________________
266
THE INDIAN ANTIQUARY.
[SEPTEMBER, 1892.
tions which are purely artificial, mingled in a variable proportion with the elements which have been directly supplied by the popular speech. A comparison of the various literary Mahârâshtris, the parallel employment of which I have already mentioned, throws a striking light upon this point of view.
Ay Professor Jacobi (loc. cit.) points out, the Maharashtri of Vararuchi and the poets differs from that of Hômachandra and the Jains in two main peculiarities. The former does not use the ya-sruti, and everywhere substitutes the cerebral n for the dental » : the latter retains the dental · at the commencement of words, and when it is doubled. It is quite natural that the origin of these divergencies has first been sought for, either in diversities of dialect, or in differences of time, but I should be surprised if anyone, with the knowledge which we are now beginning to acquire of Indian epigraphy, could persevere in this view.
So far as concerns the first point, the introduction of a y between vowels - or, according to Hêmachandra, more exactly, between two a's - which form an hiatus, I lay no stress on several circumstances, disagreement between the grammarians, disagreement between the rules of the grammar and the manuscript tradition, which à priori, appear to indicate that this rule is susceptible of arbitrary extensions and restrictions. I content myself with calling the texts of the inscriptions as witnesses. The ordinary orthography is too ready to adopt the niethods of the learned language to allow many hiatus to exist. I have, however, quoted many examples, and I could quote more; bhôa, bhóigi, páüna, chiarika, paithana, bhaasta, pulumai,
phutua ; the spellings chétiasa (Kaņh. 5), pațiasiya (Kaph.4), the terminations pavaitikád, ponakiasanda, (Kaņh. 21), bhayáa (Kaph. 27). It follows that from an epoch earlier than that of our literary authorities, the local pronunciation supported the existence of the histus in Maharashtra, as well as in the other provinces of India. It must be assumed that, there as elsewhere, but not more than elsewhere, the hiatus implied a light utterance-break analogous to the soft breathing. If this has been denoted by means of the y, whether in all, or in special cases, the choice can be explained on the one hand by the imitation of a certain number of terminations of the learned declension, and on the other by the fact that the change in every case of an original y to j, left the sign for y available for a special function. Sometimes the inscriptions apply v for this purpose, as in pulumdvisa (Nâs. 15), bhayáva vélidatáva (Kuda, No. 23), and the parallel employment in this last inscription of the spelling uyaraka, for uvaraka, clearly shews that neither the v in the one case nor the y in the other represented any actual pronunciation. They are merely equivalent expedients for concealing from the eyes a hiatus wbich the recollections of the cultivated language caused to be considered as clumsy and barbarous. It was a similar idea, and not a chimerical peculiarity of a local dialect, which has caused the employment of the ya-sruti by one school, and which has subsequently cansed it to pass into the rules of its grammars and into the usages of its books.
As for the use of the dentel n and the cerebral , the case is, if possible, still more striking. At first sight, a dialect which invariably pronounces an initial in one way and a medial n in another, should surprise us and pat us on our guard. But the question is more general, and the case is susceptible of being argued with greater precision.
I must confess that I cannot sufficiently express my surprise to see nowadays the distinc tion between the cerebral and the dental nasal taken as a basis of classification when dealing with the ancient Prâksite. It will be remembered how the form of the cerebral I is known to none of the inscriptions of Piyadasi which are couched in the Magadhi orthography. The dental 1 is alone used. If this is a peculiarity of the dialect, it is very curious that, in the literary Magadhf, the dental » should, on the contrary, completely disappear, and that the cerebral alone should be admitted. At Bharhat, the ordinary inscriptions know only one 1, the dental » ; but there is, nevertheless, one exception, and it is characteristic. The royal inscrip
• Jacobi, p. 16.-E. Müller, Beitr. Zur. Gramm, des Jainaprdkrit, pp. 8 and . * cf. Piscbel, Hémach. I. 180.