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

Khotan unsuited for nomads.

Tarim Basin not on

road of migrations.

Effects of Muhammadan conquest.

138   THE KHOTAN OASIS : ITS GEOGRAPHY AND PEOPLE [Chap. VI

great, and the grazing and produce offered by them so extremely limited, that no great host could either in modern or ancient days have attempted the passage'. What obstacles the desert route from the west offered we have had occasion to note in the previous chapter.

But what possible inducements were there for any nomadic tribe or nation to penetrate these desert defences and settle down in Khotan ? Only painstaking cultivation, based on the experience of many generations, could produce and maintain the oasis. Of ground such as the various nomadic races from the Mongolian plains and the Tibetan plateaus need for their maintenance, the region around Khotan could never in historical times offer any. Even the hardy Kirghiz, those last representatives of the Turki race retaining its original nomadic condition, who still graze their flocks in most of the mountains encircling the Tarim Basin, have given a wide berth to the inhospitable region that borders Khotan on the south and thence stretches eastwards into Tibet.

It is because the conditions here briefly indicated apply with almost equal force to all the oases fringing the great desert of Eastern Turkestan that we find each great wave of Central-Asian migration, of which historical records have come down to us, moving along the northern foot of the Tien-shan or even further north. When tracing above the general history of Eastern Turkestan, we have seen that all the tribes which followed each other on that great route of invasion westwards—Yüeh-chih, Huns, Hephthalites, Western Turks—were probably able in their time to exact tribute and a more or less nominal submission from the various small states of the Tarim Basin while Chinese power was too weak to protect them.8 But nothing leads us to suppose that this political dependence during limited periods affected the internal organization of these states and their ethnic composition, any more deeply than the presence of Chinese garrisons in Han times and again for over a century under the Tang dynasty. Nor is the brief period of Tibetan predominance likely to have been accompanied by far-reaching effects in this direction. For over a century after Tibetan power had been finally broken by the Uigurs Khotan seems to have again enjoyed independence.

It is only when we reach the time of the conquest of Khotan by the successors of Satok Boghra Khan that the possibility of a radical change in the ethnical character of the population seriously presents itself. The spread of Muhammadanism and of the Turki language testify to the powerful effects which followed the establishment of the Karluk Turk family of Satok Boghra Khan in undisputed possession of the whole of Eastern Turkestan. Do these great subversions imply that the racial character of the people brought under the new rule was also fundamentally altered ?8 Our historical records for the period which witnessed the conversion to Islam, and for the centuries immediately following, are far too scanty and vague to furnish an answer to this question. Analogies drawn from other territories of Asia, where an indigenous

7 Karim Beg, of Yarkand, an old retainer of Niaz Hakim Beg, gave me a vivid account of the difficulties encountered by the small force of three hundred horsemen which the governor dispatched in 1877 via Cherchen and Lop-Nor to join the conquering Chinese army at Kara-shahr with his submission. Most elaborate arrangements for water and fodder were needed to assure their safe passage. That the conditions of travel were not much better for caravans in old days is clearly proved by the narratives of Hsüan-tsang and Marco Polo, who both followed this route.

8 See above, p. 58.

a M. Grenard, who has devoted two interesting and suggestive chapters (ii, iii, of vol. ii) to the discussion of the

ethnical origins of the population of Eastern Turkestan generally, has treated this question of the Turkization' of the earlier inhabitants with much care and ingenuity (vol. ii. pp. 5o sqq.). His observations on the levelling influence of Islam, the successive reinforcements supplied to the Turkish element by the later invasions of the Kara-Khitai, Mongols, Jungars, &c., are very judicious. But scarcely enough allowance has been made for the considerable differences likely to exist from the beginning in the ethnical constitution of a population scattered over so vast an area, and to the strongly marked individuality of each oasis explained by differences of geographical position and historical development.