414 181. COTAN
This is also the only explanation I can think of for a document concerned with Kan-chou in Kan-su, probably dating from the 10th cent. It is written in Khotanese, but the name of Khotan is given there as Yûttinini Kûhi, i. e. *Yûtin-kûg, a transcription of the Chinese Yü-t'ien-kuo (*Jiju-d'ien-kwak), « Kingdom of Khotan » (cf. BAILEY, in BSOS, Ix, 541). The transcription betrays a Chinese pronunciation which is no longer that of the 7th cent. The final consonant of kuo (*ko°ak) is still heard, but the initial of yü (*jiu) is already the modern y-, and d'ien is rendered with a surd initial, as in the forementioned Tibetan transcriptions, which are perhaps of approximately the same date.
The ancient capital of Khotan, probably abandoned in the 11th cent., began some five miles north by west of the new Chinese town of Khotan; its remains under ground, discovered c. 1870, are within the area of the group of detached villages called « Yôtkan » (STEIN, Ancient Khotan, 191). CORDIER quotes (Y, III, 44-45) a passage of PARKER according to which « Yôtkan » probably « furnishes a clue to the ancient name of Yü-t'ien ». BARTHOLD (in EI, s. y, « Khotan ») is positive on the identity of the two names. THOMAS too says (Tibetan Texts and Documents, I, 8) that « Yôtkan » is « certainly the site, and perhaps retains a perversion of the name, of the ancient city of Khotan ». At the same time, and while speaking of « the Chinese name U-then », thus admitting that Ch. Yü-t'ien and Tib. 'U-then are one and the same name, Thomas adds that we connect that name with the word « Khotan » « naturally (but perhaps mistakenly) », and thinks of explaining « Uthen » with the name of a certain stûpa of « A-dha-ma », mentioned in a Tibetan chronicle of Khotan (Asia Major, II, 257, 270). I shall not try to reconcile three hypotheses which appear self-contradictory, since I think that they must all be abandoned. Neither the history nor the meaning of ,;ley Yôtgân (such is the true spelling; cf. GRENARD, Mission dans la Haute-Asie, III, 127; STEIN'S transcriptions do not distinguish between -q- and -k-) is known to me (does -ô- mark the slurring of a following -r?). But there is no likelihood that the specifically Chinese y- initial of Yü-t'ien should have had an independent unetymological counterpart at Khotan itself. It is still more difficult to imagine, since Yü-t'ien and Khotan are fundamentally one and the same name, that that name should have developed on the spot such divergent duplicates as Yôtgân and Khotan. In spite of a partial phonetic analogy, my opinion is that the two names are not connected.
I have said that the Altaic name of Khotan in the Middle Ages was Odon, in agreement with the form supposed by Hsüan-tsang's Yü-tun, *'Odon. There are other examples of such double names. The ancient name of Kucâ was *Kuci, duly rendered in Chinese transcriptions and in Brahmi spelling. But at the same time, there was a Turkish name of Kucâ, Küsän, which we can follow from the 10th. cent. down to the 16th, in Turkish as well as in Mongol and Chinese sources, and which may be much more ancient than the date at which it is attested in the texts. When Kâsyari compiled his Turkish dictionary in 1076, *Kuci had already passed to Kucâ (? Kücä), but, alongside of Kucà, Kâsyari gives the synonym Küsän (BROCKELMANN, 245). The case of Khotan is quite parallel, and we find in Kâsyari both « Hotan » (Khotan) and « Odon » (BROCKELMANN, 251, where the name is transcribed « Udun »; of course Arabic spelling does not permit of a distinction between u and o).
Except for the double form Tjotan and Odon in Kâsyari, all mentions of Khotan in Mussul-
IIIF Color HighRes
