The Bodhisattvãvadãnakalpalata and the Saddantãvadãna
The Bodhisattvavadanakalpalata and the Saddantavadana
The Bodhisattvavadanakalpalata) was published in the Bibliotheca Indica in two volumes. Sarat Chandra Das and Pandit Hari Mohan Vidyabhushana edited the first five fascicles of Volume 1 (Calcutta, 1888-1895, pp. xlii + 1-442) and the first five fascicles of Volume 2 (Calcutta, 1890-1897, pp. 1-480). After a long interval, publication was resumed by Das in 1906, and with Satis Chandra Vidyabhusana he edited the remaining fascicles of both volumes (Vol. I, Fascicles 6-13, pp. 443-1171, Calcutta, 1906-1913; Vol. II, Fascicles 6-11, pp. 481-1093 +13 pp., Calcutta, 1910-1913). In 1959 P. L. Vaidya reprinted the Sanskrit text in Volumes 22 and 23 of The Buddhist Sanskrit Texts.
Das’s edition is based upon a Tibetan block print which contains both the Sanskrit text in Tibetan transliteration and the Tibetan translation. According to him this blockprint consists of 620 folios and was printed in 1662-1663.2 In editing the Bodhisattvavadanakalpalata, Das has done some rearrangement of the text. In the Peking edition of the Tanjur the Bodhisattvavadanakalpalata occupies Vol. 93 of the Mdo-’grel? Story 107 ends on page 346al.
Then follows Somendra’s introduction to the last tale composed by himself: 346al-347b2 ( = Das Vol. 2, pp. 1008-1015). This tale oc-28
cupies ff. 347b2-357a7 ( = Das Vol. 2, pp. 1016-1087). Then fol- lows Somadeva’s introduction to the Bodhisattvavadanakalpalata.'. ff. 357a8-358a6. This introduction has been published by Das on pp. xxiv-xxix of his introduction. The table of contents of the Bodhisattvavadanakalpalata occupies ff. 358a6-360a5 (cf. Das, Introduction, pp. xxx-xli). This table contains 42 verses and not 43.
The 43d verse in Das’s edition is the first of the four verses of the colophon for which see Das, Vol. 2, pp. 1088-1091 ( = Tibetan translation, ff. 360a5-360b3). This colophon is followed in the Peking edition of the Tibetan translation by the colophon of the translation, ff. 360b4-361a8). The first lines of this colophon (ff. 360b4-6) are also found in the colophon of the blockprint used by Das (cf. Vol. 2, p. 1092, lines 1-7 of the Tibetan text). The same block print also contains a lengthy text edited with separate pagination (pp. 1-13) by Das at the beginning of Fascicle 11 of Volume 2. According to. Das this text contains the “concluding remarks of the last Tibetan editor.”
In the Tibetan translation the tenth pallava is called Mngal-las ’byung-ba. However, the Tibetan block print used by Das does not contain the Sanskrit text of this pallava. For this reason Das has relegated it to the end of Volume 1 (pp. 1165-1171). Moreover, Das has changed the numbers of Pallavas 11-49 to 10-48. Consequently, there is no Pallava 49 in his edition. This rearrangement of the pallavas agrees with the table of contents, which lists as the tenth pallava the story of Sundarinanda. According to this table the forty ninth story is the Saddantavadana, text and translation, which are lacking in the Tibetan block print and in the Peking edition of the Tibetan translation.
It is obvious that in the text used by the Tibetan translators one story was missing. According to Satis Chandra Vidyabhusana (Vol. 1, p. 1171, footnote) the Mngal-las ’byung-ba was evidently an interpolation introduced to make up the auspicious total of 108 pal-lavas. Tucci speculates that the forty-ninth pallava, the Saddanta-vadana, was lacking in the text on which the Tibetan translation was based and that, for this reason, the editors of the Tibetan translation compiled the Mngal-las ’byung-ba. Tucci does not explain why the editors have filled the gap caused by the absence of the forty ninth story by adding a story after the ninth with the consequence that Stories 10-48 had to be renumbered 11-49.J. W. de Jong
The Mngal-las ’byung-ba, ‘The coming forth from the womb’, is a sermon preached by the Buddha to Ananda near Campa on conception, birth, and the miseries of human life. Vidyabhusana reconstructs the Sanskrit title as Garbhakrantyavadana, but Tucci prefers Garbhavakranti. A Garbhavakranti-sutra is quoted in the Abhidharmakosabhasya and the Yogacarabhumi (ed. V. Bhattacharya, Calcutta, 1957, p. 27.5). The Abhidharmakosavyakhya (ed. U. Wogi- hara, Tokyo, 1932-1936, p. 67.1) refers to the Garbhavakranti-sutra, but the Abhidharmakosabhasya omits the word sutra. Cf. Abhidharmakosabhasya (ed. P. Pradhan, Patna, 1967, p. 24.10): saddhatur iyam purusa iti garbhavakrantau. The Tibetan translation of the bhasya renders Garbhavakranti with Mngal-du ’jug-pa.
The Mngal-du ’byung-ba is not identical with the text quoted in the Abhidharmakosabhasya and other texts. A reconstructed Sanskrit title would not be Garbhavakranti butGarbhotpatti. As to the Garbhavakranti-sutra, La Vallee Poussin refers to Chapter 11 of the Vinayasamyukta-kavastu (Nanjio 1121, Taisho 1451), to Chapter 14 of the Ratnakuta (Nanjio 23.14, Taisho 310.14), and to the Dhatuvibhahgasutta in the Majjhima-nikaya (No. 140).
Moreover, he adds that the Garbha-vakrantisutra is one of the sources of the Pitaputrasamagama which is quoted in the Siksasamuccaya, the Bodhicaryavatara, and the Madhyamakavatara. However, he has not checked whether the quotations of the Garbhavakrantisutra in the Abhidharmakosabhasya can be traced in the texts mentioned by him. He mentions only Chapter 14 of the Ratnakuta, but both the Chinese and Tibetan translations of the Ratnakuta contain two texts, entitled Garbhavakrdntinirdesa. According to the Peking edition of the Kanjur, the full Sanskrit titles are Ayusmannandagarbhavakrantinirdesa and Nandagarbhavakrantinirdesa?
Pelliot has pointed out that the Chinese translation of the Ratnakuta contains two translations (Taisho 310.13 and 310.14) which correspond to Sutras 13 and 14 of the Tibetan version of the Ratnakuta.1 However, in Taisho 310.13 the Buddha is questioned by Ananda; in the corresponding Tibetan text the Buddha addresses himself not to Ananda but to Nanda. Pelliot remarks that in an older Chinese translation by Dharmaraksa (Taisho 317) Nanda figures in the beginning but is later replaced by Ananda. Marcelle Lalou has3
pointed out that the Tibetan text was translated by Chos-grub from the Chinese translation by Bodhiruci (Taisho 310.13).8 Pelliot had already advanced the hypothesis that this text was translated from the Chinese and that the translator had substituted the name Nanda for Ananda.9 A careful comparison of both texts will be required in order to show whether this is the only substantial difference between the two texts.
As concerns Sutra 14 of Ratnakuta, the situation is more complicated. Pelliot had pointed out that Chapters 11 and 12 of the Vinaya of the Mulasarvastivadin are absolutely identical with Chapter 14 of the Chinese Ratnakuta. The Vinayaksudrakavastu was translated by I-tsing, and in compiling the Chinese Ratnakuta Bodhiruci therefore must have made use of I-tsing’s translation of Chapters 11 and 12. The Tibetan translation of the Vinayaksudrakavastu contains, according to Csoma’s analysis (folios 202-248 of Volume 10 of the Narthang edition of the Vinaya} instructions to Nanda on the conditions of existence in the womb and on the gradual formation of the human body.
Pelliot concluded that probably this sutra too had been translated from the Chinese. Sakurabe Bunkyo arrived at the same conclusion in his study of the Ratnakuta.12 Marcelle Lalou, however, compared the Tibetan translation of Chapters 11 and 12 of the Vinayaksudrakavastu with Sutras 13 and 14 of the Tibetan Ratnakuta and showed that the text of Sutra 13 is different from that of Sutra 14 and that the latter is not identical with the text of the
Vinayaksudrakavastu. This conclusion, though, does not exclude the possibility that Sutra 14 of the Tibetan Ratnakuta was translated from I-tsing’s version of Chapters 11 and 12 of the Vinayaksudrakavastu. It is quite possible that I-tsing’s translation of these two chapters is not completely identical with the Tibetan translation of the same chapters. A final solution will require a close comparison of the Chinese and Tibetan versions of Sutras 13 and 14 of the Ratnakuta with the Chinese and Tibetan versions of Chapters 11 and 12 of the Vinayaksudrakavastu.
In Sutra 14 of the Tibetan Ratnakuta, Buddha is first at Kapila- vastu and then goes to Sravasti. From Sravasti he goes to Campa, and it is here on the banks of the pond of the rsi Garga that he teaches Nanda the Garbhavakrantisutra. In Chapter 11 of the Vinayaksu-J. W. de Jong
drakavastu Buddha teaches the Garbhavakrantisutra to Nanda at exactly the same place.13 The Vinaya of the Mulasarvastivadin was well-known to the Tibetans. The fact that Buddha taught a [[Garbhavakrantisutra to Nanda on the banks of the Pond of Garga must have been in the minds of the compilers of the Mngal-las ’byung-ba, which is also set on the banks of a lotus-pond near Campa.
Although they substituted Ananda for Nanda they must have been aware of the fact that a Garbhavakrantisutra is found in the Buddhist canon in connection with the story of Nanda. This is certainly the reason why the Mngal-las ’byung-ba is placed in the Tibetan translation of the Bodhisattvavadanakalpalata before the story of Nanda, which is No. 11 in the Tibetan translation and No. 10 in Das’s edition. In the Sanskrit text of the table of contents of the Bodhisattva-vadanakalpalata no mention is made of the Mngal-las ’byung-ba.
However, in the Peking and Cone editions of the Tibetan translation the title of this text has been mentioned in an additional pada of Verse
4: gang-zhigdpal-sbas la bstan dang / / me-skyes skal-ldan du (Peking: dus) gsung dang / / mngal-nas ’byung-ba bstan-pa dang I / gang-zhig dga’-bo’i mdzes-ma la / / chags-pa dag ni ’bad-pas bsal (Peking: gsal) / /.
It is obvious that this pada has been added later in order to account for the presence of the Mngal-las ’byung-ba.
The Saddantavadana is mentioned in both the Sanskrit text and the Tibetan translation of the table of contents.
In his detailed bibliography on the Saddantajataka, Lamotte indicates that the Saddantavadana is not found in the Paris manuscripts of the Bodhisattvavadanakalpalata. However, he points out that the two Cambridge manuscripts, Add. 1306 and Add. 913, contain this avadanad5 Add. 1306 is a manuscript written in a.d. 1302.16 According to Somendra’s introduction the Bodhisattvavadanakalpa-lata was completed in the twenty-seventh year, i.e., 1051-1052. The Cambridge manuscript is therefore written 250 years after the completion of the work. Bendall has described the manuscript in detail.17 Leaves 1-174 are missing, and the manuscript begins with the last word, sahisnavah, of Verse 7 of Tale 42, Panditavadana. Bendall remarks that in the manuscript Tales 41-48 are numbered 42-49. He has changed the numbering according to the metrical table of con-32
tents. However, the numbering of the manuscripts agrees entirely with that of the Tibetan translation of the Bodhisattvavadanakal-palata in which Tale 42 is the Panditavadana. In Das’s edition this is Tale 41, wrongly called Kapilavadana. The table of contents also gives the name Pandita. If we keep the numbering of the tales as found in Add. 1306, Tale 49 (Hastakavadana) ends on f. 198b. Tale 50 (Dasakarmaplutyavadana) begins on f. 199b: namo buddhaya / ye helocchita-.
However, this manuscript contains seven extra leaves numbered 199-205. Bendall has given them the numbers 199*-205*. The Saddantavadana begins on the last line of f. 198b and occupies the leaves 199*-205*. It is obvious that the scribe completed the first part of the Bodhisattvavadanakalpalata (Tales 1-49) on f. 198b and continued with the second part on f. 199b. According to Bendall the scribe had by accident omitted this tale and copied it in afterwards.
Bendall’s conclusion was certainly justified because the table of contents lists the Saddantavadana as the Tale 49. However, with the publication of Das’s edition it has become evident that the Saddantavadana was missing in the Sanskrit text translated in Tibet. It must also have been missing in the manuscript used by the scribe of Add. 1306, Manjusribhadrasudhi. When copying the table of contents Manjusribhadrasudhi must have made the same discovery as Bendall, i.e., that the Saddantavadana is listed as Tale 49.
In order to supply this missing tale the scribe made use of another collection of tales which contains a recension of the Saddantavadana'. the Kalpadrumavadanamala. Both the Paris and Cambridge manuscripts contain the text of the Saddantavadana.^ In the Kalpa-drumavadanamald the tale is comprised of 198 verses. They are followed by several additional verses of a moralistic nature which do not belong to the story itself, and which need not be considered. The scribe of the Bodhisattvavadanakalpalata did not use all 198 verses.
He reproduced 110 verses without any alteration and added eight others, most of which were made from padas of verses of the Kalpa-drumavaddnamala recension of the story.
Feer has studied the Kalpadrumavadanamala recension of the Saddanta story together with other recensions. However, inJ. W. de Jong order to show how the scribe of the Bodhisattvavadanakalpalata made use of the Kalpadrumavadanamald recension, it is necessary to give a summary and to indicate the Kalpadrumavaddnamald verse-numbers.
Verses 1-4: Introduction. Asoka asks Upagupta to tell another tale. 5: A good man is purified by the fire of a bad man (durjanagni) just as a jewel shines after having been polished by a whetstone. 6-11: Buddha teaches the law at the Garga Pond near Campa. 12-31: Devadatta warns the ksapanakas against the Buddha. 32-37: His words provoke different reactions among them. 38-58: A ksapanaka says that he knows a way to destroy the reputation of the Buddha. He asks Canamanavika to simulate pregnancy and to accuse the Buddha of having made her pregnant. She fastens a wooden bowl under her garment. 59-88: Cancamanavika goes to the Buddha and accuses him of having made her pregnant and of having abandoned her.
The Buddha is unperturbed but the gods are greatly upset. Sakra creates two rats who cut the cord which - holds the wooden bowl. Crying “I am burnt,” Cancamanavika disappears in the flames of Hell. 89-94: The Buddha explains that she has been guilty of a grave sin in a previous existence. 95-123: The Elephant King Saddanta lived happily in the Himalayas with his two wives, Bhadra and Subhadra.
Once he played with Subhadra in the lotus pond Mandakini. Bhadra became jealous and decided to take revenge. She went to the forest where the munis live and took upon herself a fast in eight parts. She expressed the wish to be reborn as a queen and to obtain a seat of pleasure (kridasand) made from the tusks of Saddanta.
She killed herself by throwing herself from a mountain, and was reborn as the daughter of the minister Khandita (mistake for Pandita ?) of King Brahmadatta in Kasi.34
The king married her. She asked him for a seat made from the tusks of Saddanta. The king summoned an old hunter, who tried to dissuade him from killing Saddanta because he was a Bodhisattva. 124-143: The old hunter persuaded the king, but Bhadra insisted on her wish. The king summoned another hunter, who declared himself willing to kill Saddanta. 144-161: Dressed in a yellow robe, the hunter was seen by Subhadra. She told the king [Saddanta] that she was frightened, but the king explained that she had nothing to fear from someone who wears a yellow robe.
He had just spoken these words when the hunter pierced him with a poisoned arrow. Subhadra fainted, but Saddanta consoled her and asked the hunter why he wanted to kill him. 162-165: The hunter explained that Queen Bhadra desired a seat made from his tusks. 166-184: Saddanta arrived at the conclusion that he must give his tusks to the hunter, because it was impossible to disappoint someone who came with a request. He broke, off his tusks against a mountain. Five hundred elephants arrived, but Saddanta protected the hunter with his chest and sent him back with his tusks. 185-189: The hunter brought the tusks to the king, who recompensed him with gold. He sent him back to his own house.
Suddenly both his hands were cut off and fell on the ground. 190: Bhadra mounted the seat made from the tusks. Saying “I am burnt,” she fell into Hell. 191-192: Brahmadatta’s kingdom was destroyed by terrible plagues. 193-198: The dramatis personnae are identified. Saddanta = the Buddha; Bhadra = Cancamanavika; the hunter = De- vadatta’ the other elephants = monks. There are two verses on the evil behaviour of women. In the last verse the Buddha proclaims that one must speak the truth, refrain from inflicting injuries, and concentrate on santi.J. W. de Jong
The scribe of the Bodhisattvavadanakalpalata took from the Kalpadrumavadanamdla recension the following verses: 5, 59- 123, 144-161, 166-184, 190, and 193-198. In order to fill the lacunae he added five verses (A-E) between Verse 5 and Verse 59, one verse (F) between Verses 123 and 144, one verse (G) between Verses 161 and 166, and one verse (H) between Verses 190 and 193.
A-B: The Buddha preaches the law at the Garga Pond near Campa. C-E: The jealous ksapanakas say, “You must destroy the lustre (dlpti) of the Buddha by saying that you have been made pregnant by him.” The young woman simulates a pregnancy by means of a wooden bowl. F: A second hunter declares himself willing to kill Sad- danta. G: The hunter says that Queen Bhadra wants to have a seat made from Saddanta’s tusks. H: The hunter loses his hands, and Brahmadatta’s kingdom is destroyed by excessive rains.
It is obvious that the scribe of the Bodhisattvavadanakalpalata was more interested in the story of the past concerning Saddanta than in the story of the present relating to Cancamanavika. Through the omission of Verses 6-58, nothing is said of the role played by Deva- datta, although identification of Devadatta with the hunter (Verse 194) has been maintained. Moreover, verses C-E do not explain why the ksapanakas are jealous nor the identity of the young woman whom they ask to simulate pregnancy. It is equally obvious that the scribe of the Bodhisattvavadanakalpalata has made use of the Kalpadruma-vadanamald. In a long note added to the English translation of his article on the Saddanta-jataka, “Essai de classement chronologique des diverses versions du Saddanta-jataka”
{Melanges d ’Indianisme, Paris, 1911, pp. 231-248) Foucher writes that “The author of the latter collection [Kalpadrumavadanamdla} restricted himself to reproducing, without however (in any way) informing the reader of the fact, the work of Kshemendra, except that on two points he has lengthened the narrative of his predecessor, which in his opinion was too much abbreviated.” 20 I hope to be able to publish shortly the text of the Kalpa-drumavaddnamald recension of the Saddantavadana including the eight verses added by the scribe of the Bodhisattvavadanakalpalata.
It will then become absolutely clear that Foucher was wrong in assuming that the Kalpadrumavadanamdla recension is based upon the36 The Bodhisattvavadanakalpalata and the Saddantavadana
Bodhisattvavadanakalpalata recension. It is not possible to prove that the scribe of Manuscript Add. 1306, Manjusribhadrasudhi, himself took the Saddantavadana from a manuscript of the Kalpadrumavadanamala, but the similarity of the script in the Saddantavadana to that in other parts of the manuscript of the Bodhi-sattvavadanakalpalata makes this supposition highly probable.
The fact that the Saddantavadana is listed in the table of contents as the forty-ninth avadana obliges us to assume that originally the text contained this story. It was, however, already missing in the copy which was translated in Tibet in the second half of the thirteenth century.21 It is difficult to find a satisfactory explanation for the disappearance of the Saddantavadana. This is not the only problem connected with the Bodhisattvavadanakalpalata. It was completed by Ksemendra in 1052, but he did not compose Tale 108. This is surprising in view of the fact that he was still living in 1066 (when he wrote the D as avataracarita) 22 Somendra does not explain why his father, after having composed 107 tales, did not complete his work by writing the 108th. If it had been Ksemendra’s wish that his son fulfill this task, one would expect Somendra to have mentioned this.
NOTES
1. The Bodhisattvavadanakalpalata is often referred to as Avadana-kalpalata. However, according to all the colophons and the Tibetan translation the title is Bodhisattvavadanakalpalata. 2. A copy of the same blockprint edition is listed in A Catalogue of the Tohoku University Collection of Tibetan Works on Buddhism (Sendai, 1953), p. 521, No. 7034, but I have not been able to consult it. In the Cone Tanjur the
Bodhisattvavadanakalpalata occupies two volumes (Vols. 91-92: Khri-shing). The Cone edition contains both the Sanskrit text and the Tibetan translation. I have not been able to consult the Derge edition, but it also probably contains the Sanskrit text, though this is not mentioned in the catalogue of the Tohoku University: A Complete Catalogue of the Tibetan Buddhist Canons (Sendai, 1934), pp. 633-634, No. 4155. In the Narthang Tanjur the Bodhisattvavadanakalpalata occupies only one volume. Cf. Mibu Taishun, A Comparative List of the Tibetan Tripitaka of the Narthang Edition (Tokyo, 1967), p. 98, No. 3646, Vol. Ge, ff. 1-328. It would appear that the Peking and Narthang editions contain only the Tibetan translation, while the Derge and Cone editions contain both text and translation.J. W. de Jong
3. Cf. P. Cordier, Catalogue du fonds tibetain de la Bibliotheque Nationale, Troisieme Partie (Paris, 1915), pp. 419-421. 4. Tibetan Painted Scrolls, Vol. 2 (Rome, 1949), p. 613, n. 118. 5. Abhidharmakosa de Vasubandhu, Vol. 1 (Paris-Louvain, 1923), p. 49, n. 2. 6. A Comparative Analytical Catalogue of the Kanjur Division of the Tibetan Tripitaka (Kyoto, 1930-1932), p. 238. 7. “Notes a propos d’un catalogue du Kanjur f in Journal Asiatique, 1914, Vol. 2, p. 123. 8. “La version tibetaine du Ratnakutdf in Journal Asiatique, 1927, Vol. 2, pp. 240, 245. 9. Op. cit.,p. 126, No. 1. 10. Pelliot refers to Chapters 11 and 12 of the Vinayaksudrakavastu, the same text which La Vallee Poussin refers to as the Vinayasamyuktakavastu.
Cf. Taisho, Vol. 24, No. 1451, pp. 251a-263a. 11. Pelliot, op. cit., p. 125. 12. “Chibetto-yaku Daihoshakukyo no kenkyu,” Otani Gakuho, Vol. 11 (1930), p. 550. In his analysis of this article Serge Elisseef says wrongly that Sakurabe tried to prove that the whole Tibetan Ratnakuta had been translated from the Chinese. See Bibliographic Bouddhique, Vol. 2 (Paris, 1931), p.
37, No. 110). Sakurabe observed that Chapters 7, 13, and 40 were translated from the Chinese by Chos-grub and suggested that Chapters 11, 14, 17, and 20 must also have been translated from the Chinese. 13. For the Tibetan version see Lalou, op. cit., p. 242. For the Chinese version see Taisho, Vol. 24, No. 1451, p. 253al7-21. 14. In the Sanskrit text of the Mudasarvdstivadavinaya the name of the pond is Garga. Cf. Edgerton, Buddhist Hybrid Sanskrit Dictionary, p. 210.
In Pali texts the Gaggara Pond is named after Queen Gaggara. 15. Et. Lamotte, Le Traite de la Grande Vertu de Sagesse, Tome 2 (Louvain, 1949), p. 716, n. 1. According to Bendall Add. 913 is a copy of a copy, more or less direct, of Add. 1306. 16. According to Petech the date mentioned in the colophon is Sunday, April 8th, 1302. Cf. L. Petech, Mediaeval History of Nepal (Roma, 1958), p. 98. 17. C. Bendall, Catalogue of the Buddhist Sanskrit Manuscripts in the University Library, Cambridge (Cambridge, 1883), pp. 41-43. 18. Cf. Bendall, op. cit., p. 131, Add. 1590; also J. Filliozat, Catalogue du fonds Sanscrit, Fascicule 1 (Paris, 1941), pp. 14-15. For other manuscripts see Seiren Matsunami, A Catalogue of the Sanskrit Manuscripts in the Tokyo University Library (Tokyo, 1965), pp. 230-231. 19. “Le Chaddanta-Jataka,” Journal Asiatique, 1895, Part 1, pp. 31-85 and 189-223.38
20. A. Foucher, “The Six-Tusked Elephant,” Beginnings of Buddhist Art (Paris-London, 1917), p. 204, n. 1. 21. The Bodhisattvavadanakalpalata was translated by Laksmikara and the Master fron Shong rDo-rje rgyal-mtshan, at the instigation of ’Phags-pa and the Regent Sakya bzang-po. According to Cordier (op. cit.> p. 420) the translation was probably made in the year 1272 a.d. The colophon of the Peking edition does not mention a date, and it is not clear from which source Cordier took the date 1272. From the names mentioned in the colophon it is possible to deduce that the translation was made in the period 1260 to 1280. 22. Cf. Oscar Botto, IlPoeta Ksemendra e il suo Dasavataracarita (Torino,
