________________
(
120
)
Kadambart itself is another example of the same phenomenon. I have shown that the author of the Sahityadarpaņa knew that Kadambari was a redaction into prose of a tale already existing in a material form, and it is clear now that the coincidences Wilson has the credit of first observing should have been accepted by that scholar as supplying corroborative evidence of the truth of Somadeva's assertion. He preferred, however, to believe that Somadeva had borrowed from these earlier prose-writers, and that his assertion about a Vșihatkathâ, presumably prior both to him and to them, was intended to conceal his debt; and in this he was followed by other eminent scholars. The utmost respect for one of the most honoured and honourable pioneers in these studies need not prevent the obvious moral being pointed that it may sometimes be the safer course to accept statements like that made by Somadeva until their inaccuracy can be demonstrated, instead of regarding them from the beginning as tainted with fraud.
In a list of Sanskrit works prepared for Captain Wilford, and presented by him through Colebrooke to the Asiatic Society, there is an entry with regard to the Vșihatkathê by Kshemendra. This work Bühler had the good fortune to recover ;* and to him we owe the discovery that Kshemendra, like Somadeva, calls his work a traslation into Sanskrit of a work written in the Paiśâcha language by Gunadhya. "Śarva proclaimed it first : Kâņabhati heard it from the Gana (Pushpadanta-Vararuchi), and told it to Gunâdhya, who delivered it in his turn to his pupils and to Satavahana. The story which had thus come to be written in the Paiśâcha language, gave trouble to the readers. For this reason it has been rewritten in Sanskrit.”+ While as yet uncertain as to Kshemendra's date, Bühler was able to show from a comparison of the two books, whose contents are identical, that it was extremely improbable that the one was, in any way, & modification of the other, and that therefore the only reasonable supposition was that both really drew, as they said they did, from an older work in one of the low vernaculars. In the course of his brilliant researches in Kashmir, Bühler was later, able to fix Kshemendra's date as prior, by some seventy years, to that of Somadeva. He had already argued that it was impossible to admit that "the connected and clear story" given by Somadeva had been constructed out of the "short and undefined outlines," found in Kshemendra ; and, it being now clear that Kshemendra in his turn could not have copied from a poet who flourished two generations later than himself, no doubt remained as to the existence in the tenth century of a Vpihatkatha by
• There is a copy of part of the work in the Bhao Dâji collection. + Indian Antiquary, Vol. I., p. 307.
Bühler's Kashmir Tour, p. 47.
