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

e

714   THE JADE GATE BARRIER   [Chap. XIX

In the absence of any definite structural indications, the hope of solving the puzzle presented by this imposing ruin seemed to rest upon any records or other finds that might be brought to light by excavation. For such the interior of the halls seemed from the first to hold out but scant promise ; for owing to the hollows scooped out on the south by the drainage, as already mentioned, the original flooring had been washed out over great parts. Elsewhere, too, there was a striking absence of miscellaneous refuse, such as usually marks prolonged ancient occupation of stations on the Limes. In the end, our laborious clearing of the accumulations of sheep-dung and drift-sand, which covered the original floor along the foot of the north wall in places to a height of 8 feet or so, was only rewarded by seven fragmentary wooden slips found on or near the low plinth in the north-west corner of the central hall (T. xvIII. ii) within a small layer of refuse. The clearing was carried along the whole of the north side of the western and central halls, but could not be extended into the eastern one. There large portions of the north wall had completely fallen and buried the floor under heavy masses of clay, which could not be attacked with my small posse of labourers. The two records from ii included in M. Chavannes' publication, Doc. Nos. 456, 417, relate to individual soldiers from Ho-nan and Kan-su, but give neither dates nor a clue to the character of the ruin.

I had hoped to come upon deposits of rubbish in the fairly well protected narrow court that extended within the inner enclosure on the north side. But it was searched in vain until at last, on scraping the ground below the slope of the little terrace bearing the tower in the north-west corner (i), a heap of refuse was struck composed mainly of reed straw and ashes. From this forty inscribed pieces of wood and bamboo were recovered, and nearly twice as many blank bamboo slips, all much worn and repeatedly scraped, evidently representing waste paper ' which had been prepared for fresh use as palimpsest writing-material. A dozen uncut wooden sticks, of flat shape, looked as if intended to be split up into slips. Some of them were of the wood of a conifer, and had evidently been brought from a distance to be used as stationery, the nearest place where such wood could be found now being in the Nan-shan south-west of Kan-chou.4

Most of the inscribed tablets had suffered much from moisture, and it was not until the very last piece, T. xviii. i. 4o, Doc., No. 413 (Plate XII), was extracted and cleaned that Chiang Ssti-yeh discovered the precise date for which we were eagerly looking. It proved to correspond to the year 52 B.C., and thus made it certain that the ruin dated back to the early occupation of the Limes under the Former Han. The doubt about the character of the big ruin was removed when my learned secretary, after a minute scrutiny, had made sure that two among these records distinctly referred to transactions connected with a granary. In the course of the days spent here over survey and excavation the thought had already occurred to us both, as well as to Naik Ram Singh independently, that this strange pile might have been erected for the purpose of serving as a supply-store to the troops stationed or moving along the wall, and to officials and political missions travelling by the desert route.

In the light of the documentary evidence thus obtained, it became easy to recognize a simple explanation for all the structural peculiarities above noted : the big size of the halls quite unsuited for habitation, especially in the winter ; the small openings intended mainly for ventilation ; the choice of a building site conveniently accessible and near water, yet well raised above the adjoining ground to be safe from damp ; the arrangements of the two enclosures, obviously meant to facilitate the guarding of the building, not against hostile attack, but against theft. Thus we can fully account, too, for two striking features of the base of natural clay provided for the structure. Its height and steep face would help to protect the supply-stores from rodents, a possible cause of loss

For specimens of records from T. xvim i written on such wood, see Doc. Nos. 415, 4r8, Pl. XII.

Scanty finds within interior of halls.

Records from refuse in inner court.

Records relating to granary.

Structure intended for magazine.