________________
244
THE INDIAN ANTIQUARY.
(AUGUST, 1892.
treatises, but also in lay works, and that we meet it in the texts of inscriptions, this terminology has become both inaccurate and inconvenient. I propose to substitute the term 'Mixed Sanskrit,' name which will, I hope, be justified by the observations which follow.
The same caves preserve the memory of the Andhrabhrityas who were contemporary with or the immediate successors of Nahapana, - Gotamiputa Satakaņi and his descendants.66 In general (Nasik, 11 A, 11 B, 12, 13, 14, 15, 16, 22; Karli, 20, 21; Kanhöri, 4, 14, 15) the inscriptions are couched in pure Prakrit, though not without certain inconsistencies in detail (svámi beside 'samiyehi, Nás. 11 A, 15); Pulumáyi (Nâs. 13, beside Pulumdvi, Nas. 15, and Pulumdi, Nás. 12, &c.). This means that assimilation occurs everywhere, though the consonants are never written double. This does not, however, prevent us from finding at Karli (No. 22), a donation of the reign of Visithfputa Pulumayi, which has siddhari, according to the classical method; which, beside numerous genitives in asa, writes puttasya, sôvasakasya, vdthavasya, and, beside nithito, hitasughasth(Otaya. It thus unites in the same word forms which were already no longer found in the inscriptions of Piyadasi, and others which are still rare in the 2nd century of our era, to which epoch they belong ! On the other hand, at Kanhêri (No. 11)56, dedication of the reign of Vasishthipatra Satakarni, the sonin-law of the Satrap Rudradáman, is couched in pure Sanskrit, save for one single irregularity : Sátakarnisya.
Are these facts, I will not say isolated, but circumscribed in a narrow region ? Quite the contrary. It is sufficient for conviction to cast the eye over the monuments of the Turushka kings, Kanishka and his dynasty, monuments which are either exactly contemporary with those to which we have just referred, or of very little earlier date. The inscription of Sue Vihar7 is dated the 11th year of Kanishka. It may be said to be couched in Sanskpit, but in a Sansksit seriously disfigured by spellings like bhichhusya, athavi(W)ké, nagadatasya, sankhaka. tisya), yathis, yathipratithanam, &c. In the 18th year of the same reign, the stone of Manikykla, 58, however. imperfectly we may understand it, allows us clearly to recognise, side by side with the retention of the three sibilants and of groups containing an t, a number of Prakrit forms, such as budhisa, the termination aé, maharajasa, vēspasisa, chhatrapasa, &c. Mathurâ possesses, from the year 28,50 a fragment of correct Sanskrit. So also for the time of Huvishka. At Mathurâ (Growse, 2, 11; Dowson, 1, 2, 5, 7) the language of the dedications is classical; yet they present the genitive bhikshusya, and the phrase arya (or Elasya) prerváyé. On the Wardak vase, in the year 51, appear forms so much altered as thuvamhi (=stápé), bhagaé, arôgadachhinal, to speak only of those which are certain. The date of the inscription of Taxila is not fixed with certainty, but I do not think that any one can consider it as more modern than those to which I have just referred; and the name Chhtharáta, which I think I have identified at the end of the first line seems to assign it a place in about the same epoch, or in an epoch slightly earlier. Here, excepting the sibilants and a few groups (chhatrapa Thratara, vardhita, sarva, sahvatsara), everything is Prakrit, the genitive in asa, the assimilation in atha, takhasila, pratithapita, &c., and mixed up with very debased forms such as the Jocative samvatsarayé, and the dative puyaé.
It is necessary to complete this review, by noting that it is towards the end of the period of which we are treating, towards the year 75 or 80 of the Saka era, i.e. 155 to 160 A. D., that we find the first known inscription in perfectly correct Sanskrit, - the inscription of the
* The Bashkhali Manuscript, which has been published by Dr. Hoernle. * Arch. Sury. Weat. Ind. pp. 104 and ff.
16 Arch. Suru. West. Ind. V. p. 78. 6 Hoernle, Ind. Ant. X. 321 and ff. Paplit Bhagwanlal Indraji has submitted this document to an independent revision (Ind. Ant. 1882, p. 129), in which be has frequently come to conclusions different from those of Dr. Hoernle. In cases of divergency, except in certain doubtful passages in which the truth appears to me to be still undiscovered, I consider that it is Dr. Hoernle who is right.
58 Dowson, J. R. A. S. XX. p. 250. ** Growse, Ind, Ant, 1877, pp. 216 and ff. Dowson, J. R. A. 8., N. 8. V. 182 and fr. (after Cunningham).
