212 THE INDIAN ANTIQUARY. [July, 1873-. believe kera to be rather a corruption of kdryaP* So Prof. Lassen is not quite so positive as my critic represents. Prof. Weber (Hdla, p. 38), treat¬ ing of the changes of d into e, says that it changes so sometimes under the influence of a following y, as sejjd (sayyd); achchhcra (dscharya) ; maha keram (mama krite). This does not show that he is more positive than Prof. Lassen. The fact is that they are both too cautious and "too well-informed scholars to commit themselves to such a dog¬ matic statement on insufficient data. I do not know on what grounds Prof. Hoefer may have supported tho traditional interpretation, as unfor¬ nately I am not able to refer to his work. But that it is the true interpretation the modern ver¬ naculars conclusively prove. In Mar&thi the equivalent of krita is kdd, and in the Low-Hindi it is kaild (or kayald or kaild). Now keld or kaild are contractions or modifications of the Pr&krit kelao (or kelo), or kerao (or kero); and it follows that the Prakrit kerao or kero are also equivalents of the Sanskrit kritaka or krita. The interchange of r and X is so common that it needs no remark. Its extreme frequency in the modern vernaculars shows that in colloquial Pr&krit it must have been even more frequent than in literary Pr&krit. The l of kelao is a substitute for d, and d again for the Sanskrit t; namely, krita becomes kada, and kada becomes kela or kelaa. This disposes of one of the two difficulties of Prof. Lassen, which was the r in kei-aka in the place of the Sanskrit t. This assumes that the form kelao is the earlier one; but even if the other form kerao be thought the earlier, the r can be explained by the help of the modern vernaculars. The Low-Hindt has still a past part, kard for Sanskrit krita (just as mard for mrita, dhard for dhrita, etc.). Here we have r in the place of the Sanskrit t, however it may have originated. For my own part I am inclined to believe the origin to be this. In Pr&krit, roots in ri not uncommonly form tho past part. pass, with tho connecting vowel i (comp. Lassen, Inst. Prakrit, p. 363); thus bhri has bharita, dhri has dharita, etc. (I give the full phonetic ground- forms). Thus kri would form karita, that is, in Prakrit kario (or kariao), which is actually pre¬ served in the old Hindi form karyau (e. g. Chand, xxvn. 60), and in Modern Hindi is contracted to kard. Now the Prakrit forms kario or kariao would easily explain the forms kero or kerao, by the translation of the vowel i into the preceding syllable; just as achchhario contracts into ach- chhero. This disposes of the second difficulty of Prof. Lassen (p. 118), which is that the vowel a changes to e only under the influence of a follow¬ ing i or y. The difficulty, however, may not be so absolute as Lassen seems to have thought. In some instances such an influence is doubtful. The supposition is, therefore, quite allowable, that the Prdkrit past part, form kalo (in Mrichchh. Calc, edn. for kado) might be the original of the form ke¬ lao or kerao. This was my theory formerly, which was briefly stated by me on p. 133. Nevertheless my critic insinuates that I made the e of kelaka to bo a modification of the Sanskrit ri; and then he proceeds to knock down the man of straw of his own creation. (And, by the way, what are we to think of a Prdkrit form keta, to which my critic thinks krita might change P) Further on Dr. Pischel 6ays that I “ believe that in some examples keraka has become a sort of affix; if this be true, it ought not to be inflected, as it really is,” like all other adj. nouns. Now the substance of what I said was this, that in some instances keraka has no (predicative) meaning, but merely determines the case of another noun, and that in this respect it had become like an affix (see p. 130). Now this is altogether a dif¬ ferent thing from what Dr. Pischel attributes to me. That keraka is an adj. noun and treated as such, I know very well; in fact, it is the whole drift of my second essay to prove that the Hind- genitive post-positions are curtailments of Buch an adj. noun (see p. 125). Again, Dr. Pischel adduces a number of other words, aB kajjam, kichchath, etc., which he says are used exactly in the same way as I say keram or kerakam is. This is again a mis¬ understanding. What I maintain is, that keraka is used very often pleonasiically, or to form a periphrastical genitive, as amhakerao for amhdnam. Now the words instanced by Dr. Pischel are not used pleonastically; for if omitted in the sentences quoted, the sense of the latter would be incomplete or none at all; and, moreover, they are used to form a periphrastical dative, not a genitive. These means of forming a periphrastic dative are well known. Keram is one of them. But keram in this particular use was irrelevant to my purpose. Dr. Pischel will find it discussed in a future paper on the dative post-positions, which I shall try to show can be traced back to it. As regard the three words nija, gada, sandha, they are never used pleonastically, certainly not in the instances quoted by Dr. Pischel; e. g. if gadena were left out in the phrase taggadena ahild- sena, its sense would become doubtful; it might mean both “ by his desire for her” cr " by her desire.” Again if niam be omitted in the sentence • “ Hinc ldra-ka a kdrya pot ins depravatam crediderim”—Curiously, though no doubt wrongly, M. Williams, in hi3 £ak. p. 289, concludes from Lassen’s words that ho adhered to the usual derivation of keraka from the Sanskr. krita.
