gadha, the sayings and
doings of their great teacher in popular and easy-flowing
verses, which in course of time came to be regarded as the
most authentic source of all information connected with the
founder of Buddhism. The high estimation in which the
ballads and improvisations of bards are held in India and
particularly in the Buddhist writings, favours this
supposition; and the circumstance that the poetical portions
are generally introduced in corroboration of the narration
of the prose, with the words "Thereof this may be said,"
affords a strong presumptive evidence.'

Now this, from the pen of a native scholar, is truly remarkable. The
spirit of Niebuhr seems to have reached the shores of India, and this
ballad theory comes out more successfully in the history of Buddha
than in the history of Romulus. The absence of anything like cant in
the mouth of a Brahman speaking of Buddhism, the _bete noire_ of all
orthodox Brahmans, is highly satisfactory, and our Sanskrit scholars
in Europe will have to pull hard if, with such men as Babu Rajendralal
in the field, they are not to be distanced in the race of scholarship.

We believe, then, that Babu Rajendralal is right, and we look upon the
dialect of the Gathas as a specimen of the Sanskrit spoken by the
followers of Buddha about the time of A_s_oka and later. And this will
help us to understand some of the peculiar changes which the Sanskrit
of the Chinese Buddhists must have undergone, even before it was
disguised in the strange dress of the Chinese alphabet. The Chinese
pilgrims did not hear the Sanskrit pronounced as it was pronounced in
the Parishads according to the strict rules of their _S_iksha or
phonetics. They heard it as it was spoken in Buddhist monasteries, as
it was sung in the Gathas of Buddhist minstrels, as it was preached in
the Vyakara_n_as or sermons of Buddhist friars. For instance. In the
Gathas a short a is frequently lengthened. We find na instead of na,
'no.' The same occurs in the Sanskrit of the Chinese Buddhists. (See
Julien, 'Methode,' p. 18; p. 21.) We find there also vistara instead
of vistara, &c. In the dialect of the Gathas nouns ending in
consonants, and therefore irregular, are transferred to the easier
declension in a. The same process takes place in modern Greek and in
the transition of Latin into Italian; it is, in fact, a general
tendency of all languages which are carried on by