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

648   THE END OF THE CHINESE LIMES   [Chap. XVII

of China's extreme western border. This vogue is attested by seven more fragments from the same work contained in my collection.' We are thus carried appreciably nearer to the period of the usurper Wang Mang, A.D. 9-23, after which, as I shall have occasion to explain further on, the westernmost section of the Limes wall proper, comprising the series of towers T. ry-xur, was probably abandoned. All the same, there remains the notable fact that none of the great mass of records found at T. vr. b gives evidence of a date later than 56 B.c. Hence the possibility must be kept in view that, while the line of the flanking signal-stations that stretched from T. v to T. vr. d continued to be occupied as long as the western extremity of the Limes proper was held, i.e. probably down to the first quarter of the first century A.D., the station T. vr. b, lying behind the line, was abandoned half a century or so earlier. This may, perhaps, have been due to its reduced importance when there was less risk of raids from the Huns and the nomads in the mountains on the south.

Almost as important as the document relating to the first establishment of the Limes is another record, T. vr. b. i. 152, Doc., No. 63, which refers to an imperial edict about the organization of three companies named Ling-hu, Yen-hu, and Kuang-ch`ang respectively. The number of documents from T. vr. b mentioning the Ling-hu, or 'barbarian-dominating ', company is so considerable, and the local bearing of the references made to it so obvious, that it is impossible to avoid the conclusion that the station T. vr. b itself was garrisoned by that company.? The name Kuang-ch`ang is not met with again ; but the Yen-hu, or ' barbarian-subduing ', company is found repeatedly in records which indicate a special local connexion with T. vr. b and its immediate neighbourhood. In Nos. 138, 139 (T. vr. b. i. 19, 235) we have orders addressed to a certain indigenous officer charged with the safeguarding of the Yen-hu area, and No. 49 (T. vr. b. i. 91) contains a report from the commandant of the Yen-hu company which refers to ' indigenous officers '. Seeing that the same company is named in No. 268, a brief record found at T. vr. c, and that at this very place I discovered the fragment of a wooden document in Early Sogdian writing,8 there is a temptation to recognize in that neighbouring station T. vr. c the locality which was guarded by the Yen-hu company and the native auxiliaries connected with it.9

A number of documents found at T. vr. b make it very probable that the station was in administrative matters linked up with, and dependent on, Ta-chien-tu, which, as explained above,lo may be identified with the terminal western section of the Limes proper, T. iv. a-c. Thus Nos. 51, 137, 138, 168 convey direct orders from the commandant of Ta-chien-tu. There is yet another locality mentioned in the records of T. vr. b which can be determined, I think, with great probability, if not practical certainty. I mean the watch-post Pu-ch`ang .. R. The documents Nos. 58, 83, 144, 145, which mention its commandant and an indigenous officer from it, afford no topographical evidence. But all the more precise is the information which, with our knowledge of the actual ground, can be deduced from No. 95, a very clear and well preserved tablet (T. vr. b. i. 162, Doc., Plate v). It records a statement about the fatigue duty performed by three men, obviously belonging to the

Cf. regarding these text fragments, Chavannes, Documents, pp. 1 sqq. ; below, chap. xx, sec. vi.

See M. Chavannes' note on Doc. No. 38. Particularly significant in this respect are Nos. 38, 42-45, 47, 48, 64-69, all containing receipts, statements, lists, etc., relating to the internal administration of the Ling-hu company ; Nos. 54, 56 announcing verbal orders from the commandant of it; No. 7o, labels inscribed with its name and evidently intended for attachment to articles sent to, or owned by, men belonging to that company. It may be noted as negative evidence for this location that the name of the Ling-hu company is not

met with elsewhere.

8 About this interesting find and its possible connexion with the ' indigenous officers ' of Yen-hu, see below, pp. 652 sqq.

e The records Nos. 71, 136 (T. vi. b. i. 296, 25o) afford no decisive evidence, but can be reconciled with the suggested location. The first is an inventory of cross-bow ammunition issued to certain men of the Yen-hu company ; the second mentions its commandant in what seems an indorsement of a certain notification.

10 Cf. above, pp. 636, 641.

Ling-hu, Yen-hu, and Kuangch'ang companies.

'Indigenous officers' with Yen-hu company.

Post T. vi.b and Tachien-tu section.

Watch-post Pu-ch`ang.