THROUGH CHI LAS, DARFL, AND TANGYR [Chap. I
Menace to ' Gilgit Road ' from Childs.
Supplies from Kashmir through Childs.
Advantages of Childs route.
I believe that I have conclusively proved in Ancient Khotan that it designates Chitral, being probably intended to reproduce Kdshkdr, the old alternative name of that territory.6
In the same work I emphasized the obvious geographical fact that Childs was the point at which alone the line of communication from Kashmir to Gilgit and Yasin was liable to serious interference from the west, i. e. the Chitral side.' A glance at the map will explain this, and what we know of the modern history of these mountain tracts points to the same conclusion ; for it supplies abundant evidence of the danger to which the ` Gilgit Road ' of the Sikhs and Dôgras was exposed, right up to the middle of the last century, from predatory raids of the Childs people.8 These raids ceased only after Maharaja Gulab Singh's troops in 1851, operating in part by the Barai pass, had succeeded in invading Childs and temporarily reducing its chief stronghold. But the Chilasis soon regained independence, and their turbulent disposition, with the support they drew from the other Dard republics farther down the Indus, remained a source of danger to the ` Gilgit Road '. This menace was finally removed only in 1893, when after some serious fighting the territory was occupied and a garrison of Imperial Service troops permanently established in Childs Fort.
These considerations had already led me to locate in Childs the danger point referred to by the Chinese record of A. D. 749. The observations I was able to make during my visit to Childs have fully confirmed me in this view, but with a modification as regards the actual geographical position of the route in question. It appears to me now very probable that the route or routes by which those indispensable supplies from Kashmir reached the Chinese Imperial garrisons in Gilgit and Yasin did not lie, as assumed in Ancient Khotan, along the present ` Gilgit Road ' across Gurêz, the Burzil pass and Astôr, but led direct through Childs. My reasons are the following. A reference to any map of the territories between Kashmir and the Hindukush will show that the line followed by the present ` Gilgit Road ' from Kashmir to the Indus at Bûnji is far longer than the line across the Barai pass to Childs.9 Whereas the distance as reckoned in official route records between Bandipur on the Volur lake and Bûnji is 158 miles, the same from Bandipur to Childs, as tested on the route I followed, is only about 116 miles. It would be still less if instead of proceeding across the Fasat pass to Childs Fort the traveller were to follow the stream draining the Barai pass on the north straight down to the Indus at Biinar.
It deserves to be noted also that, even before the improvements referred to below, the whole of the route across the Barai Pass, though difficult in places, was passable for laden animals, and that ample grazing is found along its whole length. It was the absence of grazing almost throughout Astiir that had made the use of the Bandipur—Burzil—Bûnji route for laden transport practically impossible until the ` Gilgit Road ', a feat of modern engineering, was constructed and elaborate commissariat arrangements made under British control.10 Though some 700 feet higher, the Barai pass is not closed by snow appreciably longer than the Burzil, and the pass above Matsil which has previously to be crossed on the watershed between Kashmir proper and the Kishangangd valley is certainly easier and less exposed to danger from avalanches and snowdrifts than the Tragbal pass which corresponds to it on the ` Gilgit Road '.
6 See Ancient Khotan, i. pp. 13 sqq. ; also Serindia, i. pp. 29 sq.
7 Cf. Ancient Khotan, i. pp. 16 sq.
8 For interesting accounts of such Chilàsi raids, which being made largely for the purpose of capturing slaves caused depopulation in the Astbr valley, but were extended also to the Kishangangà and even as far as Skardo and Kashmir, cf. Drew, Jummoo and Kashmir, pp. 398, 404 sq. ; Biddulph, Hindoo Koosh, pp. 15 sq. ; Vigne, Travels in Kashmir, ii.
9 As the territory of Childs was inaccessible for survey work prior to 1893, the Atlas of India Sheets and other earlier maps offer no help for this comparison. For convenient reference, see e. g. Sheet No. 43 of the Survey of India i : i,000,000 Series, or Sheet No. 3 of the Northern Trans-frontier Series.
10 Cf. for some account of the ` Gilgit Road ' and its traffic arrangements, Ruins of Khotan, pp. 12 sqq.