________________
BUDDHIST STUDIES IN THE WEST
never be possible to fully explain the Middle Indic background of the different classes of BHS and Buddhist Sanskrit texts. In the first place the Middle Indic material at our disposal such as the Aśokan inscriptions and later inscriptions are not sufficient. Texts in Middle Indic languages were written down several centuries after Aśoka and do not allow conclusions as to their characteristic features in earlier periods. In the second place, BHS texts were submitted to a great deal of Sanskritization before they were written down; it is not possible to prove that they were originally composed orally in Middle Indic without any admixture of Sanskrit influence. Even in the case of Pali, where the problems of text editing are far less than in BHS texts, it has not been possible to determine exactly which Middle Indic dialect or dialects contributed to its formation. Both for historical and linguistic reasons western India was probably the home of Pāli, but the well-known Magadhisms in Pāli show that Pali is not based exclusively on western dialect(s). Pāli probably found its final form in western India only after having undergone the influence of Middle Indic dialects in other parts of India.
If much more work still has to be done on BHS, the same cannot be said with regard to the only extant Buddhist text in Prakrit, the Gāndhari Dharmapada as it has been called by John Brough (London, 1962). His edition contains all fragments. Previous scholars: Senart, Lüders, Franke, Bloch, Konow and Bailey had been able to study only the parts published in 1897 and 1898. The language of the text had been called Northwestern Prakrit. Gāndhārī, the name Bailey proposed, was adopted by Brough. In 1946 Bailey showed that this language has been of great importance for the history of Buddhism in Central Asia.12 Many Indian words in Khotanese, Agnean, Kuchean and other languages of Central Asia are based on Gāndhārī forms. The same language is used in the Kharoṣṭhi versions of the Aśoka inscriptions in Shahbazgari and Mansehra, later Kharosthi inscriptions,13 and in the Niya documents which were edited by A. M. Boyer, E. Senart and P. S. Noble (Oxford, 1920, 1927, 1929). This language has as typical features the preservation of all three Indian sibilants, and the preservation of certain consonant groups (tr, br)
12 Gāndhārī, BSO AS, 11, 1946, pp. 764-797.
13 Corpus Inscriptionum Indicarum, volume II, part 1: Kharoṣṭhi Inscriptions, with the exception of those of Aśoka. Ed. Sten Konow, Calcutta, 1929.
57