Sec. ii] THE LEGENDARY TRADITIONS OF KHOTAN 163
etymology ' for the name of Khotan, and in that case we must expect to find also the miraculous story about the breast of the earth ' localized in that territory. In this form the story is actually recorded by the Hsi yü-chi, while the point is lost sight of in the Tibetan version, which makes the miracle take place in India and brings Kustana into relation with Khotan only in a roundabout fashion.
It is not difficult to discern a motive for this transformation of the legend. Given the Buddhist
fact that tradition connected the foundation of Khotan with an Indian immigration about the colouring of
time of Moka, it was quite natural that the legend under Buddhist influences should take the version.
further step of deriving the first king from Moka's stock. The miraculous transfer of Moka's
exposed son to China supplied the simplest expedient for introducing this feature into the
traditional story without otherwise disturbing it in any essential point.
The Buddhistic tendency, which is so clearly discernible in this modification of the original
tradition, helps to make still more conspicuous the total absence of a similar colouring in the
legend as told by the Hsi yü-chi. There could have been nothing particularly flattering to
the Buddhists of Khotan in the legend that they were partly descended from people whom
the righteous wrath of Moka had banished from the extremity of his empire ; nor can that
part of the story which represented this Indian colony as worsted and absorbed by one from
China be reasonably assumed to have been invented with a view to pleasing Buddhist or
I think that this complete absence of Buddhist colouring in the earlier version of the story Historical
as recorded by Hsüan-tsang possesses considerable significance. It strongly suggests the prob- background
ability that the tradition was earlier than, and independent of, the introduction of Buddhism of legend.
into Khotan, and that its main features had some historical background. Distant and obscure,
indeed, the historical facts may seem which gave origin to the tradition, and we may, perhaps,
never hope to establish them in full critical clearness. But it appears to me that the assumption
of a nucleus of truth in the essential features of the tradition affords the best explanation for
certain characteristic facts in the ethnic and cultural history of Khotan as we know it in the
light of recent discoveries.
The evidence to be discussed in a subsequent chapter of the Kharosthi documents from Early use of
the Niya Site proves beyond all doubt that an Indian language closely allied to the old Indian
Prakrits of North-western India was in daily use for administrative purposes throughout the Khotanuag.e in
Khotan region about the middle of the third century A. D. Considering the character of these hundreds of documents, dealing with all the varied affairs of practical life and social organization, it is impossible to assume that their language should not have been widely, perhaps universally, known within the territory. The conclusion to be drawn from this current use of an Indian language is greatly strengthened by the Kharosthi script of the records ; for we know that within India this script was peculiar to that region of which Taxila and the adjoining Gandhara were the historical and cultural centres for centuries before and after the commencement of our era.
Neither the language nor the script of these documents can be satisfactorily accounted for by the spread of Buddhism alone, which, so far as our available evidence goes, brought to Central Asia only the use of Sanskrit as its ecclesiastical language, and the writing in Brahmi characters. But the current use in Khotan of both a Prakrit dialect and of the Kharosthi script becomes at once intelligible if we recognize a substratum of historical fact in the old local tradition heard by Hsüan-tsang, which asserted a partial occupation of Khotan by Indian immigrants from the region of ancient Taxila.