11. WS 1.12. See Zhang Wenjie 张文杰, “Bei chao jiang wu tan lun” 北朝讲武探论, Guo fang da xue tong shi jiao yu xue bao (2017.7): 38. Though having found use in this article, this author needs to disagree with the amount of time spent in it attempting to link these events with the Zhou li.
12. WS 24.609.
13. Giovanni da Pian del Carpine noted that “the men are divided, Tartars and everyone else, among the generals”: The Story of the Mongols Whom We Call the Tartars, tr. Erik Hildinger (Boston: Branden Publishing Company, 1996), 66. In his The Desert Road to Turkestan (Boston: Little, Brown, and Company, 1929), 21, Owen Lattimore described how in the process of being forced into Manchu banners, the Tumed Mongols “were hounded until they vanished.”
14. The slow, complex process of reorganization of different polities into the Khaghan’s Army is well described by Koga, “Hokugi no buzoku kaisan ni tsuite,” 62–76; the persistence of older forms of social and political organization among these groups in Matsushita Ken’ichi 松下宪一, “Hokugi buzoku kaisan saikō ― Gen Chō boshi o tegakari ni” 北魏部族解散再考―元苌墓志を手がかりに, Shigaku zasshi 123.4 (2014): 35–59.
15. See Zou, “Bei Wei de bing zhi,” 163–65.
16. For the ethnic makeup of the guard units, see Zhang, Bei Wei zheng zhi shi, 2: 691; for the troops mentioned for transfer, He, “Fu bing zhi qian,” 330, citing WS 50.1113–14. For the Murong, see He, “Fu bing zhi qian,” 323.
17. WS 40.901, 28.681. See also mention in WS 3.62 of how during a 422 progress of Mingyuan, as he crossed the Heng Mountains, “vassal lords of the four quarters each led their following to accompany” 四方蕃附大人各帅所部从 the Wei khaghan.
18. SoS 74.1912; ZZTJ 126.3963–64. Kikuchi Hideo 菊池英夫 discusses the clear separation between the inner guard units and auxiliaries in “Hokuchō gunsei ni okeru iwayuru kyōhei ni tsuite” 北朝军制に于ける所谓乡兵について, in Shigematsu Sensei koki kinen Kyūshū Daigaku Tōyōshi ronsō (Fukuoka: Kyūshū Daigaku Bungakubu [Shigakuka] Tōyō Shi Kyūken[sic]shitsu, 1957), 101.
19. It is not clear if Chinese settled to farm in the Datong Basin were also thought of as “Dai people”: according to Matsushita, Hokugi Kozoku teiseiron, 159, most at least were not, presumably because they did not give military service to the khaghan. As for military registration, this is not clearly described in our texts. There are anecdotal mentions in Wei shu, e.g., the mention of a figure who lived in Pingcheng, but “whose name was still with the garrison” 名犹在镇 to which he had been assigned (WS 91.1970; see also the bing hu 兵户 mentioned in WS 68.1520); and the mention of jun guan 军贯 in a 524 edict ending the use of military registers, in the midst of the regime’s collapse (WS 9.237).
20. The best example of release of guo ren from service is found in WS 7A.138. Other examples raised by Gao Min 高敏, Wei Jin Nan bei chao bing zhi yan jiu 魏晋南北朝兵制研究 (Zhengzhou: Da xiang chu ban she, 1998), 307, regard forgiveness given members who were not guo ren, but had in some way or another been forcibly incorporated into the garrison system: a 494 pardon for those 70 years of age, or infirm, who had been sent to the garrisons for crimes, WS 7B.174 (and see also 86.1884); as well as examples from late Wei of individuals impressed into service after conquest of their territory, who lived and worked in Pingcheng but whose “names were still on the garrison [register]” (as mentioned in the previous note): WS 91.1970, 68.1520. From the latter, and other evidence (see Scott Pearce, “The Yü-wen Regime in Sixth Century China” [PhD diss., Princeton University, 1987], 163–64), we can probably infer that guo ren assigned to the garrisons were also on those rolls, or a parallel set of them.
21. Yang Sen 杨森, “Dunhuang yan jiu yuan cang juan ‘Bei Wei jin jun jun guan ji bo’ kao shu” 敦煌研究院藏卷“北魏禁军军官籍薄”考述, Dunhuang yan jiu (1987.2): 20–24, has suggested the list came from Northern Wei. Zhang Jinlong, however, has made a good case that it probably belongs to Western Wei or Northern Zhou: Jin wei wu guan, 2: 671–74.
22. Yan, Bei Wei qian qi zheng zhi zhi du, 151; and see the comments of Di Cosmo in “State Formation and Periodization,” 18, 22.
23. Thomas Hobbes, Leviathan; or, The Matter, Forme and Power of a Commonwealth, Ecclesiasticall and Civil, ed. Michael Oakeshott (Oxford: Blackwell, 1957), 82. Edward Jones, in his “Militarisation of Roman Society, 400–700,” in Military Aspects of Scandinavian Society in a European Perspective, ad 1–1300: Papers from an International Research Seminar at the Danish National Museum, Copenhagen, 2–4 May 1996, ed. Anne N·rg·rd Jorgensen and Birthe L. Clausen (Copenhagen: National Museum, 1997), 19, adds the following, which accords well with the Taghbach state: no clear distinction between soldier and civilian (with regard to the Taghbach society itself, as opposed to the larger realm), nor between military officer and government official; the head of state is also commander-in-chief; all adult free men have right to carry weapons; a certain group or class is expected by reason of birth to participate in the army; education of the young involves a military element; symbolism of warfare and weaponry are prominent in official and private life; and warfare is a predominant government expenditure and/or a major source of economic profit. All these are important, but it seems particularly regrettable that little is recorded of how the young men of Dai were raised to take on their military duty. For military expenditure in the Wei state see Chapter 7 note 51; and also Nicola Di Cosmo’s discussion of military culture in his “Introduction” to Military Culture in Imperial China, ed. Nicola Di Cosmo (Cambridge, MA, and London: Harvard University Press, 2009), 3–4.
24. Chittick, Patronage and Community in Medieval China.
25. See the insightful comment of Zhu Weizheng 朱维铮, “Fu bing zhi du hua shi qi Xi Wei Bei Zhou she hui de te shu mao dun ji qi jie jue—jian lun fu bing de yuan yuan he xing zhi” 府兵制度化时期西魏北周社会的特殊矛盾及其解决—兼论府兵的渊源和性质, LSYJ (1963.6): 153, that coming from Inner Asian forms of organization, for the Taghbach the “soldiery and the people were one”; and Cen Zhongmian’s reminder that this was a “unity of the soldiery and the herdsman” 兵牧合一: Fu bing zhi du yan jiu, 3. Cen goes on in the same text (10) to compare and contrast armies in the north and south, stressing that under Northern Wei and its successors leadership and control of the armies was unified and centralized, while in the south it was local and dispersed.
26. Good points have been made in the last couple decades by scholars—Nicola Di Cosmo, William Honeychurch, et al.—that agriculture existed on the eastern steppe, and that these regions were not necessarily dependent for grains on the farmlands to the south. The presence of some farming on the steppe, however, does not mean that hunger never became an issue, for particular groups at particular times and places, as with the formation by Daowu of a large military force that was effectively a standing army with an insufficient base of support. In discussing the formation of the royal domain, Wei shu’s “Shihuozhi,” 110.2850, states that, “at this time the war chariots never paused, and though repeatedly there were [good] years, it was not enough to be seen for long” 是时戎车不息, 岁频有年, 未足以久瞻矣. Interesting comparisons can be made with Europe from the 1500s onward. Even if Europeans had enough food, they wished to eat it on something other than wooden plates, and taste it with added spice. And many were simply, really hungry: in E. M. Collingham’s The Hungry Empire: How Britain’s Quest for Food Shaped the Modern World (London: The Bodley Head, 2017), xv, we are told that “ ‘Bring me food’ became a persistent demand . . . a metaphor for the relationship of Britain to its empire.” Closer to home, Nicola Di Cosmo in “State Formation and Periodization in Inner Asian History,” 23, states that the “first ‘cry’ of a new inner Asian state was one of great, insatiable hunger.”
27. E.g., a campaign in 436, in which the enemy retreated. When the Wei general considered simply withdrawing, his commanders pointed out that “if we don’t have loot, then [we’ll] have nothing to provision the army, [or] reward the officers and soldiers”: WS 17.413; ZZTJ 123.3863. And see also Wang, Zhuan xing qi de Bei Wei cai zheng, 35ff.
28. WS 24.609. For another such anecdote, from the Murong point of view, JS 124.3094 speaks of how the cavalrymen of the Wei army, during the 396–397 invasion of Later Yan, “[are] provided only 10 days of grain on the backs of their horses.”
29. SoS 74.1912; ZZTJ 125.3959.
30. WS 38.867–68; Sakuma Kichiya 佐久间吉也, Gi Shin Nanbokuchō suiri shi kenkyū 魏晋南北朝水利史研究 (Tokyo: Kaimei Shoin, 1980), 364–66; He, “Fu bing zhi qian,” 348.
31. WS 79.1756.
32. He, “Fu bing zhi qian,” 347–48. This seems to have first been established on an ad-hoc basis for campaigns in the time of Xianwen (r. 465–471) (WS 60.1331, 7A.139; see Sagawa, “Bei Wei de bing zhi yu she hui,” 49), but by 493 had become regularized to the point where Xiaowen was giving relief for this tax to specific districts (WS 7B.172).
33. WS 7A.138; and see Sagawa Eiji’s analysis of this 473 edict in “Bei Wei de bing zhi yu she hui,” 50, and on p. 54, his discussion of the “rich rewards” given to families of men of the central armies who fell in battle. Regarding levies from the general military population of Dai, Yao Yanzhong suggests irregular, light taxation: Bei Wei qian qi zheng zhi zhi du, 109. Confirming this, after the move to Luoyang, in 494, an edict offered three-year remission of taxes for “households relocated from Dai” 代迁之户 (WS 7B.176). As for those within the royal domain called up for corvée duties leading to death, who were not guo ren, they had only their funeral bill paid, by the district in which they were registered 诏畿内民从役死事者，郡县为迎丧，给以葬费 (WS 7A.138).
34. See Sagawa Eiji, “Bei Wei de bing zhi yu she hui,” 50, drawing on comments made in 524, in the last years of the regime, when hereditary military service was being dismantled, about how the preexisting heng shang would be continued for those who continued to serve (WS 9.237). In early times, most of these “rewards” would have been booty, but by 524 that had not been the case for generations: see Wang, Zhuang xing qi de Bei Wei cai zheng yan jiu, 108–9.
35. Zhang Min 张敏, “Shi liu guo Bei Wei jun dui dong zhuang bao zhang ji qi dui zhan zheng zhi ying xiang” 十六国北魏军队冬装保障及其对战争之影响, Xuchang xue yuan xue bao (2003.4): 51–53. At the Luoyang Museum are two late Wei statuettes (excavated from Beichen Village 北陈村, Luoyang, in 1989) showing men in heavy overcoats, perhaps examples of this winter apparel.
36. Yan, Bei Wei qian qi zheng zhi zhi du, 164. He, “Fu bing zhi qian,” 349, quotes WS 44.996, stating that at least troops based in outposts of garrisons or provinces supplied their own food and clothing, so that they “frequently suffered from hunger and cold” 州镇戍兵，资绢自随，不入公库，任其私用，常苦饥寒. This became at times insupportable, as we see with a 467 edict ordering that fabric (the main form of money in Wei) be given to the poor in the Six Garrisons: WS 6.128.
37. Kwa and Idema, Mulan, 1; Yan, Bei Wei qian qi zheng zhi zhi du, 164, supports this association of the Mulan tale with actual Wei practice.
38. See Tang Zhangru 唐长孺, Wei Jin Nan bei chao Sui Tang shi san lun: Zhongguo feng jian she hui de xing cheng he qian qi de bian hua 魏晋南北朝隋唐史三论: 中国封建社会的形成和前期的变化 (Wuchang: Wuhan da xue chu ban she, 1993), 192, drawing on description of early recruits to the northern garrisons given in BS 16.617 (WS 18.429), that they were free of tax and could take office, which he suggests applies to the guard units as well.
39. WS 58.1287. In the same account, we are told of tigin, “sons of the [imperial] clan” (zong zi 宗子), posted to garrisons in provincial cities in the Chinese interior, who had 800 households to supply labor for their paddy fields.
40. Wang, Zhuan xing qi de Bei Wei cai zheng, 66–67. These are, of course, very tentative figures.
41. For Taiwu, see WS 5.123. For Xiaowen, WS 7B.177, 31.743; and ZZTJ 141.4430. In the time of the latter, in fact, demands from the military budget forced the monarch to reduce the stipends given to peers (WS 7B.176); and made the Prince of Pengcheng feel a need to dip into his own fortune to help bail out the state, giving back a year’s worth of “the stipend of his fief, his salary as an official, and the ‘charity’ given him as a kinsman” of the imperial clan 国秩,职俸, 亲恤: WS 21B.574; ZZTJ 141.4429.
42. WS 24.609; He, “Fu bing zhi qian,” 319.
43. As has been noted, the stirrup seems to have had a more far-reaching import in Europe than in East Asia: see Dien, “The Stirrup”; Lynn White, Medieval Technology and Social Change (Oxford: Clarendon Press, 1964). For general discussion of the stirrup in East Asia, see also Dien, Six Dynasties Civilization, 334–36.
44. Yates, “The Horse in Early Chinese Military History,” 62–63.
45. Dien, “The Stirrup,” 37.
46. See Graff, Medieval Chinese Warfare, 42–43.
47. See note 7; and discussion of the Anak Tomb No. 3, dated 357, built for a man who fled the Murong into Koguryǒ, in the region that is now northern North Korea: So Tetsu (Su Zhe) 苏哲, “Goko Jūrokkoku Hokuchō jidai no shukkōto to roboyō” 五胡十六国·北朝 时代の出行图と卤簿俑, in Higashi Ajia to Nihon no kōkogaku, ed. Gotō Tadashi and Mogi Masahiro, 4 vols. (Tokyo: Dōseisha, 2002): 2: 115–120.
48. https://commons.wikimedia.org/wiki/File:Story_of_the_Five_Hundred_Robbers_(535%E2%80%93557_CE),_Mogao_Cave_285,_Dunhuang,_China.jpg, accessed 26 August 2019; and the same image in Roderick Whitfield, Susan Whitfield, and Neville Agnew, Cave Temples of Mogao: Art and History on the Silk Road (Los Angeles: The Getty Conservation Institute and the J. Paul Getty Museum, 2000), 19. For discussion of Cave 285, with inscriptions dating it to 538, 539, in the Western Wei period (535–557), see Ma De 马德, Dunhuang Mogao ku shi yan jiu 敦煌莫高窟史研究 (Lanzhou: Gansu jiao yu chu ban she, 1996), 67–69.
49. Zhang, Jin wei wu guan, 2: 667–68. An early example would be a fellow referred to in Wei shu as Mo Ti 莫题, a man of Dai, who in the very earliest years of the regime under Daowu was a “banner commander, who commanded guard troops 禁兵”: WS 28.683. For the apparently Serbi group from which he came, see Yao, Bei chao Hu xing kao, 122–24. In terms of the decimal units: needless to say, as with the units of every army, the theoretical number rarely matched the number on the ground.
50. Zhang, Jin wei wu guan, 2: 667–68; WS 103.2291.
51. David Sneath, “Introduction,” in Imperial Statecraft: Political Forms and Techniques of Governance in Inner Asia, Sixth-Twentieth Centuries, ed. David Sneath (Bellingham: Center for East Asian Studies, Western Washington University and Mongolia and Inner Asia Studies Unit, University of Cambridge, 2006), 9–10; Di Cosmo, “Ethnogenesis,” 47. Di Cosmo suggests that the ultimate origin may have been the Persian Achaemenids.
52. For the Mongols’ units of 1,000s and 10,000s, see Rachewiltz, Secret History of the Mongols, 1: 133–34, nos. 202–3, and 2: 762–63.
53. Zhang, Jin wei wu guan, 2: 666. It will be noted, however, that the Rouran also defined their 100-man units by a flag, in a way similar to that of the Taghbach: “1,000 men were a regiment, . . . [and] 100 men a pennon; [for each] pennon one man was established as leader” 千人为军，. . . 百人为幢，幢置帅一人: WS 103.2290.
