________________
264
THE INDIAN ANTIQUARY
[SEPTEMBER, 1892.
Dr. Hoernle's opinion regarding Ardhamagadhi rests, unless I am mistaken, on but a weak basis. He has endeavoured to establish from the inscriptions of Piyadasi a geographical partition of the ancient dialects, which I have already, I believe, shewn to have little foundation. We have, as a fact, no indication of the existence, at that ancient period, of a dialect intermediate between the Magadhi and the Maharashtri. I would add that, by its name of Arsha, the Ardhamågadhi is at once classed as a literary language. It would be a strange phænomenon that we should have to take it as denoting a real idiom, - this dialect, whose sole peculiarity is the formation of the nominative singular in é, and which, in other respects, save a few insignificant exceptions, is just the same as Maharashtri. It bears clearly on its face the mark of its artificial origin, I shall indicate, later on, what we may conjecture as to its formation; and certainly, the first impression awakened by its name, the notion which that name gives of a scholastic idiom, is not one that will mislead us.
It is true that, beside these instructive names, other dialects received local titles which connected each with a definite tract of country. I do not even wish to insist on the fact that the principal dialect, the one which serves as the basis for the teaching of the grammarians, instead of habitually receiving its name of Maharashtri, is called Prakrita, the Prakrit par excellence, which manifestly contrasts it, as an artificial language, with that other learned and literary language, which is Samskļita, the Sanskrit. This detail can well have only a secondary importance, and it remains certain that several Prakrits are designated by geogra. phical names ; Maharashtri, Sauraseni, Magadhi. It is natural to conclude that they are connected respectively with the countries of the Maharashtra, of the Surasê nas, and of Magadha. But to what degree, and in what sense are they connected ?
That each borrows certain characteristio peculiarities from the popular dialect of the country of which it bears the name, is a thought which will at once occur to the mind. Several facts confirm it. Some of the phenomena attributed to Magadhî by the grammarians - the formation of the nominative of a-bases in é, the substitution of 1 for r - are also found in the official dialect of Piyadasi, and the situation of the royal residence entitles us to consider that as approximately representing the idiom of Magadha. Whatever we may be led to think of the work of regularisation and of the cutting down to measure by the grammarians, it is certain that they have taken their materials for foundation, their constituent elements, from the vulgar dialects, and the names which have remained attached to the literary idioms, when they have a definite geographical meaning, deserve to be taken into serious consideration.
Till the contrary is proved, they supply us with an historic basis, which we cannot abandon without committing a serious imprudence. So far as concerns the Maharashtri, the comparisons which the inscriptions of the western coast, in the land of Maharashtra, permit us to institute, shew that no incompatability exists between what we can identify as belonging to the popular language, and the rules of the grammatical idiom. The only thing is that we must clearly understand under what conditions these comparisons present themselves. Maharashtra, where we find at once both a long series of monuments, and, in the verses of Hâla, an ancient, probably the most ancient, instance of the application of Průksit to literature, is the tract most favourably circumstanced for us to form a clear ides, on actual evidence, of the manner in which the reform of the Prakrit grammarians was accomplished.
On a consideration of the Prakrit inscriptions of the West we have been convinced that, although they are necessarily based on the popular language of the locality, they do not give us a rigorously faithful picture of it. Their orthography is not strictly representative; but, without having that stability which can only be assured by a complete grammatical culture, it tends to get as near as it can to etymology, that is to say to the orthography preserved by the learned language. It takes as the typical ideals of its writing those instances in which the pronunciation has departed least from the primitive form. The parallel use of Mixed Sanskļit is there to prove that this conclusion does not arbitrarily attribute to the authors of the monumental orthography a predisposition which was not theirs.
