Book Reviews

Author(s): J W De Jong
Publisher: J W De Jong
Catalog link: https://jainqq.org/explore/269681/1

Page #1

REVIEWS
Reviews (0)
63
Phyllis Granoff, Philosophy and Argument in Late Vedānta. Sri Harsa's Khandanakhandakhādya (Studies of Classical India I). Dordrecht, D. Reidel Publishing Company, 1978. XXXV, 275 pp. Dfl. 100.-/ US $ 48.50.
Śrī Harsa's Khandanakhandakhadya, "The sweets of refutation", has the reputation of being one of the most difficult works in Indian philosophy. This reputation is certainly not undeserved. The main part of his work consists in a refutation of definitions proposed by his opponents. In order to understand his arguments it is necessary to know in each single instance who are his opponents and which views are being refuted. In the second place, Sri Harsa is a difficult writer because his reasoning is often abrupt and cannot be understood unless one supplies the arguments which are implied but not expressly formulated.
Dr. Granoff's work consists of two parts. In the preface (pp. 1-69) she examines Sri Harsa's refutations of the definitions of pramā, "valid knowledge", and pratyaksa, "perception", proposed by his opponents. Dr. Granoff shows how Sri Harsa refutes the views of his opponents by pointing out the contradictions in their reasonings. His main opponents are the Naiyāyikas and, above all, Udayana, but Sri Harsa shows himself also well versed in the doctrines of the Mimāmsakas, the Jains and the Buddhists. Dr. Granoff analyzes carefully Sri Harsa's arguments

Page #2

64
REVIEWS
and identifies the doctrines which he attacks. The complexity of Sri Harsa's reasoning is obvious from the fact that in the edition of his work referred to by Dr. Granoff the refutation of the first definition: tattvānubhūtih pramā, "valid knowledge is an experience of the essence of an object", takes up almost 90 pages. Dr. Granoff's analysis of this part of the Khandanakhan dakhădya does full justice to Sri Harsa's logical skill (pp. 4-30). Other definitions of pramā are refuted much more briefly (pp. 30-39). Before examining one by one the pramāna-s, "the means or instrumental causes of valid cognition", Sri Harsa refutes definitions of karana, “instrumental cause", (pp. 39-44). The first pramāna examined by Sri Harsa is pratyakşa, "perception". Many definitions given by his opponents are refuted by him (pp. 44-52). At the end of this section of the preface Dr. Granoff raises the question of the source of Sri Harşa's attacks against these opponents. According to her Sri Harsa is unlike other philosophers who also systematically refuted the arguments of their opponents, as, for instance, the Madhyamikas,
nd Jayarāsī, the author of the Tattvopaplavasimha. She suggests that, perhaps Sri Harsa owes much to Abhinavagupta's Bhedavādavidāraņa, a work which has not been preserved. In the second part of the preface Dr. Granoff examines Sri Harsa's Vedānta doctrines. She remarks that the two main points at which he radically deviates from the generally accepted tenets of non-Kashmiri Advaita are: the repudiation of a third truth value, sadasadvilaksanatva, and the assertion of the absolute identity of the visible world and knowledge.
The main part of Dr. Granoff's work is a translation of the introduction to the Khandanakhandakhādya in which Sri Harsa sets forth his own views. This introduction can be divided into four sections. The first part shows that it is not necessary to admit the existence of the means of knowledge before entering into a debate (pp. 71-110). In the second part Sri Harsa defends the doctrine of the self-awareness (svaprakāśata) of knowledge (pp. 110-137). After a brief interlude Sri Harsa develops his interpretation of the doctrine of non-duality (pp. 144 - 203). The introduction concludes with a brief examination of the khandanayukti-s, "the principles of refutation" (pp. 203-208). Dr. Granoff divides the translation of the introduction into brief sections each of which is followed by her own commentary. The notes (pp. 209– 255) contain references to other texts and, above all, discussions of the commentaries on the Khandanakhandakhādya. Dr. Granoff has not only used the commentaries which are available in print but also five unpublished commentaries. She points out that in many instances not one of the commentaries gives a satisfactory interpretation.
Dr. Granoff's translation is at the same time faithful to the text and lucid in style. This is a great merit in the case of such a difficult author as Sri Harsa. It is only very rarely that a rendering is not entirely correct and, even then, it is, in general, nothing but a minor slip. For instance: puruşatrayāvagatasyāpi ekaksaņāvagatasya purusāntarena, tenāpi ksaņāntare, bahulam badhyatvadarsanāt: "we see that what is understood even by three people at a given moment is often contradicted, either by other people, or by those very people at another moment." Dr. Granoff translates: "we see that what is understood even by three people at a given moment is contradicted by those very people or by other people at another time" (p. 83). The entire Khandanakhandakhādya was translated by Ganganatha Jha (Indian Thought Series, 4, 1915). This translation is not mentioned by Dr. Granoff and, as this work is not at my disposal, it is not possible to see whether she has been able to derive some benefit from it.
Dr. Granoff's commentary is based upon an excellent knowledge of the works of Sri Harşa opponents. It is through a close study of the works of Vācaspati Miśra, Udayana, etc. that she is able to explain fully Sri Harsa's often rather obscure arguments. In one respect, her commentary is not entirely satisfactory. In her introduction to the first part of the introduction Dr. Granoff promises a detailed comparison of Sri Harsa's solution with that of his Buddhist predecessors but the Buddhist sources are only briefly discussed in her commentary (pp. 84-85 and 87) and in the notes in which she refers mainly to Vaidya's edition of the Bodhicaryāvatāra. One has the impression that Dr. Granoff has studied the Buddhist philosophers through the eyes of a Vedāntin, as, for instance, in her discussion on the samvrtisattva. The fact that samvrti is glossed as idampratyayatāmātra (pratityasamutpāda) does not point in the direction of an

Page #3

REVIEWS
interpretation of samvṛtisattva as an existential rather than an epistemological category (p. 85). According to Candrakirti's samvṛti has three meanings: ajñana, parasparasambhavana and samketa or lokavyavahara (Prasannapada, ed. L. de La Vallée Poussin, p. 492.10-12). In Candrakirti's text the reading parasparasambhavana is doubtful but Prajñākaramati clearly indicates the pratityasamutpanna character of samvṛti: pratityasamutpannam vasturūpam samvṛtir ucyate (ed. P. L. Vaidya, p. 171.4). The samvṛti is the opposite of the paramartha in which there is neither origination (utpada) nor cessation (nirodha).
On p. 87 Dr. Granoff remarks that verse 107 of chapter 9 of the Bodhicaryavatāra raises the question of anavasthā, and further adds that it is undesirable as it would lead to the absence of final release. I am afraid that this is not found in the text (Dr. Granoff refers to p. 287; read p. 247). The opponent raises the following objection. If there is no samvṛti, there is also no paramartha and it becomes impossible to obtain merit. It is not possible to prove the existence of samvṛti by another samvṛti consisting in wrong notions (kalpanābuddhirūpā). In note 24 Dr. Granoff says that Prajñākaramati takes anya to mean another person but this is not found in the text. If samvṛti would exist through another samvṛti, no human being would obtain Nirvana (atha sapy anyasamvṛtya syat sattvo nirvṛtah kutaḥ, 107cd). Prajñākaramati explains that the reason is that a human being in the state of Nirvāṇa would be the object of a wrong notion (tasyāpi buddhyā [Vaidya: vṛddhyā!] viṣayikaranāt). The objection against proving the existence of samvṛti by another samvrti is not that this would lead to an infinite regress, but that in this way Nirvana too would become samvṛti (nirvṛtir api samvṛtiḥ syat). Also, elsewhere, Dr. Granoff attributes doctrines to the Bodhicaryāvatāra which are not to be found in it. For instance, according to note 25 that the non-existent can be a cause is discussed in the Bodhicaryavatāra, p. 186, but on this page we find only a long quotation from the Sālistambasūtra explaining the pratityasamutpada (in note 48 Dr. Granoff refers to Bodhicaryavatāra p. 183 for a proof of the fact that a non-existent can serve as a cause.) Or does Dr. Granoff explain the statement that the samskara-s are dependent on avidyā (avidyāpratyayāḥ samskārāḥ) as meaning that the non-existent functions as a cause?
Dr. Granoff has great admiration for Śrī Harṣa (cf. pp. 2 and 54). In his foreword B. K. Matilal sums up Śrī Harṣa's contributions in two points: (1) The creation of an independent philosophic method which uniquely suited the rational discussion of the monistic doctrine of the ineffable truth in the Advaita school. (2) His trenchant criticism of Nyaya categories had a salutary effect on the Indian philosophic scene, and philosophic sophistication of later authors of both Nyaya and Vedanta deepened as a result. It is interesting to note that S. N. Dasgupta expressed a completely contrary view. According to him Śri Harṣa's criticisms did a great disservice to the development of later Nyaya thought, since later Nyaya writers were mainly occupied in inventing suitable qualificatory adjuncts and phrases by which they could define their categories in such a way that the undesirable applications and issues of their definitions could be avoided (A History of Indian Philosophy, II, 1932, p. 146).
Dr. Granoff's excellent book is an important and original contribution to the study of Vedanta and other Indian systems of philosophy. In her acknowledgment she pays tribute to her Indian guru T. S. Srinivasa Sastri who explained Sri Harṣa's work to her. In a Sanskrit preface he has high praise for her achievement. We can but hope that she will continue her studies of Vedanta philosophy.1
Australian National University
65
J. W. DE JONG
NOTE
1 The book contains a few misprints, mainly in Sanskrit words. On p. 68 (note 42) "factors which mitigate against" is probably a misprint for "factors which militate against". On p. 85, line 8 read (p. 170ff.) for (p. 17ff.).

Page #4

66
REVIEWS
H. W. Bailey, Dictionary of Khotan Saka, Cambridge University Press, 1979. xvii + 559 pp. £ 80.
The long-awaited publication of Bailey's Dictionary of Khotan Saka fills a long-felt gap in the field of Iranian studies and marks at the same time the culmination of the author's life-long devotion to the language of Khotan.
The list of Bailey's publications, nearly all of which concern Khotanese in some way or other, is long. 164 items up to 1970 were listed in the bibliography in the volume in his honour (BSOAS, XXXIII. 1, 1970, IX-XIV). More than forty items have appeared since then.
The basis of Bailey's dictionary is formed by the transcriptions of Khotanese texts published by him as Khotanese texts 1-V, Cambridge 1945-63 and Khotanese Buddhist texts, London 1951. In these volumes he has transcribed most of the extant material and in numerous articles he has shed light on many of the texts contained in them and on a large proportion of the vocabulary.
Bailey's Khotanese texts VI, Cambridge 1967, which contains no texts but is, as the sub-title indicates, the 'prolexis to the book of Zambasta', was a foretaste or preview of what Bailey's dictionary would be like. It contains discussions of words found in a large Khotanese text called by him The Book of Zambasta and occasionally also of other words.
The magnitude of the task of compiling a dictionary of Khotanese was described as recently as 1974 by M. J. Dresden (Acta iranica, 3, 235): 'It cannot be expected that a complete dictionary with full textual references and a linguistic apparatus will be published in the near future. The sheer extent of the materials, which can approximately be estimated at some 40,000 printed lines, and the many unsolved problems of interpretation are still formidable obstacles. It seems, therefore, desirable that an attempt be made to collect the data required to produce a "working" dictionary - wordlist is, perhaps, a more appropriate term - of Khotanese.' He then went on to give an example of what his "working" dictionary would look like on the basis of 28 words chosen at random. Each receives two or three lines giving the essential information. I do not propose to make a detailed comparison of the items in his list with the corresponding ones in Bailey's dictionary but the difference of scale is of interest. Thus, to the one and a half lines devoted to the word paysa- in Dresden's list correspond almost one and a half columns in Bailey's dictionary. Bailey quotes five passages in which the word occurs and translates them. This occupies the first fifteen lines. The rest of the article concerns. the etymology.
Despite the obvious emphasis on etymology Bailey insists that his dictionary is not an etymological dictionary, and it is accordingly entitled Dictionary of Khotan Saka. Nevertheless, he regards it as 'one contribution to the far vaster project of the etymological dictionary of all Iranian languages' (p. vii). As for 'the fairly copious quotation of Iranian cognates, the purpose of this is 'to assure the Khotanese word and to situate it within the dialects' (ibid.).
In fact in a large proportion of the entries the space devoted to etymology greatly predominates and in some instances the Khotanese lemma seems to have been used as the title of an etymological essay. Thus, the word candarno, which occurs once only and is of uncertain meaning, is allotted more than two full columns. The word.ragai, which likewise occurs once only and is of uncertain meaning even according to Bailey, receives well over a column, and a column already contains sixty lines. The word durausa' occurs once only and is of uncertain meaning but its treatment requires 85 lines.
The etymological character of the dictionary can be seen from the fact that there is at least one entry listing only a hypothetical root *yat- 'to place properly', for which the reader is referred to the words (not all lemmata!) nistu, bistu, nyāttara-, and gista-.
A large proportion of the vocabulary occurring in most Khotanese texts consists in words of Indian origin either from Sanskrit or from Prakrit. The learned Sanskrit borrowings can usually be recognised as such, at any rate in well-written texts especially in Old Khotanese, but those from Prakrit are often difficult to detect in their Khotanese guise even for trained Indologists.

Page #5

REVIEWS
75
Das Spiel von der Himmelsfee 'Gro-ba bzan-mo ist trotz seiner religiösen Thematik und der Fülle uralter Motive auch dazu geeignet, menschliche Anteilnahme zu wecken: es schildert das Schicksal zweier noch kleiner Kinder, deren Mutter durch dämonische Missgunst gezwungen wurde, diese Welt zu verlassen. Mutterlos irren sie umher; sie sind sich darüber im klaren, dass der Grund all dieses (Unheils) darin liegt, dass wir keine Mutter mehr haben', und sie beneiden die Tiere, die zwar keine Königskinder sind, aber in der Mitte zwischen Vater und Mutter wandeln können'. Und doch ist es stets ihre himmlische Mutter, die sie - unsichtbar bzw. in anderer Gestalt - jeweils der drohenden Gefahr entreisst. Anschaulicher kann menschliches Leid einerseits und die Kraft mütterlicher Liebe andererseits kaum dargestellt werden, und wir gehen wohl nicht fehl in der Annahme, dass in eben diesen Archetypen der Schlüssel zum eigentlichen Sinn dieses Stückes liegt.
This is a valuable study which has more of a limited appeal to cultural anthropologists and folk specialists. It is not the book that everyone who reads Tibetan needs on his shelf.
Ball State University
G. W. HOUSTON
Louis Ligeti (ed.), Proceedings of the Csoma de Körös Memorial Symposium. Held at Mátrafüred, Hungary 24-30 September 1976 (Bibliotheca Orientalis Hungarica , vol. XXIII). Budapest, Akadémiai Kiadó, 1978. 586 pp. $ 51.00
The Proceedings of the Csoma de Körös Memorial Symposium contain more than forty articles. Among the contributors we find the names of eminent Tibetologists from many countries. The seed sown by Csoma de Körös has indeed produced a rich harvest! Only very few articles are disappointing as, for instance, the three pages on Buddhism in Russia and the U.S.S.R. by S. Dylykov (pp. 85-87).
Linguistic problems are studied by Betty Shefts Chang, "Tibetan prenasalized initials" (pp. 35-46), Kun Chang, "The Tibetan role in Sino-Tibetan comparative linguistics" (pp. 47-58), Roy Andrew Miller, "Is Tibetan genetically related to Japanese?” (pp. 295-312) and Eberhardt Richter - Dieter Mehnert, “Zur Struktur und Funktion der Aspiration im modernen Tibetischen" (pp. 335-352). R. A. Miller's article is mainly devoted to an analysis of three articles by T. Nishida on the relations between Japanese and Tibeto-Burman in Gengo, Vol. 5 (1976), No. 6, pp. 74-86, No. 7, pp. 64–76, and No. 8, pp. 74-83.
In "Zur Frage des sogenannten 'Akkusativs' im Tibetischen" (pp. 169-188) Rudolf Kaschewsky examines the la don case and shows that it does not correspond to an accusative: "Das Akkusativobjekt als direktes Objekt kann nur im endungslosen Kasus stehen. Bei einer la-don-Endung ist dagegen in jedem Einzelfall zu prüfen, welche der in der einheimischen Grammatik aufgeführten Funktionen vorliegt."
In "Some lexical items from the Rgyud-bzi” (pp. 101-108) R. E. Emmerick examines the meaning of several obscure words. As to bkan the Sanskrit equivalent given in Sumatiratna's dictionary (Vol. I, p. 64) is not 'utatam (p. 104) but atatam "stretched". The Mongolian translation has geteikü which Kowalewski renders as "être incliné, être couché sur le dos" (p. 2461b). Emerick renders khers-po with "puffed up". According to Sumatiratna's dictionary a Tibetan equivalent is ras-pa which is translated into Mongolian by qa yučiraysan "old, spoiled, deteriorated". Sumatiratna renders khens-po with Mong. nigsigsen "spoiled, rancid", and explains that tea and soup become khens-po when the water has not been brought

Page #6

76
REVIEWS
to the boil (khers-po ni ja-thug sogs chu bsres-nas ma khol-ba). The same thing happens with meat when the time to cook it has passed (sa-la sogs btsos-pa dus 'das-pa). Tib. gag-pa renders Skt udarda, the meaning of which seems to be a form of erysipelas. However, Sumatiratna explains it as a swelling, an obstruction deep in the throat: gre-phugs-su 'thor-ba 'dra-ba byun. ste mgrin-pa dgag-pa i rma'o. He renders 'thor-ba with Mong. göbdüriguü "elevation, roughness on the surface of an object, wart, blister, acne, pimple" (Lessing, p. 385b). The literal meaning of gag-pa is probably "obstruction", cf. gegs-pa "to hinder, prohibit, stop".
Five Tibetan dictionaries are the subject of a brief paper by J. Terjék, "Die Wörterbücher der vorklassischen tibetischen Sprache” (pp. 503-509): 1. Blo-gsal mgrin-rgyan by Alagša bstan-dar; 2. brDa'-gsar-rnir by Klon-rdol bla-ma; 3. Dag-yig mkhas pa'i byur-gnas by Canskya Rol-pa'i rdo-rje; 4. Li-si me-tog by Dkon-mchog chos-grags; 5. Li-si gur-khan by Skyogsston. Much more detailed information on three of these dictionaries (Nos. 2, 3 and 5) can be found in an extremely well documented article by Manfred Taube, “Zu einigen Texten der tibetischen brda-gsar-rñin-Literatur", Asien wissenschaftliche Beiträge. Johannes Schubert in memoriam (Berlin, 1978), pp. 169-201.
In his article "On a term of taxation in the old Tibetan Royal Annals" (pp. 357-363) A. Róna-Tas studies the term phyin-rol (d) and explains it as meaning "felt roll". Several names and titles which occur in Tibetan letters brought back from Central Asia by German expeditions are studied by Manfred Taube: "Einige Namen und Titel in tibetischen Briefen der Berliner Turfan-Sammlung" (pp. 487-502). Of special interest is Taube's study of the meaning of stag in combination with a proper name. According to him stag is "eine ehrende Anrede für Angehörige der Armee." G. Uray's article (The Annals of the 'A-ža Principality. The problems of chronology and genre of the Stein Document, Tun-huang, Vol. 96, fol. 84, pp. 541-578) is devoted to a very thorough analysis of a Tibetan manuscript which since its first publication by F. W. Thomas (JRAS, 1927, pp. 58-66) has been studied by many scholars, most recently by Yamaguchi Zuiho ("Matrimonial Relationship between the T'u-fan and the T'ang Dynasties, Part II", Memoirs of the Research Department of the Toyo bunko, No. 28, 1970, pp. 59-85; ibid., No. 35, 1977, pp. 123-130). Uray also announces a new edition of the text together with translation and detailed comments in the Monumenta Tibetica Historica. According to him the manuscript is a fragment of the Annals of the 'A-ža Principality and relates its history from the Horse year A.D. 706/707 till the Tiger year A.D. 714/715.
A rather neglected field of Tibetan studies is taken up by Dieter Schuh in "Ergebnisse und Aspekte tibetischer Urkundenforschung" (pp. 411-425) in which he distinguishes three types of documents: 1. Herrscherurkunden (bka'-sog, gtan-chig, še-bam) und administrative Erlasse; 2. Klosterurkunden (bča' yig); 3. Privaturkunden. Dieter Schuh plans to publish Tibetan documents in the third section Diplomata et epistolae of the Monumenta Tibetica Historica. He has already published one large volume containing documents from Sikkim: Urkunden, Erlasse und Sendschreiben aus dem Besitz sikkimesischer Adelshäuser und des Klosters Phodang (Monumenta Tibetica Historica, Abt. III: Band 3, St. Augustin, 1978, 413 pp.) in which 54 documents are reproduced and edited. Another volume containing edicts and letters of the Mongolian rulers of the Yüan period is announced by Schuh in his article (p. 412, n. 1).
Articles dealing with different aspects of the history of Tibet are contributed by E. I. Kychanov, “Tibetans and Tibetan culture in the Tangut state Hsi Hsia (982-1227)" (pp. 205-211), Kristina Lange, "Das Geschichtsbild des tibetischen Feudalstaates (17.-20. Jh.)" (pp. 213-222), A. S. Martynov, "On the status of the Fifth Dalai Lama. An attempt at the interpretation of his diploma and title" (pp. 289-294), Luciano Petech, "The 'Bri-gun-pa sect in Western Tibet and Ladakh" (pp. 313-325) and Pema Tsering, "Nin ma pa Lamas am Yüan-Kaiserhof” (pp. 511-540).
The concept of reincarnation is studied by Turrell V. Wylie, “Reincarnation: a political innovation in Tibetan Buddhism" (pp. 579-586). His article should be read in combination with Yamaguchi's study on the "Living Buddhas" which appeared in 1977 (""Katsubutsu" ni tsuite", Tamaki Koshiro hakushi kanreki kinen: Hotoke no kenkyū, pp. 285-302).

Page #7

REVIEWS
77
In an article entitled "A propos des documents anciens relatifs au phur-bu (kīla)" (pp. 427444) R. A. Stein analyses a Tun-huang manuscript (Pelliot No. 44) and other tantric texts relating to Phur-pa, the divinity of the phur-bu. In "Remarks on Tantristic hermeneutics" (pp. 445-458) Ernst Steinkellner studies the system of tantristic hermeneutics developed by the Arya school ('Phags lugs pa) of Guhyasamāja tantra commentators according to the introductory verses to Candrakirti's Pradipoddyotanā. Jikidō Takasaki examines several passages of Chos-grub's Tibetan version of the Chinese translation of the Lankavatara by Guṇabhadra; "Some problems of the Tibetan translations from Chinese material" (pp. 459– 467). D. Seyfort Ruegg sketches the history and future tasks of the study of Tibetan philosophy: "The study of Tibetan philosophy and its Indian sources. Notes on its history and methods" (pp. 377-391).
Various topics are dealt with in the following articles of which the contents are clearly indicated by the titles: Géza Bethlenfalvy, "Alexander Csoma de Körös in Ladakh" (pp. 7-25), F. A. Bischoff, "Padmasambhava est-il un personnage historique?" (pp. 27-33); C. Damdinsuren, "Neskol'ko slov o Kalačakre" (pp. 59-63); Eva Dargyay, "Grundherr und abhängiger Bauer in Tibet. Eine Analyse der Machtverhältnisse" (pp. 65-83); Hilda Ecsedy, "On a few traces of ancient Sino-Tibetan contacts in the early Chinese mythic tradition" (pp. 89-99); Eric Grinstead, "Tibetan studies by computer" (pp. 109-123); Walther Heissig, "Geser Khan als Heilsgottheit" (pp. 125-152); Ireneusz Kania, "Médecine tibétaine dans les collections du Musée Ethnographique de Cracovie" (pp. 153-160); G. Kara, "Uiguro-Tibetica" (pp. 161-167); Josef Kolmaš, "The aphorisms (Legs-bshad) of Sa-skya Pandita" (pp. 189-203); L. Lörincz, "Zur Katalogisierung tibetischer Märchen" (pp. 285-288); Prince Peter of Greece and Denmark, "Tibetan oracles in Dharamsala" (pp. 327-334); Rinčen Jonšebu; "Ispolniteli recitacij na ustnom mongol'skom jazyke sutr" (pp. 353-355); Hans Roth, "Die Erfassung der tibetischen und mongolischen Sachkultur an westeuropäischen Museen" (pp. 365-375); Alice Sárközi, "A thanka from Mongolia" (pp. 393-401); L. S. Savitsky, "Secular lyrical poetry in Tibet. Works of Tsangjang-jamtso (1683-1706)" (pp. 403-409); Magdalena Tatár, "Synkretistische Züge bei einem lamaistischen Kult der Mongolen" (pp. 469-472); Erika Taube, "War das Urbild des gestiefelten Katers ein Fuchs?" (pp. 473-485).
In a long article entitled "Le mérite d'ériger un stūpa et l'histoire de l'éléphant d'or" (pp. 223-284) Louis Ligeti studies two lacunae of the Mongolian inscription in small characters on the Chu-yung-kuan, the Buddhist arch of the fourteenth century which is situated sixty kilometers northwest of Peking. Although his article is mainly concerned with the Mongolian text, Ligeti refers often to parallel texts in the inscriptions in Uigur, Tibetan and Chinese. His remarks are of great importance for the study of the Mongolian and Uigur texts which, according to him are closely related (p. 234). However, Ligeti's study of the Tibetan text is not an improvement on the careful edition and translation by Nagao (Chü-yung-kuan. Edited by Jiro Murata, Vol. I. Kyoto University, 1957, pp. 230-242). The Tibetan text refers to four Buddhist sutras. The identification of the third is uncertain. According to Ligeti the Tibetan inscription contains the words Bde-ldan-par sku-[gzu]gs bkod-pa'i mdo (cf. p. 243). However, on p. 242 Ligeti writes: Bde-dan-ldan-par bkod-pai mdo. In his edition Nagao writes: bde dan Idan par [sku gzu]gs bkod paḥimdo (p. 236). According to Ligeti the sutra referred to in the inscription is the Sukhavatiguhya, tib. Bde-ba-čan-gyi bkod-pa (Ōtani Catalogue, No. 783). In a note Ligeti remarks that this identification was proposed by Huth in 1895 (JA, 1895, I, p. 371). Tib. Bde-ba-čan-gyi bkod-pa corresponds to Sukhāvativyūha and not to Sukhāvatigūhya. Nagao proposed to read Sku-gzugs bkod-paḥi mdo and to identify this text with the Chinese sutra Ta-ch'eng tsao-hsiang kung-te ching (Taishō No. 694). This sūtra contains a passage which corresponds quite well to the text of the Tibetan inscription.2 Ligeti remarks in a note that the Chinese sutra mentioned by Nagao has a certain affinity with the Tibetan sutra. This is a very curious remark because the Tibetan sūtra (i.e. the Bde-ba-čan-gyi bkod-pa) which describes the Sukhāvati paradise has no relation at all to the Chinese sutra which describes the merit of erecting an image of the Buddha. Ligeti has completely overlooked the fact that according to

Page #8

78
REVIEWS
Nagao's interpretation the words bde-dan-ldan-par do not belong to the title of the sutra mentioned in the inscription.
The Tibetan inscription contains a stanza on the merit of erecting a stupa. Ligeti proposes to read this stanza in the following way:
[gan]-Zig stoń gsum rin-chen-gyis bkan '[pha] gs-pa-rnams-la phul-ba bas / gan-gis mchod-rten gcug-lag-khan 'am skyu-ru-ra cam bžeńs-su/ bde-gšegs sku-gzugs-nas 'bru cam-žig bžugs-su bčug-pa'i bsod-nams-[kyis] / Ches-cher khyad-par 'phags-pa yin žes Khan-bu brcegs-pa'i gzuns.
According to Ligeti (p. 244, n. 36) Nagao reads:
[re și]g stoń gsum rin chen gyis bkan [dgyes ?] pa rnams la phul ba bas / gan gis mchod rten gtsug lag khan ham skyu ru ra tsam bṣen nas su / bde gšegs sku gzugs nas ḥbru tsam sig bṣugs su btsug paḥi bsod nams [ni?] / chos cher khyad [par?] ḥphags pa yin şes [?] khan bu brtsegs paḥi mdo las gzuns /
Nagao's edition is found on p. 235 and not on p. 225. In line 2 Nagao reads bṣens, in line 3 bcug, and in line 4 gsuns. In a note he remarks that chos cher is a scribal error for ches cher. In Nagao's edition of this stanza each line contains 15 syllables but in Ligeti's edition the second line has 14 syllables and the fourth line 13 syllables. The Tibetan inscription consists of 24 stanzas of four lines and of 2 stanzas of six lines. Each line contains 15 syllables. It is therefore certainly necessary to read in line 2 bźens nas-su and in line 4 brtsegs-pa'i mdo las gsuns. Accepting Ligeti's reconstructions in line 1 I propose to read the stanza as follows:
[gan]-zig ston-gsum rin-chen-gyis bkan '[pha] gs-pa rnams-la phul-ba bas/ gan-gis mchod-rten gtsug-lag-khan 'am skyu-ru-ra tsam bźens nas-su /
bde-gśegs sku-gzugs nas-'bru tsam-źig bźugs-su bcug-pa'i bsod-nams-[ni?] / ches-cher khyad-par 'phags-pa yin źes Khan-bu brtsegs-pa'i mdo las gsuns/
Line three mentions the erection of a statue of the Sugata the size of a grain of barley (nas). Ligeti does not translate the word nas which he has completely misunderstood as is obvious from his transcription (sku-gzugs-nas 'bru cam-žig), although he correctly points out that the word barley is found in the corresponding Chinese (mai), Uigur (arpa) and Mongolian (arbay) texts (p. 247). The passage of the Khan-bu brtsegs-pa'i mdo quoted by Ligeti likewise speaks of a grain of barley (sku-gzugs nas-'bru tsam), cf. p. 248. In translating this passage Ligeti renders phyogs-bži'i dge-slon-gi dge-'dun "to the universal (caturdisa) congregation (samgha) of monks (bhiksu)" with "aux moines mendiants (dge-slon) et prêtres (dge- 'dun) des quatre points cardinaux".
The stanza relating to the golden elephant has been recently studied by F. W. Cleaves (The boy and his elephant', HJAS 35, 1975). In a note (p. 252, n. 49) Ligeti remarks: "En faisant la communication au symposium le 28 sept. 1975 (sic) j'ignorais encore l'article de M. Cleaves; les références à son étude ont été ajoutées ultérieurement à mon travail." The first line of this stanza ends on the word sudur but Ligeti points out that in the facsimile the sign for d is effaced. He proposes to read su [p] ur[van-i Jasa]] u "ayant réparé le stūpa". It is interesting to note that Pelliot had already suggested the reading su[p]urxan and had also indicated that the two last letters should be Ju (cf. Cleaves, op. cit., p. 39). This is not mentioned by Ligeti. The text of the corresponding Tibetan stanza is read by him as follows:
'di-ltar snon-gyi de-bžin-gšegs-pa Dga'-Idan-gnas-nas 'phos-pa-yi //
gnis-'thun gžon-nu'i mchod-rten dkar-rcis mi žig gsos-pa'i bsod-nams-kyis // skye-žin skye-bar mtho-da'i rigs-[pa] [gser-gyi] glan-chen žon-pa-čan // ... Žin [r] gyu mthar glan-[skyon] [že] s-bya'i dgra'-bčom go-'phan thob-par gyur //
Ligeti translates: "ainsi le Tathāgata d'autrefois descendit du ciel Tuşita, / un homme répara le stūpa du jeune éléphant avec de la chaux et par ce mérite / il obtint dans chaque existence une

Page #9

REVIEWS
79
haute naissance et fil devint) propriétaire d'un éléphant d'or de monture / et (il obtint] finalement, appelé Gardien d'éléphant, le grade d'un arhat.” As Cleaves remarked a literal translation of the first line would be: "Thus, of the De-bžin gšegs-pa (Tathāgata) of olden time, who descended from Ga'-Idan (Tuşita)" (p. 25). The man repaired the stupa of the young elephant of the Tathāgata of olden time, who descended from the Tușita heaven, riding on an elephant, as is told in the forty-ninth chapter of the Hsien-yu-ching (cf. Cleaves, p. 44). The beginning of the Tibetan stanza has to be translated accordingly in the following way: "In this way a man repaired the stūpa of the young elephant (erected in commemoration of the elephant on which] the Tathagata of olden times descended from the Tuşita palace."
Ligeti announces an edition and translation of the Mongolian, Uigur and Tibetan texts of the inscriptions in small characters. It is obvious that he has a much less sure knowledge of Tibetan than of the other two languages. It is to be hoped that one of the many excellent Hungarian Tibetologists will take care of the edition and translation of the Tibetan text.
Australian National University
J. W. DE JONG
NOTES
1 The Mu-tig phren-ba which was not identified by Taube (cf. p. 176, n. 1) is found in the Tanjur (mdo-'grel, Vol. CXXXIII, 2). The full title is Mron-brod-kyi bstan-bcos sna-tshogs gsal-ba tes-pa min-gżan mu-tig phrer-ba. 2 Cf. Taisho, Vol. 16, No. 694, p. 693c7-11. The second part of this text was translated by Stephen Beyer, The Buddhist Experience: Sources and Interpretations (Encino, Cal., 1974), pp. 47-55. For the translation of the passage referred to by Nagao see p. 48, lines 12-21: "rich and honorable ........ the king of the gods."
Practice and Theory of Tibetan Buddhism. Translated and Annotated by Geshe Lhundup Sopa & Jeffrey Hopkins. London, Rider and Company, 1976. xxviii, 164 pp. £ 2.95.
The Practice and Theory of Tibetan Buddhism contains translations of two Tibetan texts. The first is a meditation manual written by the fourth Panchen Lama Blo-bzan dpal-ldan bstan-pa'i ñi-ma (1782-1853) as a commentary to a short letter in verse by Tson-kha-pa on the three principal aspects of the path to highest enlightenment. The Tibetan title of this work is: Gsun. rab kun-gyi snin-po lam-gyi gtso-bo rnam-pa gsum-gyi khrid-yig głan-phan sñin-po, which is rendered by the translators as Instructions on the Three Principal Aspects of the Path, the Essence of All the Scriptures, the Quintessence of Helping Others (cf. p. 154). The second text is the Grub-mtha'rnam-bżag rin-chen phren-ba written in 1773 by Kon-mchog 'jigs-med dban-po (1728-1791). The nature of this work is clearly indicated by the detailed title given in the colophon: Phyi-nan-gi grub-mtha'i rnam-bżag mdor bsdus-pa rin-po-che' phren-ba, “A brief exposition of the non-Buddhist and Buddhist tenets (called) a precious garland.” Several portions of this work were translated by Herbert V. Guenther in his Buddhist Philosophy in Theory and Practice (Harmondsworth, 1972). In 1977 K. Mimaki published a critical edition of the Tibetan text: Le Grub mtha'rnam błag rin chen phren ba de dkon mchog 'jigs med dbań po (1728-1791), Zinbun, number 14 (The Research Institute for Humanistic Studies, Kyoto University), pp. 55-112.
The translation of the meditation manual is preceded by an introduction (pp. xiii-xxviii) which is very helpful for a better understanding of the text. No such help is given with regard to the second work. According to the translators readers new to Buddhism will find instruction

Page #10

80
REVIEWS
on doctrinal matters in the Precious Garland of Tenets. Although the translators have added explanations, it is very unlikely that readers unfamiliar with Buddhist philosophy will derive much profit from the study of this work. The Precious Garland summarizes in a highly condensed and systematic form the main philosophical tenets of non-Buddhist and Buddhist schools. These philosophical manuals (grub-mtha', Skt. siddhanta) were very popular in Tibet in the eighteenth century. From them one can learn which doctrines Tibetan scholars of the time considered to be the main ones of the four Buddhist schools distinguished by them (Vaibhāşika, Sautrāntika, Yogäcära and Madhyamika). As an introduction to Buddhist philosophy in India and Tibet the Precious Garland and other works of the same genre are certainly not very convenient, because they can only be properly understood by specialists who know the basic Indian texts and doctrines to which they refer.
The translation is accurate though not always easy to understand. The translators render all Tibetan terms quite literally, without taking into account the Sanskrit originals. As I remarked already apropos of the translation of the first chapter of the Srags-rim chen-mo by Jeffrey Hopkins (IIJ 22, pp. 261-262), without knowledge of the Indian background it is not possible to translate Tibetan Buddhist texts whose entire philosophical terminology is derived from Sanskrit sources.
Australian National University
J. W. DE JONG
NOTES
The dates 1782-1853 are given by Zuiho Yamaguchi in his description of the biography of the fourth Panchen Lama: Catalogue of the Toyo bunko collection of Tibetan works on history (Tokyo, 1970), p. 69. According to the translators he lived from 1781 to 1852/4. . 2 For bibliographical information see the studies mentioned by Mimaki, op. cit., p. 60, n. 1.

Page #11

REVIEWS
Herman Lommel, Kleine Schriften. Herausgegeben von Klaus L. Janert (Glasenapp-Stiftung, Band 16). Wiesbaden, Franz Steiner Verlag GmbH., 1978. XX, 626 pp. DM 88,-.
The sixteenth volume of the publications of the Glasenapp-Stiftung contains a selection of the work of Herman Lommel (1885-1968). Several of his articles relating to Avestan studies were reprinted already in Zarathustra (Darmstadt, 1970) and his translations of the Gathas of Zarathustra were incorporated in a posthumous publication: Die Gathas des Zarathustra (Basel, 1971). For this reason the present volume does not fully reflect the importance of Lommel's work in the field of Avestan studies, to which he made such important contributions in the period between the first and second world wars. Most of the articles reprinted in this volume were published in the period 1939-1968, and relate to Vedic studies. In this field Lommel made much use of mythological and ethnological parallels and the later Vedic literature for the explanation of Vedic hymns. Lommel defended this recourse to later legends especially against Oldenberg, cf. for instance p. 235: "Im allgemeinen wird man ja bei Sagen und Legenden, wenn wesentliche Züge ihres Inhalts mit Angaben älterer Originalquellen übereinstimmen, darin ein Anzeichen finden, dass sie trotz legendärer Ausschmückung Glaubhaftes enthalten und, wo nicht ein Bild, doch ein Spiegelbild wirklicher Geschehnisse geben" (cf. also p. 488). With regard to mythological and ethnological parallels Lommel was aware of the fact that his colleagues did not always approve of his methods. Typical is the following passage from his article on 'Kopfdämonen im alten Indien': "Die ethnologischen Parallelen legen uns die Auffassung nahe, dass auch im Namuci-Mythos der abgetrennte Kopf, der seinem Überwinder drohend nachrollt, eine Erscheinungsform des Mondes sei. Solche Analogien können auf die richtige Einsicht lenken, aber doch nur eine Wahrscheinlichkeit ergeben; Beweise sind sie nicht. Darum mag der eine mehr, der andere weniger geneigt sein, solchen Hinweisen zu folgen, und es scheint, dass Indologen und Indogermanisten Parallelen aus anderen Kulturbereichen nicht gerne berücksichtigen" (p. 419). Perhaps Lommel has not always escaped the danger of being too much influenced by parallels in his interpretations of Vedic myths.
Lommel tried to study Avestan and Vedic concepts in a wider context. In his very speculative article on 'Mithra und das Stieropfer' he wrote: “Die kultische Tötung des Soma in Gestalt von Pflanzenschösslingen beim vedischen Opfer und der Mythos von der Tötung des Haoma-Stiers im Mithras-Mythos stehen also in einem weiten ethnologischen Zusammenhang und sind die arische Ausgestaltung eines urzeitlichen, prähistorischen Mythos und Kultus" (p. 208). It is obvious that Lommel was much influenced by the work of Walter Otto, Leo Frobenius, Ad. E. Jensen and C. Hentze, and this influence has perhaps not always been entirely beneficial. His attempts to study Vedic mythology in a wider context also led Lommel to study the relations between mythological concepts in the Near East and India, cf. his articles on 'Regenkamm und Himmelsrind' (pp. 524-550) and 'Again: the two-headed celestial cow' (pp. 551-559).
In his studies of Vedic gods (Indra, Mitra and Soma) Lommel went his own way and vigorously defended his views against those of other Vedic scholars who, in general, have remained rather sceptical towards his interpretations. It is a pity that his monograph on Indra (Der arische Kriegsgott, Frankfurt, 1939) was not reprinted in this volume, because it contains his most detailed study of a Vedic god.
Almost all the articles published by Lommel in Paideuma and in Symbolon have been reprinted in this volume. In reading these articles one cannot escape the impression that
Indo-Iranian Journal 22 (1980) 237-264.0019-7246/80/0223-0237 Copyright © 1980 by D. Reidel Publishing Co., Dordrecht, Holland, and Boston, U.S.A.

Page #12

238
REVIEWS
Lommel was more inspired when he was writing for a larger public. Sometimes his prose is animated by a lyrical tone which is rarely found in the works of contemporary Vedic scholars, cf. p. 179: "Und er (der in der Natur beheimatete Mensch] spürt den aufsteigenden Dunst, er wittert die Düfte der Nacht und lauscht dem Wehen des Windes und dem Rauschen der Flügel ziehender Vögel, er sieht die Wolken vor dem Mond vorbeiziehen, und sein Denken und Träumen durchfliegt die weiten Räume des Luftreichs, bis kühler Tau, der sich von oben senkt, ihn gegen Morgen das Feuer lebendiger schüren lässt."
Professor Janert has contributed a preface and added three indices (Sach- und Namensregister; Verzeichnis der Textstellen; Liste der behandelten Wörter, Stämme, Wurzeln) which greatly facilitate the study of this volume. Misprints have not been corrected. On p. 379 Frazers "Golden Bow" has to be corrected to Frazers "Golden Bough" and on p. 413, n. 2 in JAOS 15 (1981) S. 143 f.' 1981 is a misprint for 1891. The bibliography does not mention that Die Religion Zarathustras nach dem Awesta dargestellt (Tübingen, 1930) was reprinted in 1971 by Georg Olms Verlag in Hildesheim.
Australian National University
J. W. DE JONG
Harry Falk, Quellen des Pañcatantra, Wiesbaden: Otto Harrassowitz, 1978, IV, 205 pp. (Freiburger Beiträge zur Indologie, Band 12).
It is not within the purview of this doctoral thesis, despite its title, to reveal the sources, written and oral, of each and every frame and emboxed fable of the Pancatantra. Given our present knowledge of Sanskrit fable literature before the data of the composition of the Pañcatantra, that would be a most difficult task.' Dr. Falk has preferred to study in detail the sources of three frame and three emboxed fables, which he has traced in the Mahābhārata and in the Jā takas. He has primarily confined himself to the first three tantras (texts) of the Pancatantra as they are presented in the Tantrākhyāyika, and consequently in Franklin Edgerton's "reconstruction". By dint of a thorough and meticulous analysis of the Sanskrit texts of his source materials and those of the reworkings deliberately executed by the anonymous author of the Pañcatantra, Dr. Falk has succeeded in interpreting his findings in such a way as to enable us to understand the author's technique of composition, and thus to gain an insight into the genesis of this masterpiece of Sanskrit literature.
Dr. Falk views the Pancatantra as essentially a dharmaśāstra (and in its "reversed aspect" as a nitiśāstra). Nowhere does he mention it as a nidaršanakathā, a literary genre to which Bhoja refers in his Srngäraprakāša. Dr. Falk's interpretations are derived from those incidences in which the author of the Pancatantra has converted prior nidaršanakathās (exempla, fabulae), which may have functioned differently, into illustrative tales that inculcate principles of dharma and niti which were to be practiced by those readers who would be guided by his work, that is, by rajaputras who would eventually assume the rule of a kingdom, if we are to believe the content of the kathāmukha.
According to Johannes Hertel, each fable of the Pancatantra is to be considered a “Klugheitsfall" (case of trickery), so that the work as a whole teaches a Machiavellian doctrine, a view which Edgerton could not accept; nor can Dr. Falk, for he interprets the niti content of the Pancatantra as "defensive niti," that is, the author of the Pascatantra utilizes niti in aid of the defense of dharma. By delineating what is asādhudharma, as expressed in the niti-ślokas of Kaninka Bhāradvāja, the author of the Pañcatantra propounds in reality sādhudharma as "die Umkehrung der niti" (the reversal of niti). The arguments accompanied by illustrations which Dr. Falk advances in support of his "Umkehrungstheorie" are, in my opinion, strongly convincing, and in this theory lie the significance and value of his dissertation.

Page #13

REVIEWS
239
Equally convincing are the results of Dr. Falk's study of the parallels, which he finds in the Mahābhārata and in the Jātakas, of the Pancatantra fables to which he devotes the major portion of his work (pp. 9-164). The relationship between the initial episode of the frame fable of the second tantra and Mahābhārata V, 62 has already been discussed. The genesis of what remains is assuredly the samvāda between the cat and the mouse in Mahābhārata XII, 136, as previously suggested by Walter Ruben. Palita, the mouse of the samvāda, is described as mahāprajña (XII, 136, 21) and thus he must be considered the prototype of the nitivid Hiranyaka.
The source problem of the first tantra, LION AND BULL, is not so successfully solved by Dr. Falk. The author of the Pañcatantra, in my view, has preferred to blend several fables into one: the samvāda of Mahābhārata XII, 112 and the emboxed fable which he would later cite, LION'S RETAINERS AND CAMEL (Pañcatantra I, 9). The latter has been studied in detail by Ruprecht Geib and to him we owe the discovery of this additional source text. A further parallel may be mentioned here: Tantrākhyāyika I, 13 (Der lüstige Schakal); cf. Geib, op. cit., 137-142.
The third tantra, CROWS AND OWLS, and its second emboxed fable, BIRDS ELECT KING, are closely compared by Dr. Falk with the Ulūkajātaka and Mahābhārata X, 1. However, it must be remembered that the oral version of CROWS AND OWLS was already current in an ancient era, for it is referred to by Panini (Astādhyayi IV, 3, 125, where he quotes the compound kākolūkika).
The three emboxed fables of the Pancatantra, I, 11; II, 2 and Tantrākhyāyika IV, 3, are shown to be related to Mahābhārata XII, 135, XIII, 124 and VII, 164, respectively, and although Dr. Falk's chapter on these is a brief one (pages 145-164), it may serve as a model of source studies.
Dr. Geib's ideas on the underlying motives of the author of the Pancatantra in the composition of his work are not enthusiastically received by Dr. Falk. Nevertheless, it must be admitted that the "friendship-enmity motif” cannot be ignored. As I see it, it supports well Dr. Falk's "Umkehrungstheorie". The first tantra deals with the impossibility of friendship; conversely, the second deals with the possibility of friendship, and then, the third tantra denies this by depicting the enmity between crows and owls. On the basis of Dr. Geib's friendship-enmity theory, the germ of which is expounded by Hiranyaka in the second tantra, it is now possible for me to arrive at a different perception of the Pancatantra.
The main concern of the author of the Pañcatantra must surely have been the exposition of a theory of polity that would assure the gaining and keeping the political control through defeat of personal and state enemies as well as through union with personal and state allies. How does one recognize one's enemies? The keynote of the first tantra is found in the following prose statement from LION'S RETAINERS AND CAMEL: Saspabhujah pišitāśinas ca visamasambandhah (associations between grass-eaters and flesh-eaters are incongruous). Samjivaka and Pingalaka should never have become friends, no more than the tiger and jackal of Mahābhārata XII, 112. Friendship between vegetarians and non-vegetarians is not possible; it is an example of krtrimam vairam, and it is well expressed in these words: aham annam bhavan bhokta katham pritir bhavisyati? (I am your food; you are my eater; how shall there be friendship between us?) Nor should there be friendship between those of different genus and species, even though their daily activity may be similar; cf. verse 51 of the second tantra. O Yet, in the event of the rise of a paddharma, an emergency friendship may be established, and thus the associations of Mahābhārata XII, 136 and those of the second tantra, which are examples of "Umkehrung der niti". Cohabitation in the same tree between crows and owls is inconceivable, for how can diurnal and nocturnal creatures become allies? It is in the light of this that one may better understand the third tantra. The friendship-enmity motif of the fourth tantra, APE AND CROCODILE, however, is not as readily discernible, since its basis is sexuality rather subtly presented. Of different species, the vānara (male) and the married śiśumāra (male) form a friendly relationship that must be destroyed by the wife of the

Page #14

240
REVIEWS
sisumāra who, becoming mänini, looks upon the vānara as her rival. The second verse explains the reason for the break-up of their friendship:
kim kalatram pradha nam syad uta mitram gunādhikam kalatramitrayor núnam kalatram atiricyate (IV, 2)11
The aspects of krtrimam vairam which the Pancatantra has illustrated through four sections are now opposed to those of sahajam vairam of the fifth tantra, the enmity between mongoose and cobra (ahinakulavairam). The extensive discourse on vairalaksana (aspect of enmity) verbalized in the second tantra provides, in my opinion, the basis on which the Pascatantra is constructed and, at the same time, it strengthens considerably Dr. Falk's Umkehrungstheorie.
University of Toronto
GEORGE T. ARTOLA
NOTES
1 Artola, G. T., 'The Oldest Sanskrit Fables', Adyar Library Bulletin, Vols. 31-32, Madras: Vasanta Press, 1968, 281-312, surveys the earliest nidarśanas as they are employed in the parvans of the Mahabharata that predate the Pancatantra. 2 Edgerton, F., The Panchatantra Reconstructed, 2 vols., New Haven, 1924. Unless otherwise indicated, titles, translations and other quotations are from this work. 3 Raghavan, V., Bhoja's Śrngara Prakaša, Madras, 1963, pp. 620-621. 4 Hertel, J., Das Pancatantra, seine Geschichte und seine Verbreitung, Leipzig and Berlin, 1914, p. 10. 5 They are found in Mahābhārata XII, 138, and it may be said that they constitute a nitiśästra which deserves further study. 6 Artola, op. cit., 293-296. 7 Ruben, W., Das Pancatan tra und seine Morallehre, Berlin (Inst. für Orientforschung, Veröffentlichung 44, 1959), p. 128. 8 Geib, R., Zur Frage nach der Urfassung des Pañcatantra, Wiesbaden: Otto Harrassowitz, 1969, pp. 44-46. 9 Ibid., 72-77. 10 Edgerton has mistranslated this verse by having selected 'antelope' as an equivalent of sāranga, which here must refer to a bird, for the distinction must be made between mammals, reptiles and birds, just as it is made between gods, demons and men. 11 Should a wife take first place, or a friend that excels in nobility? Surely as between wife and friend the wife comes first.
G. A. Zograf, Morfologičeskij stroj novyx indo-arijskix yazykov. Izdatelstvo ‘Nauka', Moscow, 1976. 368 pp.
An immense amount of linguistic information, chiefly though not exclusively morphological in character, is assembled here and forms the subject of a minute, illuminating contrastive analysis. Stressing the importance of the interrelationship of linguistic units within their systems (p. 9), the author proceeds from an analysis of the grammatical forms and categories of individual languages to a consideration of NIE 'grammemes' (grammatical categories considered in their operation, range and distribution) and towards a characterisation of individual NIA languages or language groups in terms of what may be called 'allograms'. He concentrates on morphology as offering the easiest starting-point for a contrastive study. Zograf is conscious,

Page #15

REVIEWS
241
as the provisos of his introductory chapter clearly show, of the many difficulties of dealing with this disparate material. He notes, among other questions, the need not to overlook the importance of change and innovation in an essentially synchronic study, the risks involved in dealing with unevenly documented material, the possibilities (not always obvious) that morphological units which are formally identical in different languages will not be functionally identical, and the frequent non-morphological, especially syntactic aspect of morphological questions; in this connection it may be remarked that he has apparently not taken any of the opportunities offered by his material to refer to the distinctive ergative constructions of the central and western languages. Reference to these as syntactic structures is not necessary in itself to establish the east-west language division, but would have been a justifiable extension of the morphological study as illustrating the different range of function of instrumental and agentive forms in the languages concerned. Other matters dealt with in the introduction include relationships between language and dialect, and between literary and colloquial language. It is no doubt Zograf's grammatical interest that occasions his drawing the latter distinction most explicitly between the grammatically archaic sādhu bhāsă of Bengali and the Bengali colloquial calit bhāsa; comparable, if different distinctions can of course be drawn for other languages (e.g. for Hindi and Marathi in lexical and phonological terms). The introduction concludes with a survey of the NIA languages, including (p. 27; further p. 31) the recently discovered Parya (Par'ya) of Tajikistan.
The study is based on the data offered by the modern literary languages and also by various dialect or regional forms of speech, a majority of those referred to being located within the Hindi language area (p. 44). The differences between Zograf's and Grierson's presentation of the relationship between the languages etc. are largely accountable for by Zograf's synchronic perspective on the material. He sees the NIA language area as showing on the whole a gradual decrease in the number of synthetic inflectional forms in use, together with a corresponding increase in analytic and agglutinative forms, as one moves across the area from west to east. This view, arrived at on a synchronic analysis, has certain advantages. For instance, the distinctive structural differences which do after all exist between Marathi and Bengali, two of Grierson's outer languages, and which no doubt outweigh the affinities between these languages, are better suggested by a model of general west-east gradation of linguistic tendencies than by the model of the outer and inner languages. On the other hand, if the synthetic inflectional forms characteristic of the modern N.-W. languages are, with Grierson, relatively modern innovations (as is borne out by the status of some of them, e.g. those of Mūltāni (Sirāiki) occurring in ergative construction), it must be admitted that this fact is obscured by Zograf's synchronic classification. And yet at the same time there is a compensating factor here: in linking Marathi specifically with the N.-W. languages Zograf accommodates the modern tendency of Marathi to modify the older ergative construction by the use of synthetically inflected verbs (e.g. tū apler vacan pälilers 'you kept your word').
Six chapters deal with the main parts of speech. These chapters are organised in terms of grammatical categories such as gender, number and case (the noun), thematic or athematic formation (the adjective), semantic reference (the pronoun) and morphological types (the verb). The subsection of the chapter on nouns dealing with direct object forms and constructions (pp. 55 ff.) well illustrates the care taken generally in the work to amass and present facts in a balanced way, and the author's interest in proceeding beyond individual grammatical facts towards an assessment of typological implications. This subsection is entitled Kategoriya personal'nosti (oduševlennosti) 'Category of personal (animate) reference'. Zograf identifies personal reference as the factor chiefly inducing suffixation of direct objects (p. 62), and with this one is ready to agree.insofar as the category personal' may here be taken as falling within the category 'individualised' or 'of contextual importance'. Zograf however would separate the categories of personal and individualised objects (p. 57). He is perhaps partly led to do this by overlooking in his discussion on pp. 57-8, as it seems, the Hindi construction of personal objects without ko, e.g. as seen in such an example as darzi bulāo *call a tailor';

Page #16

242
REVIEWS
another example of this type of construction, illustrating what is for Zograf an individualised object, is hamārā gandhi vāpas karo 'give us back our Gandhi' (title of a Hindi novel).
A few further points are discussed briefly below: p. 57. The translation on poslal slugu 'he sent a servant' requires contextual comment if it is to be attached to Hindi usne ek gulām ko bhejā, since the primary sense of modern Hindi gulām is 'slave'. In the sense 'servant' the word has, in Hindi, a pejorative connotation (as seen, for instance, in jis launde ko apnā gulām samjhā, uskā murh na tā kegi 'she wouldn't look for support to the lad she'd considered as her servant', Prem cand, Mansarovar, I.1). But no such sense is evident in Zograf's example as it stands. p. 156 f. Since various dialect forms and usages are mentioned in the section on honorific pronouns it might have been worth stressing that many Hindi speakers use a two-term rather than a three-term honorific system. Use of ham for mais, implied to be dialectal only, is somewhat more than this, being common in standard Hindi as used in the eastern Hindi area, in particular. Perhaps this usage is due in part to some memory of the form haur < OIA aha(ka)m 'T', which is not preserved in modern E. Hindi. pp. 191 ff. The description of synchronic relationships between intransitive, transitive and causative stems might have been expanded at one point with a comment on historical development: H. lutna 'to be robbed' and Av. khicab 'to be pulled' are back-formations from preexisting transitives, as is pisnā 'to be ground' (at least to the extent of showing the NIA shortening characteristic of such formations).
Similarly the frequent semantic equivalent of modern Hindi third and fourth members of groups of verbs of related stem, which is correctly noted, could have been usefully explained at the syntactic level, as an aspect of their identical construction with se: usne gārī naukar se ladā dīladvai "he had the cart loaded by a servant'. 194. The extended form hava- of H. honā, said not to exist, can be found in such a usage as burhiyā ko kahir kuch ho-havā jāe to ... 'if anything happens to the old lady ....' (S. H. Vātsyāyan, Apne apne ajnabi, 12).
This wide-ranging and informative study deserves the attention of all persons interested in aspects of the inter-relationships of the NIA languages.
Cambridge, England
R. S. MCGREGOR
Alex Wayman, Yoga of the Guhyasa mājatantra. The Arcane Lore of Forty Verses. A Buddhist Tantra Commentary. Delhi 1977 (Motilal Banarsidass) xii + 388 pp.
The Guh yasamājatantra (hereafter: GST) is one of the major Buddhist tantras. It was first published (with a lengthy Introduction) by Benoytosh Bhattacharya in 1931 (GOS 53, reprint 1967). Giuseppe Tucci devoted a short but illuminating study to the GST in 1935 (“Some Glosses upon the Guhyasa mājatantra', Mélanges chinois et bouddhiques, vol. II, Bruxelles 1935, pp. 338-353, reprinted in G. Tucci, Opera Minora II, Rome 1971, pp. 337-348). More recently, two new editions of the Sanskrit text have been published: S. Bagchi, Buddhist Sanskrit Texts, vol. 9, Dharbhanga 1965, and Yukei Matsunaga, Koyasan Daigaku Ronshu (Journal of Koyasan University) vol. 10 (1975), pp. 1-130. The latter edition has appeared as a monograph, The Guhyasamāja Tantra. A New Critical Edition, Osaka 1978, including a most useful introduction, the two parts of which have been published previously, 'A History of Tantric Buddhism in India' in Buddhist Thought and Asian Civilization. Essays in Honor of Herbert V. Guenther on his Birthday, Emeryville, Calif. 1977, pp. 167-181, and 'Some Problems of the Guhyasa majatantra' in Studies in Indo-Asian Art and Culture, vol. 5, New Delhi 1977, pp. 109-119.

Page #17

REVIEWS
243
As far as I know, only one chapter of the GST has previously been translated, and that only in part, viz. ch. VII (1-13, 28-38) by Professor David L. Snellgrove in Buddhist Texts through the Ages (ed. E. Conze), Oxford 1954, pp. 221-224.
The author of the present volume is Professor Alex Wayman of Columbia University, New York. In the Preface, he calls it a "modest incursion into the vast Guhyasamāja lore" (p. viii). In fact it is a far-ranging study of the Buddhist tantras, based not only on the GST itself, but above all upon the vast commentarial literature in Sanskrit as well as Tibetan.
The Buddhist tantras present an intricate and difficult field of study, as anyone who has attempted to penetrate even a small sector of this vast literature will be painfully aware. Before proceeding, I would like to express my admiration for Professor Wayman's courage in attempting to present a broad survey of the basic ideas of the GST and for his learning which enables him to make extensive use of the Tibetan commentarial literature, particularly the works of Tsongkhapa. If in the following I will feel obliged to make a number of critical observations, this is only a token of the attention with which I feel his book deserves to be studied.
The book as a whole may, I feel, be regarded as a commentary or exegesis of the GST. However, it may be helpful if the textual material which is included in the book and on which it is based, is briefly surveyed. I. GST (a) ch. VI and ch. XII in extenso (translation only), pp. 23-35 (VI, 3-6 also pp.
169-170). (6) Other verses cited and translated from the GST, text emended by Wayman with
the help of the Pradipoddyotana (see below) and the Tibetan translation. A list of these verses may be helpful for readers, all the more so as Wayman's references sometimes are to the verses and sometimes only to the page of Bhattacharya's edition (in which the verses are not numbered). Below, all verses have been identified and numbered according to Matsunaga's edition. Reference to Wayman is given in parentheses. 1,4 (244); II,3 (280); III,12 (217); IV,3 (193), 9 (133), 19-21 (303-4); V,9 (307); VII, 14 (252), 33 (265), 34 (286), 35 (188); VIII, 14-15 (258); IX,14 (263), 21 (309); X,9 (249); XI,30 (255), 35 (257), 37 (256), 40-44 (282-83); XII,16 (250); XIII, 26 (292), 56 (248); XIV,39-40 (327); XV.13 (323), 22 (341), 32-34 (236-237), 125 (256), 135 (333); XVI,33 (290), 41 (292), 61 (308), 66-67 (201), 87-88 (291), 89-90 (259-60), 94 (260), 97 (358); XVII,37 (270), 50 (229), 51 (234), 62 (265), 67 (259), 72-75 (299300); XVIII, 24 (164), 25-31 (139-40), 32 (299), 39 (285), 45-47 (207-8), 69 (293), 77-78 (288), 84 (142), 107 (218), 111-112 (146), 131 (323), 135136 (156), 137 (156), 162 (290). The Pradipoddyotana (hereafter: PU), a commentary by Candrakīrti on the GST: a passage commenting on GST XII, 60-64, pp. 35-50, containing: (i) Sanskrit text, edited on the basis of an unpublished ms. belonging to the K. P. Jayaswal Institute, Patna;
(i) English translation. (b) Frequent quotations from this ms. of the PU, often embodying passages from
other texts which otherwise only exist in Tibetan translation, e.g. the Vajrosnisatantra, the Samdhivya karana, the Devendrapariprcchā, etc., always with Sanskrit
text and English translation. III. A collection of forty verses, each commenting on a syllable of the nidāna-formula
at the outset of the GST (i.e. evam mayā śrutam ekasmin samaye bhagavān sarvatathāgatakāyavākcittahrdayavajrayosidbhageşu vijahāra). The verses are originally to be found in a commentary on the GST, styled Vajramālā (hereafter: VM); as the original Sanskrit version of the VM is lost, the verses have been excerpted from quotations of the VM in the PU, as well as from the Tibetan translation of the VM in the Tanjur. The Sanskrit and the Tibetan text, with translation, is given on pp. 1-22. The collection of verses is given the title

Page #18

244
IV.
V.
REVIEWS
(a) the Pancakrama by Nagarjuna, the Sanskrit text of which edited by Louis de la Vallée Poussin, Études et textes tantriques. Pancakrama, Gand 1896. Note that what according to this edition is ch. I is in fact a separate work, the Pindikṛtasadhana; hence ch. II of de la Vallée Poussin's edition is referred to as ch. I by Wayman, etc. A list of those verses which are quoted and translated in the present volume is given below: Pindikṛtasadhana: 43-44 (192), 51-52 (159), 56-60 (209-10), 91-93 (249
P. 25
Guhyasamajanidānakārikā by Wayman. A major part of the present volume consists of the author's own commentary on these forty verses (pp. 181-331). Extensive use is made of Tsongkhapa's subcommentary on the PU (short title: mChan-'grel), as well as other works by Tsongkhapa. The Tibetan text is not always given.
Other tantric texts of which substantial portions are quoted:
P. 26
I now pass to a number of instances where I would suggest that the translation found in the present volume may be corrected or improved.
50).
Pancakrama (Wayman's numbering): 1,2 (297), 3 (198), 19-23 (72), 58 (297), 66 (74); II,5 (191), 8-13 (185), 15 (273), 16-21 (187), 23 (273), 24-25 (188), 29 (186), 30 (321), 32-34 (196), 36 (264), 37 (263), 44-47 (227-28), 50 (239), 53 (188), 57-62 (188-89), 79 (145); III,12 (330), 19 (281), 20 (297), 23 (281), 25 (330), 26-27 (267), 29-30 (297), 31 (327), 36 (293); IV,3, 5-7, 12-13, 16 (320-21), 17 (321), 25-27 (280-81), 31 (284); V,2 (312), 12 (31314), 18 (312), 25 (312), 26 (314), 34-35 (309).
(b) The chapter on the Akṣobhya-mandala in the Nispannayogavali by Abhayākaragupta, English translation pp. 126-130. The Sanskrit text is not given, as it may be found in the edition of Nispannayogavali by Benoytosh Bhattacharyya, GOS 109, Baroda 1949, reprint 1972, pp. 5-7.
from ch. VI of the GST: Wayman: "the samadhi called 'Dustless Diamond Abode"" (virajapadavajram nāma samādhim). This would presuppose *virajavajrapadam, so I suggest "The samadhi called 'Diamond of the Immaculate State"".
ch. VI,1: I suggest the following translation, "The triple vajra (i.e. body, speech and mind) of the tathāgatas, which brings about (vibhāvanam, Wayman: 'which contemplates') a state, viz. the 'Secret State', one may bring to perfection by means of these invocations (ebhiḥ pravarair, referring to the preceding three mantras; Wayman: 'by these preeminences'), (the triple vajra of which each is) distinguished by a characteristic mantra".
ch. VI,3: Wayman emends the line sadhayet pravarām siddhim manaḥ santosanapriyām to... manaḥsamtoṣaṇām priyām, and translates "should accomplish... the 'surpassing one', successful one', 'one satisfying the mind', 'beloved one"". He does not indicate to what these categories might refer. I would suggest the much simpler emendation manaḥsamtoṣaṇapriyām, and translate, "he should accomplish the supreme perfection which is dear because it satisfies the mind". ch.VI,5: kāyavākcittanidhyapteḥ svabhavo nopalabhyate; nidhyapteḥ is taken by Wayman as a gen. ("The self-existence of body-, speech-, and mind-visualization is not reached..."); surely it is an abl., "By meditation on body, speech, and mind, self-nature is not obtained". In the second half of the verse, Matsunaga reads mantramurtiprayogena na bodhir na ca bhāvanā, "nor is Enlightenment or meditational realization (obtained) by the application of the body of mantras". ch.VI,6: I would (with Matsunaga) retain vidhisamyogam (Wayman: bodhisam°), cf. VI,2 buddhan pujayed vidhivat. I translate the second half of the verse, "he

Page #19

REVIEWS
245
p. 27
p. 28
p. 29
p. 31
should bring forth mentally (a state of samadhi) created by mantras, in conjunction with the proper rites". ch. VI, 19:atha vajradharah .... sarvacaryägrasambhūto bhāsate guhyam uttamam; Wayman takes sarvacaryāgrasambhūto as qualifying guhyam uttamam, which is not possible ("... expressed the sublime secret that issues from all the best praxis"). The translation must be, “Vajradhara, ... possessed of the best of all practices, explained the highest secret". ch. VI, 24: Wayman ("The wise bodhisattva becomes dear to the Buddhas") follows Bhattacharya's reading priyo bhavati buddhānam bodhisattvas ca dhimatām. I would prefer to follow Matsunaga, who reads bodhisattvänām. In no case can dhimatam (gen.pl.) qualify bodhisattvas (“The wise bodhisattva ..."). Hence I translate, "He becomes dear to the buddhas (and) the wise bodhisattvas ...". ch.XII, introductory line: srastā "creator" is not translated by Wayman. ch.XII, 11 rudranamaskytah is not "Having made the obeisance of Rudra", but rather "Having been saluted by Rudra". Likewise XII, 15 daityabrahmendranamaskytah is not “having bowed to Brahmā, Indra, and the Daityas" (Wayman), but on the contrary, "Having been saluted by Brahmă etc.". The adept is elevated above the gods of the three worlds referred to here; he is traidhātu kamahāvajro (11) and traidhātu kamahāpujyo (15). ch.XII,35: ekalinge sivälaye is translated by Wayman "an ekalinga, or in a calm place''; however, I wonder whether it is not a reference to "a temple of Siva, having an ekalinga"? ch. XII,37: väyavamandala is translated "the windy mandala", which is less than clear; is it a reference to an 'internal mandala', created by the internal 'wind'? line 6 ff.: Wayman translates bālā and madhyamāḥ as 'children' and 'middleaged' respectively. However, they represent levels of understanding, completed by uttama buddhayaḥ in the same verse, and not stages in the human life-cycle. XVIII,112 prajnajnanam as the third initiation is translated 'insight-knowledge'. For a discussion of this and related terms I refer to my article "On the Concept of Sahaja in Indian Buddhist Tantric Literature", Temenos, vol. 11 (1975), Helsinki 1976, pp. 88-135. line 13: yum gyi bhaga rten is rendered "the mother's bhaga place". However, it is quite certain that rten is simply a gloss on the Sanskrit term bhaga, thus, "E is the mother's bhaga, i.e. support". PK 11,58: abhimănatah is translated as 'thought'; however, abhimana is a conceited thought, a high opinion of oneself. Hence I would prefer to translate as follows: "The Tathāgata ... entered the 'breath-holding' concentration with the conceited thought, 'I shall obtain buddhahood by means of the 'Great Void". (The passage goes on to relate how the Tathāgata is corrected by all the Buddhas and hence abandons this samadhi). PK 11,32-34: there are a number of printing errors in the Sanskrit text not included in the list of Errata: vayunā, read vāyunā; vāhantām, read vāhanatām; tatprakstāḥ, read prakrtih; vāyus tam prakrtim, read vāyus tam tām prao. line 5: rca'i bye brag dan rjes su mthun pa is translated by Wayman "consistent with the basic multitude (of deeds)". However, rca here means 'vein (nādi)', and bye-brag is division (bheda)', thus, 'consistent with the division into veins', the reference being to the three veins dealt with in the same passage.
p. 144
p. 146
p. 183
p. 188
p. 196
p. 197
Although dealing with detail, these notes are perhaps sufficient to show that fresh attempts at translating these tantric texts would be justified. This brings me, however, to a number of more general observations concerning the translation and interpretation of Buddhist tantric texts, Wayman makes very extensive use of texts which are traditionally styled 'explanatory

Page #20

246
REVIEWS
tantras' (vyākhyātantra), as well as outright commentarial texts (e.g. Tsongkhapa's mChan'grel). No-one would question the importance of these texts, and we must be grateful to Professor Wayman for making use of them to elucidate tantric terms and notions. However, it is by no means certain that the earlier texts should be interpreted in the light of the later commentaries at every stage of the process of study. However, this is what it would seem that Professor Wayman consistently does. I will quote only one example to show that this procedure may lead to an interpretation for which the basic text offers no justification.
p. 263
line 14, quoting the GST IX,14: dvayendriyaprayogena sarvāms tān upabhunjayet, which is translated thus: "By the union of the two organs... he would enjoy all those (goddesses)", the gloss 'goddesses' being supplied from the commentaries. However, sarvāms tān, being masc. (acc. pl.), cannot possibly refer to goddesses.
The commentarial tradition represents an exegetical effort, an effort to interpret the earlier and basic tantric texts in the light of later systematizations. The structure of the tantras, which often appears to be chaotic, is made intelligible by the commentaries by forcing the contents into certain more or less well-defined categories. Speaking of a particular tantric commentary, I have previously characterised it in the following words: "... this commentary represents a hermeneutical effort, in the course of which CG (i.e. the basic text in question) is subjected to a consistent exegesis. M (the commentary) seeks to discover... a relatively limited number of concepts and practices, forming a coherent system..." (An Anthology of Buddhist Tantric Songs. A Study of the Caryägiti, Oslo 1977, p. 17). An example of Wayman's reliance on later exegetical tradition is his analysis (p. 23) of chapters VI and XII of the GST in terms of utpattikrama and sampannakrama, an analysis for which, as far as I can see, the chapters in question offer no particular basis.
While fully aware of the dangers inherent in such a procedure, I would suggest that at the present stage of tantric studies, every effort should be made to understand and translate the basic tantric texts without reference to later tradition. Certainly many problems will remain unsolved; nevertheless, on the whole this should not be intrinsically more difficult than translating the Upanishads or a text like the Bhagavadgītā. Only then would it be possible to move on and ascertain to what extent a particular commentary interprets a basic text in terms of its own philosophic and ritualistic particularities. I grant that occasionally this ideal procedure simply becomes impossible because of the obscurity of a given text (I have personal experience of this from my study of the Caryagitikoa!) and that a commentary may occasionally be indispensable for elucidating details; still, I believe that working from the older to the newer texts (with due acknowledgement of the approximate nature of the chronology) one may be able to provide a study of a concept, a ritual, or (as in the present case) an entire textual cycle, with greater clarity and inner consistence. Thus, in spite of Wayman's imposing knowledge of tantric literature, his extensive use of the commentaries occasionally - as an unintended sideeffect - gives his book a somewhat unorganised look; his learning has tempted him to include simply too much explanatory and exegetical material.
To sum up, I would entirely agree with Wayman when he points out the difficulties of interpretation and understanding which beset the study of the tantras, and that 'unwarranted judgements' may ensue (p. 54). But to refer to the various tantric adepts of India and, later, of Tibet as 'eminent authorities' and 'later investigators' is misleading, to say the least. The basic tantras and the commentarial literature present a complex corpus of texts, presenting its own peculiar difficulties even at a very preliminary level of interpretation, but in this it is not different from many other groups of texts which have quite successfully been subjected to conventional scholarly study according to historical-philological principles. The Upanishads, or the various Gnostic traditions, or the Kabbalah, immediately come to mind, to mention but a few comparable instances. I believe that we would do a great disservice to Buddhist studies by placing the tantras in a category of their own, and accepting the attitude of esotericism,

Page #21

REVIEWS
247
the need of the guru's oral explanation' (p. 54), which certainly is a feature of the tantric movement, but which the scholar must be free to evaluate and study in terms of its historical circumstances and limitations, and hence, if need be, disregard.
On one point I must express serious disagreement with Professor Wayman. Speaking of what an introduction to the tantras should be (p. 52), he states that it should show "what the Tantra is all about, the underlying suppositions, the leading instructions" (so far I certainly agree, but then he continues) "to the extent of recreating the Tantras as a viable entity to be liked or disliked" (my italics). There is no possibility of mistaking the author's views, for he continues: "The trouble with so much of the present writing on the Tantra is that the reader is, or should be, left with a feeling of distancy or bewilderment; he is neither genuinely for or against it, because he does not understand it" (my italics). If I understand this statement correctly, Professor Wayman is saying that understanding is a means to an end, the end being to become "genuinely for or against", and that only thus can "a feeling of distancy or bewilderment" be done away with. For my part, I would strongly urge that the study of the tantras, or of any other religious tradition, must be guided by entirely different motives. The historian of religions must always exercise the most scrupulous epochē when studying and interpreting a text, a phenomenon, or an entire religion; a genuine understanding is incompatible with value judgements. This is an ideal, of course; it is a truism that complete objectivity is a goal which can never be attained, but it is a goal and an ideal which must nevertheless never be abandoned. Needless to say, there is no question of anyone approaching a subject with a mental tabula rasa. Nevertheless, I strongly feel that the only tenable position is that expressed by the eminent Swedish historian of religions, Professor Geo Widengren: "What should be the accepted, objective norm according to which such value judgements could be passed? To put the question is to answer it! Such an objective norm does not exist. As a private person who feels and thinks I may approve of certain religious phenomena and detest others, but as a historian of religions I have no possibility of objectively motivating my sympathies and antipathies, and consequently no right to present my private views as being scholarly ... An attitude of evaluation deprives the scholar of that faculty of empathy and understanding which alone can lead to the goal. It goes without saying that the historian of religions, just like the historian, depends on those standards of objectivity which historical research demands". (Foreword (p. 7) of the Swedish translation of U. Bianchi, Problemi di storia delle religiani: Religionshistoriska problem, Stockholm 1968).
Lest this review of a remarkable book should end on a critical note, I would like to conclude by repeating my admiration for the author's knowledge of the tantric literature. Perhaps the book should be regarded as a fresh commentary on the Guhyasamājatantra, as indeed the author himself suggests (p. viii).
Oslo
PER KVAERNE
J. W. de Jong (editor), Nägärjuna: Mülamadhyamaka kärikäh. Adyar (Madras), The Adyar Library and Research Centre, 1977, vi + 57 pp.
Since its appearance in separate instalments more than half a century ago the standard edition of Nāgārjuna's Mülamadhyama ka kārikās has been the one published by Louis de La Vallée Poussin as volume iv of the Bibliotheca Buddhica (St. Petersburg, 1903-13). This edition was based on three manuscripts - from Paris, Cambridge and Calcutta - described by La Vallée Poussin (LVP) as mediocre copies of an original that does not seem to have been irreproachable; and LVP went so far as to observe that he considered the Tibetan version in the bsTan 'gyur as more worthy of trust than this manuscript tradition. It presents the text of the MMK con

Page #22

248
REVIEWS
tained in Candrakirti's great Prasannapada commentary written in the 7th century, in other words about half a millenium after the time of Nagarjuna. LVP's edition was most certainly a highly meritorious achievement which has rendered excellent service to scholars. But because of the insufficiency of the manuscript materials then available its readings were sometimes doubtful or corrupt; moreover, since there were several identical lacunae in the three MSS (indicating that they derive from the same original), certain passages had to be conjecturally restored by LVP on the basis of the Tibetan translation.
Some years ago Professor G. Tucci procured from Nepal a manuscript which fortunately contains almost all the passages missing in the other three MSS; and it may accordingly be concluded that the Rome MS is independent of their immediate original. For the preparation of the present edition a photocopy of this MS has been used by Professor J. W. de Jong (J) in addition to LVP's edition and the Tibetan translation of Candrakirti's Prasannapada. And J has therefore been able either to corroborate LVP's emendations or to offer improved readings, as well as to fill up lacunae in the older edition.
J's edition comprises 447 kārikās as against LVP's 448 since what the older edition had printed as verse 7 of Chapter iii is excluded here because it is a quotation from the Ratnavali (iv. 55) omitted in all other commentaries on the MMK. (Neither editor has included the two so-called introductory verses anirodham anutpadam... vande vadatām varam in the text of Chapter i.)
Improvements in reading as compared with LVP's edition are, for example, vigamanam (ii.3, instead of dvigamanam) and ajyate (instead of the unintelligible ucyate in ii.11, 22 and 23 and uhyate in xxv.17 and 18, this latter reading corroborating our conjecture in JIP 5 [1977], p. 63 n. 67). In xviii. 7ab J corrects LVP's nivṛttam abhidhātavyam nivṛtte cittagocare to... nivṛttas cittagocaraḥ; however, Candrakirti explains that pada b gives the reason for what is said in pada a, and a Tibetan version has brjod par bya ba ldog pa ste // sems kyi spyod yul Idog pas so (I do not quite understand J's note in IIJ 20, p. 227, where he writes that pas so serves to connect the first half of the verse with the second half; such a construction usually serves to express a reason and this fits Candrakirti's explanation). In xxii. 11 LVP and J both read sūnyam... aśünyam... ubhayam... nôbhayam; that is, they (no doubt rightly) have the neuter ending, although the context refers to tathagataḥ and in his Cinq chapitres de la Prasannapada (Paris, 1949), p. 80, J accordingly translated 'Ne peut dire ni qu'il est vide, ni qu'il est non-vide, ni qu'il est vide et non-vide à la fois, ni non plus qu'il n'est ni vide ni non-vide à la fois' (see our remarks in JIP 5 [1977], p. 63 n. 55). In xxiv.13 where the type is broken in the review copy of the present edition, LVP read sa sunye. In xxiv.25 both editors read duḥkhanirodhatvāt; although the Tibetan translation appears to presuppose something else (see LVP's note ad locum), the Sanskrit text adopted in both editions accords with Candrakirti's explanation (J has no comment on this passage in his text-critical notes in IIJ 20). In xxv.14cd, instead of LVP's conjecture na tayor ekatrâstitvam alokatamasor yatha, J reads with the Rome MS tayor abhavo hy ekatra prakāśatamasor iva; this case provides an instructive example of how difficult it can be even for as competent and meticulous an editor as LVP to reconstruct a Sanskrit reading accurately in the absence of reliable manuscript evidence and on the basis only of the Tibetan translation, even if he does succeed in conveying the approximate sense. On the other hand, in xxvii.lab LVP's conjecture abhum atītam adhvānam nabhum iti ca dṛṣṭayaḥ came remarkably close to abhum atītam adhvānam nâbhūvam iti dṛṣṭayah in J (who slightly corrects the reading of the Rome MS).
The text of the MMK is followed by an index of the Kärikäs (pp. 45-57). This edition contains no critical notes or apparatus, and the reader can locate J's changes only by comparing it carefully with LVP's edition and the text-critical notes on the Prasannapadā (and the MMK) being published by J in this Journal on the base of the fourth Nepalese MS now in Rome and the Tibetan translation (see IIJ 20, pp. 25-59 and 217-52). It would have been most helpful to the reader if in his text-critical notes J had marked the verses of the MMK by a typographical device and a reference to the Kärikä-number.

Page #23

REVIEWS
249
All work on Nagarjuna's great treatise will henceforth have to make use of this convenient new text of the MMK together with J. W. de Jong's invaluable notes on the Prasannapadā and MMK.
Seattle
D. SEYFORT RUEGG
Helmut Eimer, Skizzen des Erlösungsweges in buddhistischen Begriffsreihen. Eine Untersuchung. Bonn, 1976, xvi + 184 pp. (Arbeitsmaterialien zur Religionsgeschichte, 1.)
M. Eimer présente et met en relation les unes avec les autres de nombreuses listes de termes techniques relatives au chemin de la délivrance dans le bouddhisme. Ces listes sont tirées avant tout du canon pali; mais le Grand Véhicule est aussi pris en considération. Voici les principales: (1) aṭṭhangika magga; (2) sila-samadhi-paññā; (3) les 37 bodhipakkhiya dhamma; (4) sotāpanna, sakadāgāmin, anāgāmin, arahat; (5) diverses listes tirées du Maha-Sakuludayi-sutta (Majjhima ii. 1-22); (6) les 4 carya et les 10 bhumi du Mahāvastu; (7) les 6 et les 10 pāramitā; (8) les bhūmi d'après les Prajñā-pāramita et d'après le Dasabhumika-sutra.
Lausanne
L'ouvrage de M. Eimer représente la teneur d'un travail effectué en 1967 (p. 1). Ce qui explique que la bibliographie, d'une exhaustivité méritoire, soit déjà un peu ancienne: n'y figure pas, par exemple, le tome III du Traité de la grande vertu de sagesse (1970), dont le chapitre XXXIV a apporté bien des éclaircissements sur les vimokṣa, abhibhv-ayatana, kṛtsnāyatana (cf. p. 52-53).
L'ouvrage est assez schématique, étant donné l'énormité de la matière traitée. L'idée que dans tous les cas les termes antécédents d'une liste conditionnent les termes suivants demanderait sans doute à être nuancée, même pour les écoles anciennes (p. 5). Les parallélismes semblent parfois forcés, notamment celui que M. Eimer cherche à établir (p. 37-38) entre le chemin octuple et la triade sila-samadhi-pannā.
En revanche, le travail de M. Eimer est d'une très grande valeur comme recueil de sources, notamment par les références très complètes et soigneusement ordonnées à la masse toujours un peu indigeste du canon pāli.
JACQUES MAY
The Gilgit Manuscript of the Sanghabhedavastu. Being the 17th and Last Section of the Vinaya of the Mulasarvästivädin. Part I-II. Edited by Raniero Gnoli, with the Assistance of T. Venkatacharya. Roma, Istituto per il medio ed estremo Oriente, 1977-1978, 2 vol., xxviii+ 234 pp.+ 13 pl, xi + 310 pp., facs. (Department of Archaeology of Pakistan; Istituto per il medio ed estremo Orientę.) (Serie Orientale Roma, Vol. XLIX/1, XLIX/2.)
The Gilgit Manuscript of the Sayanasanavastu and the Adhikaraṇavastu. Being the 15th and 16th Sections of the Vinaya of the Mulasarvästivadin. Edited by Raniero Gnoli. Roma, Istituto per il medio ed estremo Oriente, 1978, ix + 117 pp. (Department of Archaeology of Pakistan; Istituto per il medio ed estremo Oriente.) (Serie Orientale Roma, Vol. L.)
Le manuscrit du Vinaya des Mulasarvāstivādin trouvé à Gilgit en 1931 ne parvint pas dans son intégrité à celui qui devait en être le premier éditeur d'ensemble, Nalinaksha Dutt. (Onze feuillets avaient été édités par Sylvain Lévi dans le Journal Asiatique 220, 1932, pp. 1-45; cf.

Page #24

250
REVIEWS
XLIX/1, p. xiii.) N. Dutt publia ce qu'il avait à sa disposition. Son édition constitua le volume III (en quatre parties) des Gilgit Manuscripts: Gilgit Manuscripts, Vol. III, ed. by Nalinaksha Dutt. Part 1, 2, 3: Srinagar, 1947, 1942, 1943. Part 4: Calcutta, 1950. Dans le premier des trois volumes qui font l'objet du présent compte rendu (XLIX/1, p. xiv), la première partie de l'édition de N. Dutt est donnée comme sans date (cf. Bibliographie bouddhique, XXIXXIII, No. 107). La page de titre de l'exemplaire que j'ai sous les yeux ne mentionne en effet pas de date. Mais elle est précédée d'un feuillet (peut-être rapporté) portant un titre de collection et une date: The Kashmir Series of Texts and Studies, No. LXXI (E), Srinagar, Research Department, 1947.
Rappelons que, si l'on se reporte aux tableaux de l'édition Dutt (Part 2, pp. i-ii) et de la présente édition (XLIX/1, pp. xvi-xvii), l'édition de N. Dutt s'organise de la manière suivante: Part 1 = vastu 6, incomplet; Part 2 = vastu 7 à 10, complets; Part 3 = vastu 11 à 14, complets, plus le début du vastu 15 (Sayanasanavastu); Part 4 = vastu 1 à 5, incomplets, plus la fin du vastu 17 (Samghabheda-vastu). Le manuscrit à la disposition de N. Dutt présentait donc des lacunes dans les six premiers vastu, mais surtout une lacune considérable vers la fin: manquaient en effet la fin du Sayanasana-vastu, tout l'Adhikarana-vastu, et le Samghabheda-vastu presque entier: or ce dernier est le plus long des 17 vastu du Vinaya des Mulasarvästivādin.
Dans l'avant-propos qu'il a donné au volume XLIX/1, p. ix, M. G. Tucci raconte comment, au cours d'une mission archéologique au Swat, il a retrouvé dans une collection privée de Rawalpindi la portion égarée du manuscrit de Gilgit. Cette découverte eut lieu en 1956 (ib., p. xiv); M. E. Lamotte la signale dans son Histoire du bouddhisme indien, Louvain, 1958, p. 727, en la situant toutefois en Afghanistan.
Or, ce 'manuscrit de Rawalpindi' comble exactement la grosse lacune de celui de N. Dutt. C'est lui1 qu'éditent MM. Gnoli et Venkatacharya. Pour quelques passages endommagés ou manquants, les éditeurs recourent à des textes parallèles (Catuṣpariṣatsutra), ou fournissent la version tibétaine, tout étant clairement expliqué en lieu et place (XLIX/1, pp. 101, 151, 154; XLIX/2, p. ix, Remark; et les appendices des deux volumes). Pour avoir un Samghabhedavastu complet, les éditeurs republient la fin de cette section, bien qu'elle ait déjà été éditée par N. Dutt.
Dans le Vinaya des Mūlasarvāstivädin, le Samghabheda-vastu proprement dit, c'est-à-dire l'histoire du schisme de Devadatta, est précédé d'une corpulente biographie du Buddha depuis l'origine du monde jusqu'à la conversion d'Ananda. D'où l'extension de ce 17e et dernier vastu. (Rappelons que la version tibétaine de la biographie en question a servi de base à W. W. Rockhill pour sa Life of the Buddha, Londres, 1884, réimprimée il y a quelques années.) Cf. XLIX/1, p. xxiv. Dans la présente édition, le Samghabheda-vastu proprement dit commence à la page 68 du tome 2.
L'introduction de M. Gnoli au volume XLIX/1 contient des indications sur ce qu'étaient les Mūlasarvästivädin, leur rapport aux Sarvästivädin, leur histoire. On a tendance à faire remonter la date de leur Vinaya, et à considérer que "the Vinaya of the MSV is not, in fact, a homogeneous work, marked by a unity of conception, but rather an aggregate of different texts, laid down in different epochs, and subsequently patched up together" (p. xx).
Félicitons les éditeurs pour la publication de ces trois beaux volumes, qui donnent un texte suffisamment sûr et lisible pour une grande portion d'un des monuments de la littérature des écoles bouddhiques anciennes.
Lausanne
NOTE
JACQUES MAY
Ou plus exactement la portion relative au Vinaya: le 'manuscrit de Rawalpindi', en effet, contenait en outre un long fragment de l'Aṣṭādaśasāhasrikā Prajñāpāramitā (XLIX/1, p. ix).

Page #25

REVIEWS
251
La Saveur de l'Immortel (A-p'i-t'an Kan Lu Wei Lun). La version chinoise de l'Amṛtarasa de Ghosaka (T. 1553). Traduite et annotée par José Van den Broeck. Louvain-la-Neuve, Université catholique de Louvain, Institut Orientaliste, 1977, xii + 260 pp. (Publications de l'Institut Orientaliste de Louvain, 15.)
Le volume XXVIII du Taishō Daizō kyō se termine par huit traités d'Abhidharma (T. 1550 à 1557), d'accès plutôt malaise, car ils présentent un Abhidharma en formation, qui se cherche encore; en outre, ils n'ont pas toujours été traduits en chinois de manière satisfaisante. C'est de l'un de ces traités que M. Van den Broeck nous offre la traduction et l'étude.
L'introduction, forte de 83 pages, commence par poser quelques principes de méthodologie, d'une manière exemplairement claire et complète (pp. 1-4). Puis elle présente les problèmes historiques que soulève l'Amṛtarasa, soit: (1) celui de la date véritable de la traduction chinoise (pp. 5-11); (2) les problèmes conjoints de l'auteur et de l'époque de l'original indien.
(1) D'après la rubrique initiale reproduite dans le Taisho Daizōkyō (T. XXVIII 1553 i 966a3), la traduction chinoise aurait été faite sous les Ts'ao Wei, soit "entre 220 et 265, dans le Nord de la Chine, par un traducteur anonyme. Cette donnée est confirmée (ou peu s'en faut) par les catalogues de textes bouddhiques traduits en Chine" (p. 5). Elle est pourtant suspecte. Les traductions exécutées sous les Han et les Trois Royaumes se soucient en effet avant tout de "mettre à la portée des bouddhistes chinois des petits manuels de propagande [...] et quelques textes de Vinaya [...]" (p. 7). Il paraît anachronique de rapporter à cette époque la traduction d'un traité d'Abhidharma difficile.
Le vocabulaire chinois de l'Amṛtarasa, déjà très technique, n'est point tributaire de la méthode des équivalents taoïstes qui prévalut jusqu'à Tao-an (312-385) (p. 8). En revanche, il se rapproche de celui de la traduction d'un autre texte du même groupe, le T. 1550, Abhidharmasara ou plutôt Abhidharmahṛdaya d'après son traducteur Ch. Willemen (p. 23, n. 34), qui, lui, a été mis en chinois à la fin du IVe siècle, par l'équipe que dirigeaient Gautama Samghadeva et Hui-yüan. Enfin, il présente des incohérences (pp. 9-10). Tous ces faits donnent à penser que la traduction de T. 1553, quelle qu'en ait été la date originelle, a pu être révisée, hâtivement d'ailleurs, sur la base de celle de T. 1550 (v. notamment p. 8 et n. 16).
(2) L'auteur de T. 1553 s'appelle Ghosaka (p. 12). Trois personnages au moins portent ce nom: notre auteur; un Bhadanta 'tokharien' mentionné par Taranatha, qui le situe "après la conclusion du troisième concile et peu après la mort du roi Kanişka" (p. 13); et le fameux 'deuxième maitre' des Sarvästivädin. Au terme d'une discussion qui débat des problèmes historiques fort compliqués, et qui enveloppe l'analyse de nombreux topiques traités dans T. 1553, M. Van den Broeck conclut que l'on ne peut identifier ni notre auteur avec le deuxième maître (p. 21), ni celui-ci avec le "Tokharien' (p. 82), mais que, a priori, rien ne semble s'opposer à ce que ce dernier puisse être l'auteur de l'Amṛtarasa (ib.).
Il n'est au demeurant "pas possible de déterminer avec certitude l'époque de notre auteur, ni même le milieu dans lequel son ouvrage a vu le jour" (p. 78). Ni le troisième1 concile ni la mort de Kaniska ne fournissent de repère sûr. Ce qui ressort de l'excellente analyse du texte conduite par M. Van den Broeck (pp. 25-76), c'est que l'Amṛtarasa est une adaptation de l'Abhidharmasara de Dharmaśrī (p. 78); mais l'époque et le pays de ce dernier étant mal déterminés, on n'en est guère plus avancé (p. 78-79). L'analogie des deux textes, T. 1553 et T. 1550, s'étend donc non seulement à leurs traductions chinoises, mais aussi aux originaux indiens. Il y a dans cette littérature d'Abhidharma un ensemble dont les relations internes sont malaisées à préciser. M. Van den Broeck exprime une saine réserve à l'égard des "constructions chronologiques sur une vaste échelle" (p. 78): l'histoire de la formation de l'Abhidharma dans l'Inde du nord-ouest et de sa transmission en Chine constitue "un problème d'une complexité extrême, dépassant de loin une simple question de rapports chronologiques", et où les facteurs géographiques, notamment, jouent un rôle important (pp. 76-77).
La traduction (pp. 85-236), probe et soignée, est un bel exercice d'Abhidharma, d'autant plus méritoire qu'il s'agit, dans le T. 1553, d'une scolastique en formation, exposée par un

Page #26

252
REVIEWS
auteur qui n'est pas toujours sûr de son propos, troublée par des incohérences, des bévues, et par les séquelles d'une adaptation pas toujours réussie aux thèmes doctrinaux du T. 1550, et d'une révision maladroite de la traduction chinoise. Il fallait, pour déjouer les embûches d'un tel texte, une excellente connaissance du chinois et des logomachies de l'Abhidharma indien, doublée de beaucoup de flair critique. M. Van den Broeck est resté partout maître de son ouvrage; il convient de l'en féliciter. Une remarque au sujet du titre, où il aurait fallu traduire kan fu (ou amrta) par 'ambroisie', et non par 'immortel'.
Plusieurs index facilitent la consultation du livre. M. Van den Broeck définit les principes et les limites de ces index à la p. 4; il convient notamment de garder à l'esprit que "l'index terminologique trouve son complément indispensable dans la table analytique" très détaillée qui occupe les pp. 253 à 258. On pourra regretter maintes omissions dans l'index des noms propres: par exemple, G. Bugault et E. Lamotte sont oubliés à la p. 154, J. May aux pp. 133 et 139, T. Michael à la p. 157, G. Tucci aux pp. 154 et 205, J. Van den Broeck à la p. 205. La bibliographie est importante, les matériaux japonais peu utilisés; signalons notamment que la traduction du T. 1553 dans le Kokuyaku Issaikyo est signée d'un grand nom, Mizuno Kögen, et aurait peut-être livré des renseignements utiles.
Malgré ces quelques réserves, le livre de M. Van den Broeck constitue une excellente approche d'un Abhidharma en pleine élaboration, non encore fixé dans les grandes synthèses que seront le Kośa, l'Abhidharmadipa, le Nyāyānusara.
Lausanne
JACQUES MAY
NOTE
1 On sait que le compte des conciles bouddhiques varie suivant les traditions. Celui dont il s'agit ici est le concile dit 'de Kaniska', dit aussi de Jalandhara', auquel l'article de synthèse de B. Jinananda, Four Buddhist Councils, in 2500 Years of Buddhism, 3rd reprint, New Delhi, 1971, p. 42, assigne le numéro 4.
D. S. Ruegg, The Life of Bu Ston Rin Po Che, with the Tibetan text of the Bu ston rNam thar (Serie Orientale Roma, XXXIV), Roma, Istituto Italiano per il Medio ed Estremo Oriente, 1966, xviii + 192 p., 14 pl.
Since this work was translated so long ago by Dr. Ruegg, the question naturally arises: why another review? This is a good question and I answer it with: Professor de Jong asked me to do another review of this highly important study almost two years ago. But, better late than never, perhaps we can gleam at least one new insight after all these years.
Let us condense some of the most important data that we presently have concerning Bu ston into a few short sentences to refresh ourselves. Bu ston lived from 1290-1364 during the later period of the Sa skya pa paramountcy and was a contemporary of such notables as Jo nan pa bla ma Dol bu pas ses rab rgyal mtshan, Klon chen Tshul khrims blo gros (the codifier of the rDsogs chen doctrines), the third Karma pa pontiff Ran byun rdo rje, Tson kha pa Blo bzan grags pa, and Chos kyi 'od zer (another famous master of the Kalacakra like Bu ston). Bu ston's principal place of activity was at Za lu where he went at the age of 31 and remained as abbot for 37 years passing on his see to Sgra tshad pa Rin chen rnam rgyal (who was also the author of this rnam thar). Bu ston was mainly remembered in Tibet as a "master of the Kālacakra" (Ruegg p. 14), and as a great translator of Buddhist texts, as well as an editor, writer, teacher, builder, and as a "second Buddha". Bu ston “is one of the chief bla

Page #27

REVIEWS
253
mas in the line of transmission of the Guhyasamaja... his commentaries on the Kalacakra are of equally fundamental importance; and, as the author of the rNam thar intimates, his expositions of the Hevajra and Samputa Tantras are of particular significance for the Sa skya pa tradition descending through Grags pa rgyal mtshan." (Ruegg p. XII). And, "Bu ston was responsible for the final compilation and the manuscript 'editio princeps' of the bsTan 'gyur (Ruegg p. 35), while he was at Ža lu. Of course, his famous history (chos 'byun) needs no reminder excepting that a new translation needs to be undertaken replacing the pioneer, but defective, translation made by Obermiller.
What is not so well studied is the 'doctrine' taught by Bu ston. "According to some authorities, Bu ston is to be connected doctrinally with the Yogacara-Madhyamika-Svātantrika school which descends in Tibet from Säntarakṣita and his disciple Kamalasila. Bu ston is said to resemble this school which, having determined that all dharmas are nothing but appearance of the Mind, nevertheless held that the mind (blo) as the subjective aspect (grahaka) is not established in reality.' Klon rdol bla ma however points out that, although he considers that Bu ston may resemble Santarakṣita in this respect, it is in fact difficult to be certain since Bu ston did not clearly make a general exposition of his own method." (Ruegg p. 12). Indeed, while he founded no 'school' as such, he can well be viewed as a reformer of corrupt doctrine and his attitude was to restore proper emphasis to the doctrine of the trikaya. In this respect he resembles Klon chen pa the famous master of rDsogs chen.2 His activity has three main areas: art (sku rten), religious writings (gsun rten), and building (thugs rten), (v. Ruegg p. 39). Indeed, one crucial aspect that needed further elaboration in Ruegg's important study is precisely this: Bu ston's philosophical position regarding the doctrine of the trikaya. Since Bu ston was so active as a builder of stupas one could say that he outwardly built what was inwardly felt: rten or "support" for the Buddhist dharma.
The personal tone of this biography reminds one of the Songs of Milarepa, or the poems of the Sixth Dalai Lama.3 The ill and old Bu ston says: "He is not a man of religion who does not know that unfavourable circumstances (rkyen nan) are the friend of the bodhi; when the time has come, nothing is of any use... Though prayers are offered, death will come; and if not, if one have lived virtuously, what is the difficulty?" (Ruegg p. 124). Or, compare the pangs of Sgra tshad pa when asked to follow Bu ston as abbot at Ža lu: "I had then to accept the post, the load of a zebu being (thus) laid on a goat. The see is not for one of mediocre virtue who knows only Tibetan. How can one be asked in the Kaliyuga (rtsod dus) to occupy the seat of such a second Omniscient One - this is not a dream! I have no wide knowledge of the precepts (bslab pa), no experience of meditation (bsgoms pa'i ñams myon), no accumulated wealth and so forth as recommended by the omniscient Arya; if I consider carefully, I wonder whether the vows have been kept; my mind is confused, and the fancies of body and mind separated are awkward. But, like a torrent on a precipitous mountain, this (call) was not reversed!" (Ruegg p. 147). How would you like to follow Bu ston as a professor of Buddhism? I wouldn't. We in our century can feel close both to Bu ston and his student when reading such passages.
It is true that Ruegg's translation has some dubious readings as pointed out by Stein (BEFEO, vol. 54, 1968, pp. 635-639) and de Jong (TP, 2nd ser., 1968, vol. liv, 1-3, pp. 168-172), but we should be happy that Ruegg was able to complete this translation of Bu ston's rnam thar. Since the Serie Orientale Roma has made available the Tibetan text in Ruegg's volume, anyone that thinks he can do so is free to bring out a better translation. I agree with Stein that a technical list of terms would have increased the value of this work, but since the arrogance of most scholars keeps them from agreeing with each other anyway I doubt that it would have done any good. I would like to note in closing that the list of translations and revisions made by Bu ston, found on pp. 181 ff of Ruegg's work, is important and should be gone into by some enterprising young scholar with time on his hands.
Indiana University at Kokomo
G. W. HOUSTON

Page #28

254
REVIEWS
NOTES
1 See my dissertation: Sources for a History of the bSam yas Debate (Indiana University, Jan., 1976) soon to appear in the Monumenta Tibetica Historica series. 2 See H. Guenther's Kindly Bent to Ease Us (3 volumes, Dharma Publishing, 1975, 1976) and my forthcoming translation of Klon chen pa's History of Buddhism in Tibet to be published by Dharma Press hopefully by 1980.
See my forthcoming book: Wings of the White Crane to be published by Motilal Banarsidass.
Étienne Lamotte (tr.), The Teaching of Vimalakirti (Vimalakirtinirdesa). Rendered into English by Sara Boin (Sacred Books of the Buddhists, vol. XXXII). London, The Pali Text Society, 1976. CXVI, 335 pp. £ 22.50.
Robert A. F. Thurman (tr.), The Holy Teaching of Vimalakirti. A Mahāyāna Scripture. The Pennsylvania State University Press, 1976. X, 166 pp. $ 14.
The simultaneous publication of an English translation of Lamotte's French rendering of the Vimalakirtinirdesa and of a new translation by Thurman testifies to the growing interest in this important text. Lamotte's French translation, published in 1962, was the first translation in a Western language based upon the Tibetan version. Lamotte also consulted the three Chinese versions, especially that of Hsüan-tsang, and translated from the Chinese those passages which differed from the corresponding ones in the Tibetan version. Lamotte's lengthy introduction, his scholarly notes and the various appendices add considerably to the value of his excellent translation. This carefully done English translation will make his work accessible to a larger public. Lamotte has added a new preface and references to recent publications but, as far as I have been able to see, has made no major changes in the translation itself.
In 1970 Oshika Jisshū published an edition of the Kanjur text of the Tibetan translation (Acta Indologica I, Naritasan Shinshoji, pp. 137–240). This edition is based upon the Derge, Narthang and Peking editions. A list of important variants of the Lhasa edition was published by Bhikkhu Pasadika (A List of Important Variae Lectiones of the Lhasa Edition (Nalanda Xylograph) of the Vimalakirtinirdesa Sutra', The Tibet Journal, II, 1, 1977, pp. 77-79). Oshika published also a Japanese translation, mainly based upon the Tibetan version (Sekai no daishiso II, 2: Butten, Tökyö, 1969, pp. 203-288). In 1967 Nagao Gajin brought out a Japanese translation from the Tibetan (Sekai no meicho II: Daijo butten, Tokyo, 1967, pp. 83-194). This translation appeared separately in 1973 (Tokyo, Chuokoronsha), and a third edition of it was published in 1974 (Daijo butten, vol. 7, pp. 5-180 and 372-392)."
According to his preface Robert A. F. Thurman's goal is to present the authentic teaching of Vimalakirti, and his main focus is philosophical rather than philological. He has taken great pains with the language of the translation, which is indeed extremely readable. Although Thurman has of course carefully consulted Lamotte's French translation, he does not always follow his interpretation. It is a pity that Thurman has not consulted Nagao's excellent translation which, in several instances, is more correct than the one by Lamotte. We are fortunate to have now at our disposal three very good translations of the Tibetan version of the Vimalakirtinirdesa. It is of course not possible to compare them in their entirety but it is perhaps instructive to examine a few passages. Quotations of the Tibetan texts are made from

Page #29

REVIEWS
255
Ōshika's edition. References to Nagao's translation are to the edition published in 1973. As to Lamotte's translation, references are to the French version because it contains Lamotte's own words but, in each case, the English version has also been consulted.
Chapter five describes how through the miraculous power of Vimalakirti thirty-two thousand thrones were placed in Vimalakirti's house. The Tibetan text has: sen-ge'i khri snatshogs sum-khri nis-ston ma 'dom par son ste. The Derge edition has 'dom, the Narthang zlos (not zlogs) as stated in the English translation of Lamotte p. 140, n. 10) and the Peking 'ons. Lamotte reads ma zlogs par soms te and translates: "Ces trente-deux centaines de milliers de trônes se rangèrent sans se gêner (avivaryam prajñaptāny abhuvan)." The English translation omits 'milliers': "These thirty-two hundreds of thrones set themselves down without hindering each other." Thurman follows Lamotte's French translation: "The thirty-two hundred thousand thrones arranged themselves without crowding" (p. 51). Nagao's translation, which is based upon the Derge edition, has: "All those various thirty-two thousand thrones had plenty of space (jubun no yoyu o mochi)" (p. 83). Nagao's translation is rather free but correct. The meaning of son-ba is 'to go in, to have room in' (cf. Jäschke, p. 563b) and there is no justification for reading soms. It is also not necessary to change zlos to zlogs. Zla-ba means 'to pass, get beyond' (Jäschke, p. 491b) 'passer, passer outre' (Desgodins, p. 879b). According to the Narthang edition the thirty-two thousand thrones found place in the room without going beyond it. The meaning of 'dom-pa is 'to meet, come together'. Probably the Derge text means that the thirty-thousand thrones found place without touching each other. It is interesting to note that only Nagao correctly translated sum-khri ñis-ston with 'thirty-two thousand'.
Another instance of an unjustified change in the Tibetan reading by Lamotte is to be found on p. 134, n. 25. The Tibetan text has: lus 'di ni rtag-tu bsku-ba dan mñe-ba dan 'jig-pa dan 'gyes-pa'i chos-can te. Lamotte changes bsku-ba to 'khru-ba and translates: "Ce corps a pour loi d'être toujours baigné et massé, mais néanmoins il se brise et se détruit." Nagao (p. 27) follows Lamotte (araiotoshitari kosuttari shite mo), but Thurman translates bsku-ba correctly by 'being anointed' (p. 22). Tibetan bsku-ba (cf. skud-pa) corresponds exactly to Pali and Sanskrit ucchadana. Lamotte remarks that all modern translators have misunderstood this expression which, in its Pāli form, is anicc'ucchadanaparimaddanabhedanaviddhamsanadhammo. However, the correct explanation was alreay given in 1932 in the Critical Pali Dictionary, cf. p. 172a: "subject to impermanence, inunction and rubbing off, dissolution and annihilation" [cf. also CPD II, 8 (1973), p. 350a s.v. ucchādana].
In some passages Lamotte does not translate the Tibetan text but the Chinese text. In chapter III the Tibetan text has: dal-ba 'byor-ba yan rñed-par dka'o /mir gyur-pa yan rhed-par dka'o "To attain a favourable condition (kṣaṇasampad) is also difficult; a human destiny is also difficult to obtain." Lamotte translates: "échapper aux conditions inopportunes est difficile, acquérir une destinée humaine est difficile" (p. 182). The Tibetan translation of kṣana by dal-ba has misled Thurman who translates: "Human life endowed with leisure and opportunity is very hard to obtain. To be a human being is very precious" (p. 32). Sanskrit kṣana 'favourable condition' was translated both in Tibetan and in Chinese by 'leisure' (Ch. hsia; Jap. itoma), one of the meanings of kṣaṇa. The original Sanskrit text read probably: kṣanasampad api durlabha. Nagao (p. 47) maintains the character hsia (Jap. itoma) in his translation, but this can hardly mean 'favourable condition' to present-day Japanese readers.
In the final chapter of the Vimalakirtinirdeśa Sakyamuni tells Sakra that prince Candracchattra left his home during the life of the Tathāgata Bhaiṣajyarāja: lha'i dban-po, de-nas rgyal-bu Zla-gdugs de de-bżin gśegs-pa de bźugs-btin-du dad-pas khyim-nas khyim medpar rab-tu byun-ste "O Devendra, then prince Candracchattra left home out of faith in order to enter the homeless life of a monk while the Tathāgata was alive." A few lines later the Parinirvana of the Tathāgata Bhaiṣajyarāja is mentioned. Both Lamotte and Thurman omit the words bźugs bźin-du. Lamotte translates: "Alors, ô Devendra, le prince Candracchattra quitta la maison dans l'excès de sa foi afin d'entrer dans la vie religieuse" (pp. 383-4). Thurman has: "Then, prince of gods, the prince Candracchattra, out of his great faith in the Tathāgata, left

Page #30

256
REVIEWS
the household life in order to enter the homeless life of a monk" (p. 100). Only Nagao (p. 163) correctly translates this passage (nyorai ga mada kono yo ni aru aida ni).
The examination of these passages shows that all three translations have to be consulted in studying the Tibetan version of the Vimalakirtinirdesa. Lamotte's work will always be indispensable on account of the wealth of information given in the introduction, the notes and the appendices. However, in several places both Nagao and Thurman follow the Tibetan translation more strictly. Nagao's translation is meant for the general reader and the notes are mainly limited to the elucidation of Buddhist terms and concepts. Thurman's notes are also in the first place destined for the general reader although he discusses points of interpretation and points out a few differences between the Tibetan version and the Chinese versions. Lamotte's very learned commentary is not primarily concerned with philological problems. In comparing these three translations one is forcibly struck with the impression that a much more detailed philological commentary is required in order to make clear the problem encountered in interpreting the text. Let me quote one example. Chapter one mentions that the bodhisattvas are all perfectly skilled (niryāta) in the ten pāramitās. Lamotte translates: "accédant aux perfections du don, de la moralité, de la patience, de l'énergie, de l'extase, de la sagesse, de l'habileté dans les moyens, du voeu, de la force et du savoir (dā naśīlaksantiviryadhyānaprajñopayakausalya. pranidhanabalajñānaparamitāniryāta)" (p. 99). However, the Tibetan versions adds three expressions between dāna and sila: sbyin-pa dan dul-ba dan mi-'gyur-ba dan yan-dag-par sdompa dan tshul-khrims. Lamotte omits all three without indicating it in a note. Thurman translates: "the transcendences of generosity, subdued, unwavering, and sincere morality...”(p. 10). Nagao translates dul-ba, etc. as nouns "(three kinds of sila, i.e. discipline, unchangeableness and restraint (sanshu no kai - suna wachi kiritsu to fuhen de aru koto to yokusei)" (p. 8). However, Nagao does not explain these three kinds of sila. Of these three the first (dul-ba) and the third (yan-dag-par sdom-pa) are also found in the Chinese translations by Chih Ch'ien and Hsuantsang (Taisho no. 474, p. 519a14 and no. 476, p. 557c14). In chapter two, sbyin-pa is followed by dul-ba and yan-dag-par sdom-pa: sbyin-pa dan dul-ba dan yan-dag-par sdom-pa las skyes-pa "né du don (dāna), de la discipline (vinaya) et de la maîtrise de soi (samyama)" (Lamotte, p. 139). The same three are found in the Chinese translations by Chih Ch'ien and Hsuan-tsang (Taisho no. 474, p. 521b17-18 and no. 476, p. 561a14). This seems to indicate that these three concepts form a triad which occurs twice in the Indian text translated by Chih Ch'ien and Hsuan-tsang. In the Tibetan version this triad is found also in chapter two but, in chapter one, it is expanded by the addition of mi-'gyur-ba. Lamotte gives as Sanskrit equivalent for dul-ba vinaya but dul-ba certainly renders dama. The triad dāna, dama and samyana is common in the Pāli nikāyas (cf. Pali Tipitakan Concordance, s.v. dama). It is also found in an unidentified Sanskrit abhidharma text edited by Sudha Sengupta: katamesāṁ trayāņām yaduta dānasya damasya saryamasya ("Fragments from Buddhist Texts', R. C. Pandeya, ed., Buddhist Studies in India (Delhi, 1975), p. 195.6-71.2 This example shows sufficiently that it is necessary to scrutinize the text much more carefully than has been done so far. The philosophical meaning of the text has been brought out clearly by these three translations but the philological problems are far from being solved or even pointed out. New translations of the Tibetan version of the Vimalakirtinirdesa will certainly be welcome but only if more attention is paid to the text itself.
Australian National University
J. W. DE JONG
NOTES
Nagao published a review of Thurman's translation and of the English version of Lamotte's French translation. The Eastern Buddhist, N.S. XI, 1 (1978), pp. 109-112. 2 Cf. also Jātakamälä (ed. H. Kern), p. 100.11: dänadamasamyamadibhih.

Page #31

REVIEWS
257
Ariane Macdonald et Yoshiro Imaeda, Choix de documents tibétains conservés à la Bibliothèque Nationale complété par quelques manuscrits de l'India Office et du British Museum. Tome jer. Paris, Bibliothèque Nationale, 1978. 27 pp., 304 pl.
Le premier tome du Choix de documents tibétains conservés à la Bibliothèque Nationale contient une reproduction de presque soixante manuscrits sur 304 planches. Un deuxième tome reproduira presque cent-vingt manuscrits. Le fonds tibétain de Tun-huang compte plus de 2500 manuscrits (cf. p. 11). L'inventaire de Marcelle Lalou (Paris, 1939, 1950, et 1961) décrit 2216 manuscrits. Dans son introduction Mlle Marie-Rose Séguy ne dit pas si un catalogue des manuscrits qui n'ont pas été décrits par Marcelle Lalou est en préparation. Espérons qu'un quatrième volume de l'inventaire des manuscrits tibétains de Touen-houang sera publié sous peu. Nous nous permettons d'exprimer le voeu que ce volume contiendra une bibliographie aussi complète que possible des travaux relatifs aux manuscrits tibétains de Tun-huang à Paris. Dans sa préface M. R.-Stein remarque que les premières éditions et traductions des manuscrits de Tun-huang à Paris et à Londres sont dues surtout à J. Bacot, F. W. Thomas, et M. Lalou. Depuis la guerre le nombre des travaux s'est multiplié et il devient de plus en plus difficile de se tenir au courant des publications dans ce domaine.
Les documents à reproduire ont été choisis par Mme A. Macdonald, Mme A.-M. Blondeau et M. R.-A. Stein, tibétologues dont les travaux sur les manuscrits tibétains de Tun-huang sont bien connus aux spécialistes. M. Stein écrit: "Il [le choix des documents) a été dicté par le souci de Fendre accessibles avant tout les textes qui peuvent servir à découvrir ou à éclaircir des domaines inconnus ou peu connus de la civilisation tibétaine. C'est pourquoi on a donné la priorité aux textes non bouddhiques. Mais on a retenu les documents qui sont d'un intérêt exceptionnel pour le bouddhisme. Pour gagner de la place, on a dû omettre certains textes dont les photographies ont été publiées" (p. 7). Les savants mentionnés ci-dessus n'ont pas seulement choisi les manuscrits mais ont aussi ajouté une analyse sommaire des matières contenues dans les manuscrits reproduits (pp. 17-18) qui donne une idée des richesses de la collection de Paris. Comme l'écrit M. Stein dans sa préface, les documents de Tun-huang ont véritablement permis pour la première fois de révéler les débuts de la civilisation tibétaine, les institutions et les conceptions de la période royale qui étaient restées jusque-là plongées dans une brume épaisse (p. 5). M. Stein fait bien ressortir l'importance de ces manuscrits et la nécessité de les étudier dans l'original. La qualité des reproductions est excellente et, probablement, ce n'est que dans de rares cas qu'il faudra avoir recours aux originaux. La reproduction des manuscrits a posé des problèmes difficiles pour les éditeurs, Mme Macdonald et M. Imaeda. Pour plusieurs manuscrits il a fallu procéder à une nouvelle analyse qui, parfois, a nécessité un nouvel arrangement des folios ou des morceaux d'un rouleau. C'est ainsi que les éditeurs ont apporté des corrections ou des précisions aux analyses données par M. Lalou dans son inventaire. On trouvera des renseignements à ce sujet dans les notes préparées par les éditeurs (pp. 19-27). Ces notes énumèrent aussi les travaux relatifs aux documents reproduits dans ce tome. Elles ont été rédigées en 1972 et ne mentionnent donc pas les publications parues depuis. La communication sur An old Tibetan version of the Rāmāyana mentionnée p. 27 a paru depuis dans le T'oung Pao (LVIII, 1972, pp. 190-202). Les manuscrits de Paris et de Londres ont été édités dans The Tun-huang manuscripts of the Tibetan Ramayana (IIJ 19, 1977, pp. 37-88). Malheureusement, les dernières linges du manuscrit C ont été omises lors de l'impression (cf. p. 53). Il faut lire:
(38] [na) s // yul lan ka pu ra 'i lha dan myi (ya {n} brlag ste // srin po 'i sde tshan gis gan bkan nas // mda' sa 'gri ba (39) rgyal po byed do //
(40) // mda' sa 'gri ba rgyal po byed do //// de bżin gśegs pa dgra bcom ba yan (dag) (41) par rtogs pa'i sans rgyas sag kya thub pa la phyag 'tshal lo // (cf. pl. 284).
Nous nous réjouissons de voir que les deux volumes de reproduction sont dédiés à la mémoire de Jacques Bacot et de Marcelle Lalou. Jacques Bacot compila un premier inventaire des manuscrits tibétains de Paris. Marcelle Lalou a consacré pendant de nombreuses années une grande partie de son énergie au travail immense d'arranger les manuscrits et de les analyser. C'est

Page #32

258
REVIEWS
grâce à leurs labeurs que de nombreux savants ont pu utiliser des manuscrits et qu'il a été possible d'effectuer ce choix. Tous les tibétologues sauront gré aux savants qui se sont chargés du choix et de la publication de ces documents dont l'importance s'accroît à mesure que les études progressent.
Australian National University
J. W. DE JONG
Essais sur l'art du Tibet. Paris, Librairie d'Amérique et d'Orient, 1977. X, 188 pp. F 96,
Mme Ariane Macdonald et M. Yoshiro Imaeda, les éditeurs de ce volume, ont réuni des articles dont "toute la valeur tient dans la précision et la délimitation du champ d'investigation, et dans la capacité des auteurs à lier l'oeuvre d'art à la tradition écrite ou orale qui s'y rapporte" (avantpropos).
L'article de Mme Anne-Marie Blondeau "Le Tibet, Aperçu historique et géographique" (pp. 1-22) sert d'introduction aux articles contenus dans ce volume. Dans une "Mise au point concernant les éditions chinoises du Kanjur et du Tanjur tibétains" (pp. 23-51) M. Yoshiro Imaeda étudie les différentes éditions du Kanjur et du Tanjur imprimées à Pékin. La première édition du Kanjur date de 1410. M. Imaeda écrit que l'existence de cette édition est attestée par un article de M. Sakai paru dès 1943. Toutefois, la date et l'existence de cette édition étaient bien connues longtemps avant la parution de l'article de M. Sakai en 1944.' L'histoire de cette édition est assez obscure. M. Imaeda attire l'attention sur quelques textes chinois signalés par des savants japonais et conclut que "il est hautement vraisemblable que la copie manuscrite du Kanjur a été exécutée au Tibet à la demande des envoyés de l'Empereur, Fazun et autres à l'occasion de l'invitation de De bzhin gshegs pa en 1403, apportée en Chine à une date imprécise mais postérieure à 1403, et qu'elle a servi de base à l'impression chinoise du Kanjur tibétain réalisée aux alentours de 1410." L'édition de 1605 n'est pas pas une nouvelle édition mais un tirage fait des planches préparées pour l'édition de 1410. M. Imaeda remarque que "on fit le tirage cette fois-ci avec de l'encre noire contrairement à l'édition Yongle qui était à l'encre rouge et aux autres éditions que l'on va passer en revue qui sont toutes en rouge également." Toutefois, von Staël-Holstein mentionne un Kanjur imprimé en encre noire: "Another (incomplete) copy of the A.D. 1700 Kanjur, of which I possess a number of leaves, is, however, printed with black ink."2 Selon M. Imaeda, une nouvelle édition du Kanjur fut préparée en 1684-1692. De cette édition il y avait des tirages comportant des corrections et des additions en 1700, en 1717/1720 et en 1737.
L'article de M. R. A. Stein "La gueule du makara": un trait inexpliqué de certains objets rituels" (pp. 52-62) s'occupe du symbolisme de la gueule de makara du phur-bu. Dans "Tibetan costume, seventh to eleventh centuries" (pp. 64-81) Mme Heather Karmay étudie des illustrations de costumes tibétains dans des peintures du septième au onzième siècle, en analyse les caractéristiques principales et compare le costume tibétain à celui des peuples avoisinants. M. John Lowry étudie un carnet de croquis du quinzième siècle dans "A Fifteenth Century Sketchbook (Preliminary Study)" (pp. 83-118). Le carnet contient des portraits de maîtres tibétains, de siddhas, des dix-huit arhat, etc. La plupart des portraits peuvent être identifies par les inscriptions tibétaines et des inscriptions en écriture newarie M. Lowry écrit "The Newari inscriptions in A 5 and 32-40 appear to give the Sanskrit names of the Arhats" (p. 88). Il n'en est rien car ces inscriptions ne font que reproduire la prononciation des noms tibétains des Arhat. Ainsi, par exemple, illustration A 34 dépeint Klu'i-sde. L'inscription newarie se lit

Page #33

REVIEWS
259
Luyit alors que le nom sanskrit est Nägasena. Quant à la série des dix-huit Arhat M. Lowry dit qu'elle correspond en général avec la liste d'Atisa (cf. G. Tucci, Tibetan Painted Scrolls, Roma, 1949, p. 565) à l'exception de l'omission du numéro 13 (Bharadvaja). L'ordre des Arhat est tout à fait différent dans l'ouvrage d'Atisa alors que l'on trouve exactement le même ordre en ce qui concerne les seize Arhat dans un ouvrage tibétain écrit par spyan sna Chos grags rgyal mtshan et revisé par Sa skya pa snags 'chan Nag gi dban po Kun dga' rin chen au quinzième siècle (cf. Tucci, op. cit., p. 560).
L'article écrit par Mme Ariane Macdonald avec la collaboration de Dvags-po Rinpoche et de Yon-tan rgya-mtsho "Un portrait du cinquième Dalai-lama" (pp. 119-156) est la pièce de résistance de ce volume. Mme Macdonald prend comme point de départ une petite statue décrite par M. Pratapaditya Pal comme le portrait d'un lama rNin-ma-pa nommé Zil-gnon bzad-pa rtsal du dix-huitième siècle. Elle rappelle que Zil-gnon bžad-pa rtsal est un des noms employés par le Cinquième Dalai-lama. L'inscription sur la statue comporte deux parties. La première est un poème écrit par le moine de Zahor (le Ve Dalai-lama) pour Nag-dban ses-rab le Chef des Offrandes. La seconde partie est rédigée par Nag-dban ses-rab qui fit exécuter la statue. Cette courte inscription est élucidée par Mme Macdonald dans un long commentaire historique sur la vie du Ve Dalai-lama, les activités du régent (sde-srid) Sans-rgyas rgya-mtsho et la composition de la Maison Privée du Dalai-lama. L'article de Mme Macdonald montre bien l'importance d'une étude approfondie des sources historiques relatives à l'histoire du Tibet au dix-septième siècle.
Le dernier article est une description détaillée du Jo-khan, la “Cathédrale de Lhasa" par M. H. E. Richardson "The Jo-khang "Cathedral of Lhasa" (pp. 157-188). Pendant son long séjour à Lhasa M. Richardson a visité souvent le Jo-khan et a fait des notes sur le Jo-khan et son contenu. M. Richardson s'est servi aussi du dkar-chag, écrit par le Ve Dalai-lama, et des descriptions données par des visiteurs précédents et a consulté des fonctionnaires et des savants tibétains. Sa description du Jo-khan qui est illustrée de plans et photographies met en lumière son importance comme monument historique. Aux travaux mentionnés par M. Richardson on pourra ajouter le Voyage au Tibet (Chibetto yuki) par Aoki Bunkyo (Tokyo, 1920) qui a consacré plusieurs pages au Jo-khan.3
Les Essais sur l'art du Tibet montrent l'importance de l'étude des textes tibétains pour une meilleure compréhension de la signification et de l'histoire de l'art du Tibet. Les éditeurs de ce volume annoncent la publication d'un volume consacré à La peinture et la statuaire tibétaines qui, sans aucun doute, contiendra d'autres travaux importants sur l'art du Tibet.
Australian National University
J. W. DE JONG
NOTES
1 Eugen Pander, 'Geschichte des Lamaismus', Verhandlungen der Berliner Gesellschaft für Anthropologie, Ethnologie und Urgeschichte (Berlin, 1889), p. 201; Albert Grünwedel, Mythologie du Buddhisnte au Tibet et en Mongolie (Paris, 1900), p. 75; Berthold Laufer, Review of Hermann Beckh, Die Handschriften-Verzeichnisse der Königlichen Bibliothek zu Berlin, Band XXIV (Berlin, 1914), JRAS (1914), p. 1130, n. 1. Ajoutons que le titre de l'article de M. Sato est 'Les canons bouddhiques au Wu-t'ai-shan dans le Nord de la Chine.' M. Imaeda fait de Hua-pei 'Le Nord de la Chine' le nom d'une province. Le Wu-t'ai-shan est situé dans la province de Shansi. ? On a Peking edition of the Tibetan Kanjur which seems to be unknown in the West (Peiping, 1934), p. 12. 3 Chibetto yūki, sec. ed.: Chibetto (Tokyo, 1969), pp. 128-134.

Page #34

260
REVIEWS
Tibetan Tripitaka, Sde dge edition, Bstan hgyur. Dbu ma. Compiled and edited by K. Hayashima, J. Takasaki, Z. Yamaguchi and Y. Ejima with table of contents and bibliographical notes. Published by the Sekai Seiten Kanko Kyokai Co. Ltd. for the Faculty of Letters, University of Tokyo. 1977-Price per volume Yen 8.000.
The Department of Indian Philosophy and Indian Literature of the University of Tokyo has undertaken to publish a photomechanical reprint of the Derge Tanjur. The first section to be published is the Dbu ma (Madhyamaka) section which comprises seventeen volumes. Each volume of the reprint contains one volume of the blockprint. Two folios (four pages) are reproduced on one page. The number of the text in the Tohoku Catalogue of the Derge Tripitaka (Sendai, 1934; sec. ed., Tokyo, 1970) is indicated in the margin. Roman figures in the margin refer to the numbers of the chapters of a work. On the top of the page the folio numbers are given. The size of the pages is 210 x 296 mm. Each volume is bound in blue and carries on the spine the number of the volume: Sde dge TIBETAN TRIPITAKA Bstan hgyur dbu ma 1, etc.
The editors of this reprint have added to each volume a detailed table of contents containing the number of the text, the Tibetan title, the Sanskrit title, a Japanese rendering of the title, the names of the author, the translator(s) and the reviser(s), and the titles of the chapters in Tibetan and Japanese, and in Sanskrit if the work is available in Sanskrit. The table of contents of each text is followed by a detailed bibliography divided into five sections: 1. Editions of the original Sanskrit text; 2. Editions of the Tibetan version; 3. The Chinese version (number and title in the Taisho Daizõkyo); 4. Related works such as commentaries in the Tibetan and Chinese Tripitaka; 5. A selection of modern translations and textual studies.
The bibliographical notes will be extremely welcome to scholars, especially to those in the West, who experience great difficulties in being informed about the work done by Japanese scholars. Madhyamaka philosophy has attracted the attention of many Japanese scholars, and the number of Japanese publications mentioned in the eleven volumes published so far is impressive. The bibliographical information contained in the first four sections is meant to be exhaustive. Very little seems to have escaped the editors. Volume one contains the Tibetan translation of Buddhapalita's Mülamadhyamakavștti. No mention is made of Tachikawa Musashi's critical edition of the first part of chapter 2 (verses 1-6) on the basis of the Peking, Derge and Narthang editions: 'A Study of Buddhapālita's Mulamadhyamakavrtti (1)', Nagoya daigaku bungaku-bu kenkyū ronshū 63 (1974), pp. 16-18. Tachikawa's article contains also an. English translation of the same section (pp. 2-8). Hirano's article on the relation between the Akutobhayā and Buddhapalita's commentary is mentioned in the bibliographical notes on the Akutobhayā but no reference to it is found in the bibliographical notes on Buddhapalitā's commentary. Volume seven contains the Tibetan translation of the Prasannapada (pp. 1-100, ff. 1-200b). Chapter Twenty-seven ends on f. 19865. Then follow fourteen verses (f. 198b5200a3). The Sanskrit text of these verses is missing in the manuscripts used by de La Vallée Poussin for his edition (Bibliotheca Buddhica, IV), but it is found in a manuscript discovered by Professor Tucci. The Sanskrit text and the Tibetan translation were published in 1962: 'La Madhyamakaśāstrastuti de Candrakirti', Oriens Extremus 9, pp. 47-56. In one instance the information given by the editors is not entirely correct. Volume 11 contains Vibhūticandra's commentary on the Bodhicaryāvatāra. Ejima writes: "At the beginning of this commentary (193a3-1943) a biography of Santideva is introduced. For the first time Ejima drew attention to this fact in his article (5) - 1 and assumed that this biography is based on a text which is very similar to the Nepalese Sanskrit manuscript reported by Haraprasad Šāstri." However, Ejima's article does not contain any indication that this biography is based on a text which is very similar to the Nepalese Sanskrit manuscript. In his article Ejima wrote: "If it (this biography) has been inserted by the author himself, then it will be even older than the old biography of the fourteenth century reported by H. P. Shastri" (moshi kore ga chosha jishin no kakageta mono de aru to sureba, H. P. Shastri ni yotte hökoku-sareta jushi seiki no kodenki

Page #35

REVIEWS
261
yori mo yori furui koto ni naru de aro), cf. Indogaku bukkyōgaku kenkyū, vol. XIV, 2 (1966), p. 192. Ejima's article was mentioned and discussed in 'La Légende de śāntideva' (IIJ 16, 1975, p. 164). In the same article the identity between the Nepalese Sanskrit manuscript and the biography in Vibhūticandra's commentary was clearly demonstrated (pp. 168-172).
This excellent photomechanical reprint of the Derge Tanjur will be extremely useful. The Japanese reprint of the Peking edition of the Kanjur and Tanjur has been a great stimulus to Buddhist studies. However, it is often not possible to rely on only one edition of the Tibetan Tripitaka. Many scholars have praised the excellence of the Derge Kanjur and Tanjur. Only very few libraries possess a copy of the Tibetan Tanjur from Derge. This photomechanical reprint will make it much easier to consult this edition of the Tanjur which was prepared with great care in Derge from 1737 to 1744.2 The bibliographical notes greatly add to the value of this edition. It is to be hoped that it will be possible to publish the entire Tanjur, which comprises 213 volumes, and 1 volume for the index. From September 1977 one volume has been published each month. If this rate of publication is maintained, the entire Tanjur will be available by September 1994.
Australian National University
J. W. DE JONG
NOTES
'1 Cf. Josef Kolmaš, The Iconography of the Derge Kanjur and Tanjur (New Delhi, 1978), pp. 31-32 2 Cf. Hadano Hakuyū, 'Chibetto daizõkyo engi (2): Derge-han daizõkyo, Suzuki gakujutsu zaidan kenkyū nenpo 8 (1972), p. 25.
Tsong-ka-pa, Tantra in Tibet. The Great Exposition of Secret Mantra. Introduced by His Holiness the Fourteenth Dalai Lama. Translated & edited by Jeffrey Hopkins (The Wisdom of Tibet Series, 3). London, George Allen & Unwin, 1978. 252 pp. £ 5.95 hbk; £ 2.95 ppr.
Tson-kha-pa's two major works are the Lam-rin chen-mo and the Snags-rim chen-no. The complete title of the second work is Rgyal-ba khyab-bdag rdo-rje 'chan chen-po' lam-gyi rimpa, Gsan-ba kun-gyi gnad rnam-par phye-ba 'Exposition of the essence of all secrets. The stages on the path of the Conqueror, the Lord, the great Vajradhara'. The Snags-rim chen-mo is Tsonkha-pa's most important work on Tantrism and this translation of the first chapter is therefore very welcome. Jeffrey Hopkins's translation of the Ratnāvalī was not free from mistakes (cf. IIJ 20, pp. 136-140). The Ratnāvali is not an original Tibetan text and it is impossible to translate a work like this without a good knowledge of Sanskrit and of Indian Buddhism. Hopkins's knowledge of Buddhism seems to be based entirely upon the study of Tibetan texts and the explanations given by Tibetan masters. In view of this fact it is not surprising to see that his translation of the first chapter of the Snags-rim chen-mo is by far superior to his rendering of the Tibetan version of the Ratnāvali. It is only in rare cases that his translation is completely wrong. The most striking instance is his translation of dal-ba by 'leisure', cf. p. 166: "He [a qualified spiritual guide) will teach the ways in which leisure is meaningful and difficult to find.” Dal-ba is used in Tibetan to render Skt. ksana 'birth under favorable circumstances', cf. Edgerton, Buddhist Hybrid Sanskrit Dictionary (New Haven, 1953), s.v. Hopkins made a similar mistake in translating mi-khom by 'without leisure' and 'non-leisure'

Page #36

262
REVIEWS
(cf. IIJ 20, p. 138). Tson-kha-pa quotes many texts and Hopkins has taken great trouble in tracing these quotations and indicating their sources in the notes. His insufficient knowledge of Indian Buddhism shows itself in the fact that he has not tried to trace a quotation from the Saddharmapundarika in which one finds the following line: sans-rgyas theg-pa dman-pas 'khrid mi-mdzad //. Hopkins translates: "It is the vehicle of Buddhahood. I do not lead with a low vehicle" (p. 101). The Sanskrit text has: na hinayanena nayanti buddhah (cf. Kern-Nanjio's ed., p. 46.14) "Buddhas do not lead with a low vehicle."
Apart from a few mistakes Hopkins's translation is, on the whole, excellent. He is consistent in his renderings of technical terms, which are listed in English, Sanskrit and Tibetan in a glossary (pp. 220-224). His English equivalents are not always very well chosen and, for instance, one wonders whether any non-specialist would have any idea of the meaning of 'ground' which renders sa (Skt. bhumi). In one passage Hopkins renders sa with 'level' (p. 95). However, even the choice of such equivalents would not have prevented a careful reader from understanding the meaning of Tson-kha-pa's text if Hopkins had been more helpful in providing explanations. Tson-kha-pa's work is written in the traditional style of Tibetan Buddhist works and is full of technical terms, the meanings of which were well known to his Tibetan readers. Moreover, he quotes many texts and refers to wrong interpretations which are subsequently refuted. The first chapter of his Snags-rim chen-mo is an introduction in which he discusses the differences between the different vehicles and the division of the tantras in four classes. Tsonkha-pa treats briefly difficult philosophical problems which he has discussed at great length in other works, as, for instance, the opinions of the Buddhist schools on the non-existence of a person (cf. p. 93). Such difficult problems cannot be understood without a detailed commentary. The Dalai Lama discusses the 'selflessness of persons' in his introduction (pp. 39-42) but he fails to explain fully Tson-kha-pa's arguments and apparently finds it difficult to understand Tson-kha-pa (cf. p. 40: "He seems to be saying..."; p. 41: "Tsong-ka-pa here and in other places seems to say..."). In his supplement Hopkins studies the same problem in a chapter entitled 'Emptiness'. His exposition is based upon two Grub-mtha's written by "Jamdbyans bzad-pa Nag-dbań brtson-'grus (1648-1722) and dKon-mchog 'jigs-med dban-po (1728-1791). These works give a systematic survey of the Buddhist schools but are not very useful for explaining Tson-kha-pa's views.
Let us hope that Hopkins will also translate the remaining chapters of the Snags-rim chen-mo and so make this important work accessible to the Western public. However, without more detailed explanatory notes, a translation of such a difficult technical work will remain largely incomprehensible for the Western reader and will only be useful for scholars in the field. One would also wish Hopkins to consult the work done by non-Tibetan scholars, to which he makes no reference at all. It is no doubt of great importance to make use of works written by Tibetan scholars and of oral explanations by Tibetan masters but it is also important to take into account the fact that Tibetan scholars have very little understanding of a historical perspective in studying Buddhism and know the Indian Buddhist sources of Tibetan Buddhism only in Tibetan translations. It is in these fields that much work has been done by non-Tibetan scholars since the beginning of the nineteenth century.
Australian National University
J. W. DE JONG
NOTE.
1 Cf. Katsumi MIMAKI, 'Le Grub mtha' rnam bźag rin chen phren ba de dKon mchog 'jigs med dbań po (1728-1792)', Memoirs of the Research Institute for Humanistic Studies, Kyoto University, Zinbun 14 (Kyoto, 1977), pp. 55-112; Herbert V. Guenther, Buddhist Philosophy in Theory and Practice (Harmondsworth, 1971), pp. 53-61, 78-83, 104-112, 130-136,
143-149.

Page #37

REVIEWS
263
Manfred Taube, Tibetische Handschriften und Blockdrucke, 1-4 (Verzeichnis der orientalischen Handschriften in Deutschland, Band XI, 1--4). Wiesbaden, Franz Steiner Verlag GmbH, 1966. XVI, VII, VII, VII, 1 296 pp., 8 pl. DM 360, -
Taube's catalogue describes 3000 Tibetan manuscripts and blockprints held in six German collections in Berlin (Deutsche Staatsbibliothek, formerly Preussische Staatsbibliothek), Halle (Bibliothek der Deutschen Morgenländischen Gesellschaft), Dresden Sächsische Landesbibliothek and Staatliches Museum für Völkerkunde), Herrnhut/Oberlausitz (Archiv der Brüdergemeine) and Altenburg (Landesarchiv). The most important collection is found in the Deutsche Staatsbibliothek. In 1929 Erich Haenisch bought a complete set of the texts printed in the Sung-chung-szu monastery. The collection brought together by Haenisch contains also texts printed in other Peking monasteries. Manfred Taube points out the importance of this collection for the study of Lamaism in Peking, of Tibetan-Chinese relations in the period from the seventeenth to the nineteenth century and of the influence of the Tibetan lamaist priests on the Mongols during the Manchu period. This is the great importance of a collection which has been systematically brought together. The Deutsche Staatsbibliothek obtained by exchange since the war Tibetan block prints from the State Library in Ulaanbaatar. According to Taube all or most of the Tibetica of the Sächsische Landesbibliothek were collected by the Herrnhut missionary H. A. Zwick among the Kalmuks. The collection in the library of the Deutsche Morgenländische Gesellschaft in Halle contains copies of Tibetan blockprints and manuscripts which belonged to H. Wenzel (1855-1893). Taube makes no mention of the collection in the Landesarchiv in Altenburg which comprises 181 items (cf. p. 1293). However, 180 items are part of a Peking blockprint of a gZuns-bsdus printed in 1729 which belonged to either HansConon (1807-1874) or Georg von der Gabelentz (1840-1893), cf. No. 203. It is not clear why Taube attributes 181 texts to this gZuńs-bsdus (p. 1137). According to his catalogue the only other text in the Landesarchiv is a manuscript of a Tibetan letter probably addressed to Samuel Turner (No. 2801).
Taube's description of the texts is much more detailed than the one given in the catalogue of the Tohoku University collection of Tibetan Works on Buddhism (Sendai, 1953) which gives only the number of pages, the Tibetan title, a Japanese translation of the title and a note in English on the contents. Taube indicates carefully the size of the pages, the number of lines, the presence of miniatures, etc. Then follow the title, indications on the contents, the author and the scribe, the beginning of the text and the colophon. The proper names found in the colophon are printed in italics. The fact that the colophons have been reproduced is especially welcome because they contain often many interesting historical details. Finally, Taube gives references to editions and translations and to other catalogues. Another difference with the Sendai catalogue consists in the fact that Taube describes the texts contained in gSun-'bums, gZuńs-bsdus and Chos-spyod rab-gsal editions systematically according to their contents and not as part of the collection to which they belong. However, the appendix lists the texts in the order in which they are found in these collections. Moreover, Taube gives a complete concordance of the four gZuns-bsdus editions (pp. 1137–1144). The collections described by Taube contain twelve gSun-'bum. Among the authors one finds several famous scholars such as Thu'u-bkvan Nag-dban chos-kyi rgya-mtsho, Can-skya Qutuytu Rol-pa'i rdo-rje, Klon-rdol Bla-ma Nag-dban blo-bzan, Thu'u-bkvan Blo-bzan chos-kyi ni-ma and Jaya pandita Blo-bzan 'phrin-las.
Taube has divided the texts into eight groups: A. Kanonische Texte und Kommentare; B. Esoterischer Buddhismus; C. Vinaya-Exegese; D. Wissenschaften; E. Geschichte und Ortsbeschreibungen; F. Lieder; G. Sammlungen; H. Fragmente. By far the largest group is group B (pp. 228-850). The appendix contains - apart from the lists of the works in the twelve gSun'bum, the four gZuns-bsdus editions and the four Chos-spyod rab-gsal editions - a description of the Narthang Kanjur, the names of the persons (danapati, revisor, scribe, translator, author) mentioned in the colophons, the titles of the texts in Tibetan, Mongol and Sanskrit, the brief

Page #38
264 REVIEWS titles given in the margin in Tibetan, the Chinese characters in the margin of Peking blockprints and, finally, the catalogue numbers of the texts in the libraries. It would have been useful if Taube had added a list of the places in which the texts were printed. In order to know, for instance, which texts were printed in Peking one is obliged to go through the entire catalogue. It is only in recent years that the enormous extent of Tibetan literature has become obvious. Probably no other literature in the world is still so unexplored. It will take a long time before a history of Tibetan literature can be written. It is above all necessary to catalogue and describe the existing collections. At present this is a difficult and time-consuming task because only very few Tibetan texts have been adequately studied and analyzed. Even to indicate briefly the contents of a Tibetan text is often far from easy. However, Taube's catalogue will be of very great use in the study of other collections which contain the same editions as those described by him. It is to be hoped that more collections will be described in the same careful and detailed way but it will be difficult to find scholars willing to spend so much time and energy on the cataloguing of Tibetan texts. We must be extremely grateful to Manfred Taube for having undertaken this laborious task and for having carried it out in such an exemplary manner. [Note. The publication of a review in 1980 of a work published in 1966 is not due to any fault on the part of the publisher or the reviewer but is due to the fact that the copy sent to the reviewer in 1966 never reached him.] Australian National University .J. W. DE JONG