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

746   HISTORY AND RECORDS OF THE TUN-HUANG LIMES [Chap. XX

r

Military subdivisions on Limes.

Officers subordinate to lu-wei.

of the section of Hsiin-ch`êng 1 tt. The name is not met with elsewhere and looks curious, as it literally means ' along the [Great] Wall '.8 In what relation the ' superintendent of Yii-mên barrier ' mentioned in some éocuments of T. xtv stood to the ' to-wei of Yii-mên', I am unable to make out.°

We find yet another to-wei of Tun-huang mentioned in the interesting and fortunately complete document No. 592, T. xii. a. 3, of A.D. 21, which contains the pay account of a certain soldier, and incidentally furnishes us with definite indications as to the successive grades of the military hierarchy on the Limes. Corporal Wang, to whose origin and claim for pay we shall have occasion to refer further on,10 is described as ' subordinate to the officer commanding a watch-post of the Ping-wang barrier, which depends on Pu-kuang, sub-section (ch`ü ;;;) and residence of the [tu-]wei, in [the command of] Tun-tê', i. e. Tun-huang." We have already seen that the local name Ping-wang

was applied to that portion of the Limes which extended from the Jade Gate eastwards to

T. xxii. c and perhaps further still.12 Of Pu-kuang „   M. Chavannes shows that it is described

in the CIz`ien Han shu as a military subdivision (hou-kuan   0) with fortified headquarters in

which the to-wei of the central section had his official residence." It seems, therefore, to follow that the section of the Limes comprising Ping-wang and extending east of the Jade Gate was under the administration of the military commandant, or to-wci, of the ' central section ' of Tun-huang.l4 The term hone-kuan which has just been mentioned, and the application of which during Han times M. Chavannes has discussed at some length," is also met with in our documents. In the Han Annals the term is applied to administrative subdivisions subject to military authority and having fortified headquarters. The references in the documents are too brief to furnish clear evidence on the point. But it is of interest to find the term coupled with the names of Yii-mên and Ta-chien-tu.16

The fact that the great mass of our Limes documents has been recovered at small watch-posts explains why references to officers subordinate to the tze-wei are far the more numerous. Among such the Chien Han shu, as M. Chavannes points out, specially names ' chiefs of a thousand men ',

ch`ien jên   A, and ssû-ma   ,f6," and, no doubt, they represented the higher ranks in the military
hierarchy below the to-wei. The ssû ma are met with in about half a dozen documents without their function being made clear to us." The former were evidently graded higher, and may have exercised functions corresponding to those of a battalion commander. It is significant that the three documents in which they are mentioned were found at T. xiv, the Jade Gate headquarters.19

8 The order is addressed to the commander of the Yen-hu company, which, as ive have seen above, p. 648, probably garrisoned the watch-station T. v1. c, and it was actually found at T. vt. b. Was Hsiin-ch'êng at any time the designation of the extreme western section of the Limes, or is it possible to assume that what was meant by the `section along the Wall' was really the YU-men barrier section ? Ta-chien-tu at T. tv was certainly comprised in this according to other documents; see above, pp. 689, 691.

9 Cf. Doc. Nos. 315-17. The specific title is shown by

M. Chavannes, Documents, p. 231, to be attested also in the commentary of the Chien Ran shu.

10 See below, p. 757.

" For the name Tun-11—   given to Tun-huang by
Wang Mang, cf. Chavannes, Documents, p. 13o.

12 Cf. above, pp. 691, 699, 72o, and for the companies which furnished pickets for its watch-posts, pp. 717, 719. Though frequently mentioned (see Documents, Index), P'ing

wang is not specifically designated as a ' barrier ' elsewhere.

13 Cf. Chavannes, Documents, pp. 13o sqq. Doc. No. 623, T. xxvut. 37, mentions Pu-kuang as a place where a halt was made for the night.

" M. Chavannes is doubtful about a possible reference to yet another section commanded by a tu-wei which may be contained in Doc. No. 168, T. vi. b. i. 142.

16 Cf. Documents, pp. 130 sq.

1° Cf. Doc. Nos. 166, 458. In the latter record, referring to Yti-ma, M. Chavannes translates hou-kuan by forteresse'. The term is found also in Doc. Nos. 165, 261, 302, 542. No. 302, about which see p. 748 below, distinctly points to a hou-kuan comprising a cultivated area.

About the position and character of Ta-chien-tu, cf. above, pp. 636, 691.

17 Cf. Documents, p. 72 in note on Doc. No. 305. 1e See Doc. Nos. 55, 6o, 275, 438, 461, 605.

19 In Doc. No. 305 a certain chienu n, having the title of