________________
No. 13.]
SOME UNPUBLISHED AMARAVATI INSCRIPTIONS.
259
collection of impressions made in 1907 is represented by Nos. 3-20, 33-45 and 49-52, and that of 1913 by the rest. No. 58 has been copied by me from a stone evidently also removed from Amarāvati and not copied before.
The most notable feature of these short epigraphs is the different varieties of the Brahmi alphabet used in them. Four such varieties are clearly distinguished :
(1) Nos. 1-20 are engraved in archaic Brahmi characters. This fact has already been noticed by Dr. Stea Konow in his article on "Epigraphy "in the Director-General's Annual Report for 1905-06. He writes :
"Still greater importance must be assigned to the discovery of inscriptions in ancient Brahmi at Amarāvati. Up to the year under review there was nothing to show that the stūpa there was older than the second or third century; and Bühler, in his Indian Paleography, came to the conclusion that the alphabet used in the inscriptions of the Amarāvati and Jaggayyapēta stūpas was developed out of the ornamental Brabmi known from the Western Dekkhan and the Konkan, in the third century A.D. We know, however, from the epigraphs of the Bhattiprola stupa that the Brāhmi alphabet had been used in the Kistna district as early as the third centory B.C. Mr. Rea's recent discovery, an account of which will be found above, has now added considerably to the materials available for the history of the alphabet in that part of India. It will be seen that he found a number of sculptured stones and also several plain slabe and pillars, many of which carried inscriptions. Those incised on sculptured stones are of the same kind as the epigraphs previously found, and it is doubtful whether any of them can be dated before the Christian Era. The inscriptions found on the plain slabs, on tho other hand, are inscribed in characters which must be of the Maurya period and probably go back to the second, or more likely to the third, century B.C. There are at least eighteen such, of which impressions have been sent to me. They contain no historical information and very few proper names. Two of them ascribe the stone to the Dhamakataka and Dhamakadaka nigama, respectively. This wame of Amaravati has long been known. Tārānàth informs us that Nāgarjuna built a railing round the great shrine of Dhảnyakataka. Dhankutaka is the regular Pāli form corresponding to Dhānyakataka, and the Dhammakadaka, with the weakening of i to d, probably represents the vernacular name of the Kistna district in the third century B.C. The change of a ţ between vowels into a d, which occurs already in the Asoka euicts, is common in all the Prakrits, and its occurrence in Amaravati does not, therefore, tench us any. thing about the affiliation of the Aryan dialect spoken in the Kistna elistrict in those early days. The language of the old inscriptions is, on the whole, identical with the PĀli of Buddhist literature. The form Dlumnakataka, i.e. Dhannakalaka, well agrees with this, because the change of ny into añ, according to Prākṣit grammarians, does not belong to other Prakṣit dialects than Māgadhi and Paisobi, with which forms of speech Páli agrees in this and in several other features" (pp. 165-166).
Dr. Konow's statement that “ up to the year under review there was nothing to show that the stūpa there was older than the second or third contury" is due to oversight. Inscription No. 4 published in Burgess's Archeological Survey of Southern India, Vol. I, Plate LVI, is in archaic Brāhmi characters and appears with the following vote:
"On a small fragment of stone found in the south-east quadrant, where also the granite pillars and most of the earliest sculptures occurred, was the following fragment of an
This is not quite accurate. Bühler anys thunt the mure ornamental alphabet found in the baggayyapota inscriptions and in some Amarivati inscriptious (uoted at the foot of the pago)" was developed out of the ornamental variety of Westeru Dekkban and the Konkan. But regarding forr-fifths of the Amaravati inscriptions published in 4. 8. 8. I., Vol. I, ho observes: "It is, therefore, certaia tint during the 2nd century A.D. all those three varieties were used promiscuously in the Western Dekkliai, and the inscriptions from the Amaravati stupa prove that they occurred ulso on the Eastern coast of India." (Indian Antiquary, Vol. XXXIII, APP, p. 43 and note 5.)
