National Institute of Informatics - Digital Silk Road Project
Digital Archive of Toyo Bunko Rare Books

> > > >
Color New!IIIF Color HighRes Gray HighRes PDF   Japanese English
0330 Innermost Asia : vol.2
Innermost Asia : vol.2 / Page 330 (Color Image)

New!Citation Information

doi: 10.20676/00000187
Citation Format: Chicago | APA | Harvard | IEEE

OCR Text




locations, which general topographical considerations suggest to me, must be considered as partly conjectural.

Location of   I should be inclined to seek Chi-cho Kuan, the ` inn ' OR of Chi-cho 7 4, near Chilan,

Chi-cho   a point which the road was bound to pass at all times, and which, owing to its assured water-supply,

Kuan and

Yeh-chê   must always have formed an important halting-place on a stretch of the route passing arid wastes

Kuan.   on either side. The mention of the old city of Ta-kan A tL beyond it suggests a site even then

abandoned to the desert. The distance of 120 li indicated to the next stage, Yeh-chê Kuan

i *   (` the visitors' inn '), and the significant term of Kuan   occurring in its designation
seem to point to the site of Chong-tim on the ancient and more direct road line. If this location

were right, the ` city of Chü-shih-tê'   k it, situated 6o li farther on, might well be sought
to the south-west of the ruins of Lâl-tâgh, where ` Tati ' remains indicate the former existence of a settlement of some size.23 The distance recorded would agree well with this location ; for the direct distance between the two last-named sites is, as the Map No. 7. B, C. 4, shows, just one half of that from Chilan to Chong-tim. But the reference made to Chü-shih-tê as a place ` on the boundary

line of Kuchâ', i. e. Ch`iu-tzû   #, is rather puzzling. It is difficult to believe that the territory
of Kuchâ could have extended so far westwards, or that, if reference to Chü-shih-tê as a border town were intended, the text would have been worded as it is ; for the usual practice in these itineraries is to mention the first place in the new territory reached and not the last of the territory left behind.238 If it were permissible to assume that some mistake has crept into the text here one might suggest that the name of Ch`iu-tzû (Kuchâ) has replaced that of Ch`ü-sha or Chia-sha

I   a rendering of the local designation of Kâshgar recorded by Hsüan-tsang and the

  • Tank-shu.24

Mention of   That the Tang itinerary has here brought us to the immediate vicinity of the hills east and

` Red   north-east of Marâl-bâshi is made quite clear by what it next tells us about ` Yü-t`ou Chou

KâshgarRive i. e~ on the Ku-shih Hill   at on the north bank of the Ch`ih Ho   tff (Red River)'. It appears
daryà.   to me very probable that this refers to the extensive site to the north-west of Tumshuk, marked by

the ruins of large Buddhist shrines, an ancient circumvallation and numerous dwellings found on, and close to, the southern end of the rocky hill chain known as Chöl-tâgh.25 Through the wide gap between it and the smaller Tumshuk-tâgh to the south there passes a dry river-bed, which finds its continuation partly in the Ghôra-akin, and which undoubtedly represents an old terminal branch of the Kâshgar river. Considering the general direction which the ancient route must have followed past Marâl-bâshi and the isolated hills to the east of it, there seems to me to be little doubt that the Kâshgar-daryâ is meant by the ` Red River '. The identical name, in the form of Kizil-su, is still borne nowadays by the main branch of the Kâshgar river, which passes to the south .of the ` Old Town ' of Kâshgar and by the river as a whole higher up (Map No. 2. D. 2).

` Mount   We have another definite topographical indication as to the line of the ancient road where we

Ch`i'   are told that it ` crosses the Red River and passes Mount Ch`   . I believe this mountain
with ified may safely be identified with the Mazâr-tâgh, the highest and by far the most conspicuous of

Mazàr-   the hills which rise above the flat alluvial plain in this region. Its rugged mass stands up to a height

tagh.   of over 2,500 feet above the Marâl-bâshi oasis, and culminates in two easily distinguished peaks,

for which our clinometrical readings indicated elevations of 6,33o and 5,910 feet respectively. These peaks attract the traveller's attention from a very considerable distance, and obviously


23 See above, i. p. 78.

23a Cf. Chavannes, Turcs occid., pp. 6, 9, 13 ; Serindia, iii. p. 1335.

24 See Chavannes, Turcs occid., p. 121 ; for other Chinese

renderings, cf. Anc. Khotan, i. p. 48.

25 See Map No. 8. B. s. Regarding this site, repeatedly visited by European travellers and partially explored by M. Pelliot, cf. Serindia, iii. p. 1309.