________________
120
THE INDIAN ANTIQUARY.
[APRIL, 1876.
see. In the following remarks B means barytone, O oxytone, pro-examples making for the theory, con--those against it. On pp. 7, 8, there are quoted 23 B pro; on pp. 8, 9, are 17 O pro; on p. 10 are 11 O con; pp. 11, 12, are 70 con; pp. 12, 13, are 5 B con, 8 0 pro, 1 0 con; on p. 14 are 8 O con; p. 15, 2 B con ; pp. 15, 16, 4 O con; pp. 16, 17, 11 B pro, and again 9 B pro; on p. 18 only one example of the infinitive is quoted, because they are too numerous to cite all; they are all and every one con; but I will content myself with counting only as many as instances pro are quoted on pp. 16, 17, that is, 20 B con;* on p. 21 are 6 O con, 2 0 pro, and 1 B pro; p. 22 are 6 0 con, 20 pro, 1 B pro; pp. 23, 24, are 3 B pro, 1 0 pro, 60 con, and again 3 O pro; on p. 25 there are 6 examples: of these one is a tatsama, another does not occur in Hindi, and of two more the accent is not mentioned; of the remaining two lis B pro and 1 B con; on p. 26 are 2 B con, 2 B pru, 40 con, 10 pro. Now add. ing up these examples, it will be seen that they are altogether 168, among which there are 85 (i.e.5i Band 34 0) which make for the accent-theory, and 83 (ie. 30 B and 53 0) which make against it. Mr. Beames himself says that his rule does not apply to Tatsamas: hence, strictly, about 16 examples ought to be excluded (e.g. vikh, p. 8; dipd, p. 9, unless this is a misprint for diyd ; 8may, keray, ray, p. 10; kațin, p. 13;jatan, praśn, supun, darpan, p. 16; daśan, brothan, roshan, endn, p. 17, &c.). But as they are about equally (9 pro, 7 con) divided for and against, this slightinaccuracy may be passed over as not affecting the general result. The latter is that as many instances (83) may be cited against the accent-theo y as there are in support of ii (85). There is no reason to suppose that any more extended collection of examples would alter this result materially; for all practical purposes Mr. Beames's collection of examples is quite sufficient, and it is a perfectly fair one. But I may be allowed to point out that 80 far I have confined myself to a consideration of Hindi alone; if I had taken into the range of the present examination the other Gausian languages too (Marathi especially), the result would have gone (as Mr. Beames himself seems to foel, see p. 9) still more decidedly against the accent-theory. But even taking the result as we have found it above,---unless it can be accounted for in some way-it is, to my mind, fatal to that heory; for it is founded on an induction which is not only partial, because it only includes about 50 per cent. of phenomena, but one-sided, because it has the other 50 per cent directly against itself. Mr. Beames indeed makes an attempt to account for this adverse result; but I think on reconsi.
* The examples quoted on pp. 16, 17, and the infinitives, have alike, according to Mr. Beames's theory, bases in
deration he will see that his explanation involves a petitio principii. For example, on p. 10 a list of eleven words is given which, being oxytones, ought to terminate in a; but in reality they end in ă, as if they were barytones. Mr. Beames accounts for this failure of the accent-theory by the conjecture" that though the learned accentuated the last syllable of stems of this small class, the masses did not at any time observe this distinction, but treated them as barytones." But what ground is there for this supposition ? Is it not merely the fact that those words end in a instead of a ? That is, the fact of their ending in ă is explained by their being supposed to have been used as barytones, and the supposition of their having been thus used is based on the fact of their ending in a.
Thirdly, all words formed with the suffix aka have the heavy termination d (p. 29). The Ka-theory explains their termination and that of the previously considered words by the same phonetic process; whereas, if the accent-theory be accepted, two different causes must be assumed to account for an identical result. This offends against the logical rule of economy. For example, from the stem ghotala is derived in Hindi ghord, and from the stem anla (oxytone) the Hindi word anda ; accordding to the accent-theory the identical Hindi termination d is accounted for by ghotaka ending in ka and andd being oxytone; the Ka-theory, on the other hand, derives an lá not directly from the stem anrld, but from the-as regards meaning-identical stem andaka, and thus accounts for the identical termination d by an identical cause, viz. both ghotaka and an laka ending in ka.
Fourthly, even if it be allowed that the accenttheory accounts for the difference of some nouns ending in d, and others in a, it affords no help-80 far as I can see-towards understanding the origin of the termination of the oblique form of Gaurian nouns in di.e. Hindi d ore; Marathi ya or va, &c.). On the other hand I contend that the Ka-theory explains both. I cannot ask for space to prove this here, and therefore must refer to my Essays (IV., V.) in the Jour. Beng. As. Soc. But, if my contention is well founded, this is clearly another point in favour of the Ka-theory.
But I am confronted with the demand, "If all nouns might and did take this ka, why do not all nouns of the a stem end in d? why do some end in ă (pp. 30-37)P" I admit the question is legitimate. It cannot be ascribed to caprice.' There must be a reason for it, as for everything else in the world. But I do not see that this question, whether or not it can be answered, affects at all the truth of the Ka-theory. I suppose it will ana; if my theory of the infinitive be accepted, they make equally against the accent-theory.
