150 THE KHOTAN OASIS : ITS GEOGRAPHY AND PEOPLE [Chap. VI
in character, must relate to objects of ordinary life, suggest that they belong to a language which had once been commonly used in this region. The character of the unknown' language which is found in a number of manuscript texts and fresco inscriptions from Dandan-Uiliq and Endere, and which may well prove to be the source of those non-Indian terms in the older Kharosthi documents, points to the same conclusion. For Dr. Hoernle, who has carefully analysed a number of similar texts, probably translations of Sanskrit originals into the language of the country, has from the general character of the language derived the impression ` that its identity has to be looked for in the direction of the monosyllabic Tibetan rather than of the Turki or Mongol languages '$'. Finally, we may note in passing that the hypothesis just indicated would furnish also the best explanation for a number of ancient local names of the Khotan region which suggest Tibetan origin, and which yet are anterior to the historical period of Tibetan invasions $S.
I need scarcely point out that most, if not all, that relates to the ethnic evolution of the Khotan people previous to the Muhammadan period, must in the present state of our knowledge remain distinctly conjectural. Yet I believe that the comparison we have just concluded between the actual anthropology of the oasis and the extant data concerning its earlier inhabitants may help us in interpreting correctly historical records and archaeological facts.
E See Report on Central-Asian antiquities, ii. p. 13. " See Grenard, Mission D. de Rhins, ii. p. 54.