________________
SEPTEMBER, 1892.]
THE INSCRIPTIONS OF PIYADASI.
269
I now come to the second object of inquiry.
Monumental Prakrit and the Literary Prakrits start from the same source. Their main difference consists in this, that they have been unequally cultivated. The latter possess a character more stable, their mode of writing is more perfect. Is this to be explained by indifference to these particulars on the side of the former? Certainly not. The part which it plays as the official language of the inscriptions, the general level which it knows how to retain above the more altered locai dialects, allow us to recognise in it an idiom already refined, and with an inevitable tendency, as is universal in India, to establish itself as a fixed and regular language. How could we believe, if there already existed, in the Literary Prakrits, a parallel model of better regulated and more complete orthography, that the writers could have, when using the language for inscriptions, neglected to profit by it, and to utilize its experience ?
Bat general considerations are not sufficient. Whatever it be worth, the demonstration, to be conclusive, must be connected with precise and characteristic phænomena. The facts relating to the graphic representation of double consonants have afforded us valuable assistance for establishing certain essential points in the comparative history of Classical and Mixed Sanskrit, and the data of the same order are no legs instructive in the new ground on which we tread at present.
The Literary Prakrits observe every doubling without exception. There does not exist a single Prakrit text which departs from this rule, or a single grammarian who does not explicitly teach it, or shew by evidence that he assumes it. The strictness with which it is uniformly introduced in all the dialects shews that we have here a rule which has from the very commencement exercised its influence on the grammatical work,
This mode of writing seems, in itself, to be perfectly simple; it is only the expression of the actual pronunciation. But the matter is not so easy as that. Not only does the most ancient orthography, that of the edicts of Piyadasi, abstain from observing it, but we have seen that Mixed Sanskrit, in spite of the tendency which led it to approach historically older forms, adopted it slowly, and, as I have admitted, under the influence of Classical Sanskrit. It is no less a stranger to the Prakrit of the monuments throughout the whole period with which we are now dealing. We are entitled to affirm this as a general fact, though I shall shortly point out certain exceptions, which, far from weakening the rule tend to emphasize its correctness.
This graphic usage of the literary Prakrits, which is inseparable from their very elaboration and from their grammatical establishment, was, therefore, not borrowed by them from earlier established customs. It is not met in epigraphy, nor in the current practice which epigraphy certainly reflects. It can only have been borrowed by them, as it was borrowed by Mixed Sanskrit, from the pre-existing orthography of Classical Sanskrit. I have just shewn that it was à priori more than probable that the very idea of refining the local dialects into literary tongues, and still more probably the principles under which the latter were elaborated, must have had their source in the existence, in the employment, and in the rule of profane Sanskrit. This orthographical peculiarity lends to this view a new and positive foundation in fact, and certain data borrowed from epigraphy shew it in its full light.
I have said that the Prakrit of the inscriptions does not double its consonants. It remains, in this respect, faithful to the ancient tradition. This fidelity is not invariable, and does not endure to an indefinite period. From & certain epoch, we find some examples of doubling appearing sporadically. The last inscription of Vasithfputa Palumâyî (4. S., IV. p. 113, No. 21) has sétapharaṇaputtasya. The termination asya, which is repeated in sövasakasya, abuldmavathavasya, clearly shews that the engraver employed this doubling in a moment of Sanskritizing imitation. In the purely Praksit texts of Madhariputa Sakasêna, we meet
Amongst the neo-Aryan languages, Sindhi, re-adopting in its case the primitive inexactness of the Hinda orthography, noglects to note these doublings: but it none the less faithfully observes them in pronunciation.
