国立情報学研究所 - ディジタル・シルクロード・プロジェクト
『東洋文庫所蔵』貴重書デジタルアーカイブ

646   THE END OF THE CHINESE LIMES   [Chap. XVII

impressed me. Ends of brushwood, reeds, layers of dung, and the like, cropping out on the gravel-strewn slopes of the little eminence occupied by the ruined watch-tower, were sufficient evidence of them. An experimental scraping, made on the evening of our arrival at a point where a few wooden posts protruded some dozens of yards to the north-east of the tower (Fig. r 72), soon brought to light, from the lower edge of a large deposit of refuse, over two score of Chinese records on wood of the usual slip' size (T. vi. b. i. 1-46). Most of them were complete, but had suffered much decay by moisture owing to exposure near the surface and to the close vicinity of a shallow drainage channel scooped out by the rare rain that this arid ground may have seen during many centuries. The find was encouraging, especially as some of the records read there and then bore dates from 63 to 57 B.C. Yet it in no way prepared me for the big haul which was waiting here to be gathered next morning. While looking after the sinking of a well at our camping-place, a little over a mile to the south-east, I had sent Chiang Ssû-yeh ahead to continue the clearing. When I rejoined him an hour later, I found him triumphantly guarding for me nearly a hundred fresh Chinese wooden documents (T. vi. b. i. 47-140), most of them complete ' slips ' and, in spite of slight damage from damp, still legible. There were close on a hundred more, either blank or completely effaced. The whole had been recovered from an area scarcely more than two feet square ; in Fig. 172 the splintered piece of timber held by a labourer exactly indicates the place. None of these records lay deeper than about a foot from the surface, where the natural gravel was reached under the covering layer of reed-straw, chipped wood, and similar refuse.

By clearing the stratum of rubbish lower down the slope, where it gradually thinned out until the natural surface, hard gravel, emerged, the records T. vi. b. i. 141-88 were discovered. Subsequent excavation and careful search of the refuse round the place of the main find brought the total number of inscribed and still legible pieces to 310. Of blank slips, too, and of others which had completely lost their writing, another hundred or so were found. It was quite clear that at this particular spot the contents of a small official archive had been thrown down together on the rubbish-strewn slope. As M. Chavannes' careful examination has since established the fact that the very numerous dated records among them are all comprised within the years 65-56 B.C., we can approximately gauge the rate at which the ' waste papers ' of the ancient office established at this station had grown during that period of ten or eleven years.

The rest of the rubbish-heaps on the slopes below the watch-tower, extensive as they were, added very little to the collection of documents. The slips T. vi. b. i. 311-20 were the only ones found above and below the main deposit just described. From another large layer of refuse, about a dozen yards to the west of the tower, came the records T. vr. b. iv. 1-3 and two fragments of small rectangular tablets, once apparently inscribed, T. vi. b. iv. ooi, 002. A curious discovery was made in a shallow layer about sixteen yards to the north-west of the tower. Here was found a great mass of wooden ' shavings ' covered with Chinese characters, probably over a thousand in all. It might have passed for a great find—if Chiang Ssû-yeh had not at once noticed that the writing was obviously by the same hand and the phrases constantly recurring. He was, no doubt, right in concluding that these were chips from improvised tablets which some officer or clerk, eager to improve his penmanship after the wont of the present-day lileratus, had used again and again for writing exercises, planing them down with a knife each time to obtain a fresh surface. The material, roughly cut from tamarisk and Toghrak branches, was such as the jungle close by on the fringe of the marshy basin would furnish in plenty.

The abundance of the written records which the remains of T. vi. b have yielded is best illustrated by the fact that, among the 708 ancient Chinese documents which M. Chavannes by

3 This fold of the slope is visible in the background of Fig. 172, behind the figure of the labourer.

Wooden records from refuse-heap.

Relics from office archive, 6556 B. C.

Chinese writing exercises on wood.

Abundance of Chinese records.