SHAN-SHAN BETWEEN TANG AND HAN TIMES 323
Thus the important description of the three routes to the West compiled about A.D. 607 by the
Chinese general and diplomat Pei Chu. clearly indicates that the southern route passed through Shan-shan, south of Lop-nor, on to Yu-t'ien or Khotan.' As we are distinctly told that Pei Chu collected his information from the foreigners visiting Kan-chou, it can scarcely be doubted that the application of the name Shan-span to the Lop tract was actually known locally and was not merely the result of learned conjecture.
The same conclusion is justified in the case of the narrative of Sung Yun, who reached the town of Shan-shan in A.v. 519 after having travelled three thousand five hundred li westwards through the territory of the Tu-yii-hun.2 As this people held the Koko-nor region and the elevated plateaus west of it, Sung Yün must have travelled along the route which still leads from the Koko-nor across Tsaidam down to Mirân and Charkhlik.3 ' The kings whom this town had [previously] owned have been conquered by the T'u-yii-hun ; at present the ruler in this town is the second son of the [king of the] T'u-yii-hun ; he has the title of the General Pacificator of the West and commands three thousand men whom he employs to stop the western Hu (barbarians).'
Sung Yün's account, brief as it is, is of considerable interest for our inquiry. It shows in the first place that the designation of Shan-shan was in actual use in the early part of the sixth century for the Lop tract ; for a reference to the map and the study of his subsequent march westwards to Tso-mo or Charchan show that the route he followed from the Koko-nor region could not have debouched from the mountains elsewhere than upon Miran and Charkhlik. The mention of the T'uyü-hun conquest, which a notice of the Pei shih confirms for about A. D. 540 and extends to Charchan also,* is of further interest because it may yet help to explain the appearance,-more than a century later, of that new name of the territory which Hsüan-tsang's Na fu j5o is intended to transcribe, and which we have traced through the Nob of the Tibetan records down to the modern Lop. Perhaps also the new name of Cher-chen, which first appears in the Tibetan records of M irân,6 may be attributed to the change wrought by this T'u-yü-hun occupation. But whereas Charchan at the time of Sung Yün's passage counted only a hundred families,6 the ' town of Shan-shan ' must still have been a place of some size to serve as a royal residence, and the resources of its cultivated area must have been considerable to maintain a garrison of three thousand men. The friendly relations which prevailed in the early part of the sixth century between the T'u-yü-hun and the Northern Wei dynasty of cognate origin ruling in China,' explain why the T'u-yü-hun established at Shan-shan might be considered, in the eyes of a Chinese official mission, as an outpost of the Empire against the ' Western barbarians'.
We may here conveniently take note of the account which Fa-hsien, the earliest of the Buddhist pilgrims from China to the Western regions, has left of Shan-shan. He and four other monks, in the autumn of A. D. 400, started from the frontier territory of Tun-huang, ' in the suite of an envoy'. The prefect of Tun-huang ' had supplied them with the means of crossing the desert [before them], in which there are many evil demons and hot winds. Travellers who encounter them perish all to
Pei Chu calls Lop Shan-chan.
T'u-yü-hun occupa- tion of Charkhlik.
' See Ritter, Asien, v. p. 563, quoting from Neumann, Asiatische Studien, p. x96 ; also Richthofen, China, i. p. 53o, note.
2 Cf. Chavannes, Voyage de Song Yun, p. 12 ; concerning the Tu-yü-hun, cf. ibid p. r x, note 5, and now also M. Pelliot's note in/ Asia., 1912, xx. pp. 520 sqq.
3 See e.g. the Royal Geogr. Society's map of Tibet and the surrounding regions,1904. The same route must have been followed also by the Indian Buddhist monk Jinagupta, but in the reverse direction, when about A.D. 557 he proceeded from Khotan
to Hsi-ning ; cf. T oungpao, 1905, p. 34 E. But his lamentably brief narrative does not specify the intermediate stages.
4 See Voyage de Song l'un, p. 12, note 7 ; also above, p. 298.
6 Found in document M. 1. xxviii. 2.
e See Chavannes, Voyage de Song Yun, p. 13 ; also above,
See Chavannes' article, finagupta, T'oung pao, 1905,
See Legge, Travels of Fd-hien, p. 11 sqq.
T t 2