Sanskrit Fragments Of Kasyapaparivarta

Author(s): J W De Jong
Publisher: J W De Jong
Catalog link: https://jainqq.org/explore/269275/1

Page #1

SANSKRIT FRAGMENTS OF THE KĀSYAPAPARIVARTA
by J. W. DE JONG, Canberra
In 1938 Kuno Höryū edited two fragments of a manuscript of the Kāśyapaparivarta'. They had been sent to Hoernle by P. J. Miles in 1903. According to Hoernle the fragments had been found in Khadalik. The two fragments (Hoernle No. 143 S.B. 38 and No. 143 S.B. 39) are at present in the India Office Library to which institution I am obliged for having put at my disposal excellent photocopies. Kuno had no difficulty in showing that they belong to one and the same leaf. The text corresponds to sections 128-136 of the edition of the Kāśyapaparivarta published by A. von StaëlHolstein (Shanghai, 1926). In von Staël-Holstein's edition each of the sections 128–133 consists of a prose part and a verse part, but in the fragments the verses are missing. However, the fragments do contain the first words of section 136: atha khalu bhagavām tasyä[m] velā(y)[ām imām gāthām abhāşata). Kuno pointed out that of the four Chinese versions the two versions dating from the periods of the Chin and Ch'in dynasties, correspond more closely to the Sanskrit text of the fragments. He concluded that this text must have been in existence in the 3rd-5th centuries A.D. Comparing the fragments with the corresponding prose parts in von Staël-Holstein's edition, Kuno tried to reconstruct the missing parts of the entire leaf. His readings of the manuscript are not always correct and his reconstruction does not take into account the exact extent of the missing portions. Even more important is the fact that Kuno was not aware of the fact that a fragment of the same leaf was edited twenty years before by J. N. Reutero. The fragments, published by Reuter, were brought back from his expedition to Central Asia and North China in 1906–1908 by Colonel Baron Gustav Mannerheim. The third fragment contains a passage of the Kāśyapaparivarta corresponding to sections 130-135. It exactly fills one gap in the leaf, edited by Kuno, between lines 3 to 8 of the recto and lines i to 6 of the verso. The following edition of the three fragments of this leaf is based upon a photocopy of the two fragments in the India Office Library and a photocopy of the Manner
1 Saiiki shutsudo bukkyō bonpon to sono seiten shiron-jo chii (jo). Daihoshakkyo to Zoagonkyō no genten, I. Uten shutsudo Daihöshakkyo bonpon to sono kachi',
Bukkyo kenkyū, II,3 (1938), pp. 71-110. 2 J. N. REUTER, 'Some Buddhist Fragments from Chinese Turkestan in Sanskrit and
"Khotanese", Journal de la société finno-ougrienne, 30 (1913-1918), pp. 1-37 (Reprinted in: C. G. Mannerheim, Across Asia from West to East in 1906-1908, vol. II, Helsinki, 1940).

Page #2

248
J. W. DE JONG
heim fragment which Professor Pentti Aalto has been so kind as to send me at my request. (M) indicates the beginning and end of the Mannerheim fragment. Missing syllables are indicated with and missing letters with.. As in von Staël-Holstein's edition a single oblique stroke represents a dot, a pair of oblique strokes two upright strokes. Moreover, a colon stands for a colon in the manuscript, a punctuation mark which was not recognized as such by von Staël-Holstein3. Akṣaras which can only be read partially are put between round brackets and restored aksaras between square brackets. O is used to indicate the circle round the hole in the right half of the leaf.
RECTO
I. m eva käsyapaikatyä śramaṇabrāhmaṇā bahümn darmām paryāpya na rāg.m(o) hatrsnä vinodayamti /tte dharmarnavanohya
manne klaiśa (tr)ṣṇayā kā(la) -
2. tigămino bhavamti // tadyathä käśyapa vaidya ausa|da|bhrasta grhi(t)---utpadyeta / na ca tam vyädhi / saknuya
cikitsittu/evam eva kāśyapa bahuśrutasya-
3. drastavyah yaḥs tena śrutenna na knoty ātmānam klaiśavyädhim citsiirarthakam (tasya) tam śruttam bhavatt(i) (M) // tadyatha kasyapa) glanapu[ru]s[o] r (M) äjärham bhaişajyamm upayujyattäsamvatsarena kālam - -
4 evam eva käiyapa bahuśrutasya klaiśavyādhi drasta yabs tenäsamvatsareņa kā (M) lam karotti // tadyathā kāśyapa man[i] (M) ratnam ucāre patita akäryopagam bhavaty evam elva) --
5. pa bahuśrutasya läbhasatkäro(carapa) - Ostavyaḥ niskimcana devaman(u)..e (M). yasu // tadyathä käśyapa mṛtasya mälä (M) / evam eva kasyapa dušilasya käs(alyandra) -
innakeśana
6. vyah // tadyathā kāśyapa susnätasya suvili- (sya) khasyavada(ta) [M] vastraprävṛtasya pravaracandanänuliptasya śre (M) stiputrasya sirse canpakamälä evam eva kasya ---
7. lavato (balhuśrutasya kāṣāya(dha)ra----// catvāraḥ ime [M] kāśyapa dušila silavapratirüpakāḥ kata (M) m[e] catvara iha käsya-katyo bhiksu --
8. mok(sa) sam ---- rto bhavati / ācā - vadyesu bhayadarsi samādāya (śilkṣa (M) ti sikṣāpa(de)
- an (M) treṣv api (iśuddha)
3 F. WELLER, Zum Kasyapaparivarta. Heft 2. Verdeutschung des sanskrit-tibetischen Textes (Berlin, 1965), p. 63, n. 3.

Page #3

Sanskrit Fragments of the Kāśyapaparivarta
249
VERSO
.I. ---- ga(t)o (v)iharati pa(ri)su ------------(M) yam
kāsyapa prathamo duśila silavapratirūpakaḥ // (M) (p)unar apa(ra)m
kāólya) - (i)haikatyo bhikṣu vvina(ya) -- 2. - vati pravịta - nayo v(i)naya - pto---- (ya)drsțim
cāsy(a nuca (M) litam bhavati : ayam kāśyapa dvitiyo duśila sila
(M) vaprattirūpakaḥ // puna param kāśyapa (i)haikat - - 3. kşu maitrāvihāri bhavati / satvārambana ------manvāgatto
bhavati / (a) (M) jātim ca sarvvasaņskārāņām śrutvā : utrasati sam (M)
trasati samtrāsam āpadyate / ayam kāśyapa třti -- 4. silah silavapratirūpakaḥ // pu(na)r apara(m) O kāśyapa ihaikatyo
bhikṣuḥ dv (M) ādaśa dhuttaguṇān samā - ya varttatte / (M) upalambha
drșţikaś ca bhavati / ahamkāramamamkāra --- 5. ayam kāśyapa caturtho dusilaḥ śilavapra O tirūpakaḥ // ime kāśyapa
ca (M) tvāro duśila silavapratirūpakā (M) silam silam itti kāśyapa ucyate / yatra nātmā (n) - - yam : na kriyā nākriya / na karaṇam nākaraṇam : na cāro nācāro na pracār. (M) na nāmarūpam / na nimittam : na śamo (M) na prasamaḥ na
graho notsargah na grāhyam : na satvo na ---- 7. ptih na vā na vāprajñaptih na citam na citaprajñapti / na lo(klo
nālokaḥ na nis[ra]yo nān(i)śrayaḥ nātmaś -- tka . (șa) --
paräkarmmasilyapamnsanā: na silamannyana / na si ---- 8. nā : na samkalpanā : idam ucyate kāśyapa aryāņām silamm /a ----
------äpagatam // sarvvaniśrayavigatam // atha khalu bhagavām tasyā velāly)
In line 5 of the recto the word devamanuşyesu has to be read. Fragment 143 S.B. 38 has devaman and the vowel e. M contains the subscript y and the syllable șu. Reuter read xyanu, x indicating a deleted consonant. In line 7 of the verso Reuter read only the word (du] sile). It is possible to read ilo - r -.. The fragment 143 S.B. 39 has ātmas -- tka . - . ā (sa) - . It is therefore possible to reconstruct the word ātmaśīlotkarşaņā. The scribe has the habit of doubling the t and the n, cf. recto (1) tte, -ohyamānne; (2) cikisittu; (3) śrutenna, śruttam, etc. Several syllables have been omitted by him, cf. recto (3) knoty for saknoty; (4) citsi – for cikitsi –; verso (2) puna param for punar aparam. In recto (3) and (4) the scribe wrote asamvatsarena which must be a mistake for asamvareņa, cf. Weller's translation of the corresponding passage in the Chin version: "Gerade wie wenn ein kranker Mensch die wunderbare Arznei eines Königs einnimmt, sein Ende erreicht, (da er) sich nicht an die Regel hält, so, Kāśyapa, verhält es sich wiederum auch (damit) so, daß es vielfach Sramaņa, Brahmanen gibt, (die) das der Lehre nicht Gemäße ausführen, (sondern) alle Krankheiten der Bindungen aufkommen lassen, (und sie nach ihrem) Ende auf dem schlim

Page #4

250
J. W. DE JONG
men Wegen (wieder) geboren werden." It is difficult to know how far the scribe has correctly reproduced the language of the text. In samdhi between vowels, a hiatus usually occurs, but in the first line of recto the scribe wrote käśyapaikatyä. In reconstructing the text of the leaf I have normalized the spellings and the samdhi, although I am aware of the fact that the language of the original may have been more irregular. Apart from this aspect, it does not seem too difficult to reconstruct the text of the leaf. It is possible that the original had na pracãro näpracäraḥ instead of na pracāraḥ and na nāma na rūpam instead of na nāmarūpam (cf. $ 135).
RECONSTRUCTED TEXT OF THE LEAF
128. evam eva käśyapaikatyāḥ śramaņabrāhmaṇā bahūn dharmān paryāpya na
rāgatrşņām vinodayanti / na dveşatrşnām na mohatļşņām vinodayanti / te dharmārņavenohyamānāḥ kleśatņšņayā kālagatā durgatigāmino
bhavanti / 129. tadyathā kāśyapa vaidya auşadhabhastrām gļhitvānuvicaret / tasya
kaścid eva vyädhir utpadyeta / na ca tam vyādhim saknuyac cikitsitum / evam eva kāśyapa bahuśrutasya kleśavyādhir drașțavyo yas tena śrutena na saknoty ātmanaḥ kleśavyādhim
cikitsitum / nirarthakam tasya tac chrutam bhavati / 130. tadyathā kāśyapa glānaḥ puruṣo rājārham bhaișajyam upayujyāsam
vareņa kälam kuryāt / evam eva kāśyapa bahuśrutasya kleśavyādhir
draştavyo yas tenāsaņvarena kalam karoti / 131. tadyathā kāśyapa maņiratnam ucсāre patitam akāryopagam bhavaty
evam eva kāśyapa bahuśrutasya lābhasatkāroccārapatanam drașțavyam/
nişkimcana devamanuşyeșu / . 132. tadyathā kāśyapa mstasya mālā / evam eva kāśyapa duḥšilasya
kāṣāyam drastavyam/ 133. tadyathā kāśyapa susnātasya suviliptasya suchinnakeśanakhasyā
vadātavastraprāvịtasya pravaracandanānuliptasya śreșthiputrasya śirse campakamālā evam eva kāśyapa duḥsilavato bahuśrutasya kāṣāyadhāraṇam drastavyam/ catvāra ime käśyapa duḥšilāḥ śilavatpratirūpakāḥ / katame catvāraḥ / iha kāśyapaikatyo bhikṣuḥ prātimokşasamvarasamvịto bhavati / ācāragocarasampanna aņumātreşv api vadyeşu bhayadarsi samādāya siksate śikṣāpadeșu parisuddhakāyavāňmanaskarmaṇā samanvāgato viharati parisuddhäjivaḥ sa ca bhavaty ātmavādi / ayam kāśyapa prathamo duhšilah silavatpratirūpakaḥ //
+ Kāśyapaparivarta nach der Djin-Fassung verdeutscht, MIO, XII (1966), p. 419.

Page #5

Sanskrit Fragments of the Kāśyapaparivarta
251
punar aparam kāśyapehaikatyo bhikṣur vinayadharo bhavati pravịtavinayo vinayaguptau sthitaḥa satkāyadsstiś cāsyānucalitā bhavati / ayam kāsyapa dvitiyo duhšilaḥ śilavatpratirūpakaḥ // punar aparam kāśyapehaikatyo bhikṣur maitrāvihari bhavati / sattvārambaņayā karuņayā samanvāgato bhavati / ajātim ca sarvasamskārāņām śrutvā / uttrasati samtrasati sastrāsam āpadyate / ayam kāśyapa tỉtiyo duḥsilah silavatpratirūpakaḥ // punar aparam kāśyapehaikatyo bhikṣur dvādaśa dhutaguņān samādāya vartate / upalambhadssţikaś ca bhavati / ahamkāramamakārasthitaḥ / ayam kāśyapa caturtho duḥsilah silavatpratirūpakaḥ //
ime käśyapa catvāro duḥśilāḥ śilavatpratirūpakāḥ // 135. silam silam iti kāśyapocyate / yatra nātmā nātmiyam / na kriyā nākriyā /
na karaṇam nākaraṇam / na căro nācāro na pracāraḥ / na nămarüpam na nimittam / na samo na praśamah / na grāho notsargaḥ / na grāhyam na sattvo na sattvaprajñaptiḥ / na vāň na vākprajñaptiḥ / na cittam na cittaprajñaptiḥ / na loko nālokaḥ / na niśrayo näniśrayah / nātmaśilotkarşaņā na parasilapamsanā / na silamanyanā / na silavikalpanā (?) / na samkalpanā / idam ucyate kāśyapāryāņām silam anäsravam aparyāpannam
traidhätukäpagatam sarvaniśrayavigatam // 136. atha khalu bhagavāms tasyām velāyām
a Cf. Tib. 'dul-ba'i tshul-la gnas-pa and the Indikatusāya copper plaque no. 67: vanaya-gupto sthitah (S. Paranavitana, 'A Note on the Indikatusäya copper plaques', Epigraphica Zeylanica, vol. IV, pt. 5, 1939, p. 241). Thanks are due to Mr. G. Schopen for drawing my attention to this article.
In 1957 V.S. Vorob'ev-Desjatovskij published two fragments of Sanskrit manuscripts of the Kaśyapaparivarta. The first fragment is an almost complete leaf. It carries the number three and forms part of the manuscript published by von Staël-Holstein. Weller is of the opinion that this leaf does not agree in all details with the manuscript published by von Staël-Holstein. As his only reason Weller adduces the fact that this leaf contains the words: idam uvāca bhagavāms which are absent in the Tibetan version. This only proves that the Sanskrit original on which the Tibetan version is based is different. Vorob'ev-Desjatovskij points out that the dimensions of the leaf are the same as those of the other leafs of the manuscript. His conclusion that this leaf is the formerly unknown third leaf of the manuscript is undoubtedly
5 'Vnov' najdennye listy rukopisej Kašyapaparivarty', Rocznik Orientalistyczny, 21
(1957), pp. 491-500. 6 Zum Kaśyapaparivarta, p. 63, n. 3.

Page #6

252
J. W. DE JONG
correct. The Institute of Oriental Studies has kindly put at my disposal a photocopy. Vorob'ev-Desjatovskij's transliteration has to be corrected in the following points: recto (3) väksyaparamaḥ, read: vakyaparamaḥ, verso (a) dhäritvä, read: dharitva, (3) taparo; read: ca taparo; (4) vartate; read: vartamte; (5) -visaṁvädanataya; read: -visamvadanatayd. Recto (1) Vorob'ev-Desjatovskij reads adiptaširas ce [la..kru) and adds in a note that he is unable to restore the missing word. The Tibetan version has mgo-am-gos-la me bar-ba bźin which renders adiptaširaścelopama, cf. Edgerton, Buddhist Hybrid Sanskrit Dictionary s.v. adiptaširaścailopama. Recto (2) árutal for śrutas and verso (2) pratipattiya for pratipattiya are misprints as is obvious from Vorob'ev-Desjatovskij's notes. At the beginning of verso (1) Vorob'ev-Desjatovskij reads.. y... taḥ. I fail to see any trace of a y on the photocopy. Also the photocopy does not show the syllable ka which Vorob'ev-Desjatovskij adds at the end of recto (5): läbhasatkäraśiloka.
The second fragment corresponds to sections 14-19. The text does not contain the verses which are to be found in von Staël-Holstein's text. The Institute of Oriental Studies has kindly given me a photocopy of this fragment also. In recto (6) read bhlutagu/nah for bhluta gujnah, (7) read cabhiśraddhadati for cabhiśraddadhāti; verso (2) read satpāramitā for satparamita-, (2-3) read dharmabhāṇakadar/śa) (b31 namm for dharmabhāṇakadarśa) (b3] năm; (4-5) read sarvvadṛstikr(tand] (bs) m for sarvvadṛstikṛtalna) (bs) m, (6) read niskuhakasy/dra/nyavasaḥ for niskuhasy[äralnyavāsaḥ, pratikārāpratikamkṣiņas for pratikarakratikaṁkṣiņas; (7) read saddharmaparyeştim for saddharmamaryastim.
Weller is of the opinion that this fragment does not entirely agree with the Chinese Chin version, as Vorob'ev-Desjatovskij had maintained. This is not quite correct because Vorob'ev-Desjatovskij said only that "of all Chinese versions only the Chin version reproduces exactly marasamatikramand dharmah in section 18 and that this represents a distinctive peculiarity of this version" (p. 498).
Although Weller agrees with him on this point, I fail to see in what way the Chin version is any closer to the Sanskrit text of the fragment than the Han and Ch'in versions. None of the three versions translates only the word mära. According to Weller's translations of the three versions, the Han version has "die Welt Maras", the Chin version "alle mara" und the Ch'in version "maras Angelegenheiten".
7 Op. cit., p. 10.
8 'Kasyapaparivarta nach der Han-Fassung verdeutscht, Buddhist Yearly 1968/69 (Halle, 1970), p. 114; 'Kasyapaparivarta nach der Djin-Fassung verdeutscht', MIO, XII (1966), p. 394; 'Kasyapaparivarta nach der Tjin-Übersetzung verdeutscht', Wiss. Z. der KarlMarx-Universität Leipzig, 13 (1964), Gesellschafts- und Sprachw. Reihe, Heft 4, p. 775.

Page #7

Sanskrit Fragments of the Kaśyapaparivarta
253
Again according to Weller, the fragment agrees with the Chin version in section 19: śrutārthātrptatā sarvvakusalamūlasamudānayanļāyal". The Chin version has: "Die Lehre zu suchen gibt es (für ihn) keinen Uberdruß (und kein) Genügen, weil (er) jedwede Wurzel des Heilsamen aufsammelt." However, "die Lehre zu suchen" does not correspond to śrutārtha but to saddharmaparyeștim ārabhya which belongs to the third category in this section. It is interesting to note that in this instance the text of the fragment corresponds better to the Sung version and the Tibetan version than the text of von Staël-Holstein's manuscript. The latter has atịptitā as against śrutārthātņptata which is confirmed by the Tibetan version: thos-pa dan don-gyis noms mi-myon żin and the Sung version: "Der Sinn (der Lehre) zu hören findet er kein Genügen"10. Another example is nişkuhakasyāraṇyavāsaḥ in section 19. In this case the fragment agrees with the Tibetan version (tshul-'chos-pa med-par dgon-pa na gnas-pa) and von Staël-Holstein's manuscript (nişkuhakasyāraṇyavāsābhiratiḥ) with the Sung version: "Er freut sich, im Walde zu weilen, (ist) völlig still und zurückgezogen''11. In section 14 von Staël-Holstein's manuscript has bhūtakalyāṇamitrāņi as against kalyānamitrāni in the fragment and in the Tibetan version (dge-ba'i bấes-gñen). Von Staël-Holstein's manuscript adds after each category the word samvartate which is missing in the fragment and in the Tibetan version. The oldest Chinese version does not contain any verses in sections 136–137. In spite of this, it is not certain that the Sanskrit original, on which this version is based, was written in prose. The Chinese version is very primitive and it is possible that the translator rendered the original Sanskrit verses into prose. It is noteworthy that section 136 in the Han translation begins with a solemn statement which is not found anywhere else in the text: "At that time the Buddha spoke, saying". As to the date and the author of this translation, both Pelliot and Õno Hõdõ have drawn attention to the fact that, although the oldest extant Chinese catalogue, the Ch'u san-tsang chi-chi (about 515 A.D.), states that the text was translated in 179 A.D. by Lou-chia-ch'an (Taisho no. 2145, P. 6b17), this is very doubtful12. Pelliot supposes that Lou-chia-ch'an's translation of the Asțasāhasrikāprajñāpāramitāsūtra was confused with the translation of the Kāśyapaparivarta. However, Pelliot points out that the translation judging by its archaic aspect could well have been made in the Han period: Ono declared peremptorily that the language of the text proves that it was translated during the Eastern Chin. It is impossible to adhere to his point of view.
9 Zum Kaśyapaparivarta, p. 10. 10 'Die Sung-Fassung des Kāśyapaparivarta', Monumenta Serica, 25 (1966), p. 240. 11 Ibid., p. 239. 12 P. Pelliot, Compte rendu de The Kāçyapaparivarta, etc., T'oung Pao, 32 (1936),
pp. 68-76; ONO Hõdo, Daijo kaikyō no kenkyū (Tokyo, 1954), pp. 98, 101-102. Cf. also Taishō no. 2145, pp. 19519 and 29017.

Page #8

254
J. W. DE JONG
Already von Staël-Holstein remarked that the translation uses the same equivalent for bhagavat as Lou-chia-ch'an's translation of the A-ch'u fo-kuo ching (Taisho no. 313). The transliterations of cakravartin, upāyakausalya and abhijñā are the same as those found in Lou-chia-ch'an's translation of the Așțasāhasrikāprajñāpāramitasūtra13. It is therefore probable that this translation of the Kaśyapaparivarta is indeed due to Lou-chia-ch'an.
According to von Staël-Holstein the manuscript edited by him was probably written in the ninth and tenth centuries. His main argument seems to be the agreement of the text with the Chinese Sung version. This version was made in the year 98614. Whatever the date of the manuscript may be, the text must already have been more or less the same at the time of the Tibetan translation, which dates from the beginning of the ninth century. It is possible to push the terminus ad quem even further back. The Kāśyapaparivarta is quoted in many texts. Von Staël-Holstein listed quotations in the Sikşāsamuccaya, the Mahāyānasūtrālamkāra and the Bodhicaryāvatārapañjikā. Japanese scholars have pointed out quotations in the Prasannapadā, the Madhyāntavibhāgaţikā, the Lankävatārasūtra and the Ratnagotravibhāga and in texts, preserved only in Chinese and/or Tibetan translation, such as the Mahāyānasamgraha, the Daśabhūmivibhāṣā (Taisho no. 1522), the Prajñāpāramitopadeśa (Taisho no. 1509), the Mahāyānāvatāra (Taishō no. 1634), the Fo-hsing lun (Taisho no. 1610), the Chi chu-fa-pao tsui-shang-i lun (Taishō no. 1638), the Anuttarāśrayasūtra (Taisho no. 669) and the Mahāparinirvāṇasūtra (Taisho nos. 374-377)15. Tsukinowa drew attention to the fact that the four verses of section 71 are quoted in the Prasannapadā pp. 156–15716. As Sthiramati, the commentator of the Kāśyapaparivarta, did not yet know a text which contained verses apart from those in sections 136-137, he concluded that the text, as transmitted in von Staël-Holstein's manuscript, the Tibetan version and the Sung version, was established in the period
13 Cf. F. Weller, Buddhist Yearly 1968/69, pp. 91, 85 and 75, L. R. LANCASTER, An
Analysis of the Astasāhasrikāprajñāpāramitāsūtra from the Chinese Translations (The University of Wisconsin, Ph. D. 1968), pp. 393, no. 198, p. 388 no. 144 and p. 392, no. 185. Cf. also PAUL PELLIOT, Pāpiyān > Po-siun', T'oung Pao, 30 (1933),
p. 92. 14 Cf. MOCHIZUKI SHINKO, Bukkyo daijiten, vol. 8 (Tōkyō, 1958), p. 259. 15 SHIOMI TETSUDŌ, 'Ryūju shoin no Daijo kyõten no ni-san ni tsuite', Shūkyo kenkyū,
N.S. IX, 6 (1932), pp. 1031–1044; TSUKINOWA KENRYŪ, Kohon Daihoshakkyo ni tsuite', Bukkyōgaku no shomondai (Tokyo, 1935), pp. 849-869 | = Butten no hihanteki kenkyū (Kyoto, 1971), pp. 393-407); KUNO, op. cit.; ONO, op. cit.; HASEOKA KAZUYA, Jūjūji-basharon ni okeru Kaçyapaparivarta no inyo ni tsuite', Indogaku Bukkyōgaku kenkyū, 2 (1954), pp. 553-556; AMANO HIROFUSA, 'Höchōkyō ni tsuite', ibid., 4 (1956), pp. 464-465; HACHIRIKI HIROKI, 'Purasannapada no inyo kyõten (2).
Ratnakūtasutra no inyo ni tsuite', ibid., 15 (1967), pp. 720-723. 16 Op. cit., p. 863.

Page #9
Sanskrit Fragments of the Kasyapaparivarta 255 between Sthiramati and Candrakirti. This is certainly possible, although it must be pointed out that other quotations in the Prasannapada do not entirely agree with the text of von Stael-Holstein's manuscript. This, however, is only to be expected in the case of a text which has been transmitted over centuries and which has been continually expanded. In any case, the quotations of the Kasyapaparivarta are capable of throwing much light on the history of the text. For instance, it is very interesting to see that the comparison of the bodhisattva with the kalavinka in section 84 is absent in the Chin and Ch'in versions, but is already quoted in the Prajnaparamitopadesa which was translated about 404 A.D.17. It is interesting to note that this comparison is also absent in Sthiramati's commentary which was translated into Chinese in the period 508-535, and in a hitherto unnoticed Chinese version of the Kasyapaparivarta, chapter 7 of the Ratnameghasutra (Taisho no. 659, vol. XVI, pp. 276-283). This latter version was recently discovered almost simultaneously by Takasaki Jikido18 and by Nagao Gajin19. Nagao shows that this version is closely related to the Ch'in version. According to him the translator of it was not Man t'o lo hsien from Funan, but Subodhi from Funan who worked as translator during the Ch'en dynasty (557-589). The Sanskrit fragments in London, Helsinki and Leningrad confirm the evidence which can be gained from the Chinese versions and the quotations of the Kasyapaparivarta in other texts. The Kasyapaparivarta, in which the verse parts are later than the prose parts, offers an interesting example of a text in which the verses, written in Buddhist Hybrid Sanskrit, are definitely later than the prose parts, the language of which is much closer to standard Sanskrit. 17 Taisho no. 1599, p. 266c. Cf. KUNO, op. cit., p. 92. The same comparison is already found in the Ratnakarandasutra, cf. Siksasamuccaya (ed. CECIL BENDALL, Bibliotheca Buddhica, vol. I, 1897-1902), p. 6.11-15; Taisho no. 461, p. 45405-9, no. 462, p. 468b25-29. Taisho no. 461 was translated by Chu Fa-hu in 289, cf. Taisho no. 2145, p. 70. 18 Cf. Nyoraize shisa no keijo (Tokyo, 1974), p. 449. 19 <<"Kashohon no shohon" to "Daihoshakkyo" seiritsu no mondai', Suzuki gakujutsu kenkyu nenpo, 10 (1973) (published in 1974), pp. 13-25.