| || || || || || |
326 HISTORICAL NOTICES OF LOP, SHAN-SHAN, AND LOU-LAN [Chap. IX
by Li Tao-yuan, or rather the older authority he followed, to have been an earlier capital of Shanshan or Lou-lan. However, its situation must have lain near the ancient capital of the kingdom ; otherwise the establishment of a Chinese military colony there could not have served its avowed object of rendering the position of the new king secure. Thus the presumption is created that the ancient capital also, like I-hsün itself, lay to the south of the terminal course of the Tarim.
Before proceeding to record the data which enable us more definitely to fix the place of that earlier capital, the commentator makes a long digression to record an anecdote which has no direct bearing on the question immediately before us, but which it will be necessary to consider hereafter. It concerns the foundation of a military colony at Lou-lan by a certain Chinese commander, So Man, who was believed to have secured the necessary irrigation by means of a dam across the Chu-pin River, and to have miraculously protected this dam against the river's flood, after the fashion of the story of Achilles and the river Xanthus. Whatever historical facts underlie this anecdote, their time and place must be looked for elsewhere ; for, on the one hand, inferential evidence shows that the alleged founder of this colony belonged to the Later Han period, while, on the other, geographical considerations point clearly to some locality higher up on the Tarim as the site of the dam which figures in the story. We shall have to return to it when discussing the ruined site north of the Lop desert.12
Resuming his geographical account, Li Tao-yuan proceeds to tell us : ` This river [of Chu-pin] empties itself eastwards into the lake. The lake is to the north of Lou-lan, at the town of Yü-ni i a. This [town of Yü-ni] is the one which is usually designated as the " old eastern town "
- . It is 1,600 li distant from the Yang barrier ; towards the north-west it is distant
| || || || || || |
1,785 li from Wu-lei ,( ; on the east it is 1,8651i from the kingdom of Mo-shan w ; towards
the north it is distant 1,890 li from Chü-shih . 3 (Turfan).' The distances and bearings here indicated, together with a brief description of the land and its products, are directly taken from the account of the kingdom of Shan-shan or Lou-tan 13 in the Former Han Annals. Hence the discussion of them will best be left until we come to examine that account. Here it suffices to note that they are easily reconciled with the location of Yü-ni to the south of the terminal marshes of the Tarim.
The point of essential importance is that the tradition of Li Tao-yuan's time knew Yü-iii, which in the Former Han Annals figures as the capital of Shan-shan or Lou-lan, as ` the Old Town', and placed it to the east of I-hsün, the actual capital of the territory. It must further be kept in view that whereas I-hsün is described as situated to the south of the united course of the Charchan and Tarim rivers, Yü-ni is placed by Li Tao-yuan to the south of ` the lake ' which receives them, and apparently not far from it. If we now compare with these data the geographical facts regarding the areas of early oci=upation in the Lop tract as well as the archaeological evidence detailed below, we are, I think, necessarily led to the conclusion that I-hsün must be located at the present Charkhlik, while the position of Yü-ni or the ` old eastern town' is marked by the earlier remains of the Miran Site. It will be seen further on that these ruins must have been abandoned well before Li Tao-yüan's time, and that their distance from the nearest Lop-nor marshes northward is even now, in spite of the latter's obvious shrinkage, scarcely more than twelve miles.
The main argument for these identifications lies in the fact that the relative position assigned to I-hsün and Yü-ni by our texts corresponds exactly to that actually observed in the case of Charkhlik and the Miran Site. As regards the identity of the latter with Yü-ni, important confirmatory evidence is supplied by the fact that the excavations described below prove the old
" See below, chap. xi, sec. x.
Is Cf. Wylie, Notes on the Western Regions, in J. Anlhrop. Insl., x. p. 24.