1. Li Wenxin, "Liaoyang," p. 39, fig. 31. See also Fairbank and Kitano, "Han Mural Paintings."

2. LDMHJ, vol. 1, pp. 73-75.

3. For example, Sanguo zhi (Wei shu), juan 16, pp. 512-13; ( Shu shu ), juan 35 PP. 928-29, juan 42, pp. 1033; ( Wu shu ) , juan 54, pp. 1285, juan 57, pp. 1334.

4. "Jingfudian fu," Wen xuan, juan 1, pp. 240-41.

5. LDMHJ, vols. 21, p. 83; 2.2, p. 72.

6. Fan Wenlan, Zhongguo tongshi, vol. 2, chap. 3; Rafe DeCrespigny, The Last of the Han, translates chaps. 58-68 of Sima Guang's Zizhi tongjian. For an excellent summary of events, see Chen, Hsün Yüeh, esp. chaps. I and 2.

7. Biography of Zhuge Liang, Sanguo zhi (Shu shu), juan 35, pp. 911-37. See also Chi Li, "Changing Concept of the Recluse in Chinese Literature," pp. 239-40.

8. The claim is tenuous but a good sign that the clan's more immediate ancestry was less than illustrious. For this, and its local status, see Rafe DeCrespigny, The Biography of Sun Chien, p. 55, n. 2.

9. It is doubtful, for example, that he controlled the marches of Gansu; control of the territories to the northeast—Liaoyang, Lelang, etc.—was most certainly a sometime thing.

10. For a recent study of the capital, see Paul Steven Levine, "The Development of the Medieval City in South China: Chien K'ang from the Second to the Sixth Centuries A.D. "

11. Chen, Hsün Yüeh, pp. 22-23.

12. I use the poor abused word ideology in its OED definition: a system of ideas concerning phenomena, especially those of social life.

13. For the dilemma of one Confucian, see Chen, Hsün Yüeh.

14. Confucianism survived, of course, "in the mystique of imperial rule, and . . . as the heritage of culture in the new elite circles, a cultural continuity sustained by classical education, family traditions, and the clan mores of the great aristocratic households" (ibid., p. 164).

15. For a general account of political philosophy of the period, see Kungchuan Hsiao, A History of Chinese Political Thought, vol. 1, chap. 11.

16. R. P. Kramers, translator, K'ung Tsu Chia Yu: The School Sayings of Confucius.

17. For the biography of Wang Bi, see Sanguo zhi (Wei shu), juan 28, pp. 795-96. For recent translations of his Dao dejing, see Paul J. Lin, A Translation of Lao Tz'u's "Tao Te Ching" and Wang Pi's "Commentary"; Ariane Rump and Wing-tsit Chan, Commentary on the "Lao Tzu" by Wang Pi. For interpretations of Wang Bi's commentaries as attempts to reconcile the philosophies of Confucianism and Daoism, see Yung-t'ung T'ang, "Wang Pi's New Interpretation of the I-ching and Lun-yü"; Arthur F. Wright, "Review of A. A. Petrov, Wang Pi (226-249): His Place in the History of Chinese Philosophy."

18. Wing-tsit Chan has noted that ziran appears as a term in the Laozi text five times, whereas it occurs twenty-five times in Wang Bi's commentary (Rump, Wang Pi, p. xvii). break

19. "Jingfudian fu," Wen xuan, juan 1, pp. 235-36. Translated in Hsiao, Chinese Political Thought, vol. 1, p. 618, n. 42. See also Richard B. Mather, "The Controversy over Conformity and Naturalness during the Six Dynasties," p. 164.

20. "Whatever policies it adopted would spring spontaneously in response to what each situation called for. They would come forth from the Nonactual ( wu ), the substratum of all actual events ( yu ) [ you ]" (Mather, "Controversy," p. 164). And Guo Xiang's commentary: "One must not fail to look carefully into the term non-action. The one who wields the empire . . . thereby undertakes the action of wielding. . . . Such action, however, is that of (the ruler's) own fulfillment for he fully accords with the natures of things, and for that reason it is called non-action" (Hsiao, Chinese Political Philosophy, vol. 1, p. 616). Legitimacy thus requires no longer the Mandate of Heaven but the Concordance with Nature. For wu/you, see A. C. Graham, "'Being' in Western Philosophy Compared with Shih/Fei and Yu/Wu in Chinese Philosophy."

21. For jiupin studies, see Donald Holzman, "Les Débuts du système médiéval de choix et de classement des fonctionnaires: Les Neufs Catégories et l'Impartial et Juste"; Miyakawa Hisayuki, "Chusei * seido no kenkyu * " (in Rikucho * shi kenkyu: seiji shakai hen, chap. 4); Miyazaki Ichisada, Kyuhin * kanjin ho no kenkyu: kakyo zenshi; Tang Changru, "Jiupin zhongzheng zhidu shishi." Of considerable interest for this study are the remarks of Yoshio Kawakatsu, "L'Aristocratie et la société féodale au début des Six Dynasties."

22. With the dislocation brought on by political and military upheavals, it was impossible to utilize the Han principle of relying on officials who were native to the district and who could therefore be expected to know all the local families. For the high rank of the examiners, see Holzman, "Débuts du système médiéval," pp. 406-7.

23. E.g., the complaint of Liu Yi (d. 285), Jin shu, juan 45, p. 1276. For "the old days," see Holzman, "Débuts du système médiéval," p. 391; Tang Changru, "Jiupin," pp. 89-91. For the historical significance of the new system, see Holzman, "Débuts du système médiéval," p. 414.

24. Sanguo zhi (Wei shu), juan 1, p. 32. See also Tang Changru, "Jiupin," pp. 96-97.

25. "Men of superior intelligence, outstanding talent / Supported the Mandate for the imperial house" (Zuo Si, Wei Capital Rhapsody, translated by David Knechtges, Wen xuan, or Selections of Refined Literature, vol. 1, p. 465).

26. Tang Changru has noted this ("Jiupin," pp. 97-98).

27. Biography in Sanguo zhi (Wei shu), juan 21, pp. 617-29. For a translation of Renwu zhi, see J. K. Shryock, The Study of Human Abilities: The "Jen Wu Chih" of Liu Shao. See also Mou Zongsan, Caixing yu xuanli, pp. 43-66. It is interesting that the treatise was written after Liu Shao's retirement from the Wei court, for his biography states that during the Jingchu period (237-239) the emperor commanded him to draft criteria for the recommendation of officials. It further states that Liu Shao, although submitting a statement of seventy-two items, nevertheless demurred that he was not qualified to carry out the command (pp. 619-20). For other studies of ability known to have been composed in the third century, see Mather, "Controversy," p. 167; Shryock, Human Abilities, p. 17. break

28. Renwu zhi: shang, p. 3a, zhong, pp. 2a, 9a; translated by Shryock, Human Abilities, pp. 98-99, 120, 132.

29. Renwu zhi: zixu, p. 1, xia, p. 6b; translated by Shryock, Human Abilities, pp. 2, 148ff.

30. Guo Xiang's commentary to the Zhuangzi, cited in Hsiao, Chinese Political Thought, vol. 1, p. 612. For a recent study of Guo Xiang's philosophy, see Isabelle Robinet, "Kouo Siang ou le monde comme absolu."

31. Guo Xiang, cited in Hsiao, Chinese Political Thought, vol. 1, p. 612.

32. Mou Zongsan, Caixing, p. 59.

33. Renwu zhi, shang, p. 3a; Shryock, Human Abilities, p. 98.

34. Mou Zongsan, Caixing, pp. 55, 59, 63-64.

35. For its history, see Shryock, Human Abilities, p. 27ff. It is listed in Liu Shao's biography, as well as in later dynastic histories. In the fifth century, one Liu Bing prepared a commentary to the text.

36. "The ability of the man of sublime behavior is manifested by his appearance and bearing, and issues in virtuous actions. It shines forth even when not in use . . . therefore even before he is in power, the people unite in recommending him" ( Renwu zhi, zhong, p. 3b; Shryock, Human Abilities, p. 122).

37. " Having a knack for gimmicks and one-liners is essential to the wouldbe film person. With the intense competition for such crucial stakes, a person usually has only fleeting moments to impress their uniqueness upon the movers and shakers of the industry" (Letter from Kenneth D. Merriman, Merriman Productions, in the Los Angeles Times, March 14, 1987).

38. Tang Changru, "Qingtan yu qingyi" and "Jiupin," esp. pp. 96-97; Holzman, "Début du système médiéval," pp. 388, 401-2; Chen Yinke, "Tao Yuanming zhi sixiang yu qingtan zhi guanxi," pp. 180-81.

39. Shishuo xinyu [jiaojian]; see also Richard B. Mather, Shih-shuo Hsinyü: A New Account of Tales of the World, hereafter cited as SSXY, followed by juan and anecdote numbers (the reference applies to both the Chinese and the English editions). Unless otherwise specified, all translations are from the Mather edition.

40. For xuan xue, see, for example, Mou Zongsan, Caixing, pp. 67-99; E. Zürcher, Buddhist Conquest, vol. 1, pp. 86-87.

41. SSXY 4.18. The same tale is told of other members of the Wang family, including Yan's cousin, Wang Rong, and of other members of the Ruan family. See Mather, SSXY, p. 101; also, Chen Yinke, "Tao Yuanming," p. 181. Ruan Xiu was Ruan Ji's nephew. For the use of this example as evidence of late Western Jin attempts to harmonize the two schools, see Chen Yinke, "Tao Yuanming," esp. pp. 181-88. I agree with Chen that the weight of the response is toward the affirmative.

42. For the use of such aid-words and their stylistic import, see Kojiro * Yoshikawa, "The Shih-shuo hsin-yü and Six Dynasties Prose Style," esp. pp. 137-38.

43. Sanguo zhi (Wei shu), juan 27, p. 748. The emperor's full list of requirements makes one wonder if any such paragons were ever located.

44. Note here the changing connotation of "pure." In Latter Han, qing was associated with morality, as in the case of those who shunned political activity on moral grounds. In that sense, anyone with moral scruples, regard- soft

      less of rank or station., could be "pure." Indeed, one's "purity" was a means to social mobility. Here, however, the emperor links "purity" to other qualities—literary distinction, refinement, cultivation—that are not evidence of morality. In spite of Ming Di's assertion, these latter qualities were class-bound and would become more so in the future.

45. Biographical information in Sanguo zhi (Wei shu), juan 9, p. 292; for sharpness of mind, ibid., juan 29, p. 819. For discussions of his political philosophy, see, e.g., Hsiao, Chinese Political Thought, vol. 1, pp. 607-19; Mather, "Controversy," pp. 163-65.

46. Wei lue (third century?), in Sanguo zhi (Wei shu), juan 9, p. 292. For the use of cosmetics and perfumes in this period, see Wang Yao, "Wenren yu yao," pp. 21ff; Joseph Needham, Science and Civilization in China, vol. 5.3, p. 126.

47. Sanguo zhi (Wei shu), juan 9, p. 292. Since He Yan is purported to have been a natural son of Cao Cao's, Wen Di's epithet may have other connotations as well.

48. Ibid., p. 283.

47. Sanguo zhi (Wei shu), juan 9, p. 292. Since He Yan is purported to have been a natural son of Cao Cao's, Wen Di's epithet may have other connotations as well.

48. Ibid., p. 283.

49. Donald Holzman, "Literary Criticism in China in the Early Third Century A.D. ," p. 122.

50. "Dianlun lunwen," Wen xuan, juan 5, p. 22. Translated by James J. Y. Liu, Chinese Theories of Literature, p. 12. The entire essay is translated by Holzman, "Literary Criticism," pp. 128-31. For other translations and discussions, see John Timothy Wixted, "The Nature of Evaluation in the Shihp'in (Gradings of Poets) by Chung Hung ( A.D. 469-518)," p. 251, n. 19.

51. Wen xuan, juan 4, p. 140; Sanguo zhi (Commentary, Wei lue ), juan 21, p. 608. Translated by Holzman, "Literary Criticism," p. 123.

52. Mather has demonstrated this ("Controversy").

53. R. H. van Gulik discusses the sources and their value for the Seven Sages in his Hsi K'ang and His Poetical Essay on the Lute, chap. 2. On the basis of what he considers to be accurate, he has constructed a tentative biography of Xi Kang.

54. These latter remnants, although important, must be used cautiously, for by the time of their writing events had altered the view of the past—so much so, that these later writers do not always agree with one another in their recording of events (for but one example of mythmaking—the recluse summoned by the ruler—and contradiction in the sources cited in Pei Songzhi's commentary to the Sanguo zhi, see the biography of Ruan Ji's father Yu, [ Wei shu ] juan 21, p. 600). Morever, their works have survived only in the commentaries of men who lived much later, in the fifth century. That is to say, they were selected from the original, earlier works in order to make a point, and we cannot know what the later commentators chose to omit (for discussion of omissions, see, e.g., Rafe DeCrespigny, Records of the Three Kingdoms, p. 30). It seems a small matter, but it will be useful when we turn to later traditions.

The extant literary works of the Seven Worthies are collected in Quan Han Sanguo Jin Nanbeichao shi: Quan Sanguo shi; Quan Jin shi and in YKJ.

55. For the works of Xi Kang, see XKJJ; Gulik, Hsi K'ang; Robert G. Henricks, Philosophy and Argumentation in Third-Century China; Donald Holzman, "La Poésie de Ji K'ang." LDMHJ states that he was also skilled continue

      in calligraphy and painting and lists two paintings as still extant (vol. 21, pp. 91-92; vol. 2.2, p. 73).

56. Sanguo zhi (Wei shu), juan 21, p. 605.

57. Ibid.,juan 28, p. 786.

56. Sanguo zhi (Wei shu), juan 21, p. 605.

57. Ibid.,juan 28, p. 786.

58. "Sijiu fu," Wen xuan, juan 2, pp. 81-83. Translated by Burton Watson, Chinese Rhyme-prose, pp. 61-63.

59. Sanguo zhi (Wei shu), juan 20, p. 583 (citing the Xishi pu ).

60. "Yu Shan juyuan juejiaoshu," Wen xuan, juan 4, pp. 156-62. All quotations are from the elegant translation by James Robert Hightower, "Hsi K'ang's Letter to Shan T'ao," pp. 162-66.

61. Wen xuan, juan 4, p. 157; Hightower, "Letter," p. 162.

62. Xi Kang's literary works contradict the last point.

63. Wen xuan, juan 4, p. 158; Hightower, "Letter," p. 163. Note the similarity of the idea to Liu Shao's views of human nature: being lazy by nature, he could not correct the waywardness. There was no place to go but down. See, for similarity to Liu Shao, Henricks's discussion of Xi Kang's essay "On Wisdom and Courage" in Philosophy and Argumentation, p. 126. Also, in the letter: "What is esteemed in human relationships is the just estimate of another's inborn nature, and helping him to realize it . . . you would not want to pervert its heaven-given quality, but rather see that it finds its proper place" ( Wen xuan, juan 4, pp. 160-61; Hightower, "Letter," p. 165).

64. Wen xuan, juan 4, p. 160; Hightower, "Letter," p. 165. The reference is to the Confucian doctrines upheld by the Sima. Elsewhere in the letter, however, he implies admiration for Confucius, and in the "Essay on Kuan and Ts'ai" he praises and defends the Duke of Zhou (Henricks, Philosophy and Argumentation, p. 123).

65. Wen xuan, juan 4, p. 157; Hightower, "Letter," p. 162.

66. Wen xuan, juan 4, pp. 158, 159, 161; Hightower, "Letter," pp. 163, 164, 165.

67. Vervoorn, "Eremitism," pp. 208-9.

68. Wen xuan, juan 4, p. 161; Hightower, "Letter," p. 166. Compare Holzman's translation of the last two lines of Xi Kang's poem "Parti à boire": "Il ne me reste plus qu'a m'adonner tout entier aux cordes de ma cithare / Et confier ainsi mon coeur à mes amis intimes" (Holzman, "La Poésie," p. 172; XKJJ, pp. 72-73.

69. Kenneth DeWoskin, A Song for One or Two, p. 57. DeWoskin's splendid study deals with many aspects of music in the early periods that cannot be discussed here. It is essential reading for those interested in the early art of China. See also DeWoskin, "Early Chinese Music and the Origins of Aesthetic Terminology." For the qin, see also R. H. van Gulik, The Lore of the Chinese Lute and Hsi K'ang.

70. DeWoskin, Song, p. 113.

71. Of interest to art historians is Xi Kang's account of the painted decorations of the qin: "It shows figures of dragons and phoenixes, and of antique worthies: ones sees Po-tzû-ya playing his lute, and Chung-tzû-ch'i listening to him, brilliant and shining in full colors. . . . Ling-lun adjusts the sonorous tubes, T'ien-lien composes his melodies" (Gulik, Hsi K'ang, pp. 88-89).

72. Ibid., pp. 108-9. break

71. Of interest to art historians is Xi Kang's account of the painted decorations of the qin: "It shows figures of dragons and phoenixes, and of antique worthies: ones sees Po-tzû-ya playing his lute, and Chung-tzû-ch'i listening to him, brilliant and shining in full colors. . . . Ling-lun adjusts the sonorous tubes, T'ien-lien composes his melodies" (Gulik, Hsi K'ang, pp. 88-89).

72. Ibid., pp. 108-9. break

73. Wen xuan, juan 2, p. 136. Gulik's translation, which differs slightly from mine, appears in Hsi K'ang, p. 112.

74. Wen xuan, juan 2, p. 137; Gulik, Hsi K'ang, p. 120.

75. SSXY 2.1.

76. Ibid. For the powder, see Needham, Science and Civilization, vol. 3, p. 44; vol. 5.2, pp. 287-89; Yu Jiaxi, "Hanshi san kao"; Lu Xun, "Wei Jin fengduji wenzhang yu yao ji jiu zhi guanxi"; Rudolf G. Wagner, "Lebensstil und Drogen im Chinesischen Mittelalter."

75. SSXY 2.1.

76. Ibid. For the powder, see Needham, Science and Civilization, vol. 3, p. 44; vol. 5.2, pp. 287-89; Yu Jiaxi, "Hanshi san kao"; Lu Xun, "Wei Jin fengduji wenzhang yu yao ji jiu zhi guanxi"; Rudolf G. Wagner, "Lebensstil und Drogen im Chinesischen Mittelalter."

77. Needham, Science and Civilization, vol. 5.2, pp. 287-89.

78. Lu Xun, "Wei Jin fengdu," p. 495.

79. See, for example, Holzman, "La Poésie," pp. 340ff., 347-51.

80. Resonances with other poets of the period can be found in all the poems grouped by Holzman as autobiographical, as Holzman notes (ibid., pp. 344-72).

81. Gulik, Hsi K'ang, p. 96.

82. As both Chen Yinke and Donald Holzman have argued (Chen Yinke, "Tao Yuanming," pp. 182ff; Holzman, "La Poésie," pp. 340ff).

83. Sanguo zhi (Wei shu), juan 21, p. 604.

84. Biography in Sanguo zhi (Wei shu), juan 21, pp. 599ff. See also Holzman, Poetry and Politics, pp. 3-5. Ruan Yu was one of the deceased friends to whom Cao Pi alludes in his letter.

85. Wen xuan, juan 4, p. 159; Hightower, "Letter," pp. 163-64. For Ruan Ji's association with the Sima clan and his political activities, see Holzman, Poetry and Politics, pp. 14-17.

86. Holzman, Poetry and Politics, pp. 49-50.

87. YKJ: Quan Sanguo wen, juan 53.1-3. For Ruan Ji's reply, see juan 45.2-3.

88. Holzman, Poetry and Politics, p. 83; see also pp. 82-87.

89. " . . . as far as the technique of the thing is concerned you could be a pure Confucian" (ibid., p. 84).

90. Ibid., nos. 21, 58, 38, respectively. For essays by Ruan Ji that appear to be Daoist in orientation, see ibid., chap. 4; chap. 5 for those of a seemingly Confucian persuasion; for the complete translation of Ruan Ji's reply to Fu Yi, see ibid., pp. 85-86.

89. " . . . as far as the technique of the thing is concerned you could be a pure Confucian" (ibid., p. 84).

90. Ibid., nos. 21, 58, 38, respectively. For essays by Ruan Ji that appear to be Daoist in orientation, see ibid., chap. 4; chap. 5 for those of a seemingly Confucian persuasion; for the complete translation of Ruan Ji's reply to Fu Yi, see ibid., pp. 85-86.

91. Many of the numerous critical interpretations of Ruan Ji's life and works throughout the centuries are noted by Holzman. I have never examined any library edition of Ruan Ji criticism that has not been splattered by anonymous vandals with loathsome stars, exclamation marks, and comments proclaiming disagreement with the author's interpretations.

92. Holzman, Poetry and Politics, p. 1.

93. He Qimin, Zhulin qixian, pp. 115-31; Mather, SSXY 4.17, nn. 1 and 4. See also Robinet, "Kouo Siang," for the conclusion that Xiang Xiu's and Guo Xiang's commentaries were different (p. 73, n. 1).

94. For Xi Kang's original essay and the two replies, see XKJJ, pp. 143-60, 161-67, 168-95, respectively; Henricks, Philosophy and Argumentation, pp. 21-30, 31-37, 38-70. respectively.

95. Henricks, Philosophy and Argumentation, p. 24; XKJJ, p. 146.

96. Henricks, Philosophy and Argumentation, pp. 35-37.

97. Jiude song. The text survives, attributed to ZLQXL, in the Commen- soft

        tary to SSXY 4.69 (from which this translation by Richard B. Mather is quoted), and in Wen xuan, juan 4, p. 261.

98. SSXY 4.69; Wen xuan, juan 4, p. 261.

99. Chen Yinke says that Shan Tao was related to the Sima clan, but this is disputed by Holzman (Chen Yinke, "Tao Yuanming," p. 187, citing the Jin shu biography; Holzman, Poetry and Politics, p. 250, n. 33). The Jin shu is a compilation of the Tang dynasty and relies heavily on the same fourth- and fifth-century sources I shall use in the following chapter. Hence I do not draw on it here. For the critical historiographic issues in re: official biographies, Denis Twitchett's "Problems of Chinese Biography" is most pertinent.

100. Following Mather's translation of the following titles: libu shangshu, puye, taizi shaofu, situ. Cited in the Commentary to SSXY 3.5.

101. By Yuan Hong (328-376); cited in SSXY 8.12 (Chinese edition only). Charles O. Hucker translates this title as "Gentleman Cavalier Attendant" and says it was originally an honorific conferred upon favored officials ( Dictionary of Official Titles in Imperial China [Stanford: Stanford University Press, 1985] no. 4837). The Jin shu biography states that Ruan Xian served as the grand warden ( taishou ), first of Wu, then of Shiping District.

102. Jin zhugong zan by Fu Chang. Cited in the Commentary to SSXY 20.1.

103. Yu Yu's Jin shu, cited in the Commentary to SSXY 2.23 (Chinese edition only).

104. Gaoshi zhuan by Huangfu Mi, juan 1, p. 31. It is almost identical with the version in Shuo Yuan, juan 17, p. 170 ( Congshu jicheng ed., vol. 528).

105. Dai Mingyang collects and discusses the purported and frequently reedited fragments in XKJJ, pp. 397-426. The story ofRong Qiqi appears on p. 405. Both Shen Yue's Song shu ( juan 93, p. 2280) and the Commentary to SSXY 9.80 refer to a Gaoshi zhuan by Xi Kang, clearly in circulation in the fifth century.

106. Gulik, Hsi K'ang, p. 82.

107. The exchange of essays on geomancy are translated by Henricks, Philosophy and Argumentation, pp. 144-99. Ruan Kan and the exchange of poems are discussed, and the poems translated, by Holzman, "La Poésie," pp. 161-71.

108. Weishi chunqiu, in Sanguo zhi (Wei shu), juan 21, p. 606.

109. SSXY 4.94.

110. Shuijing zhu, juan 9 p. 301.

111. He Qimin has examined in exhaustive detail the question of their meeting ( Zhulin qixian, pp. 1-15). See also Holzman, "Les Sept Sages," p. 327, and Poetry and Politics, p. 8.

112. For population decrease, see, for example, Sanguo zhi (Wei shu), juan 8, p. 262; juan 16, p. 499.

113. Lien-sheng Yang, "Notes on the Economic History of the Chin Dynasty," pp. 126ff.

114. See, for centers of commerce in the South, Étienne Balazs, "Le Traité économique du Souei-chou, " p. 232, n. 215.

115. Mather, SSXY, p. xv.

116. I have relied extensively for my understanding of the economy of the period on Lien-sheng Yang's "Notes" and Balazs, "Le Traité économi- soft

      que," esp. pp. 135-75. Balazs, "Le Moyen Âge: les six dynasties et les empires barbares," offers a useful summary of the period (in Henri Maspero et Etienne Balazs, Histoire et institutions de la Chine ancienne ).

117. Balazs, "Le Traité économique," p. 136 and n. 60; Yang, "Notes," pp. 128ff. For a recent study of these dependents, see Tang Changru, "Gi Shin Nanbokucho * no kyaku to bukyoku."

118. See, for example, Tang Changru, "Sun Wu Jianguo ji Hanmo Jiangnan de zongbu yu shan Yue," pp. 3-29; Wang Zhongluo, Wei Jin Nanbeichao shi, vol. 1, pp. 105-9; Kawakatsu, "L'Aristocratie et la société féodale."

119. For the development from individuals as shi to clans as shi, see Chen Chi-yun, "Zhongguo zhonggu 'shizu zhengzhi' kaolun zhi yi."

120. In his study of the oligarchy, David Johnson found that every reported case of exemption from corvée was based on a person's official rank or that of his immediate ancestors ( Medieval Chinese Oligarchy, p. 15).

121. See Shigeaki Ochi, "Thoughts on the Understanding of the Han and Six Dynasties," p. 44 (in English).

122. Ochi, "Thoughts," pp. 51, 42. These "hereditary" appointments, although only at the local level of administration, were more advantageous than they seem. A local appointment at a high level qualified the occupant for advancement in the central hierarchy. See Miyazaki, Kyuhin * pp. 125-30.

123. It is perhaps more pertinent to say that the Sun upstarts sealed their military and political alliances by marrying into prominent families—a path to social eminence I shall discuss further (Wang Zhongluo, Wei Jin Nanbeichao shi, vol. 1, p. 106).

124. Fan Wenlan, Zhongguo tongshi, vol. 2, p. 456.

125. Biography in Jin shu, juan 65, pp. 1745-54. Fan Wenlan credits Wang Dao's genius with the successful forging of a new dynasty ( Zhongguo tongshi, vol. 2, pp. 454-62).

126. Settled on these "lodged" lands ( qiaojun, qiaoxian ), the most powerful of the clans ruled them as private kingdoms and often found themselves in conflict with the military and civil governments of their district. (See Yang, "Notes," for private enclosure of mountains, p. 134; J. D. Frodsham, The Murmuring Stream, vol. 1, p. 66). For a study of the conflict between powerful émigrés and state authority, see Yasuda Jiro * , "Shin so * kakumei to Yo * Shu * [Jo * Yo] no kyomin."

127. For a conjectural figure of the number of officials appointed in 318, see Zürcher, Buddhist Conquest, vol. 1, p. 85.

128. Xin Tangshu, juan 199, p. 5677. Note Liu Fang's contrast, in which he states that during the Han dynasty, office was honored; in Jin and Song, family was honored (:bid.). An excellent summary of this material is found in Balazs, "Le Moyen Âge," esp. pp. 101ff. See also Johnson, Medieval Chinese Oligarchy, p. 99, for a survey of Chinese terms for these genealogies.

129. See Johnson, Medieval Chinese Oligarchy, p. 5; Chen Chi-yun, "Shizu zhengzhi."

130. The three are not identical; the possession of one does not automatically provide the other two: hence the difficulty of defining the precise characteristics that qualified a family for "greatness." For important studies of the subject, see Mao Hanguang, Liang Jin Nanbeichao shizu zhengzhi zhi yanjiu; Wang Yitong, Wuchao mendi; Patricia Buckley Ebrey, The Aristocratic Families continue

      of Early Imperial China; Dennis Grafflin, "The Great Family in Medieval South China"; Johnson, Medieval Chinese Oligarchy.

131. Jin shu, juan 99, p. 2597; for the genealogy, see Jin shu, juan 74, p. 1939 (Biography of Huan Yi).

132. For the claim: "Ming zi" (Charge to My Son), translated by A. R. Davis, T'ao Yüan-ming ( A.D.   365-427), vol. 1, pp. 26-27; Chinese text, vol. 2, pp. 22-23. For the "fact": Chen Yinke goes so far as to conclude that Tao Kan (259-334) was not even a Han Chinese (Chen Yinke, " Wei shu Sima Rui zhuan Jiangdong minzu tiao shizhengji tuilun," p. 80).

133. Guhuapinlu, p. 29; LDMHJ, vol. 2.1, p. 24; vol. 2.2, p. 65.

134. See, for example, the fifth-century anecdote about Gu Kaizhi and Huan Xuan in the Commentary to SSXY 21.7.

135. Jin shu, juan 99, pp. 2592, 2594; LDMHJ, vol. 1, pp. 116-17; vol. 2.1, p. 49, n. 9.