from more orthodox Buddhists in their joint possession of the teachings known as the Great Fulfillment (rdzogschen), in their separate possession of groups of tantras that were excluded from the later Tibetan Buddhist Canon, and in their tardy development of monastic orders, both preserving a long tradition of noncelibate religious practice" (p. 404). Buddhism was introduced to Tibet both from China and India. According to Snellgrove the period of Chinese influence was limited to the short period during which the Tibetans controlled Central Asia and northwest China. After the disintegration of the Tibetan empire from 842 onward India became the dominating factor. As to the famous council held either

Page #30
224 REVIEWS in Lhasa or bSam-yas Snellgrove points out that the Chinese teacher Mahayana taught doctrines derived from the teachings of the Mind Only school and similar to those found in Yoga Tantras promulgated in Tibet from the eighth century onward. He describes very well the complicated religious situation in Tibet during the eighth and ninth centuries. It is only in recent years that due to the study of the Bon religion, of the teachings of the rNin-ma-pa school and of the Tun-huang manuscripts that it hast . become clear that later Tibetan historians have misrepresented the history of Buddhism during the first diffusion. For the later period as from 1000 much useful material is to be found in the Tibetan religious histories and Snellgrove quotes abundantly from the Blue Annals written by 'Gos Lo-tsaba gzon-nu-dpal (1392-1481). In this chapter Snellgrove has given us a masterful expose based on original sources and the work done by scholars in this century. Especially for the period of the seventh to the tenth century Snellgrove's work is without doubt the best available at present. He is in full command of his sources and his long experience with Tibetan sources and with living Tibetan Buddhism in Nepal and Ladakh inspires his writing. On pp. 446-449 Snellgrove quotes extensively from a letter written by the Indian scholar Buddhaguhya to king Khri Sron-lde-brtsan. He considers the letter to be genuine but doubts as to the authenticity remain. In an article published in 1968 Hadano maintained that the letter is a late forgery. Snellgrove's translation is based upon the Peking and Narthang Tanjurs. He does not mention Siglinde Dietz's dissertation Die buddhistische Briefliteratur Indiens (Bonn, 1980) in which she studies the same letter. Dietz's dissertation has been published in a revised version as volume 84 of the Asiatische Forschungen (Wiesbaden, 1984) to which we refer. Dietz's edition of the text of the letter is based upon the Peking, Narthang, Derge and Chone Tanjurs. On p. 448 Snellgrove translates: "One's later knowledge is commingled with existing) knowledge. The essential property of mind is. wealth of knowledge, I say." In Dietz's edition the text reads: rdzas kyi ses pa ses pa'i nan na dbul / sems kyi mtshan nid ses pa phyug ces mchi'o "The knowledge of material things is poverty among knowledges. Knowledge of - the essential nature of the mind is richness. So it is said." Siglinde Dietz translates: "Man biete sein Wissen uber materielle Dinge unter Weisen dar, (dann wird man reich (an Wissen [hinsichtlich des Charakters des Geists genannt (werden)" (p. 371). Dietz is correct in reading rdzas and not rjes as found in the Narthang and Peking texts but her translation is wrong. Snellgrove remarks that the Tibetan translation of this letter is filled with strange obscurities. Further study of this important document is certainly required.

Page #31
REVIEWS 225 In translating passages from the Blue Annals Snellgrove does not always correct mistakes made by Roerich. For instance, on p. 476 he follows Roerich in restoring Tshad-ma rgyan to Nyayalamkara. Tshad-ma rgyan translates Sanskrit Pramanalamkara of which the full title is Pramanamahabhasyavarttikalamkara.12 In another passage Roerich's translation is correct and Snellgrove has misread the text. Roerich has: "Better than this trance, is the trance of the gods of the rupa and arupa dhatus who are able to meditate throughout an entire cosmic period" (Roerich, The Blue Annals, II, p. 455). Snellgrove mentions a trance devoid of sensations (p. 495) but the Tibetan text has: mi 'chor-ba'i tin-ne-'dzin 'a trance which does not pass away' and not mi 'tshor-ba'i tin-ne-'dzin. Snellgrove's book contains an extensive bibliography but many books and articles mentioned in the notes, often without sufficient details, are not mentioned in it. Information on texts is not always correct. For instance, Snellgrove says that the editions of the Catuhaataka by Bhattacharya and Vaidya contain the edited Sanskrit and Tibetan texts for chapters VIII--XVI and that the Sanskrit is missing for the earlier chapters (p. 85, n. 69). There are fragments of the Sanskrit text of both the verses and Candrakirti's commentary for all sixteen chapters. Both Bhattacharya and Vaidya have not only edited the Sanskrit text of the verses found in the fragments but have also tried to reconstruct the text of the verses not found in them. Snellgrove states that Emmerick has translated the Tibetan version of the Surangamasamadhisutra supplemented by a fragmentary Khotanese version (p. 632). However, Emmerick has only translated from the Tibetan passages missing in the Khotanese and has not translated the sections to which nothing corresponds in the Khotanese. To sum up. Snellgrove's book is an important contribution to the study of Indo-Tibetan Buddhism and especially the two long chapters on Tantrism and The Conversion of Tibet deserve the highest praise. Minor flaws in the philological study of the texts do not impair the value of this work and have only been pointed out because Snellgrove himself (and rightly so) sets a high value on philological exactitude. NOTES Cf. Asia Major 17 (1971), p. 110. 2 Akira Hirakawa, "The Rise of Mahayana Buddhism And Its Relationship to the Worship of Stupas', Memoirs of the Research Department of the Toyo Bunko 22 (1963), pp. 57-106. 3 Cf. Takushu Sugimoto, Studies on Buddhist Stupa-cult in India (English title] (Kyoto, 1984), pp. 40-41. + 'Sur la formation du Mahayana', Asiatica, Festschrift Weller (Leipzig, 1954), p. 379. 5 Tibetan Studies (Budapest, 1984), pp. 231-232.

Page #32
226 6 Sata-pitaka Series, Volume 269. 7 Sata-pitaka Series, Volume 262. 8 Delhi, 1987. REVIEWS Cf. N. Iyanaga, 'Recits de la soumission de Mahesvara par Trailokyavijaya', Tantric and Taoist Studies in honour of R. A. Stein, Volume three (Bruxelles, 1985), p. 657. 10 Ed. Horiuchi, Kanjin, Vol. I (Koyasan, 1983), p. 352. 11 Cf. Z. Yamaguchi, Tibetan Studies (Oriental Studies in Japan: Retrospect and Prospect 1963-1972, Part II-17) (Tokyo, 1975), pp. 4-5. For a survey of Tibetan studies in Japan in the period 1973-1982 see Sh. Matsumoto, Tibetan Studies in Japan 1973-1983 (Asian Studies in Japan, 1973-1983. Part II-18). Tokyo, 1986. 12 Cf. Rahula Sankrityayana (ed.), Pramanavartikabhashyam of Prajnakaragupta (Patna, 1953), p. 648. Australian National University J. W. DE JONG Donald S. Lopez, Jr., A Study of Svatantrika. Ithaca, New York USA, Snow Lion Publications, 1987. 483 pp. $35.00. Indo-Iranian Journal 32: 1989. The division of the Madhyamaka school into Prasangikas and Svatantrikas is well known, but until now no detailed study of the Svatantrika system had been undertaken. Mimaki writes that it seems that the terms Prasangika and Svatantrika were used for the first time by Pa tshab Ni ma grags. (1055-7). In the beginning of the ninth century Ye ses sde used the terms Sautrantika-maadhyamika and Yogacara-madhyamika for the first time.2 Tibetan scholars have made many different classifications of the Madhyamaka schools, but since Pa tshab Ni ma grags all agree in considering Bhavaviveka as the founder of the Svatantrika school. In his introduction Lopez enumerates the main works of the Svatantrika school by Bhavaviveka, Jnanagarbha, Santaraksita and Kamalasila (p. 21). Bhavaviveka is the author of two important works, the Madhyamakahrdaya and its autocommentary, the Tarkajvala, and his commentary on Nagarjuna's Mulamadhyamakakarikas, the Prajnapradipa. The latter work is said by Lopez to be a terse commentary. However, it is more extensive than the commentaries by Buddhapalita and Candrakirti. In the Derge Tanjur Bhavaviveka's commentary occupies 214 folios, those by Buddhapalita and Candrakirti respectively 122 and 200 folios. Of Bhavaviveka's commentary only two chapters have been. rendered into Western languages. In 1936 Rahula Samkrtyayana discovered a Sanskrit manuscript of the Madhyamakahrdaya and since 1958 several chapters of this text and its commentary have been edited and translated.4 Lopez's study of the Svatantrika system is not based primarily on the works by Bhavaviveka and other Indian scholars but on the works of Tibe

Page #33
REVIEWS 227 tan doxographers. His main sources are the Svatantrika chapter in the Grub-mtha' chen-mo by 'Jam-dbyans bzad-pa (1648-1721) and the Svatantrika chapter in the Grub-pa'i mtha'i rnam-par bzag-pa by iCan-skya rol-pa'i-rdo-rje (1717-1786). Both authors rely heavily on the works of Tson-kha-pa (1357-1419). Lopez remarks that in his work he attempts "to present ... some of the major concerns of the Svatantrika school as understood by the Ge-luk doxographers" (p.33). He is aware of the dangers of his enterprise and writes that "to conclude that the image of Svatantrika created by the Ge-luk doxographers is complete, obviating the independent study of their Indian sources, would be a serious error" (p. 31). Lopez considers his work as a preliminary study, an attempt to outline the Ge-luk tradition of interpretation of Svatantrika, a grid through which the works of the Indian masters may usefully be viewed" (p. 33). The first part of Lopez's work is a systematic study of the Svatantrika system as described by 'Jam-dbyans bzad-pa and ICan-skya rol-pa'i-rdo-rje with frequent references to Tson-kha-pa's works. The notes give exact page references. Apart from written works Lopez refers also to oral commentaries by Tibetan scholars. Lopez does not explain how far these scholars have studied the works of Bhavaviveka, etc. He points out that dGe-lugs-pa scholars would obtain their knowledge by studying the Grub-mtha's of 'Jam dbyans-bzad-pa and ICan-skya rol-pa'i-rdo-rje and textbooks on the structure of the path to enlightenment based on the Yogacara-Svatantrika and Haribhadra's commentaries on Maitreya's Abhisamayalamkara. Lopez adds that it would be a rare scholar who had devoted extensive effort to the study of the original Svatantrika texts in Tibetan translation (p. 23). What one finds in Lopez's work is an excellent description of some of the main aspects of the Svatantrika system as can be obtained from contemporary Tibetan scholars and from the sources they used. The second part of Lopez's work comprises a translation of the Svatantrika chapter in ICan-skya rol-pa'i-rdo-rje's Grub-pa'i mtha'i rnam-par bzagpa. Lopez has used an edition published in Sarnath in 1970. Another edition was edited by Lokesh Chandra in 1977. In this edition the Svatantrika chapter is found on pages 419-659. The translation is faithful and clear though one would have liked to have more explicatory notes. The text is not always easy to understand and references to the first part of Lopez's work would have been very helpful. It is a pity that Lopez seems to have felt obliged to follow Hopkins' example writing Tibetan names and titles in phonetic transcription and in translating Arhat by "Foe-Destroyer", paraprasiddha by "renowned to others", gnod-byed by "damaging", dal-'byor by "leisure and fortune", etc.

Page #34
228 REVIEWS His knowledge of Sanskrit is not very reliable as can be seen from such spellings as Ghandapada (p. 251), Khangavinanakalpa (p. 339), Paramapitsbuddha (p. 263; Tib. dam-pa sars-rgyas), etc. Lopez does not seem to have consulted Sanskrit texts. For instance, ICan-skya rol-pa'i-rdorje often quotes the Lankavatara but no reference is made to the Sanskrit text. In one instance, Lopez points out that the Tibetan translation of stanzas from it cited by ICan-skya rol-pa'i-rdo-rje and by Santaraksita differs from that found in the Peking edition (chapter 13, note 7), but does not quote the Sanskrit text which has: cittamatram samaruhya bahyam artham na kalpayet / tathatalambane sthitva cittamatram atikramet // cittamatram atikramya nirabhasam atikramet/ nirabhasasthito yogi mahayanam na pasyet // (ed. Bunyiu Nanjio, pp. 298-299). Lopez is very inconsistent with regard to the identification of quotations, the sources of which are indicated in some cases but not in others. Sometimes an identification is too vague to be useful. On p. 291 ICan-skya rolpa'i-rdo-rje quotes Maitreya. Lopez adds that this quotation comes from the Mahayanasutralamkara but without giving a precise reference. I have been unable to trace this quotation in the Mahayanasutralamkara. It will be a task for the future to confront the opinions of Tson-kha-pa and later Tibetan scholars with the original Svatantrika texts. It is to be hoped that Eckel's translations of the eighteenth and the twenty-fourth chapters of Bhavaviveka's Prajnapradipa will soon be published. A complete English translation of Buddhapalita by Akira Saito will soon be published by Motilal Banarsidass. A complete translation of Bhavaviveka's commentary is an urgent desideratum. NOTES Cf. Katsumi Mimaki, Blo gsal Grub mtha' (Kyoto, 1982), p. 45. 2 Op. cit., p. 40. 3 Kajiyama, Vuichi, 'Bhavaviveka's Prajnapradipah', WZKSOA 7 (1963), pp. 37-62; 8 (1963), pp. 100-130 (chapter 1); M. David Eckel, 'Bhavaviveka's Critique of Yogacara Philosophy in Chapter XXV of the Prajnapradipa', Indiske Studier 5: Miscellanea Buddhica (Copenhagen, 1985), pp. 25-75. 4 Cf. V. V. Gokhale, The Vedanta-Philosophy described by Bhavya in his Madhyamakahrdaya', NJ 2 (1958), pp. 165-180; V. V. Gokhale and S. S. Bahulhar, 'Madhyamakahrdayakarika Tarkajvala, Chapter I', Indiske Studier 5: Miscellanea Buddhica, pp. 76-108; Ejima, Yasunori, Chugan shiso no tenkai. Bhavaviveka kenkyu, Tokyo, 1980; Lopez, pp. 437-8.

Page #35
REVIEWS 229 5 Lokesh Chandra (ed.), Buddhist Philosophical Systems of iCan-skya rol-pa'i rdo-rje (Satapitaka, vol. 233). New Delhi, 1977. Australian National University J. W. DE JONG Jeffrey Hopkins, Emptiness Yoga. The Middle Way Consequence School. Ithaca, New York USA, Snow Lion Publications, 1987. 510 pp. $35.00. Can-skya rol-pa'i-rdo-rje's Grub-mtha'i rnam-par beag-pa consists of four chapters, the first dealing with the tenets of nine non-Buddhist schools, the second with the two Hinayana schools, Vaibhasika and Sautrantika, the third with the Cittamatra school and the fourth with the Madhyamaka schools. The fourth chapter is by far the longest and occupies folios 419- 873 in Lokesh Chandra's edition. This chapter is divided into two sections, dealing with the Svatantrika school (ff. 419-659) and the Prasangika school (ff. 661-873). Hopkins's book comprises a translation of the first part of the section on the Prasangika school (ff. 661-739). According to note 128 the other parts of this section will be translated by Guy M. Newland, Jules B. Levinson and S. Brian Daley. The section on the Sautrantika school has been translated by Anne C. Klein. We may expect that in due course ICan-skya rol-pa'i-rdo-rje's entire work will be available in translation. Hopkins's book comprises a biography of ICan-skya rol-pa'i-rdo-rje (pp. 15-35), a detailed commentary on ICan-skya rol-pa'i-rdo-rje's text which includes a complete translation of the text (pp. 36-354), the translation without commentary (pp. 355--428), bibliography (pp. 431-447), notes (pp. 448-485) and index (pp. 486-510). Hopkins introduces his commentary with the following words: "I will give explanation as gleaned from oral and written traditions of Tibetan Buddhism and, occasionally, from my own attempts over the last twenty-three years to probe and to internalize the meaning" (p. 35). Hopkins's explanations of Indian Buddhist philosophical problems are based' entirely upon the Tibetan Buddhist tradition as represented by the dGe-lugs-pa school. As such his commentary is certainly very useful. As to Hopkins's method of translating it is superfluous to repeat remarks made in a review of his Meditation on Emptiness. We still find grags-pa translated as "renowned". For instance: "that what is renowned in the world ('jig-rten-gyi grags-pa) is not refuted by reasoning" (p. 408). However, in some places grags-pa is rendered by "approved", cf. p. 360: "approved by that very other partly"; p. 361: "approved by the other Indo-Iranian Journal 32: 1989.

Page #36
230 REVIEWS party". It is sometimes necessary to consult the Tibetan text in order to understand Hopkins's renderings. On p. 417 "the definitive great scholar" renders Tibetan res-pa don-gyi mkhas-grub "the Perfect one who knows the definitive meaning". However, such mistranslations are rare. Hopkins has an excellent knowledge of Tibetan Buddhist texts and his works could be so much more readable if he would be willing to give up some of his idiosyncratic renderings. NOTES 1 Cf. Donald S. Lopez, A Study of Svatantrika (Ithaca, 1987), p. 229. 2 Mind and Liberation: The Sautrantika Tenet System in Tibet. Ann Arbor: University Microfilms 1981 (cf. Hopkins, p. 455). 3 JIABS, vol. 9, no. 1 (1986), pp. 124-128. Australian National University J. W. DE JONG Dieter Michael Back, Rig pa no sprod gcer mthon ran grol. Die Erkenntnislehre des Bar do thos-grol (Freiburger Beitrage zur Indologie, Band 18). Wiesbaden, Otto Harrassowitz, 1987. IX, 142 pp. DM 58, . Back's book comprises a critical edition and translation of a text which belongs to the corpus of the Bar do thos-grol (BTG) literature. In his introduction Back studies the composition of different collections of BTG texts. According to Back the original collection consisted of eight texts (p. 9). Other texts were added later and a collection of 17 texts was formed (p. 10). Larger collections of texts contain commentaries on these texts or texts which share the same world of ideas (p. 10). In the English summary Back distinguishes an "edition for use" which comprises the texts necessary for the ritual of death and the guidance of a dead person to salvation or a good rebirth and a "library edition" which adds texts which deal with the ideas and doctrines of the BTG (p. 135). Only a detailed study of the texts in the different collections will enable us to clarify the history of their composition. According to Back the Rig pa no sprod gcer mthon ran grol reflects the fundamental ideas of the Vijnanavada school. Although the text contains idealistic theories it is not possible to detect any definite influence of the doctrines of the Vijnanavada school. Back's edition of the text indicates the variant readings found in different editions. The edition and text are Indo-Iranian Journal 32: 1989.

Page #37
REVIEWS 231 followed by a commentary. The text was first translated by W. Y. EvansWentz with the help of two Tibetan scholars: Lama Karma Sundhon Paul and Lama Lobzang Mingyur Dorje. In his notes Back criticises EvansWentz's translation. This translation is often very free and not devoid of imperfections which, perhaps, are due rather to Evans-Wentz than to his Tibetan informants. The text is written in verses of nine syllables in a kind of telegraphic style omitting particles, etc. This seems to have caused difficulties to Back which he could have avoided by paying more attention to Evans-Wentz's work. For instance, p. 66, 7 the Tibetan text reads: dros po gsis kyi gnas lugs ma mthon bas. Back translates: 'Weil man die Dinge nicht an ihrem naturlichen Ort sieht." Evans-Wentz has: "Unless one knows or sees the natural state of substances for things)" (p. 229). It is obvious that Evans-Wentz's translation is the correct one. Back mistranslates gnas-lugs, although he renders dros po'i gnas lugs (p. 46, 13) correctly by "die Existenzweise der Dinge". In quite a few places it is difficult to accept Back's translations. He has made a meritorious effort to render the text as closely as possible but without sufficiently taking into account the syntactical relations of the words. It is rather surprising to see that Back mistranslates some well-known expressions. P. 38, 7: 'khor 'das yors la khyab pa'i sems gcig po. Back translates: "Obwohl das einzigartige Denken, das den Samsara ubersteigt und ihn) vollstandig ausfullt". Evans-Wentz has: "All hail to the One Mind that embraces the whole Sangsara and Nirvana" (p. 203). In other places Back always translates 'khor-'das correctly. He does not know that mtshan and dpe-byad designate the laksana-s and anuvyanjana-s. P. 134, 1: lons sku mtshan dan dpe byad lhan pa'i brda. Back translates: "ein Zeichen voller Schonheit und Zeugnis fur den Sambhogakaya". P. 40, 19: gzun 'dzin 'object (grahya) and subject (grahaka)'. Back translates it by "Zielvorstellungen". His translation of ma-mthon by "Blinde" is also strange: p. 74, 7: ma mthon rgyan gi gtam rgyud bsnad pa bzin. Back translates: "So gleichen sie doch Blinder, die eine lange Uberlieferung erklaren." EvansWentz has: "like harkening to flying rumours concerning a distant place one has never visited" (p. 237). Sometimes Back translates a reading which is found in a footnote and not in the text. For instance, p. 48, 23: rtag pa. Back translates rtog pa. The edition is not free from misprints. P. 38, 15: gyal bas; read rgyal bas. P. 40, 12: pyi rol; read phyi rol. P. 72, 10: pyogs su; read phyogs su. P. 131, line 18 from below: ma gton pa; read ma gtogs pa. P. 132, No. 2, line 7: tshul la yin pa'i; read thsul ma yin pa'i. P. 133, line 4 from below: rdo rje semd; read rdo rje sems.

Page #38
232 REVIEWS It is to be hoped that Back will continue to edit texts belonging to the BTG corpus. Very welcome would be critical editions of such basic texts as the Chos nid bar do and Srid pa bar do which have been translated several times already. A critical comparison of these translations could also be very useful. NOTE The Tibetan Book of the Great Liberation (Oxford, 1954), pp. 202--240. Australian National University J. W. DE JONG Rene de Berval (ed.), Presence du Bouddhisme. Paris, Gallimard, 1987. 816 pp., 151 illustrations. 360 FF tc. En 1959 Rene de Berval publia a Saigon un numero special de la revue France-Asie. Ce volume intitule Presence du Bouddhisme comptait environ 885 pages (non 1025 pages comme le dit l'editeur de cette nouvelle edition). L'intention de Rene de Berval etait de presenter le bilan de 2.500 ans d'action et de culture bouddhiques. Le volume contenait une soixantaine de contributions de valeur tres variee, la plupart ecrites en francais, d'autres en anglais. La reedition se distingue de l'edition originale par l'omission d'un grand nombre d'articles consideres, d'apres l'avertissement de l'editeur, comme perimes et depasses. Les articles en anglais sont traduits en francais. On n'y trouvera que trois articles ecrits specialement pour la reedition: Jean Perrin: Le Bouddhisme en Birmanie (pp. 469-487); Louis Gabaude: Le Bouddhisme en Thailande (pp. 489-514) Rhi Ki-yong: Le Bouddhisme en Coree (pp. 591-614). Tres peu d'articles ont ete revus par leur auteur. Par exemple, dans l'edition originale, Pierre-Bernard Lafont traitait de l'introduction du bouddhisme au Laos en quatre pages (pp. 889-892). Dans la reedition son article sur le bouddhisme au Laos occupe huit pages et donne beaucoup de renseignements neufs (pp. 515-533; onze pages contiennent des illustrations). Par contre, l'article de Phouvong Phimmasone sur la litterature bouddhique lao a ete omis completement. L'article que Gaston Renondeau avait contribue sur le bouddhisme au Japon pour la premiere edition avait ete remanie par celui-ci, et, apres sa mort, revu et augmente par Bernard Frank. C'est dommage que d'autres articles n'aient pas ete remanies et revus de la meme facon. Indo-Iranian Journal 32: 1989.

Page #39
REVIEWS 233 La premiere edition contenait plusieurs contributions qui ne meritaient certainement pas d'etre reimprimees. Par exemple, l'article de Valisinha sur le bouddhisme en Inde est tres maigre et aurait du etre remplace. On peut faire la meme remarque a propos de l'article de Migot sur le bouddhisme en Chine. L'article est insuffisant en ce qui concerne l'histoire du bouddhisme en Chine, mais ce qui frappe surtout le lecteur en 1988 sont les remarques sur l'aide materielle apportee par le gouvernement pour l'entretien et la restauration des sanctuaires bouddhiques. Depuis que Migot visita la Chine en 1957 beaucoup d'eau a coule dans le Fleuve Jaune. Combien de temples et de statues n'ont pas ete detruits pendant la revolution culturelle! Le texte de la couverture dit que la reedition, mise a jour et renouvelee dans ses textes et son illustration, est appelee a demeurer le classique de reference. On ne peut guere souscrire a cette declaration. Une complete mise a jour aurait pu faire de la Presence du Bouddhisme un livre de valeur sinon le classique de reference. Il faut apprecier l'effort fait pour renouveler * l'illustration et pour completer la bibliographie, mais tout cela n'empeche pas que cette reedition est loin de representer la somme d'ensemble sur tous les pays bouddhiques dans leurs traditions et leur realite actuelle mentionnee par le texte de la couverture. La bibliographie detaillee qui occupe presque soixante-dix pages est la bienvenue car depuis la disparition de la Bibliographie bouddhique il est difficile de se renseigner sur les publications parues depuis 1958. Les compilateurs de la bibliographie ont porte une attention speciale aux ouvrages parus en francais durant les dernieres vint-cinq annees. Il y a plusieurs erreurs a corriger. La plus grave se trouve a la page 757 ou plusieurs ouvrages ecrits par A. Foucher ont ete attribues a J. Fontein: Etudes sur l'art bouddhique de l'Inde, etc. L'article de Durt sur "The Counting Stick" (p. 754) n'est pas paru dans le volume 1 des Indian and Buddhist Studies mais dans le volume 23 du Journal of Indian and Buddhist Studies, pp. 470(28)-- 646(34). L'article de Schopen sur le culte du livre dans le Mahayana se trouve dans le volume 17 de l'Indo-Iranian Journal (p. 754). C'est Caroline Rhys Davids et non T. W. Rhys Davids qui a ecrit Gotama, the Man, etc. (p. 733). L'auteur de l'Essai sur la legende du Buddha est E. Senart et non E. Senart (p. 733). Note (non Notes) on the Kashgar Manuscript of the Saddharmapundarikasutra de Toda fut publie en 1977 et non en 1907 (p. 742). Il faut corriger Samahinirmocanasutra en Samdhinirmocanasutra (p. 739). Australian National University J. W. DE JONG

Page #40
234 REVIEWS Carol Meadows, Arya-Sura's Compendium of the Perfections: Text, translation and analysis of the Paramitasamasa (Indica et Tibetica 8). Bonn, Indica et Tibetica Verlag, 1986. X, 371 pp. DM 72.00. In 1946 Alfonsa Ferrari (1918-1954) published the text and an Italian translation of the Paramitasamasa.' Her edition is based upon a copy of a manuscript in the library of the Maharaja of Nepal. The colophons of both the Sanskrit manuscript and Vairocana's Tibetan translation attribute the work to Sura or Aryasura. Ferrari accepted this attribution on account of similarities in topic, style and vocabulary between the Jatakamala and the Paramitasamasa. She also pointed out that a number of verses of padas are either identical or almost identical in both texts (p. 101). Carol Meadows refers to Ferrari's list of correspondences and adds some corrections and additions (p. 121). Both Ferrari and Meadows mention that PS (Paramitasamasa) 1.9 JM (Jatakamala) VIII. 43. However, the only word they share is krtaghne! Carol Meadows writes that the second pada of PS 1.48 the third pada of JM V. 15, but they have in common the words dane nidane only. In the first chapter of the introduction Carol Meadows compares in detail the PS and the JM and points out many differences in the treatment of the paramitas; she arrives at the following conclusion: "My investigation has uncovered no positive link between the two works strong enough to decide the issue in favor of common authorship" (p. 19). She supposes that the PS was probably written around the 7th or 8th century A.D. Meadow's edition of the PS is based upon a manuscript written in Newari which was originally in the collection of the Rajguru, Hemraj Pandit, and at present in the National Archives of Nepal. Carol Meadows examines the relationship between this manuscript (K) and the copy used by Ferrari (F co.) and carefully collects the evidence for and against F co. having been copied from K. She concludes: "on the basis of the evidence I have in hand I believe K was the ms. from which Ferrari's copy was made" (p. 31). There is no doubt that both manuscripts are closely related to each other but this is not sufficient to suppose that F co. is copied directly from K. The text of K is definitely superior to that of F co. According to Carol Meadows her edition differs in 150 instances from the text of Ferrari's edition (F); 115 changes have been made on the basis of K and in the remaining 35 cases Carol Meadows has adopted readings found neither in F nor in K. Carol Meadows lists all the 88 cases in which her edition (CM) differs from F and F co. (pp. 32-34). These 88 cases include such variants as samgacittam / sangacittam and tvakpatra- / tvakpattra- and a number of Indo-Iranian Journal 32: 1989.

Page #41
REVIEWS 235 obvious misprints in F. In a further 19 cases K and Fco. have different readings and the emendations in F are not supported by K which is followed in CM. In 8 cases K and Fco. have the same readings. F offers emendations which are rejected in CM in favour of K. It is a pity that Carol Meadows does not list the remaining 35 cases in which she rejects the readings found in K. Carol Meadows remarks that in 166 cases Ferrari emended a faulty reading in Fco. to one which K supports, and which she feels is the correct reading. She adds: "this speaks very well for the quality of Ferrari's work on the PS with the material that was available to her" (p. 36). Ferrari remarked that Vairocanaraksita in his Tibetan translation of PS often renders the general meaning of a stanza and sometimes rearranges padas. Carol Meadows does not say anything about the Tibetan translation apart from remarking that "All substantive emendations (those affecting meaning) were made by me on the basis of the Tibetan" (p. 36). Undoubtedly, the text as edited by Carol Meadows is almost everywhere to be preferred to the text of Ferrari's edition. However, there are quite a few stanzas where the text needs further emendation. On account of the unreliability of the Tibetan translation it is not easy to propose emendations and in many instances one can only point to the difficulties in the text. On some other occasion I hope to deal with some textual problems which would take up too much space in a review. The edition of the Sanskrit text is preceded by a preliminary note on the system of annotation used by the editor (pp. 151-155). One must object strongly to the way Carol Meadows has established the critical apparatus. In the first place, the abbreviation Fco. for the copy used by Ferrari is rather clumsy. Why not use a single letter such as C? Carol Meadows states that since she takes K as her basic text, the reading given is that found in K unless otherwise indicated in the notes. So far, so good. However, in 166 cases Fco. has a reading which Ferrari has emended and the emendation is supported by K. In these cases Carol Meadows puts in her note the reading of Fco. but omits to mention that K confirms Ferrari's emendation. In another 45 cases both K and Fco. have the same reading but emended by Ferrari and this emendation is accepted by Carol Meadows. In all these cases Carol Meadows mentions in the notes only the readings of K. The absurdity of this system of annotation becomes even more obvious from the fact that she is forced to adopt a category of readings which she calls presumed readings of the copy used by Ferrari and indicated by (F co.). For instance in 1.4d K and CM read sambodhimargavaranam. The note has: "1.4d sambodhimargacaranam F, (F co.)" (p. 153). Carol Meadows adds

Page #42
236 REVIEWS that if the reference to Fco. were left out, the reader would assume that it was in agreement with K according to her system of presentation! The emendations which Carol Meadows proposes are not always acceptable. For instance in 2.9d K and Fco. read etan prayatnena: samasatah silam idam vadanti yah samvarah kayavacomanastah / kartsnyena catraiva yatah sa tasmad etan prayatnena visodhayec ca // Carol Meadows reads etany ayatnena and translates: "In short, they call this morality (sila) restraint in body, speech, and mind; therefore, fixed in just this (morality) in its entirety, he would also effortlessly bring about the purification of these [the body, speech and mind)." In pada c sa refers to samvara: "because it (i.e. restraint] resides entirely in them, one must purify them [body, speech, and mind) with effort." The Tibetan translation of cd is quite different: tshul-khrims ma-lus 'dul-ba'i gai yin-pas / de-phyir 'di-dag rnam-par sbyan-bar gyis, "morality is the basis of the discipline; therefore one must purify them." In 4.14b Ferrari emended phalam svamuni to phalantvamuni which is confirmed by K. Without any justification Carol Meadows changes this to phalanty amuni: tatha hi karunyavisuddhabuddhih sarvajnabhavaya phalanty amuni/ punyani lokasya caracasyety evam sa tany arabhate susattvah // In two instances Carol Meadows separates words connected in Ferrari's edition without mentioning this in the notes. The first case is 3.16: loko 'yam atmabhinivesamudhah (not -nivesasamudhah!) sesan paran ity abhimanyamanah / tad viprakarair abhibhutaceta (h) ksamaviyogat parikhedam eti // Carol Meadows translates tad by 'therefore'. One must read tadviprakarair, i.e. tesam viprakarair, cf. Ferrari: "con l'animo sopraffatto dalle loro ingiurie." In 6.43d Carol Meadows separates prajna and gunamatyasanathata: na yad virupam prakrtim vrajanti prajna gunamatyasanathata sa // One must read prajnagunamatya-: "this is due to the fact that they [i.e. the

Page #43
REVIEWS 237 bodhisattvas) have as ministers the virtues of insight, cf. Ferrari: "questo e (dovuto al fatto che) hanno come ministro la virtu della gnosi." In 1.75 K has the correct reading uddharatat but Carol Meadows keeps the reading of F co.: mamsarthino mamsam idam harantu majjanam apy uddharanat tadarthi / Carol Meadows's translation of pada b is completely unacceptable: "by drawing out even my marrow, let the one who wants it stake it]." If on the whole the edition of the text by Carol Meadows marks progress on Ferrari's edition thanks to the superior readings found in K, the same cannot be said of the translation which is full of elementary mistakes. Carol Meadows does not refer even once to Ferrari's translation which she does not seem to have consulted. In many instances Ferrari's translation is by far to be preferred. It is obvious that Carol Meadows was insufficiently qualified to undertake a translation of the PS. It is impossible and useless to point out every error made by Carol Meadows but the following instances may suffice. pitrnam api putresu putranam ca pitrsv api / pritisarvasvabhutesu suhrtsu suhrdam api // gunapracayabaddhasya vyudhesu samaresv api/ darsitasthairyasarasya snehasetor vidarinah // 5.22-23 Carol Meadows translates: "They break the bond of affection, which expresses the very essence of constancy, that fathers feel for their sons and sons for their fathers, that friends feel for friends with whom they share all joys, and that one is bound by a mass of virtues feels even for armies arrayed in battle." Carol Meadows completely misunderstood verse 23: "They (sensual enjoyments) destroy the dam of affection which is formed by the mass of virtues and which has shown the strength of its firmness even in the case of armies arrayed for battle." Carol Meadows translates setu by "bond', although the dictionaries give this meaning only for the Rgveda. In a note on p. 206 she reproaches Ferrari for stating that abhyasanti is a 4th Class verb here following the conjugation of the 1st Class by referring to the 1st Class conjugation noted for as- by Whitney in his Roots, etc. (p. 5). Whitney here notes one case in the Rgveda of the 1st Class conjugation for as-! Carol Meadows is very uncritical in her use of dictionaries. For instance in 1.60 she translates dainya as 'covetousness', a meaning taken from Wilson's dictionary by Monier-Williams. Another wrong use of the dictionary is found in her translation of tadrk in 2.20:

Page #44
238 REVIEWS divaukasam ca priyatam yad eti satyapriyas citram idam na tadrk / "It is not so strange that a lover of truth becomes beloved by the gods." Carol Meadows translates: "And that such a lover of truth." In a note she remarks that tadrs connotes a holy person (p. 281). In another note relating to tadrgvidhajnana she again refers to tadrs "holy, religious' (p. 302). Let me conclude by drawing attention to a number of mistranslations in the first chapter. In 1.21 lajjamayasyaiva ca bhusanasya' is translated as a modest ornamentation'. In 1.36 it is said that the Buddha does no harm to beings while giving: dadati notpidanaya parasya. Carol Meadows translates: "That one (the bodhisattval does not give ... under pressure from another." In a note on p. 273 she translates nasti sattvotpidanadanam as "There is no giving under pressure from sentient beings." In 1.42 the text says that the bodhisattva does not look at the face of others while giving: nasau mukhollokanaya dadati, Carol Meadows seems always to translate instrumentals in the same way and renders this as follows: "He does not give because he is looked up to (with admiration." In 2.54 the text has 'parananollokanakatarah syat, "he would be afraid to look at the face of others." Carol Meadows translates: "He would be reluctant to have the faces of others looking up at him sin admiration)." 3 In 1.50 Carol Meadows completely mistranslates the word sadharana: sadharanam raksyam atarpakam ca: "[possessions are common to all, must be guarded and do not give satisfaction". Carol Meadows renders sadharanam by 'ordinary possessions'. This use of sadharana is well known from the Jatakamala, cf. V.15ab: vidyullatanrttacale dhane ca sadharane . naikavighatahetau. See also Saundarananda V.23 and Buddhacarita XI.26 with Johnston's note (here sadharana is used of the passions). In 1.53 Carol Meadows misunderstood padas c and d: prayo viyogo hi parigrahebhyo / danad bhavaty abhyudayo yasas ca // "For, as a general rule, one is separated from one's possessions, but from giving proceed well-being and glory." Carol Meadows translates: "For, as a general rule, separation from property through giving turns into well-being and glory." In 1.54 Carol Meadows translates klesaparigraha "possession of the defilements' by 'possession by the defilements'. Carol Meadows must have spent many years in studying the Paramitasamasa. In her preface she writes that the translation and analysis presented in this book were done as part of her 1976 Ph.D. dissertation for Columbia University. Both were subsequently changed and modified on the basis of the new edition prepared from the Kathmandu manuscript. Undoubtedly,

Page #45
REVIEWS 239 she has brought together valuable materials in her extensive analysis but it is a great pity that no competent scholar seems to have read her translation and to have advised her against publishing it in its present form. NOTES I 'Il "Compendio delle Perfezioni" di Aryasura', Annali Lateranensi X (1946), pp. 9-102. 2 Is there any evidence showing that manuscripts from the library of the Maharaja were later transferred to the collection of the Rajguru? 3 ullokana is not mentioned in Edgerton's dictionary. It is also not mentioned by Willem B. Bollee, Pali mukham ulloketi (oloketi)', KZ 83 (1969), pp. 243-255. It would be useful to compile an index verborum of Buddhist Hybrid Sanskrit words in the Paramitasamasa. Australian National University J. W. DE JONG Richard Gombrich, Theravada Buddhism. A Social History from Ancient Benares. to Modern Colombo. London and New York, Routledge & Kegan Paul, 1988. X, 237 pp. PS7.95. Gombrich's book on Theravada Buddhism is one of the most original and inspiring works on Buddhism published in recent years. He has a first-hand knowledge of Buddhism in Sri Lanka and knows well the history of Theravada Buddhism both in India and in Ceylon. More important, however, is the fact that Gombrich's work is not one of those many introductions to Buddhism which are content with repeating well-known facts. Gombrich concentrates on the problems which he considers important and does not hesitate to state clearly his personal opinion. His own position regarding the problems involved in writing a social history of Buddhism is set out in the introduction. Gombrich advises readers not interested in theoretical issues to skip this chapter. However, this chapter is of essential importance for the better understanding of Gombrich's approach to the social history of Buddhism. The introduction deals with many questions such as the limitations of Marxist and Weberian views of religion, Popper's views on determinism, the unintended consequences of our actions and the conspiracy theory of society. As to the reliability of the Pali sources Gombrich believes that 'the main edifice', i.e. the content of the main body of the four Nikayas and of the main body of monastic rules is the work of one genius. He concludes that we should provisionally accept tradition till we have something to put in its place. However, he does not explain how it would be possible to replace tradition by something else. Gombrich says Indo-Iranian Journal 32: 1989.

Page #46
240 REVIEWS that it should not be our only aim to try to restore original texts and to establish the original meanings intended by their authors and that modern scholarship should begin by examining the tradition itself. However, when one tries to know what the Buddha himself taught, it is not possible to maintain this position and Gombrich repeatedly speaks of developments which must have taken place after the Buddha's lifetime and which were not intended by him (cf. pp. 70, 109 and 120). One of the most interesting sections of Gombrich's introduction is the one which deals with Buddhist identity (pp. 23-31). Gombrich distinguishes between two kinds of religion, one being soteriology adherence to which is defined by assent to its doctrine and entry into the membership of which is formalized by a declaration of faith, the other communal religion which is primarily a pattern of action providing rules for the ordering of society. He explains the different implications for Buddhist identity in ancient India, on the one hand, and in Sri Lanka and other Theravada countries, on the other hand. The second chapter 'Gotama Buddha's problem situation' deals with A. Vedic civilization and B. The social conditions of his day. Gombrich remarks that the Buddha spent much of his time in cities, whereas the Vedic civilization depended on an entirely rural, village-based society, He writes: "Thus the Buddha's period saw not only urbanization, but the beginnings of what one might call states. The period also saw the first use of money and the beginnings of organized trade" (p. 51). "Thus there arose two major new professions, both presumably urban-based: state officials and traders" (p. 55). Gombrich remarks that there is some evidence that the Buddha's message appealed especially to town-dwellers and the new social classes. and refers to Gokhale's analysis of the social composition on the basis of the commentary on the Thera- and Theri-gatha.' The commentary dates from the fifth century A.D., but according to Gombrich it rests on a far older tradition and its information may be authentic.2 Akanuma Chizen has used the same sources besides many others in his study of the four samghas (monks, nuns, male and female lay followers) published in 1928.3 Akanuma's results are not substantially different from those obtained by Mrs. Rhys Davids and B. G. Gokhale. It is difficult to know to what extent the figures obtained reflect the social composition of the samgha but it is perhaps not surprising that Buddhism appealed to an urban elite. In the third chapter Gombrich discusses the Buddha's Dharma and stresses the importance of the ethicization of the doctrine of karman. Gombrich remarks: "Since ethical value lies in intention, the individual is autonomous

Page #47
REVIEWS 241 and the final authority is what we could call his conscience" (p. 68). It is obvious that such a doctrine is not addressed to the "downtrodden" who are more likely to be converted to a religion in which bhakti is the main element. Gombrich points out that the Buddha's recorded sermons to the laity deal mainly with morality and he explains why the Buddhist ethics appealed to businessmen. Particularly interesting are the two sections of chapter three, entitled "Buddhism as religious individualism" and "An ethic for the socially mobile". I believe that Gombrich has given many good reasons for explaining the success of Buddhism among the wealthier and more urban sectors of society. In this respect one must also stress the fact that the duties of the lay followers were not very onerous. It is, however, much more difficult to understand why members from the urban elite should abandon everything in order to strive for salvation. The final section of chapter two puts the question to whom did the Buddha's message appeal. Gombrich remarks that the view that life is suffering had not been current in India before Buddhism and that it is his task to explain why it became acceptable when it did (p. 59). I have not been able to find a convincing answer to this question in his book. Urbanization does not necessarily lead to spiritual malaise as Professor Ghosh seems to believe. Gombrich adduces as a possible contributory cause the connection between cities and morbidity but the Buddhist texts do not give any indication of an increase in morbidity at the time of the Buddha. Buddhism may well have been an attractive religion for urban lay followers but why members of the urban elite came to the realization that life was suffering and decided that for them there remained no other solution than to become monks remains a problem which probably cannot be explained due to lack of sufficient source materials. Gombrich is one of the few Western scholars who have an intimate knowledge of Buddhism in Sri Lanka in the past and the present. Three chapters deal in detail with the history of Sinhalese Buddhism, the rise of "Protestant" Buddhism in the nineteenth century and with contemporary trends and problems. It must be left to specialists to discuss details of interpretation but there is no doubt that these chapters make extremely stimulating reading. Let me end with a few minor points. On p.-59 Gombrich refers to Frauwallner's hypothesis that the very format of the first sermon and its Four Noble Truths follows a medical model. This idea was first suggested by Hendrik Kern. Recently A. Wezler has discussed this question in great detail. He arrives at the conclusion that there is not the slightest evidence for the assumption that this fourfold division of the science of medicine inspired the Buddha to his Four Noble Truths.4

Page #48
242 REVIEWS On p. 71 Gombrich states that the Canon was first written down in the first century BCE. This statement is based upon the Dipavamsa and the Mahavamsa. In 1928 Friedrich Weller tried to prove that the verses which mention the writing down of the Canon are a later interpolation. As far as I know, nobody has tried to refute Weller's opinion. On p. 74 Gombrich states that the etymological meaning of upanisad is 'sitting down by'. It is strange to see that Gombrich seems to be unaware of the different explanations of the meaning of upanisad by Schayer and Renou. On p. 109 Gombrich mentions the rule according to which laymen were not to be present at the uposatha ceremony. He adds: "this suggests to me that originally they were and it was felt to be embarrassing for them to hear the confessions." It seems to me that the reason for this rule is rather to be sought in the feeling that the community of monks formed a separate entity. This feeling must have arisen when the community of monks became formally established. NOTES IB. G. Gokhale. "The Early Buddhist Elite', Journal of Indian History 43 (1965), pp. 391402; 'Early Buddhism and the Brahmanas', Studies in the History of Buddhism (Delhi, 1980), pp. 67-80. Gokhale does not mention that Mrs. Rhys Davids had already used the commentary in order to analyse the social origin of the monks in the Theragatha, cf. Psalms of the Brethern (London, 1913), p. xxviii. 2 Bechert is less inclined to accept the information given by the commentary, cf. Buddhismus, Staat und Gesellschaft in den Landern des Theravada-Buddhismus, Dritter Band (Wiesbaden, 1973), p. 130. 3 Akanuma's article is reprinted in his Genshi bukkyo no kenkyu (Nagoya, 1939; Kyoto, 1981), pp. 383-430: Shakuson no shishu ni tsuite. 4 'On the Quadruple Division of the Yogasastra, the Caturvyuhatva of the Cikitsasastra and the "Four Noble Truths" of the Buddha', Indologica Taurinensia XII (1984), p. 323. 5 'Die Uberlieferung des alteren buddhistischen Schrifttums', Asia Major 5 (1928), p. 164 - Kleine Schriften L(Stuttgart, 1987), p. 136): "Nach den Zeugnissen beider Chroniken haben wir also in dem Doppelverse uber die schriftliche Abfassung des Tripitaka samt der Auslegung dazu einen jener haufigen Falle vor uns, dass ein Verspaar ursprunglich am Rande eines fertigen Manuskriptes nachgetragen wurde, das nun von einem Benutzer dieser Handschrift an der einen, vom anderen dagegen an der anderen Stelle eingeschoben wurde." 6 St. Schayer, "Uber die Bedeutung des Wortes upanisad', RO 3 (1927), pp. 57-67; L. Renou,"Connexion" en vedique, "Cause" en bouddhique', Dr. C. Kunhan Raja Presentation Volume (Madras, 1946), pp. 1-5 - L'Inde fondamentale (Paris, 1978), pp. 149-153); L. Renou, L'Inde classique I (Paris, 1947), p. 299. See also Gunilla Gren-Eklund, 'Causality and the method of connecting concepts in the Upanisads', Indologica Taurinensia XII (1984), p. 117; Harry Falk, 'Vedisch upanisad', ZDMG 136 (1986), pp. 80-97. Australian National University J. W. DE JONG

Page #49
REVIEWS 243 Jens-Uwe Hartmann, Das Varnarhavarnastotra des Matrceta. Herausgegeben und ubersetzt (Abh. d. Ak. d. Wiss. in Gottingen, Phil.-hist. Kl. Dritte Folge Nr. 160; Sanskrittexte aus den Turfanfunden XII). Gottingen, Vandenhoeck & Ruprecht, 1987. 372 pp. DM 198, Fragments of the Varnarhavarnastotra were edited by Hoernle in 1916, Siegling in 1922, Shackleton Bailey in 1950, Pauly in 1964 and Couvreur in 1966. In 1968 Schlingloff published facsimiles of the Berlin fragments. Apart from these publications Hartmann has been able to make use of 71 newly discovered fragments in the collections in Berlin and London. According to Hartmann, Shackleton Bailey had at his disposal about 58% of the original text. The publications by Pauly and Couvreur and the newly discovered fragments make it possible to recover about 82% of the original text. Shackleton Bailey had tried to reconstruct the lost parts as far as possible on the basis of the Tibetan translation. Hartmann shows that often many different Sanskrit words are rendered by one and the same Tibetan word and that it is therefore difficult to reconstruct the original Sanskrit text. He is very critical of Shackleton Bailey's reconstructions. Shackleton Bailey remarked that "the supplements here put forward may always be taken as fairly representing Matnceta's meaning" (BSOAS XIII, p. 673), and it is certainly useful to see how an editor has understood the Tibetan translation even if his reconstructions can never be more than an educated guess. Hartmann writes that in 1964 Pauly made the Varnarhavarna fragments of the Pelliot collection accessible in transcription and facsimile. This is not quite correct. Pauly's article in the Journal asiatique (CCLII, 1964, pp. 197--271) was not accompanied by facsimiles. Hartmann refers to my article on the Varnarhavarnastotra (IIJ 10, 1967-8, pp. 181-183) and remarks that I had not consulted the facsimiles (cf. p. 45). However, the facsimiles were only added later by Pauly in 1965 in a separate publication of the fragments which was not available to me when I wrote my comments on Pauly's publication. In his introduction Hartmann discusses the sources for the biography of Matnceta without adding anything new to the information already given by Shackleton Bailey. In the second part of his introduction he lists the works attributed to Matncela, and quotations of the Varnarhavarna (VAV) and Prasadapratibodbhava (PPU). Hartmann has been able to collect no less than 27 quotations. He points out the importance of the quotations for determining Matnceta's date. Hartmann has discovered a long quotation in the Mahaprajnaparamitopadesa' (T 1509, p. 222c22ff.; Lamotte, Traite Indo-Iranian Journal 32: 1989.

Page #50
244 REVIEWS III, pp. 1378-1381). This quotation is written in prose and does not entirely agree with the wording of Matrceta's verses. The Upadesa usually seems to quote verses in the form of verses and it is therefore possible that both Matrceta's verses and the passage in the Upadesa are based upon another text. Verse 5.22 which states that the Buddha speaks with a single sound (ekasvara) is quoted both in the Upadesa and in the Mahavibhasa, but in the latter text this verse is quoted as a gatha. It is a pity that it is very difficult to determine the date of the Mahavibhasa. In his Histoire du bouddhisme indien (Louvain, 1958, p. 424) Lamotte placed it in the second century, but in volume III of the Traite (Louvain, 1970, p. XXI) he writes that all we know is that the Mahavibhasa was composed after the time of Kaniska who is mentioned in it. In the introduction (p. 52) and in his commentary on VAV 8.23 Hartmann writes that this verse does not prove that Matoceta was a follower of Mahayana. He disagrees with Shackleton Bailey who saw in this verse clear proof of it (BSOAS XIII, p. 947; The Satapancasatka of Matrceta, Cambridge 1951, p. 237). Hartmann remarks that this verse only shows that Matrceta knew the concept of Mahayana. However, it is difficult to imagine that somebody who does not adhere to Mahayana would use the expression mahayanayauvarajye as in this verse: abhisikta mahayanayauvarajye 'pare vare. In an article mentioned in Hartmann's bibliography NARA Yasuaki has studied the doctrinal position of Matrceta. As indications of his belonging to Mahayana he mentions apart from this verse the treatment of the six paramitas (PPU chapter two), the praise of sunyata as the chief of the tattvas (VAV 3.21) and its mention in VAV 2.37 and 5.28. Nara concedes that in other respects Matrceta's ideas are closer to those of Hinayana schools such as the Mahasamghikas as, for instance, in his references to Buddha's dharmakaya. Nara remarks that Matrceta was a poet and not a thinker and that he reflected the ideas of his epoch. He characterizes him as belonging to a Buddhastotra Vehicle.? Probably Matrceta belongs to an early stage of Mahayana. Hartmann's edition of the text is excellent. He is very careful in not trying to reconstruct missing padas and his additions do not go beyond one or two syllables. Only in very few cases is it possible to suggest another reading. In 7.1cd Hartmann reads: yam sarve nativartante prthaglokah sadevakah. In a note he remarks that Tib. separates prthag and lokah, but that here prthaglokah is used in the meaning of prthagjanah. I believe that it is necessary to separate prthag and lokah. The expression prthagloka is entirely unknown whereas the texts often mention the lokah sadevakah (see Hartmann's note ad 11.13). The text of 11.5 reads:

Page #51
REVIEWS 245 parakanksavighatartham ya mukhad abhinirhsta / saratsamdhyabhralekheva vaktracandravaghatini // Hartmann remarks that avaghatini is not found in the dictionaries but that ud-ghat "to open" occurs in the Petersburg dictionary. Probably avaghatini is a scribe's error for avagunthini due to the presence of vighata in pada a. In 12.10 avaghunthitah is rendered in Tibetan with khebs-pa which in 11.5 renders avaghatini. In 11.28 the subject which is not mentioned is Buddha's tongue (prabhuta): varnan alamkarotiva varnair mandayativa gam / adeyataratam prokta mukham gamayativa te // The Tibetan translation of cd reads: bsad pa mdzad na khyod kyi zal lsin tu sba bar mdzad pa bzin //. Hartmann is puzzled by prokta and suggests tentatively changing it to praudha. He translates bsad pa mdzad na as "wenn eine Erklarung gegeben wird" without explaining why mdzad-pa would have a passive meaning here. It is not likely that Matoceta would have used an incorrect form proktva. Read adeyataratam uktva? As to Tibetan sba this is probably an error for spa, cf. Hartmann's note ad 12.14. 12.13-14 describe the glory of the pravacana. The text reads (square brackets and parentheses in 13b and 13d have been omitted): nivrstta]man(d)a[p]e(yena) mlayamanagunaujasah / janitavranadosasya kalpantaramatantaraih // gata pravacanasyasya tvadrte vyustir anyatam / hrtanagasya sarasah srir ivaciranasini // Hartmann's reconstruction of 13ab is based upon the Tibetan translation: 'tsham pa'i snin po dan bral bas / yon tan gzi yan mi spa la / The meaning of 'tsham pa'i snin po is not clear but it seems here to translate mandapeya. Tibetan bral-bas, however, does not justify the reading -peyena and it is certainly preferable to read nivrttamandapeyasya in line with the genitives in padas b and c. Hartmann has edited the Tibetan translation on the basis of the Cone, Derge, Narthang and Peking Tanjurs. He points out that the Cone text is not based upon the Derge text (p. 45) as is the case for other texts. This shows that it is dangerous to make general statements on the relationships between the different recensions of the Tanjur after having compared the differences in a single text. In his notes Hartmann draws attention to differences between the Tibetan translation and the Sanskrit original and to

Page #52
246 REVIEWS problems in the interpretation of the Tibetan translation. In 2.4 he notes that the meaning of Tibetan mi sro ste is problematical. In PPU 39 bsros renders Sanskrit vyakhyatam and the Bod-rgya tshig-mdzod chen-mo (Beijing 1985) explains sro-ba with yid-kyis bgran-bser byed pa 'to compare in one's mind'. In 3.6 Tibetan dgra zin renders Sanskrit samasama. Hartmann remarks that Tibetan zin is not understandable but according to the Bodrgya tshig-mdzod chen-mo zin has the meaning gnen 'friend'. In 6.29 Tibetan lam mal than chad smra ba rnams renders Sanskrit ekamsavadinah. The meaning of lam mal than chad is obscure. Hartmann remarks that than chad is unknown, but Chos-grags explains than-chad-pa as meaning naldub-pa 'to be tired'. In 10.10 Tibetan ston ka'i chu bo sun 'byin pa renders Sanskrit saratsalilahasini. Hartmann renders the Tibetan text as "mit den Herbstfluten uberstromen" (?), but sun 'byin-pa means 'to refute, to disapprove'. Probably the Tibetan translator thought that hasini here means 'to laugh at', 'to mock'. In 10.19 kham pa renders ardhabimba. According to Hartmann kham pa is unknown in this meaning. However, zla-ba kham-pa is found in the dictionaries, cf. Jaschke p. 491a, Das p. 1099b. . Shackleton Bailey's edition of the VAV is superseded by Hartmann's edition but the same cannot be said of his translation. Where both editions have the same text, it is interesting to compare the translations by Shackleton Bailey (SB) and Hartmann. In quite a few instances, the former is to be preferred. For instance, 2.14: anataya namas tubhyam namaskararhate 'rhate bhavato 'stu bhavabhavabhavanabhavitatmane / Shackleton Bailey translates: To thee, unbending one, the Arhat deserving of homage, whose soul was developed by meditation on existence and non-existence, in reverence hail! Hartmann has: Dem Ungebeugten, dir, dem Verehrungswurdigen, dem Arhat, sei vom Existentiellen her Verehrung, dem, dessen Selbst entfaltet ist durch die Entfaltung der Nichtexistenz des Werdens! Hartmann is correct in his translation of bhavabhava but what is the meaning of "vom Existentiellen her Verehrung"? Shackleton Bailey follows the Tibetan translation which renders bhavato by gus par "in reverence". As Hartmann indicates there is a play on words in bhavatas - bhavaabhava - bhavana - bhavita but this does not imply that bhavatas has to be translated in accordance with the meaning of abhava. In 2.70 yate is a vocative "Oh ascetic!" (SB) and not a dative "dem Asketen". In 9.22 vyasanaprabhavan bhavan are not "Existenzen, die aus

Page #53
REVIEWS 247 Leid entstehen" but "existences, spring of misery" (SB). In 11.23 it is said of Buddha's tongue: aho gocaracaritvam aho nyayaikatanata. Hartmann translates: "Welche Anhanglichkeit gegenuber dem Stall, welches Ausgerichtetsein auf die Regeln!" In a note Hartmann remarks: "Tib. ubersetzt gocara, "Weide, Stall", mit dem ublichen Aquivalent spyod pa (in Pada c), wodurch der Bezug zum Pferd, der an den vorausgehenden Vers anknupft, verloren geht." The small Petersburg dictionary gives the meaning "Weideplatz fur Rinder" without giving any reference. The only places quoted in MonierWilliams's dictionary are Apsr. i, 2, 4 and (R. iv, 44, 80). SB has: "How wonderful its range of action, how wonderful its undeviating adherence to rule." The translation "range of action" can very well refer to the horse. In 9.21 Hartmann translates mahaprapatandhakaraparidaha by "Finsternis und ausserordentliche Qual tiefer Abgrunde." SB has "(suffering) anguish amid the darkness of the great abyss." 9.26cd has: rathatirathasamkhya tu ka klesavijite jane. Hartmann translates: "die Zahl der Haupthelden unter den Helden aber, wie (gross kann die sein) in einer Welt, die von den Befleckungen besiegt ist?" Hartmann has misinterpreted the meaning of samkhya ka, cf. SB: "but among people vanquished by impurity what reckoning can there be of warriors' excellence?" In 12.6 Hartmann translates sarvakarabala by "Macht uber alle Erscheinungen", SB has "every kind of strength". The meaning of it is "force (which is complete) in all aspects". In quite a few cases SB has adopted more suitable equivalents than Hartmann, cf. 5.13 praksepapanaya; Hartmann "Aufbauschen und Verstummeln"; SB "inserting (and) excluding". Better would be "adding and omitting"; 10.22 surambhoruhagarbhabham angam te; Hartmann "dein wie der Kelch eines Gotterlotos scheinender Korper"; SB "Your body, resembling the calyx of a celestial lotus" (cf. Tib. gser gyi padma'i snin po Itar); 11.25 punyaslokamayena; Hartmann "die aus den verdienstvollen Lauten besteht; SB "that which consists in the glory of merit." Hartmann never translates the word dharma which has many different meanings in the VAV.4 This results in such rather clumsy renderings as "Den den Dharma der bestandigen (Unverwirrtheit besitzt" for nityasammosadharmaya (2.35c). Hartmann has added to his book a pada index of both the VAV and the PPU and an index of selected words. He announces a complete word index for both texts which will be very useful. NOTES Mahaprajnaparamitopadesa is probably the original title of T. 1509 and not Mahaprajna

Page #54
248 REVIEWS paramitasastra, cf. Lamotte, Traite III (Louvain, 1970), pp. VII-VIII; Lamotte, 'Les sources scripturaires de l'Upadesa et leurs valeurs respectives', Cahiers d'Extreme-Asie 2 (1986), p. 1. 2 VAV 7.17-22 are quoted in the Upadesa as verses, cf. Lamotte, Traite I (Louvain, 1944), pp. 83-84. 3 Also Nakamura and Tsuji consider Matrceta to be a follower of Mahayana, cf. Nakamura Hajime (ed.), Butten I (Tokyo, 1966), pp. 425-426; Tsuji Naoshiro, Sansukuritto bungakushi (Tokyo, 1973), p. 19. 4 For a study of the different meanings of the word dharma in the VAV see Kanakura Ensho, Memyo no kenkyu (Kyoto, 1966), pp. 104-116 (first published in Bunka 20.4, 1956). Australian National University J. W. DE JONG Pandit Usharbudh Arya, Yoga-sutras of Patanjali with the Exposition of Vyasa. A Translation and Commentary. Volume I: Samadhi-pada. Honesdale, Pennsylvania, The Himalayan Institute of Yoga Science and Philosophy of the U.S.A., 1986, XXI, 493 pp. $16.95. According to the translator "the purpose of the present translation and commentary is to open the Yoga-sutras to the serious English-reading, aspirant" (p. 16). He writes that a translation and explanation of the Yoga-sutras will have to fulfill three conditions: I. to address its subject matter from within the initiatory tradition of the Himalayan masters; II. to incorporate the essentials of information and exposition given in the exegetical tradition of all past commentators whose works are available; III. to remain true to the basic tenets of the formal Sankhya-yoga school of philosophy (p. 4). Dr. Arya refrains from discussing the scholarly problems or even dates of the authorship of the Yoga-sutras and Vyasa's Bhasya. He writes that "like much of the tradition in India we regard the works of Patanjali and Vyasa as a single composite whole" (p. 8). Dr. Arya is very critical of the works of most scholars who have written on the Yoga-sutras: "The works of Western scholars such as Ballentyne [i.e. Ballantyne), Boissenain [i.e. Boissevain), Deussen, Eliade, Feuerstein, Frauwallner, Garbe, Hauer, Hopkins, Hultzsch, Jacobi, Janacek, Johnston, Judge, Koelman, von Mangoldt, Pensa and others, as well as the westernized Eastern scholars such as Dasgupta and Takagi, were found to have no bearing on the purpose of this translation" (p. 16). According to him their interest is study, his interest is the practice. However, Dr. Arya does not reject academic methods and remarks that "the academicians will find that all the scholarly norms and rules have been carefully observed, that convictions of the tradition have not been permitted to overcloud the clarity Indo-Iranian Journal 32: 1989.

Page #55
REVIEWS 249 of academically established facts, and that the translation as well as the discussion on each sutra will serve as a useful source of information to all" (p. 17). It is obvious that after these declarations the reader cannot expect to find in this work, for instance, a critical analysis of Vyasa's bhasya which would try to determine in how far Vyasa correctly explains the sutras. Dr. Arya translates both the sutras and the bhasya. This is followed by a paraphrase of Vyasa's commentary and a discussion in which he refers to the opinions of the commentators. Vyasa's views are considered authoritative and only in a single instance is his interpretation rectified. In his commentary on drstavisaya in sutra I.15 Vyasa mentions women, food, drink and power (striyo'nnapanam aisvaryam). Dr. Arya remarks: "The word "women" from Vyasa's commentary is being paraphrased here as "the opposite sex", out of consideration for contemporary concerns. These texts were composed and taught in monasteries by yoga masters for whose male disciples the attraction of women must have been a common problem. Although there have been many great women yogis (yoginis) known to the tradition, it is thought that men are not as strong an attraction to aspiring women as women are to aspiring men. This is the only explanation that can be offered at this point, and wherever such allusions appear, they will be translated as "the opposite sex"" (pp. 206-207). One wonders whether the "Himalayan masters" would approve of this interpretation of the word "women". Dr. Arya has read almost all of the commentators on the sutras and Vyasa's bhasya and points out their numerous disagreements on important points. This fact proves sufficiently that the exegetical tradition is contradictory in many respects and cannot guarantee a correct interpretation of the Yogasutras. It shows also that the "experiental tradition" to which Dr. Arya appeals is not reflectled in the commentaries. It is certainly useful to read Dr. Arya's comments and his references to the commentaries which he has consulted to a far greater degree than previous translators but it is difficult for an "academician" to accept his magisterial pronouncements as the final word. Australian National University Indo-Iranian Journal 32: 1989. J. W. DE JONG Gareth Sparham (tr.), The Tibetan Dhammapada. Sayings of the Buddha. A translation of the Tibetan version of the Udanavarga. London, Wisdom Publications, 1986. 235 pp. PS7.95/$11.95. The title of this work is rather misleading for, as the subtitle explains,

Page #56
250 REVIEWS Gareth Sparham's work is a translation of the Tibetan version of the Udanavarga, a work different from the Pali Dhammapada. The Tibetan Udanavarga was first translated by W. W. Rockhill in 1883.' Gareth Sparham discovered Rockhill's translation when he had nearly finished checking his translation against the Commentary by Prajnavarman. He decided to publish his own translation which he considered to be an improvement upon the one that Rockhill had done because the extensive training in the scholastic tradition of Buddhism that he had received made him more familiar with both the technical language of the text and the specifics of the mental development being presented (p. 12). Gareth Sparham's translation was first published in 1983. It is only after the publication of his translation that he discovered the existence of a critical edition of the Sanskrit text. However, Gareth Sparham did not consider it necessary to compare his translation with it and to make the necessary revisions. He declares that it has become apparent in the last twenty years that serious Buddhist scholarship must, where possible, base itself primarily on Tibetan sources and use extant Sanskrit texts as an aid, not vice versa. It is obvious that Gareth Sparham has not the slightest notion of the philological problems connected with the translation of a Tibetan version of a Sanskrit text. Knowledge of Tibetan and assistance from Tibetan lamas are not sufficient to arrive at a correct understanding. In the case of the Udanavarga a translator of the Tibetan version is at present in a much more favourable position than Rockhill who could only make use of parallel verses in the Pali Dhammapada. As Lambert Schmithausen has shown, there are two different recensions of the Udanavarga.2 The first is the Vulgate which is represented by most manuscripts and which is the basis of Bernhard's edition.3 The second formed the basis of the Tibetan translation of the Udanavarga. Because the Tibetan version represents a text different from the Sanskrit text, it would be useful to translate both the Sanskrit text and the Tibetan version. Although the Sanskrit text represents a different recension, it is not possible to arrive at a satisfactory translation of the Tibetan version without consulting the Sanskrit text. Very important in this respect also is Prajnavarman's commentary which Gareth Sparham has consulted but without making sufficient use of it. To mention only a single example: in verses 14 and 15 of chapter four the text mentions a byis-pa'i sin-rta which Sparham translates as 'an infant's cart'. However, Prajnavarman explains that sin-rta stands for sin-rta-pa 'a waggoner' and a byis-pa'i sin-rta is 'a foolish waggoner' as Rockhill correctly translated. The Sanskrit text differs but contains the word sakatika which, undoubtedly, figured in the text translated into Tibetan. In other instances, a better translation would

Page #57
REVIEWS 251 have been possible even without consulting the Sanskrit text if Sparham were familiar not only with the technical language of the text but also with Sanskrit Buddhist terminology. In chapter 19, verse 2, one finds the expression rig dan rkan-par Idan. Sparham translates: "with insight and mobility" and explains that insight provides the 'mobility', literally "legs' of ethics. Rig dan rkan-par ldan translates Sanskrit vidyacaranasampanna "perfected in wisdom and good conduct and the translation of carana with the word rkar-pa 'foot' or 'leg' is due to the fact that Sanskrit carana means both 'foot' and conduct and has nothing to do with insight providing mobility. There is no point in enumerating the numerous mistranslations found in Sparham's translation. As long as it is believed that serious Buddhist scholarship must base itself primarily on Tibetan sources and use extant Sanskrit texts as an aid, we can expect to see similar works continue to be published. Needless to say, Sparham is unaware of the fact that the Tibetan Udanavarga has been published in 1911 by Beckh and the Tibetan Udanavargavivarana in 1984 by Michael Balk (cf. IIJ 30, pp. 236-237). There is, however, need for a critical edition of the Tibetan version for Beckh's edition was based upon a single edition of the Kanjur and a single edition of the Tanjur. This would be very useful for the study of the Udanavarga, a text which presents many still unsolved problems. Australian National University J. W. DE JONG NOTES | Sparham mentions only the 1892 edition. 2. Zu den Rezensionen des Udanavargah', WZKS 14 (1970), pp. 47-127. 3 Franz Bernhard, Udanavarga. Gottingen, 1965-1968. + Sparham has used the Narthang and Derge editions of the Kanjur and does not seem to have discovered that the Udanavarga is also found in the Tanjur preceding the Udanavargavivarana. Gerhard Ehlers, Altturkische Handschriften. Teil 2: Das Goldglanzsutra und der buddhistische Legendenzyklus Dasakarmapathavadanamala (Verzeichnis der Orientalischen Handschriften in Deutschland, Band XIII, 10). Stuttgart, Franz Steiner Verlag GmbH., 1987. IX, 170 pp., 40 pl. DM 128,-. According to the preface by Klaus Rohrborn, the two collections of Old Indo-Iranian Journal 32: 1989.

Page #58
252 REVIEWS Turkish manuscripts from Turfan in West Berlin will be catalogued in four volumes. By far the greatest collection of manuscripts is found in the socalled "Mainzer Sammlung" of the Staatsbibliothek Preussischer Kulturbesitz. The second collection belongs to the Museum fur Indische Kunst and comprises some very interesting items. The first volume, which is yet to be published, describes Sanskrit - Old-Turkish bilingual and other manuscripts written in Brahmi and Tibetan script. The third volume will be devoted to manuscripts of the Maitrisimit. The second volume describes 269 fragments in the Staatsbibliothek Preussischer Kulturbesitz. Of those 183 fragments belong to the Altun Yarok, the Turkish translation of I-tsing's Chinese version of the Suvarnaprabhasa. 166 fragments have been identified and the remaining numbers (167-183) are still to be identified. Fragments 184-269 belong to the Dasakarmapathavadanamala. Most of these fragments were published by F. W. K. Muller and A. v. Gabain in Uigurica III and IV (Berlin, 19221931). In his introduction Gerhard Ehlers indicates the origin of the manuscripts and describes the arrangement of the catalogue. The fragments of the Altun Yarok are arranged according to the Petersburg edition of the text. Gerhard Ehlers points out that the Berlin manuscripts are in general closer to I-tsing's translation than the Petersburg manuscript. As to the fragments of the Dasakarmapathavadanamala Gerhard Ehlers distinguishes five different manuscripts (N, D, S, C and Z). The text is divided into ten chapters dealing with the ten karmapathas. Gerhard Ehlers indicates which fragments belong to each of the ten karmapathas. Klaus Rohrborn remarks in his preface that of all the collections of OldTurkish manuscripts from Turfan until now only one collection has been described in "eine Art von Katalog", i.e. that in the Ryukoku Library in Kyoto. For the collections in London, Stockholm and Paris there are only unpublished lists available and nothing is known about the collections in Leningrad and in Berlin (DDR).' It is to be hoped that the publication of this very carefully compiled catalogue may act as a stimulus and example for catalogues of the collections in other places. NOTE There must also be a collection in Peking. Australian National University J. W. DE JONG

Page #59
REVIEWS 253 Jens Peter Laut, Der fruhe turkische Buddhismus und seine literarischen Denkmaler (Veroffentlichungen der Societas Uralo-Altaica, Band 21). In Kommission bei Otto Harrassowitz, Wiesbaden, 1986. X, 228 pp. DM 60, In his preface the author stresses the importance of the Old Turkish literature for Indian and Buddhist studies. He points out that among the fragments of Uigur texts are Buddhist texts which are not known elsewhere or which differ greatly from versions in other languages. As examples he mentions the Uigur Dasakarmapathavadanamala which shows a structure in ten Karmapathas not found in any other collection of legends and the Maitrisimit which, although based on the Tokharian Maitreyasamitinataka, differs considerably from it. Laut concludes his remarks as follows: "In jedem Fall ist sicher, dass auch die Turkologie einen eigenstandigen Beitrag zur buddhistischen Literaturgeschichte leisten kann." In the introductory chapter Laut discusses the "Sogdian hypothesis" and points out that the discovery of a Sogdian inscription near Bugut in Mongolia renders it probable that the Turks used the Sogdian script in the sixth century for writing Turkish and that the Sogdians have been active in propagating Buddhism under the Uigurs in Mongolia. Laut remarks that the "Sogdian hypothesis" is supported by Old Turkish Buddhist texts which contain a greater number of loan-words from Sogdian than other texts and which show the characteristics of the n-dialect, the language of the Manichaean Turkish texts. These texts are clearly differentiated from the great mass of Old Turkish texts written in the language of the Old Turkish koine and in which loan-words from Tokharian are predominant. The two Old Turkish "pre-classical" texts which Laut examines in his book are the Sakiz Yukmak Yarok sutra, a Turkish version of a Chinese apocryphal text probably compiled in the first half of the 8th century, and the Maitrisimit. Before examining the loan-words and the dialectical features of both texts Laut reports on the history of the study of Maitrisimit. The discovery in 1959 of 293 folios of a manuscript in Hami and the publication of several chapers of this manuscript enable the author to identify many fragments and to analyse differences between the manuscripts from Hami and Sangim. As concerns the Sakiz Yukmak Yarok Sutra, the author does not make use of the edition published by W. Bang, A. von Gabain and R. Arat in 1934 because it does not faithfully reproduce the readings of the London scroll, but quotes from the scroll itself. The main part of Laut's book is devoted to a study of the orthography and language of these two texts and to the study of the loan-words in them. Indo-Iranian Journal 32: 1989.

Page #60
254 REVIEWS Both texts show characteristics of the n-dialect, the London scroll even more than the Sangim manuscript of the Maitrisimit, and the Hami manuscripts to a lesser degree than the Sangim manuscript. Of particular interest is Laut's study of the loan-words. He examines in both texts the Indian loan-words taken from Sogdian and gives a list of the Indian loan-words in the Maitrisimit which were borrowed from Tokharian. Laut points out that all Old Turkish texts contain a number of religious terms which constitute a Buddhist basic vocabulary. He lists the words of this basic vocabulary in the Sangim manuscript of the Maitrisimit and in the London scroll. Finally Laut remarks that the London scroll shows a higher proportion of Indian loan-words from Sogdian (22 out of 31) than the Sangim manuscript (65 out of 380). The results obtained by Laut are important not only for the study of the Old Turkish manuscripts but also for the history of Buddhism in Central Asia. It is to be hoped that once all fragments of the Hami manuscript of the Maitrisimit are published a new translation of this text will be prepared. Further study of the Maitrisimit and other "pre-classical" Old Turkish texts will help to give a better insight into the history of Old Turkish translation of Buddhist texts and in this way contribute to our understanding of the history of Buddhism among the Turks.2 NOTES ! On this text see Juten Oda, 'Remarks on the Indic "lehngut" of the Sakiz yukmak yaruq sutra', Sprachen des Buddhismus in Zentralasien (Wiesbaden, 1983), pp. 65-72; R: A. Stein, 'Tibetica antiqua I, BEFEO LXXII (1983), pp. 214-215... 2 In the bibliography of Laut's book two corrections are to be made. The author of 141 is the Dutchman Hendrik Kern and not Heinrich Kern. Werner Thomas's article on the Maitreya-samiti-nataka (192) was published in volume VIII (not VII) of Kindlers Literaturlexikon. Australian National University J. W. DE JONG