The Emperor Ch\'ien-lung and The Larger Śūramgamasūtra(pp. 136-146)

Harvard-Yenching Institute The Emperor Ch'ien-lung and The Larger Śūramgamasūtra Author(s): Baron A. von Staël-Holstein Source: Harvard Journal of Asiatic Studies, Vol. 1, No. 1 (Apr., 1936), pp. 136-146 Published by: Harvard-Yenching Institute Stable URL: http://www.jstor.org/stable/2718044 . Accessed: 03/08/2013 01:42 Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at . http://www.jstor.org/page/info/about/policies/terms.jsp . JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org. . Harvard-Yenching Institute is collaborating with JSTOR to digitize, preserve and extend access to Harvard Journal of Asiatic Studies. http://www.jstor.org This content downloaded from 211.95.61.35 on Sat, 3 Aug 2013 01:42:26 AM All use subject to JSTOR Terms and Conditions THE EMPEROR CH'IEN-LUNG AND THE LARGER SURAMGAMASUTRA BARON A. VON STARL-HOLSTEIN HARVARD UNIVERSITY SINO-INDIAN INSTITUTE, PEIPING According to the catalogue of the Chinese Tripitaka published by Bunyiu Nanjio, there are two Buddhist siatras, the titles of which con- tain the characters '-&1, (m-ramgama). The full Chinese titles of the two works are the following: - (NANJIo No. 399), and 1 (NANJIo No. 446). I call these two works, which have practically nothing in common, the smaller (NANJIo No. 399) and the larger (NANJIo No. 446) Sgzram- gama respectively. The smaller M-tramgama fills about 16 pages in the Taish3 Tripitaka, and is said to have been proclaimed by the Buddha on Mount Grdhrakcta near Rajagrha. The larger o2ramgama covers about 49 pages in the Taish3 Tripitaka and is said to have been pro- claimed in Andthapindika's park near 9rdvastI. According to Nanjio, Chinese tradition ascribes the translation of the smaller orframgama to Kumdrajiva lY*lff, while the larger 1 The original Sanskrit title of Nanjio No. 399 is evidently (Buddhabhdsita-) ?urarmgamasamddhisitra. Cf. Bibl. Buddh. I, page 91, and SAKURABE'S catalogue No. 800. Nanjio (No. 446) gives us the following as the full Sanskrit title of the larger ggrarmgama: Mahabuddhosnisa-tathdgata-guhyahetu-saksatkrta-prasannartha-sar- vabodhisattvacarya-stlrdiagarma-sitra. I prefer asram4gama or surangama to siradgama. Cf. BMN1. Buddh. I, pages 8 and 91, and SAKARI's edition of the Mahdvyutpatti, pages 40, 54, 63 and 104. In the XVIII century quadrilingual edition of the larger gararmgama we find the following Sanskrit title: Sarvata- thdgatasya guhyasiddhertha abhisamayahetu sarvabodhisatvasya caryd samutra- degabuddhausnisaguranga nama mahdyanasi4tra. The reading samutradesa (for the somewhat less incorrect samutdega) is probably due to the negligence of a scribe, not to the ignorance of the translators. The following is the corresponding Tibetan title: De bshin ggegs pahi gsaft ba sgrub pahi don mnfon par thob pahi rgyu / byan chub sems dpah thams cad kyi spyod pa rgya mtsho ston pa / sfts rgyas kyi gtsug tor dpah bar hgro ba shes bya theg pa chen pohi mdo. The Sanskrit title given above has evidently been translated from the Tibetan title. 136 This content downloaded from 211.95.61.35 on Sat, 3 Aug 2013 01:42:26 AM All use subject to JSTOR Terms and Conditions THE LARGER SURkfAGAMASUTRA 137 9furamgama is supposed to have been translated in A. D. 705 by the sramana Pan-la-mi-ti 2 in collaboration with the sramanas Mi-ch'ieh-shih- chia WOMM and Shih-huai-ti R'M7. The Sanskrit text of the smaller S9framgama seems to be lost, but the fact that it existed in the past is proved by the Sanskrit text of the Sikcsasamuccaya (BENDALL'S ed., pages 8 and 9), which twice refers to it. The passages referred to in the oikcsasamuccaya are both found in the Chinese (Taisho Trip. 15, 638 b and c) and in the Tibetan (A. D. 1700 Kanjur, vol. THU, 314 and 315) versions 3 of the smaller 6ftraingama. In the Pei Liang translation of the Mahaparinirvancas2tra 92ramgama- sftra MOMIZ) is mentioned, but it is evidently the smaller grtram- gama, which is referred to there, not the larger one. Cf. Taisho Trip. 12, 388 b, 15, 640 a, 12, 390 a and 15, 640 a-b. Volume THU of the A. D. 1700 Kanjur contains a Tibetan translation of the smaller 9ftrarmgama, which is attributed to 9dkyaprabha. The latter is known as the translator of some other works, and it seems to be certain that the smaller oifraingama was translated by him from Sanskrit into Tibetan. (Cf. Ann. du Musee Guimet 2, 399.) No Sanskrit text of the larger jiframgama exists, and the s-dtra, as far as I know, is never quoted in Sanskrit books or in Chinese or Tibetan works undoubtedly translated from Sanskrit. The A. D. 1700 Kranjur edition catalogued by Sakurabe contains only two fragments of the larger Siframgama, but no complete version. These fragments are, according to BECKH'S Verzeichniss (Berlin, 1914, p. 52, note 3), the Tohoku Index and Sum-pa's Dpag-bsam-ljon-bzaA (Calcutta, 1908, page 414), translated from Chinese.4 The larger fragment (Sakurabe No. 902) contains a con- 2 Nanjio (column 443) thinks that 1|A is a transliteration of Pramiti. According to the HIbogirin (Fascicule Annexe, page 137) Paramiti is the original Sanskrit name. According to an authority mentioned in the Hs& Tsang Ching (tao No. 21, ts'6 No. 1, p. 53a shang), [Shih-]huai-ti translated the larger Sgramgama in collaboration with an Indian Sramana whose name was unknown. 8In an article, which has already been printed and which will form part of the forthcoming third volume of the Harvard Sino-Indian Series, I describe a copy of the Kanjur, which must have been issued in A. D. 1692. The pagination of this copy agrees with the pagination of the A. D. 1700 K'ang-hsi Kanjur (catalogued by Sakurabe) except as far as the volumes OM, ZA, HI and SHI are concerned. 4The title of Sakurabe's catalogue (Kyoto, 1930-1932) is: W-Jj1 R A The title of the Tohoku catalogue (Sendai, 1934) is: g RN. The date (A. D. 1747) given by Sarat Chandra Das (introduction, page iii) for the Dpag-bsam-ljon-bzai! is wrong. Cf. PELLIoT, JA mai-juin 1913, This content downloaded from 211.95.61.35 on Sat, 3 Aug 2013 01:42:26 AM All use subject to JSTOR Terms and Conditions 138 BARON A. VON STAEIHOLSTEIN secutive translation of a part of the ninth chapter 5 and the entire tenth chapter of the larger 9furamgama. The smaller fragment (Sakurabe No. 903) contains numerous passages belonging to the ninth and tenth chapters of the larger &2dramgama. A considerable part of the larger &drarmgama has been translated from Chinese into English by the Rev. Samuel Beal (Catena pp. 286-369). The fact that no Indo-Tibetan translation of the larger 92ramcgama exists, is one of the arguments used by many ancient and modern authori- ties, who want to prove that the larger g2rargama should not be regarded as authoritative (tshad ma) by pious Buddhists. The Emperor Ch'ien- lung is acquainted with the sceptics' view, but he does not share it. In an imperial introduction issued on the 25th day of the 7th month of the 35th year of Ch'ien-lung (A. D. 1770), we find the following passage: Safis rgyas kyi gtsug tor chen po dpah bar bgro babi mdo glegs bam yofas rdsogs bshugs pa hdi s~a phyihi mkhas pa mani pos nub phyogs kyi dpe ma rned zer nahand / / deki nafi gi bde ggegs gtsug tor gyi gzunls rgya gar gyi dpe dani gin tu hgrig pas / / mdo kdi tshad ma yin par mfion / / des na mdo hdis theg pa chen poki lam kbras kyi lus yofis su rdsogs pa gsal bar rigs pa yafi dag gis bsgrubs pas na / / gus par bya bahi gnas su. cihi phyir mi hdsin.6 p. 651. According to page 347a of a xylograph of the Dpag-bsam-ljon-bzaa, Sum-pa's work was composed in A. D. 1748 (rab byutA bcu gsum pahi sa pho hbrug lor). 5In the ninth chapter of the larger 9ilramgama (Taisho Trip. 19, 149c, A. D. 1692 Kanjur vol. DSU, p. 301b) true and false, male and female Buddhas are mentioned (i ; yaa dag pahi sails rgyas rgyu mahi safis rgyas skyes pahi saois rgyas / bud med kyi sails rgyas; read sgyu instead of rgyu). The Tibetan version of a commentary on the Saddharmapui00arikastatra (CORDIE1R, Fonds Tibe'tain, III, 372), which is said to be translated from Chinese, says that there are times when false Buddhas (safis rgyas sprul pa) appear in the world and others when they do not. Cf. the Choni Tanjur, Mdo, vol. DI, p. 227a. 8 The passage occurs in an Imperial Introduction (rgyal pos mdsad pahi gser gyi bkah hgyur rin po che gsar bsheas kyi kha byaf), which is found in the first volume of a " golden Kanjur " belonging to the Peiping Palace Museum. This " golden Kanjur " is hand-written (golden Tibetan characters on dark blue paper) and most beautifully got up. The title-page of the first volume is adorned with pearls and other precious substances. This particular golden Kanjur, which now reposes in the vaults of a Shanghai bank, evidently contains a copy of the Tibetan version of the larger ?iiramsgama (cf. the words glegs barn yons rdsogs bshugs pa hdi in the passage quoted above). I am not in a position to ascertain the fact because the golden Kanjur is no longer accessible. I had a photographic This content downloaded from 211.95.61.35 on Sat, 3 Aug 2013 01:42:26 AM All use subject to JSTOR Terms and Conditions THE LARGER SURA4IGAMASUTRA 139 The imperial introduction affirms that a certain dhdrani, which forms part of the Chinese version of the larger oiiramgama, entirely agrees with the Indian text of the same dhdrani.7 From the fact that one important part of the larger Adra-ngacma is undoubtedly based upon an Indian original the Emperor draws the conclusion that the entire larger A3uramgama must be authentic (tshad ma). A critical examination of the same dhdrany has quite lately led Mr. Li I-shao 4EJ*V-J to an entirely different conclusion. He thinks that the dharani is badly transliterated in the larger S6ramgama and believes that this fact militates against the authenticity of the entire work.8 Mr. Li compares our dhdran.i as it appears in the Chinese version of the larger Sklramgama with the transliteration of it by Amoghavajra (Taisho Trip. 19, 100-102) and finds that the version found in the larger &duramgama is most unsatisfactory ( M fZL J14ft). It is certainly true that the S69aramgama version of the dhirani contains many mistakes. The form par (Taisho Trip. 19, 135 c), for instance, is evidently wrong (yakesabhyah). Amoghavajra (Taisho Trip. 19, 101 c) has 8a (yakesebhyah). The character 4 represents both va, in bhavatu, and bha, in bhaya (Taisho Trip. 19, 135 a 15 and 16). Amoghavajra (Taisho Trip. 19, 101 a 13) has If for va and - for bha in the same instances. When considering such inconsistencies in the larger A23ramgama we should not forget that even great authorities, who worked when the art of trans- literating Indian sounds with Chinese characters was still in its infancy, reproduction of the Imperial Introduction made when the collection was still in Peiping. 7The only dhdrazn which occurs in the larger giaramgama is a very long one (it occupies nearly three pages in the Taish4 Trip. 19. 134-136). The Emperor calls it Bde-g~egs-gtsug-tor-gyi-gzuiis [Sugatosnisadhdrani], but the text of the same dhdrani in Indian characters, which we find in the Taish6 Trip. (19. 102- 105), bears the following Chinese title *iiP kItJZ Mahabuddhosnisa- mahadhiracn. The dhdrani is known under a number of other designations. Cf. SAKURABE'S catal., No. 202, and HOERNLE'S Manuscript Remains, I, 53. 8 he essay Fo-hsiieh Wei Shu Pien Lfteh i in which Mr. Li discusses the larger ?izramgama, appeared at Nanking in 1934 in the Kuo-li uhung-yang Ta-hsiieh Wen-i Ts'ung-k'an 1 * $ tfIJ vol. 1, part 2, pp. 7-46. I have to thank Professor Y. K. Tschen [Ch'6n Yin-k'o] for having drawn my attention to this essay. 8a The Chinese characters enclosed within square brackets would normally be of a very small font. 10 This content downloaded from 211.95.61.35 on Sat, 3 Aug 2013 01:42:26 AM All use subject to JSTOR Terms and Conditions 140 BARON A. VON STAEL-HOLSTEIN were not always consistent. HsUan-tsang (Taish5 Trip. 20, 404 b 3) uses the character A for dha in Dhanada and for d in Rudra. Accord- ing to Nanjio (columns 435, 443 and 444) Hsiian-tsang began trans- lating Indian works in A. D. 645, and Amoghavajra in 746. The Chinese version of the larger Skirargama is said to have been compiled in A. D. 705. A dhdrani consisting of nothing but meaningless rows of Chinese characters, like certain supposedly " foreign " names occurring in Taoist books,9 could be used to prove that the work, to which it belongs, had been compiled in China independently of an Indian original. But the dhdrani, which forms part of the larger Siframpmgam, notwithstanding its shortcomings, is evidently based on an Indian original,10 and we cannot agree to the use which Mr. Li makes of it in his argumentation. Neither can the Emperor's view be accepted in its entirety (the dhdrani proves the authenticity of the larger Rtram gam-a as a whole), but we must admit that the 9ftramgama- (or Sugatosncsa-) dharan'i makes the thesis of the ultra-sceptics (the larger Skirarngama is a Chinese forgery from beginning to end) equally untenable. Another argument advanced by Mr. Li against the authenticity of the larger 9ftramgama is the fact that a stanza which occurs in lsUan-tsang's translation of Bhavaviveka's Mahayanatalaratna astra (Nanjio No. 1237) is also found in the larger ,itramgama. Mr. Li seems to be quite certain that the stanza was first translated by Hsiian-tsang (about 647 A. D.) from Bhavaviveka's Sanskrit and that HsUan-tsang's Chinese stanza was later (about A. D. 705) taken over by the compiler of the Chinese version of the larger 9ftram gama. Mr. Li evidently thinks that the compiler slightly changed the Chinese stanza before taking it over, in order to make the loan from Esiian-tsang less apparent. 9 Professor Forke says: " Ti-a-sha [the name of a Taoist god] " is intended to sound like a foreign name. The third heaven is styled " the extremely fine Po-lo-ju heaven with the deep dark celestial king Yfin (cloud), personal name Kuei-ling." "Po-lo-ju means nothing, but it sounds like the Chinese rendering of some Sanskrit name." Cf. The World-Conception of the Chinese, London, 1925, pp. 141-142. Professor Hu Shih jnJ very kindly draws my attention to the fact that some of the supposedly Indian names attributed by Chinese writers to certain Buddhist patriarchs are evidently not based upon Sanskrit originals, but invented in China. Cf. Taish6 Trip. No. 2079. 10 No Chinese work which the compiler could have used for his dharani is known to exist or to have existed. All three Chinese transliterations of the dhlrani mentioned by Sakurabe (No. 202) are of a later date than the larger gararmgama. The latter is mentioned in the K'ai-yiian Lu (Taisho Trip. 51. 603 a), which was composed in A. D. 730. Cf. NANJIO, Introduction, p. xxvii. This content downloaded from 211.95.61.35 on Sat, 3 Aug 2013 01:42:26 AM All use subject to JSTOR Terms and Conditions THE LARGER SFURAMGAMASUTRA 141 The following is the form in which the stanza appears in Hsiian-tsang's translation (A) and in the Chinese version of the larger ,iframgama (B). A. 'W-9A B. itfin ifJu_~ Sta6 Cf. Taish5 Trip. 30, 268 b and 19, 124 c. An argument which militates still more strongly than this comparison of the two stanzas against the purely Indian origin of the larger Aialrarmgama has been suggested to me by the late Mr. Huang Chien WLO one of my former pupils. In Kumarajiva's translation of the Saddharmapundarikasfitra (Tokyo Meiji Trip., vol. &, fasc. I, page 49 b) we find the following passage: X~htO ++t~~tWS, M~fieW$to eJOC (mil Wt jX X lp~tt) f84$J12o The bracketed part of the passage has no equivalent in the Sanskrit text edited by Kern and Nanjio, in the Tibetan translation (Sakurabe No. 781), or in the oldest Chinese trans- lation (by Dharmaraksa, Nanjio No. 138). Therefore I regard the bracketed part of the passage as one of the explanatory notes, which Kumdrajiva '11 so often added to his translations. The bracketed part of the passage contains an etymology (Gil) of the name T@:W, and the fact that this etymology is also found (Tokyo Meiji Trip., vol. lit fasc. I, page 21 b) in the larger S2ramngamca (compiled about 11 Professor Lilders says: Ofter werden [von Kumarajiva] auch erklirende Zusatze gemacht. Cf. Britchstiicke der Kalpandman.d1itika des Kimdralata, Leipzig, 1926, pp. 58-59. 12 Rev. Samuel Beal, who translates our passage from Chinese into English (Catena, pp. 389-390) says in a note: "This explanation [the eleven bracketed characters] is wanting in the French version (i. e. Burnouf's translation of the Saddharmapundarikasiltra from Sanskrit into French)." This content downloaded from 211.95.61.35 on Sat, 3 Aug 2013 01:42:26 AM All use subject to JSTOR Terms and Conditions 142 BARON A. VON STAEE-HOLSTEIN three hundred years after Kumdrajiva's death) proves that the compiler did not rely exclusively on purely Indian material.13 The imperial introduction (dated the 18th day of the 10th month of the 28th year of Ch'ien-lung = A. D. 1763) to the quadrilingual edition 14 (Chinese, Manchu, Mongol and Tibetan) of the larger ft2ramn- gama does not discuss the authenticity of the work, but it contains a number of more or less plausible data connected with the s-atra's history. The quadrilingual introduction covers altogether twenty-four pages (la-13a, there is no page numbered lb), seventeen of which are repro- duced on plates 1-9 below. The following is a translation of the his- torical portion of the introduction, which occupies pages la-9a and a part of page 9b of the original 1 xylograph: AN IMPERIAL INTRODUCTION TO THE TRANSLATIONS OF THE MAHAYANASUTRA ENTITLED BURAMGAMA.16 All the sacred texts [of Buddhism, which are designated as] the Trtptacka and [as] the Dvadasaifgapravacana [la] originated in the holy land, and spreading [beyond the boundaries of the holy land] gradually reached this middle empire (China). During their progress from West [2a] to East [the sacred texts] first penetrated into the country of Dbus-Gtsafi.17 The holy land is India, and Dbus-Gtsafa is [one of the names of] Tibet. Owing to this [i. e. to the sacred texts having penetrated first into Tibet] all the sacred texts [2b], which now exist in China translated [into Chinese], are complete in Tibet [in Tibetan translations]. The Mahayawnasftra called Sftraingama alone is 13 The name Bitt seems to be a translation of *Avalokitalokasvara. This form of the name has, as far as I know, never been found in ancient books written with phonetic characters. The forms Avalokitalokeivara and Avalokitasvara, however, have been used in India and in Central Asia respectively. Cf. the YCHP No. 17. 36. 14 The Yung Ho Kung lamasery of Peking possesses a block-printed copy of the quadrilingual edition. It has ten volumes. The leaves are made of thick white paper, and their size is 8/2 X 28 inches. The letters are red. l5 The Chinese version of the A. D. 1763 introduction occasionally differs from the Tibetan version, on which my translation is mainly based. I regard most of those differences as insignificant and discuss only some of them in my notes. 16 The bracketed figures inserted into this translation refer to the ends of the pages of the original xylograph. 17 Read brgyud na (instead of brgy