1. Mencius, 117.

2. There were a few exceptions to the principle of open access, namely certain base groups such as actors and entertainers, Guangdong boat people, Jiangnan beggar communities, and Anhui bond servants who were disqualified from the examinations. Ping-ti Ho 1962, 18-19.

3. Weber 1951, 107-108.

4. Balazs, 6.

5. Weber 1951, 13-20; Balazs, 66-78.

6. Hsiao-t'ung Fei 1953, 74.

7. Eberhard 1971, 71-75. Eberhard is unusual in seeing this much continuity in the Chinese elite. Most scholars see a significant shift from an aristocratic to an examination-based elite beginning in the mid-Tang dynasty (618-906); and usually the term "gentry" is reserved for the late imperial elite of the Ming and Qing dynasties.

8. T'ung-tsu Ch'ü, 2. This figure actually includes 1,282 counties ( xian ) and 154 departments ( zhou ). Because departments were virtually undistinguishable from counties in size and administrative level, we will normally refer only to counties. "County elites" means "county or department elites."

9. This figure was supplied by Robert Forster at the Banff conference. For a general picture of the French state under the ancien régime, see Mousnier, esp. vol. 2.

10. Wittfogel (1953), on the contrary, advanced the idea of a supremely strong Chinese state in his theory of "Oriental despotism" built on Marx's notion of an "Asiatic mode of production." Wittfogel's state based its power on control over waterworks and dominated a fragmented peasant society. Eberhard (1970) directly challenged this exaggeration of state power, pointing out that local leaders took responsibility for much of the water control.

11. T'ung-tsu Ch'ü, 168.

12. Chung-li Chang 1955, 71-141. The distinctions between upper gentry, lower gentry, and commoners have been debated by Chang, Ping-ti Ho and T'ung-tsu Ch'ü. For a summary of views on this issue and of the Chinese terms translated as gentry, see Min Tu-ki, 22-32.

13. Ping-ti Ho 1962, 52.

14. Marsh, 187-188.

15. Huang Ming tiaofa shilei zuan (Categorized substatutes and regulations of the Ming dynasty), cited in Oyama Masaaki, 130-131. On gentry landownership, see also the articles by Tanaka Masatoshi, Tsurumi Naohiro, and Shigeta Atsushi in Grove and Daniels. Mori Masao (1975-76) provides a quite comprehensive introduction to this literature. For more recent English summaries, see Mori Masao 1980; Grove and Esherick. Skocpol (p. 49) observes that competition between imperial states and landed elites to control peasant labor and surpluses from agriculture and trade occurred in Russia and France as well as China.

16. Shigeta Atsushi, 335-386. Shigeta treats essentially the same problem as English historians of the eighteenth century: how did the ruling gentry class maintain its dominance as a wage economy eroded the personal dependence inherent in more paternalistic relations between masters and manual laborers? (See E. P. Thompson 1974). We shall return to this problem in the concluding chapter.

17. Mori Masao 1980, 35-37, 47; Shigeta Atsushi, 337, 351.

18. Hymes 1986a.

19. Kuhn 1970, quotes from 213, 223.

20. Wakeman 1975a, 4, 8 and passim.

21. Mann 1987.

22. For a classic example, see Kung-chuan Hsiao.

23. Lapidus, 42.

24. Yung-teh Chow, 158-172, 220-225. Cf. Hsiao-t'ung Fei 1953; Hsiao-t'ung Fei and Chang Chih-i.

25. Beattie 1979a, 4.

26. Meskill 1979.

27. Schoppa 1982.

28. Rankin 1986.

29. Among the many collections of articles on capitalist sprouts are Nanjing daxue lishixi, ed. 1981 and 1983. More recent articles stressing the limits of capitalist development and a consolidation of the merchant and gentry classes include Huang Qichen and Ye Xian'en 1987. On some socioeconomic effects of Southeast Asian trade on Fujian and Guangdong, see Lin Xiangrui and Luo Yixing.

30. Naquin and Rawski; Ping-ti Ho 1954.

31. Gates, 241-281.

32. See Weber 1958, 180-195.

33. E. P. Thompson's (1978, 146-150) comments on class in eighteenth-century England would seem to, apply very well to China.

34. Swartz, 6, 8-10. The concept of "arena" is sometimes used interchangeably with "field" in the anthropological literature.

35. Bourdieu 1977, 171-183.

36. For an introduction to recent thinking on these matters in social anthropology, see Ortner, esp. 130, 145; and Vincent 1978 and 1986.

37. Kuhn 1970, 64-87, 102-104, chap. 4.

38. See Skinner 1977c. In a revision of this original scheme, Skinner (1985, 273) split Jiangxi from the Middle Yangzi to form a macroregion in its own right.

39. Detailed accounts of eighteenth-century society in each of Skinner's macro-regions appear in Naquin and Rawski, 138-216. For a theoretical discussion distinguishing economic cores and peripheries, see Schoppa 1982, chap. 1.

40. E.g., 23.6 percent of all jinshi came from the two Lower Yangzi provinces of Zhejiang and Jiangsu during the Ming and Qing. Ping-ti Ho 1962, 227-228, 246-247.

41 Gu Yanwu, Rizhi lu (1695), 10: 15a, cited in Oyama Masaaki, 103. Tenancy rates were lower in other parts of the Lower Yangzi, but still usually well above 50 percent.

42. For a picture of the Jiangnan elite of the Ming, see Wakeman 1985, 92-126. For the Qing, see Naquin and Rawski, 55-72, 147-158.

43. Wakeman 1985, 99n; Naquin and Rawski, 151; Ebrey 1983 and (for an earlier period) Walton.

44. Rankin 1986, 2-8, 45-46, 61-62.

45. Ibid., 3, 92-135; Mann, 94-120.

44. Rankin 1986, 2-8, 45-46, 61-62.

45. Ibid., 3, 92-135; Mann, 94-120.

46. Rankin 1986, 202-309; Schoppa 1982.

47. The Japanese literature on this process is extensive. See especially Muramatsu Yuji; Suzuki Tomoo. A forthcoming study by Kathryn Bernhardt will soon introduce this subject in English.

48. See Imahori Seiji 1956.

49. The classic works on this area are Freedman 1958 and 1966. On frontier conditions and lineage formation, see Freedman 1966, 162-166; Pasternak, 551-561. On the influence of official Confucian models, see Faure, 142-144, 149-165.

50. Lin Xiangrui, 61-72.

51. Rubie Watson 1985, 90.

52. Freedman 1966, 68-76, 82-85; Faure, 23-26, 111-113, 128-140.

53. Jing Su and Luo Lun, 106-153.

54. E.g., Shan xianzhi , 2:77a-77b.

55. For a discussion of the Shandong gentry, see Esherick 1987, 28-37. Sources for this calculation are indicated in that volume on pp. 347-348, n. 11. Life expectancies for juren are based on Chung-li Chang 1955, 122-125. For a similar calculation for the eighteenth century, see Naquin 1981, 29-32.

56. Philip Huang, 224-233; Esherick 1987, 238, 242.

57. Niida Noboru, et al., eds., 4:506. A communist organizer in Shandong similarly stressed ties of the local elites to the county officials during the war against Japan. Wang Yu-chuan, 87.

58. Perdue 1987, esp. 168-170, 226-227.

59. In twentieth-century Zhejiang local magistrates often pressed innovations in the periphery in contrast to elite leadership in the cores. Schoppa 1982, 102, 132-134, 187.

60. Ibid., 130-131; Yongkang xianzhi , 2:20b, 3:8a-11a.

59. In twentieth-century Zhejiang local magistrates often pressed innovations in the periphery in contrast to elite leadership in the cores. Schoppa 1982, 102, 132-134, 187.

60. Ibid., 130-131; Yongkang xianzhi , 2:20b, 3:8a-11a.

61. For a breakdown of types of frontiers and frontier cycles of development, see Von Glahn, 215-220. See also Rowe 1985, 251-252; and for the eighteenth century, Naquin and Rawski, 199-205, 226-227.