________________
182
THE INDIAN ANTIQUARY.
[APRIL, 1908.
In this respect Saurasênt differs widely from Mabarishtrt, which in the rich system of inflexional forms and the frequent occurrence of provincial words agrees with eastern languages, especially with Ardhamagadbi.
On the other hand, there is a certain relationship between Saurasêni and the so-called Magadhi. Vararuchi, XIII., 2, declares that the praksti or base of that latter dialect is Sauraseni, and similar statements are made by other grammarians. And in reality, both dialects often seem to agreo very closely, in inflexional forms and in vocabulary. On the other hand, there are also important points in which they differ. These points would probably be more numerous if we knew a little more about Magadhi. Our knowledge of this dialect is, however, very limited. It seems certain that several vernaculars were considered to be related to Magadbi. We may perhaps assign the whole eastern part of India to that language, and it is probable that the particular dialect which was described as Magadhî by the Prakrit grammarians was one of the westernmost which had been largely influenced by the language of the Doab. It should be remembered that the Brahmanical civilisation of Western India came very early to the Videha country, which is to this day a stronghold of Brahmanism, and Vidêha in early times formed part of the Magadha kingdom. Brahmanical civilisation early pervaded the whole of NorthEastern India. The prevalence of the Ganda-Riti is to a great extent due to the influence of an old civilisation which had been forced into narrow forms sanctioned by old tradition. The Vidarbha-Riti, on the other hand, is closely connected with the country of Maharashtra. It is, therefore, perhaps allowable to infer that the apparent connection between Sanrasên and Magadbi was due to the influence of Sanskrit in the East, and that the vernaculars of the people were different. The modern dialects of the Magadha and Vidêhe countries seem to show that the old vernaculars of those districts were more closely connected with Maharashtrt and Ardhamågadhi.
This latter dialect may be described as a knk between Maharashtri and Magadbi, more closely connected with the former than with the latter.
It is very difficult to make a precise statement of the mutual relationship of the various Prakrit dialects, the more so because all later Indian authors usually confound them. An author like Rajasekhara, for instance, who was & native of the Maratha country, freely mires Mâhârâshtri forms and words in his Sauraserat. And some of the most striking differences between the various dialects are by no means local variations, but simply different stages in the same development.
Everyone who has the most superficial knowledge of the Prakrits will remember that they apparently may be divided into two groups, Maharashtri and Ardhamagadhi on one sido, and Saurasêni and Magadhi on the other. Both groups are distinguished by the different treatment of single consonants between vowels, and by the different inflexional systems. I have already pointed out that the latter characteristic cannot be urged so long as our knowledge of Magadhi is so limited. With regard to the former the facts are as follows.
The Prakrit grammarians teach that an unaspirated mate consonant, if not a cerebral, is generally dropped between vowels in all Prakrit dialects, and a faintly sounded y, or, in the Case of p and b, a v, is substituted for it. This y is not, however, written in other than Jains manuscripts. It seems certain that this rule of the grammarians was a generalisation of stray occurrences or of a phonological tendency, and did not exactly represent the actual facts of the genuine vernaculars. The tendency to drop consonants in such positions must, however, have been rather strong, as we find its results largely prevalent in modern dialects. Compare Marathi kumbhar, Sanskrit kumbha (k) dra, potter; tall, Sanskrit tada(9), a tank; niy, San krit sú (ch), a needle ; nonang, Sanskrit na-6) andmi, I don't know; &, Sanskrit bra, seed; sam (bhar), Sanskrit sa(t) ani, hundred; ply, Sanskrit pd(a)a, foot; k&l, Sanskrit ka(d)all, a plantain, and so on.
