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

Sec. v]   OFFICERS AND SOLDIERS OF THE LIMES   751

stations (T. Iv. b, xti. a, xiv, xv. a, xvr, xvii, xvIlr, xxvrt), the occupation of which,•as we have seen,

continued down to later periods."   •   .

From this we may be tempted to draw the reasonable, if conjectural, conclusion that during the early times of the Limes extension the body of the border troops had to be made up by soldier colonists brought from other parts of the Empire, while later on the ranks could be filled more easily with local men drawn from the north-west marches themselves. I think that this conclusion receives a distinct and interesting support from another observation also. Among the twelve men from distant provinces not less than eleven are mentioned with the special designation of ` garrison soldiers',

shat-tsar   7,'0 whereas we never find this term applied to any of the men from Tun-huang
or Kan-su. The true significance of the distinction becomes evident in the light of what M. Chavannes himself has rightly observed about the probability of the men from a distance having been convicts deported for border service. In the document No. 263, T. vI. b. iv. I, we find in fact eight men, otherwise of unspecified origin, distinctly spoken of under the exact juridical term which, as M. Chavannes shows, was applied during Former Han times to convicts condemned to forced labour." As Chinese historical texts definitely indicate the construction of defensive walls and the guard service on the frontier as the main kind of forced labour, we may conclude that these ` garrison soldiers ' were convicts deported for service at the Limes.

Impressment of criminals for military service beyond the northern borders of the Empire appears to have been a method regularly resorted to in connexion with those distant expeditions which the Emperor Wu-ti's Central-Asian policy necessitated.52 The historical records show what terrible losses and hardships they often implied. Permanent service on such lines of the Limes as that of Tun-huang, passing through absolute desert, must have been equally trying. The very interesting specimens of Chinese poetry which M. Chavannes at the end of his Iantroduction has reproduced and translated from the literature of the Tang period `i3 give eloquent expression to the dread with which those expeditions and forced service on the border alike were remembered even centuries later. It seems very probable that the majority of the men moved into the early ` agricultural colonies ' were deported convicts, and that this class continued to be drawn upon to some extent later, too, as a supplementary source in addition to the men furnished by the local border population. But the documents do not reveal to us the changing proportion of the convict contingents brought from afar for the Limes service of Tun-huang. Nor could the dry details of petty routine, which form the bulk of their contents, be expected to give us such vivid glimpses of the personal trials and sufferings undergone by the exiles as those poems just referred to. We have, however, evidence of an incident such as must have been common under these conditions of recruitment and service in No. 439, which orders the pursuit of deserters by the commandant of a certain watch-post.

That some of the soldiers stationed at the watch-towers were mounted men is distinctly mentioned in a number of records." The abundant remains of horse-dung found in the refuse-

Garrison soldiers', i.e. convicts, as border colonists.

Criminals impressed for Limes service.

Mounted soldiers on Limes.

" Among the fourteen men who are there named, seven are natives of Tun-huang (Nos. 392, 574, 579, 58o, 592), two of Kan-chou (Nos. 342, 417). Only three are from Ho-nan (Nos. 416, 434, 456), and two more from Shan-hsi (No. 550).

a° In the case of the twelfth man (Doc. No. 183) the incomplete record leaves a doubt as to his employment as a soldier or otherwise.

61 Cf. M. Chavannes' note on Doc. No. 263.

i2 Cf. references in Ssû-ma Ch`ien's chap. cxxiii, as trans

lated by Kingsmill, f.R.A.S., 1882 (reprint), pp. 16, 24. Parallels in the history of more than one country of modern Europe are not difficult to find.

n See Chavannes, Documenls, pp. xvii—xxiii. The last poem is of special interest to us because it shows that the demands of military service might claim the whole family of a settler on the border. It thus helps to support what has been stated above, p. 748, about the numerical relation between the lui and the fing which maintained it.

64 Cf. Doc. Nos. 279-83, 286, 416, 417.