________________
34
THE INDIAN ANTIQUARY.
tators is Sanskrit, and nothing else is entitled to the name. The young pandit, then, is expected to master the system and to govern his Sanskrit speech and writing by it. This he does, with immense pains and labour, then naturally valuing the acquisition in part according to what it has cost him. The same course was followed by those European scholars who had to make themselves the pupils of Hindu teachers, in acquiring Sanskrit for the benefit of Europe, and (as was said above) they did so to their very great advantage. Equally as a matter of course, the same must still be done by any one who studies in India, who has to deal with the native scholars, win their confidence and respect, and gain their aid. They must be met upon their own ground. But it is a question, and one of no slight practical importance, how far Western scholars in general are to be held to this method whether Pânini is for us also the law of Sanskrit usage; whether we are to study the native Hindu grammar in order to learn Sanskrit.
There would be less reason for asking this question, if the native grammar were really the instrumentality by which the conserving tradition of the old language had been carried on. But that is a thing both in itself impossible and proved by the facts of the case to be untrue. No one ever mastered a list of roots with rules for their extension and inflection, and then went to work to construct texts upon that basis. Rather, the transmission of Sanskrit has been like the transmission of any highly cultivated language, only with differences of degree. The learner has his models which he imitates. He makes his speech after the example of that of his teacher, only under the constant government of grammatical rule, enforced by the requirement to justify out of the grammar any word or form as to which a question is raised. Thus the language has moved on by its own inertia, only falling, with further removal from its natural vernacular basis, more and more passively and mechanically into the hands of the grammarians. All this is like the propagation of literary English or German; only that here there is much more of a vernacular usage that shows itself
[FEBRUARY, 1885.
able to override and modify the rules of grammar. It is yet more closely like the propagation of Latin; only that here the imitation of previous usage is frankly acknowledged as the guide, there being no iron system of grammar to assume to take its place. That such has really been the history of the later or classical Sanskrit is sufficiently shown by the facts. There is no absolute coincidence between it and the language which Pânini teaches. The former, indeed, includes little that the grammarians forbid; but, on the other hand, it lacks a great deal that they allow or prescribe. The difference between the two is so great that Benfey, a scholar deeply versed in the Hindu science, calls it a grammar without a corresponding language, as he calls the pre-classical dialects a language without a grammar. If such a statement can be made with any reason, it would appear that there is to be assumed, as the subject of Hindu grammatical science, a peculiar dialect of Sanskṛit, which we may call the grammarians' Sanskrit, different both from the pre-classical dialects and from the classical, and standing either between them or beside them in the general history of Indian language. And it becomes a matter of importance to us to ascertain what this grammarians' Sanskrit is, how it stands related to the other varieties of Sanskrit, and whether it is entitled to be the leading object of our Sanskrit study. Such questions must be settled by a comparison of the dialect referred to with the other dialects, and of them with one another. And it will be found, upon such comparison, that the earlier and later forms of the Vedic dialect, the dialects of the Brahmanas and Sutras, and the classical Sanskrit, stand in a filial relation, each to its predecessor; are nearly or quite successive forms of the same language: while the grammarians' Sanskrit, as distinguished from them, is a thing of grammatical rule merely, having never had any real existence as a language, and being on the whole unknown in practice to even the most modern pandits.
The main thing which makes of the grammarians' Sanskrit a special and peculiar language is its list of roots. Of these there are reported to us about two thousand, with
Einleitung in die Grammatik der vedischen Sprache, 1874, pp. 3, 4.