________________
NOTE ON THE DRAVIDIAN CASES
JUNE, 1911.]
refutation by the former would have no effect on his kosa." Vasubandhu's hope, that this literary production of his genius would be immortal, was amply realized, because the study of this kosa was so universally popular in the first half of the seventh century that, "even devout parrots expounded it." Bâna says:---
farewed: vetrarmâ qåde uraaurangrait: aird agaftafe
171
Harsha-charita, VIII, p. 317.
Bombay Sanskrit Series Edition.
Here the word kosa is explained by the commentator, Sankara, as antet utafagredì veCELENCI, Bâpa is misunderstood and mistranslated by Prof. Macdonell, when he tells his readers that "pious parrots expounded a Buddhist Dictionary" (History of Sanskrit Literature, p. 333). This testimony of the Brahman poet Bâns to the immense popularity enjoyed by the Buddhist author Vasubandhu and to the fact that to explain the Abhidharma-kosa was a very common attainment in the first half of the seventh century is very important. We need not, therefore, be surprised that the rhetorician Vâmana has preserved for us the historical fact that Vasubandhu enjoyed the patronage of Kumaragupta. The interesting half-verse, which Vâmana has rescued from oblivion, is evidently taken from some lost Guptavamsamahakarya, in which the name of Vasubandhu is directly mentioned or which was composed by Vasubandhu himself, to congratulate Kumâragupta on his accession to the throne, as the word 'samprati in the verse shows. It may be hoped that manuscripts of this Guptavamsamahakarya, or whatever it may have been really called, may yet be recovered in Kâśmir, where Vasubandhu spent many years of his life.
NOTE ON THE DRAVIDIAN CASES.
BY P. SESHACHAR, ESQ.; GOKARAM.
In the very interesting contribution to a Comparative Grammar of the Dravidian languages' on Dravidian Cases, by Mr. K. V. Subbayya, M.A., L.T., M.R.A.S. (above, May 1910), we are informed that the primitive Dravidian termination of the accusative was am, found in an unaltered form in old Kanarese,' I am inclined to believe that the primitive termination in Kannada was not am, but un, as seen from the samdhi rules applied to substantive accusatives in combination with words having an initial vowel; cf., Nripanan-abhimanadhananan-atiśáya viś dla kirtidhvajanu (Kaviraja-marga II, 16), palavuman-odagadire (II, 18), Kandosedan banadolage Janakatanagatan. Anuvan (II, 38), padangalan-amardire (II, 83). This is true of all genders and numbers. In the same paragraph we have In Mid, and New Canarese the m of am is softened to n and the transformed termination takes a final euphonic u, thus becoming anu or anna.' This is true without the softening' (?), since the primitive an can enphonically become anu or annu. For instance, we have bhagavanu, accusative of bhagava, god? I have not been able to trace this bhagava or bhagavanu to any period of the Kannada language unless as in Modern or New (so-called) Kanarese, both forme be regarded as accusatives of the Sanskrit bhaga, which evidently does not signify 'god.' The proper nominal theme in this instance would be bhagavanta from the Sanskrit bhagavan [see Sabdanusasana-sutra 129; Sabda-mani-darpana 86].
In quoting the Sabda-mani-darpana 115, we have to remember, that it is not the ge of the dative that is optionally doubled, but the g of the ge termination,
In connection with the augment in of the genitive, apparently Mr. Subbayya uses Sabda-manidarpana Sairas 108 and 109; but there is contradiction in (1) and (2) with regard to words ending in consonants which is not explained. Comparative study of the fórms would render (1) untenable