________________
148
THE INDIAN ANTIQUARY.
[WAY, 1892.
local predominance of a special orthographical system, or rather of special orthographical tendencies.
The observations which still remain for me to make are of a kind to add further proof to these conclusions.
The orthography of Kapur di Giri, as in Sanskrit, distinguishes the three sibilants, 8, 6, sh, Is it really the caso that the dialect of this region retained a distinction which, if we are to judge from the parallel versions, was lost everywhere else? It is sufficient to record the irregularities accumulated in the distribution of these sibilants, to convince the reader that nothing of the sort occurred.
We read é instead of sh in manuka (II, 4; 5) beside manusha (XIII, 6), and in the futures which are formed in sati for shyati. We have s'for sh in yésu (XIII, 4), arabhiyisu (1, 2), beside mikramishu, &c., in abhisita, which is never written abhishita ; and for sin anusichano (XIII, 2); sumachariya (XIII, 8), ssta (1, 2): sh for sin pashshashu (ILI, 6), shashni (XIII, 8); < for sin anusasana (IV, 10), antiabisciati (ibid.). It cannot be imagined that this confusion may be referred to the real usage of the local diáleet. It can only be accounted for by one theory, the only one which explains analogous mistakes, whether in manuscripts or in more modern Sashkpit inscriptions. The error of the engraver or of the scribe arises in both cases from the fact that he has before him a learned spelling, in the application of which he cannot be guided by the usage of the current dialect, because the distinctions he has to deal with are strangers to it. The locative pashshashu, '& clamsy imitation of locatives in éshu, is very characteristic as illastrating the way in which the sibilants were used at Kapur di Giri.
The fact mast not be lost sight of that this method of writing is not an isolated example ; it is borne witness to by other parallel ones, which leave us in no doubt as to what conclusions we are to draw from it. It is certain that the distinction between the sibilants did not exist in the dialect of the western coast; yet that does not prevent as finding all three at Násik (Nos. 1 & 2, A. 8. iv, 114), in dedications, which in every other respect are couched in pure Prakrit, not even in mixed Sanskrit. As at Kapar di Giri, a mistake, saka sa for sakasa, is there to warn as as to the trae character of this use. It is the same in No. 27 of Kanhêri (A. $. v. 85), in which the pretension to learned orthography leads to such forms as dunhanaik, sárvvabatvinas.
In the instances whicb we have just passed in review, we may perhaps be allowed to hesitate as to the origin of the spelling, though not as to the sound which it represents or is intended to represent; the problem becomes more thorny when we consider certain orthographical phenomena, which express accurately neither the learned form, nor the form adopted in popular usage; which can, in some respects, be considered as intermediate between these two poles of linguistic movement.
Dr. Pischel has correctly pointed out that, at Kapur di Giri, the words which I have, according to precedent, transcribed as dharma, darsi, darśana, karmaye, varsha, purva, &c., are really written dhrama, draiana, &c., the r being joined to the consonant dh, d, &c. He adds that hore, as in the coin-legends which observe the same method of spelling, this writing certainly represents a dialectio peculiarity, and that the people for whom the tables of Kapur di Giri were inscribed, actually pronounced the word as dhrama, pruua, &c. At this point I am unable to agree with his daductions.
He bases his argument specially on certain readings, such as mruga, equivalent to mriga, in the first édiot of Kapar di Giri, graha and dridha, equivalent to griha and dridha in the 13th, pariprickha, equivalent to pariprichchha in the 8th, vrachhé, equivalent to vriksha, in the 2nd edict of Girnar. He compares the forms ru, ri, rá, taken by the vowel 'pi in several modern dialects.
• Götting. Gel. Anteigen, 1881, p. 1916.
