123. CASCAR 201
that of Ch. Shu-lo or Sha-lo, nor of Tib. Su-iig (BAILEY speaks of « Tib. Su-Iig »; it is a slip or a misprint). We should rather expect Sulya to be a designation of the Sogdians.
A last explanation of Shu-lo or Sha-lo has been proposed by HERRMANN in Sven HEDIN, Southern Tibet, viii, 448 : that name of Kâsyar would originally represent «Sogdak (Sogdian-City) ». It is true that Suydi had become Suli when Hsüan-tsang passed through ancient Sogdiana in the 7th cent., and attempts have been made to trace to the name of the Sogdians that of the Çulika of Indian texts. But we must not forget that the form Shu-lo goes back to the 2nd cent. B. c., and that the forms Suyd, Suydak (or Soyd, Soydak) are the only attested ones, particularly by Chinese transcriptions, down to c. A.D. 500. On the other hand HERRMANN'S hypothesis does not find the slightest support in what we know of the history of Kàsyar.
Several other names have been said to be names of Kâsyar without sufficient reason. Such is the {hp * , Ch'ieh-shê-lo (*G'ia-,ia-IA) of the Shui-ching chu equated with Kàsyar by HERRMANN, in Southern Tibet, viII, 438. The name must have been known in China in the second half of the 4th cent., but it is of doubtful form, being also written Chia [40]-shê-lo (*Ka-§ia-11) and even Ch'ieh-shê-lo-shih [] (*G'ia-sia-1A-ziäi), abbreviated to Lo-shih (LA-ziäi); cf. Shui-ching chu, ed. of WANG Hsien-ch'ien, 2, 5 a. The location is no more certain than the name, but the description in the Shui-ching chu, quite independent of that devoted further on to Shu-lo (2, 10 a), makes it very unlikely that Shu-lo was meant.
The same objection applies to - 70 Ch'i-sha, in which CHAVANNES proposed to see another name of Kâsyar (BEFEO, In, 432-433). In CHAVANNES'S translation, the pilgrim Chih-mêng had at Ch'i-sha, with the bowl of Sâkyamuni, the same miraculous experience which Kumàrajiva had with the same bowl, at about the same time, at Sha-lo, i.e. Kâsyar. The coincidence appeared striking enough to carry STEIN'S conviction, who declared that CHAVANNES'S argument left no room for doubt (Ancient Khotan, I, 67). I hold, however, a different view. The name Ch'i-sha (*G'jie-a) is probably a transcription, though it has the disadvantage of having too obvious a meaning «Marvellous Sands » in Chinese. A kingdom of Ch'i-sha is mentioned in the Pei shih (97, 10 a-b; Wei shu, 102, 8 a), in connection with other unknown names of Western countries. When China created in 659-661 a new administrative nomenclature for the countries situated to the west of Chinese Turkestan and resorted for that purpose to old geographic names which she applied in a more or less haphazard manner, the name of Ch'i-sha was given to a district corresponding to Gûzgan, between Mary and Balkh (cf. CHAVANNES, Doc. sur les Tou-kiue, 71, 278) . The authors of that new nomenclature were more probably indebted for the name to the notice in the Pei shih (as in many other cases) than to Chih-mêng's little-known account. We have no direct indication as to the true location of Chih-mêng's Ch'i-sha, which occurs in the text incidentally, among place-names of northern India, not of Chinese Turkestan. If CHAVANNES thought of Kâsyar, against what the text, as he translated it, would seem to imply, it was only because an identical legend was attributed to Sha-lo (Kàsyar) in Kumârajiva's life and to Ch'i-sha, according to CHAVANNES, in Chih-mêng's fragmentary account. Nothing could be built on such a coincidence. The text of Chih-mêng's account which is given in his biography of 519 is only a summary, clearly corrupt in some cases (there are many independent quotations of the original account, which would be worth collecting); on the other hand, Kumârajiva's biography abounds in legendary
IIIF Color HighRes
