38 Proceedings of the Asiatic Society. [Feb. 



it. It would be a mere statement without any reliable evidence to 

 support it, no more than to support the theory that the Sanskrit 

 grammar was elaborated at Taxila and not elsewhere in the Panjab, 

 or even in Brahmavarta. 



" Mr. Thomas assumes that the Brahminic Arians first constructed 

 an alphabet in the Arianian provinces out of an archaic type of Phoeni- 

 cian, which they continued to use, until they discovered the superior 

 fitness and capabilities of the local Pali. He states that he has been 

 collecting proofs of this for some time past, and each fresh enquiry 

 more and more confirms his early impression. It is a matter of regret 

 that the published report of his lecture does not give any of his 

 evidences, and I am at a loss, therefore, to know on what grounds he 

 takes the Arian alphabet to have been elaborated in the Arianian 

 provinces before the Brahmins came to India. That alphabet may 

 be a Bactrian adaptation from the Phoenician, but the question is, 

 when did the Brahmans first use it ? The oldest Arian record is 

 long subsequent to Buddhism ; none that I know of dates before the 

 Pali edicts of As'oka ; and there is nothing to bridge over the gap of 

 at least some thirteen hundred years between that time and the period 

 when the Brahmans dwelt in Bactria. 



'* Then as to the Pali, it is evident that it existed in the country long 

 before the time of As'oka. The different shapes under which the 

 same letters of the Pali alphabet appear at Junaghur and Dhauli are 

 marked and peculiar, and they cannot be accounted for by any candid 

 enquirer, except on the supposition that long usage had brought on 

 local peculiarities. The allusions to alphabetic writing in Panini and 

 other purely Indian pre-Buddhist authors point likewise to an 

 Indian, and not to a Bactrian alphabet. Again, the oldest Sanskrit 

 inscription that has yet been found is recorded in the Pali (the 

 Junagarh inscription of As'oka) and not in the Arian letters ; indeed 

 no Sanskrit inscription has yet been met with in the Arian characters. 

 The Pali, besides, is a vernacular form of the Sanskrit — the first stage 

 in its transition to the Prakrit — and the alphabet used to write it 

 down may more reasonably be taken to be its legitimate vehicle, and 

 not that of the Dravidian, of which no inscription of any kind, either 

 • old or new, has yet been discovered in the Pali character. Indeed, 

 I can see no connexion whatever between the Dravidian languages 



