________________
DECEMBER, 1873.]
is so important and interesting that I trust you will allow me space for a few remarks on the reply of Prof. Hoernle, published in the July number of your valuable periodical. As regards my view on the different kinds of Prâkrit, I agree with Mr. Beames, that none of the Prakrits was ever a spoken language, and that in order to learn what was the spoken language of the Aryans we must turn principally to the modern vernaculars. I have never had any other opinion on this subject, and in this respect there is no controversy at all between Prof. Hoernle and myself. But I am sorry to see that Prof. Hoernle still adheres to the error which I had already pointed out in my review of his essays. It is perfectly erroneous to say that Vararuchi's sûtras are founded upon the plays, or that the plays are founded upon Vararuchi's sutras. The language of the plays is Sauraseni, and the language taught by Vararuchi in the first nine sections is Mahârâsatri, of which dialect comparatively few instances occur in the plays. Now it is clear that a man who teaches the Mahârâshtri will not derive the rules for that language from the Sauraseni. It is true that Vararuchi, XII. 32, distinctly says sesham Maháráshtrivat, and that on the whole he does not make many exceptions from the principal Prâkrit. But this is only one of his numerous blunders. Later Prakrit grammarians, especially Râmatarkavagisa and Markandeya Ravindra, who treat more carefully of the lower dialects, have a good many more rules, which are confirmed throughout by the plays. Vararuchi's rules in the first nine sections are derived from works like the Saptasati and the Setubandha, which were written in Maharasthrt and composed in verse. This is clearly proved by the corresponding rules of Hemachandra, who adds numerous examples which are exactly like the poems of the Saptasati, and several of them already to be found in Prof. Weber's edition. Hence it is ridiculous to affirm that the Prakrit of the plays has been grammarized by Vararuchi and his successors.
MISCELLANEA AND CORRESPONDENCE.
The imaginary participle kunno can by no means be used to explain the Gujaráti postpositions. That the colloquial has many forms which in the literary language are restricted to poetry is an old story, but those words are then of frequent occurrence in either the colloquial or the poetry; kunno, however, is not yet found, and I have not met with it, though I am in possession of extensive materials drawn from manuscripts. Prof. Hoernle is very partial to words formed according to analogy; but such words never prove anything; if the participle kunno had given rise to the Gujaráti postpositions, it ought to be found very often. The principal question, however, is that concerning
367
the genitive postpositions in Bangali and Oriya. I think still that it is very easy to prove that Prof. Hoernle is in error. In fact there are no post positions at all in Bangali and Oriya, and these two languages must be separated at once from all the rest. Prof. Hoernle remarks that my statements as to the use of keraka have no particular bearing on the question whether the Bangâli er is a curtailment of keraka or not. My arguments already intimated in my review, where I have tried to state them as briefly as possible, are as follows-Firstly, the word kera is the original of the word keraka, and hence it follows that kera has not been curtailed, but, on the contrary, has been lengthened. The word kera or keraka is found in the Mahârâshtri, the Saurasent, and the Magadhi; it is found in the various Apabhrañáâs as well as in the vernaculars. In the Sinhalese language, as Prof. Childers informs me, it is used to form the locative of a certain class of words. Prof. Kern has lately called attention to the very common use of this word in the language of the gipsies; but even there kero has not been changed in the least, but has remained unaltered to the present day, as stated by Prof. Pott, Paspati, and other authorities. The word, though not noticed by Vararuchi, is well known to the later Prakrit grammarians. Hemachandra, VIII. 2, 147, has a special sûtra running thus: || idamarthasya keraḥ ||
idamarthasya pratyayasya kera ity âdeso bhavati | yushmadiyah tumhakero | asmadiyah amhakero na cha bhavati | malapakkho | panini. Since Hemachandra in the following sutra: || pararajabhyam kkadikkau cha || expressly mentions the two words para and rdjan, I am inclined to suppose that the use of kera was originally restricted to the same words which, according to Pânini, may assume in Sanskrit the suffix kiya. This question I shall discuss at full length in my edition of Hemachandra's Grammar. A sutra corresponding to that of Hemachandra occurs in Markandeya, fol. 28 b; and in the Trivikramavritti II. 1, 8, we have: || kera idamarthe || idamarthe vihitasya chhapratyayasya kera ity âdeso bhavati and now Trivikrama, as usual, gives the same examples as Hemachandra. Simharaja, fol. 43 a, has the same sûtra. Hemachandra mentions the word again in the section on the Apabhrañsa, VIII. 4, 422: || sambandhinaḥ keratanau || gaaii su kesari piahujalu niéchimtai harinaim | jasu kerem humkâradem muhahu padamti trinkim]. The same is given by Trivikrama, III. 3, 51, and means in Sanskrit: gatas sa kesari pibantu jalam niśchinta harinah yasya (sambandhina) humkarena mukhât patanti tripâni: "The lion is gone; without fear may the antelopes drink the water; (the lion) by