* An earlier version of this article was presented in the symposium “Chinese factions: conflict and consensus” at the American Anthropological Association 1974 Annual Meeting. I am grateful to G. William Skinner for his helpful critique at that time and, for their useful comments on the earlier draft, to Andrew Nathan, Philip Staniford, Michael Pillsbury, and Alan Landers.

1. The most detailed analysis of “face” remains Hu Hsien-chin's “The Chinese concepts of face,” American Anthropologist, Vol. 46 (January–March 1944), pp. 45–65. For discussion of harmony and conflict avoidance see Derk Bodde, “Harmony and conflict in Chinese philosophy,” in Arthur Wright (ed.), Studies in Chinese Thought (Chicago: University of Chicago Press, 1953), pp. 19–80; Mousheng Hsitien Lin, “Confucius on interpersonal relations,” Psychiatry, Vol. 2 (1939), pp. 475–81; Richard Solomon, Mao's Revolution and the Chinese Political Culture (Berkeley: University of California Press, 1971), Parts I, II; and Wen-shing Tseng, “The concept of personality in Confucian thought,” Psychiatry, Vol. 36, No. 2 (May 1973), pp. 191–202.

2. The terms “impression management,” “front/back region,” and “front-/back-stage” come from Erving Goffman, The Presentation of Self in Everyday Life (Garden City: Doubleday, 1959). “Public/closed sphere” is from Harald Eidheim, “When ethnicity is a social stigma,” in F. Barth (ed.), Ethnic Groups and Boundaries (Boston: Little, Brown and Co., 1969). Gerald Berreman expands upon these concepts in “Behind many masks: ethnography and impression management,” in Hindus of the Himalayas: Ethnography and Change (Berkeley: University of California Press, 1972), pp. xvii–lvii.

3. Andrew Nathan, “A factionalism model for CCP politics,” The China Quarterly (CQ), No. 53 (1973). See also Tang Tsou, “Prolegomenon to the study of informal groups in CCP politics,” CQ, No. 65 (March 1976), pp. 98–114; Andrew Nathan, “Andrew J. Nathan replies,” CQ, No. 65 (March 1976), pp. 114–17, and Peking Politics, 1918–1923: Factionalism and the Failure of Constitutionalism (Berkeley: University of California Press, 1975); Philip Bridgham, “Factionalism in the Central Committee,” in John W. Lewis (ed.), Party Leadership and Revolutionary Power in China (Cambridge: Cambridge University Press, 1970), pp. 203–35; Michel Oksenberg, “The pursuit of interest through strategy,” unpublished paper prepared for the Seminar on the Pursuit of Political Interest in China, under the sponsorship of the Joint Committee on Contemporary China, New York City, 27–29 December 1973; Yung Wei, “Elite conflicts in Chinese politics: a comparative note,” Studies in Comparative Communism, Vol. 7 (Spring/Summer 1974), pp. 64–73; and Edwin Winckler, “Political institutions and political outcomes in China: evaluating the evaluators,” unpublished manuscript.

4. Tsou, “Prolegomenon,” pp. 98–99.

5. See, for example, Morton Fried, “Anthropology and the study of politics,” in Sol Tax (ed.), Horizons of Anthropology (Chicago: Aldine, 1964), pp. 181–190; M. Swartz, V. Turner, and A. Tuden (eds), Political Anthropology (Chicago: Aldine, 1966), and Local-Level Politics (Chicago: Aldine, 1968); Edwin Winckler, “Political anthropology,” in B. Siegel (ed.), Biennial Review of Anthropology (Stanford: Stanford University Press, 1969), pp. 301–86; and Robert LeVine, “Anthropology and the study of conflict: an introduction,” Journal of Conflict Resolution, Vol. 5, pp. 3–15.

6. Tsou, “Prolegomenon,” p. 113.

7. This is surveyed in Yung Wei, “A Methodological critique of current studies on Chinese political culture,” Journal of Politics, Vol. 38, No. 1 (February 1976), pp. 114–40. For work utilizing this approach, see Lucian Pye, The Spirit of Chinese Politics: A Psychocultural Study of the Authority Crisis in Political Development (Cambridge: M.I.T. Press, 1968); R. J. Lifton, Revolutionary Immortality: Mao Tse-tung and the Chinese Cultural Revolution (New York: Vintage Books, 1968); Richard Wilson, Learning to Be Chinese: The Political Socialization of Children in Taiwan (Cambridge: M.I.T. Press, 1970); Richard Solomon, Mao's Revolution and the Chinese Political Culture (Berkeley: University of California Press, 1972); Sheldon Appleton, “The Political socialization of college students on Taiwan,” Asian Survey, Vol. 10 (October, 1970), pp. 910–23, and “Regime support among Taiwan high school students,” Asian Survey, Vol. 13 (August 1973), pp. 750–60; and Roberta Martin, “The socialization of children in China and on Taiwan: an analysis of elementary school textbooks,” CQ, Vol. 62 (1975), pp. 242–62. The political culture approach has been adopted for analysis of factional conflict in Alan Liu, Political Culture and Group Conflict in Communist China (Santa Barbara, California: Clio Press, 1976). See also Andrew Nathan, “The analysis of political conflict structures: a foundation for theory,” unpublished paper prepared for the Seminar on the Pursuit of Political Interest in China, under the sponsorship of the Joint Committee on Contemporary China (New York City, 27–29 December 1973).8. See, for example, Margaret Mead and Rhoda Métraux (eds), The Study of Culture at a Distance (Chicago: University of Chicago Press, 1953); and Ruth Bunzel, Explorations in Chinese Culture (New York: Columbia University Research in Contemporary Cultures, unpublished manuscript, 1950). This literature is reviewed in Alex Inkeles and Daniel Levinson, “National character: the study of modal personality and sociocultural systems,” in G. Lindzey and E. Aronson (eds), The Handbook of Social Psychology, 2nd Edit. Vol. 4 (Reading, Mass.: Addison-Wesley, 1969), pp. 418–506. I am indebted to Dr Mead for her guidance in developing the field-work methodology on which my research is based.

8. Wei, “A methodological critique,” p. 117.

9. Ibid. p. 134.

10. Ibid. p. 137.

11. “Face” as used in this article conforms to Hu's definitions of mien-tzu and lien. Hu defines 31 “varieties of meaning,” some concrete but most figurative. “Face” in this article's title corresponds to Hu's second concrete usage, face as in piao-mien (outer appearance). Elsewhere my usage generally corresponds to Hu's figurative varieties of meaning. The body of data presented here is intended to shed light on the nature of face as a behaviour-modifying force but does not purport to articulate new theoretical insights on face per se.

12. By the mid-1930s the government of the Republic of China estimated China's Muslims to number almost 30 million, one-tenth the existing population of China. See Ha Kuo-tung, “Mohammedanism,” in The Chinese Year Book (Shanghai: Commercial Press, 1935–36). Recent Peking sources list 11 “national minorities” among whom Islam has been the dominant tradition and state their total population to be only 10 million. For general discussion of Chinese Muslims see my Cohesion and Cleavage in a Chinese Muslim Minority (Ph.D. dissertation, Columbia University, 1973), and “The Hui of East Asia (China),” in Richard Weekes (ed.), An Ethnographic Survey of the Muslim World (Westport, Connecticut: Greenwood Press, 1978).

13. The same assertion could not be made for Muslims of north-west China who are of Turkic ethnicity and not sinified like the Hui.15. The ROC requires that all non-governmental organizations be registered with the Ministry of the Interior. This entails approval of three items: a constitution, a list of officers and a list of a minimum of 45 members. It is illegal to meet in an unregistered organization. To maintain confidentiality, pseudonyms have been employed here for the Muslim organizations and Muslims still living on Taiwan.

14. F. G. Bailey, Stratagems and Spoils: A Social Anthropology of Politics (New York: Shocken, 1969), p. 145.

15. Ibid. p. 12.

16. Interesting parallels exist between Taiwan's Muslim “big men” and the wealthy Chinese in overseas Chinese communities who accepted Kapitan positions in order to seek in the larger society rewards above and beyond those the communities that supported them could provide. See G. William Skinner, “Overseas Chinese leadership: paradigm for a paradox,” in G. Wijeyewardene (ed.), Leadership and Authority (Singapore: University of Malaya Press, 1968), pp. 191–207.

17. David Jordan reports a similar phenomenon from a Taiwanese village he calls Bao-an. “Politically, Bao-an regards itself as a unitary entity, almost entirely free of factions. (Yet) … one can distinguish two important factions. …” See Gods, Ghosts, and Ancestors: Folk Religion in a Taiwanese Village (Berkeley: University of California Press, 1972), p. 9.

18. K. Wolff (transl.), Conflict (Glencoe, Ill.: Free Press, 1955), p. 17. See also Lewis Coser, The Functions of Social Conflict (Glencoe, Ill.: Free Press, 1956), p. 33; and Robert North, et al., “The integrative functions of conflict,” Journal of Conflict Resolution, Vol. 4, No. 3 (1960), pp. 355–84.

19. Shih Tse-chou (Khalid T. C. Shih), “T'uan-chieh ying-kai chu-i ti t'ai-tu” and “I-ssu-lan chiao pu ying-kai fen p'ai,” in I-ssu-lan chiao-i wu-shih chiang (Fifty Talks on Islamic Doctrine) (Taipei: Taipei Mosque, 1951), pp. 91–93, 137–41.

20. See Derk Bodde's excellent analyses, Japanese Infiltration among Muslims in China (Washington D.C.: Office of Strategic Services Research and Analysis Branch, 1944, No. 890.1), and “Japan and the Muslims of China,” Far Eastern Survey, Vol. 15, No. 20 (9 October 1946), pp. 311–13.

21. A noted example is Ta P'u-sheng who was active in Kuomintang (KMT) Muslim activities before rising to high positions in religious and minorities work in the People's Republic of China (PRC). As in Taiwan, PRC Muslims have played important “international arena” roles in cultivating friendship with Islamic countries. See, for example, the China Islamic Association, Chinese Moslems in Progress (Peking: Nationalities Publishing House, 1957). Ta's career is outlined in Donald Klein and Anne Clarke, Biographic Dictionary of Chinese Communism, 1921–1965 (Cambridge: Harvard University Press, 1971), Vol. 2, pp. 794–96.Certain Tapei Muslims recall some of their leaders “waiting it out” in Hong Kong in 1950 to see “which way the winds would blow” before deciding whether to go to Taiwan, return to the mainland, or depart for the west. Taipei Muslims who claim formerly close relationships with people like Ta P'u-sheng or Pai Shou-i (prominent in Peking's minorities work) say they still greatly admire those men but think it a pity they “took the Communist road.”

22. The word ahung derives from a Persian root meaning “to instruct.” To become an ahung a Chinese male spent as many as 13 years studying Islamic law and theology, Arabic and Persian.

23. His family claims its prominence can be traced back to an eminent ancestor who was one of several Muslim military leaders that assisted Chu Yüan-chang in establishing the Ming dynasty. Many Muslims claim, in fact, that Chu himself was a Muslim but had to conceal this in his strategy to become emperor.

24. Ma had been to the Middle East several times and had enjoyed much prestige as an important ahung and director of the former Islamic Cheng Ta Normal School in Peking. Taipei Muslims say he understandably felt the Taiwan position beneath his status and so left to serve the Muslims in Hong Kong. From there he was “persuaded” to return to Peking, a member of his family allegedly having had close connections with Kuo Mo-jo, president of the Chinese Academy of Science. In the PRC he served as deputy chief of the Chinese Islamic Association and the Hui People's Cultural Association and as deputy superintendent of the Chinese Islamic College. He was purged in 1958 anti-rightist campaigns as a “politically ambitious person of anti-communist inclination parading under the cloak of religion.”

25. For example, mosques maintained schools, provided charity, and owned rent-generating land and buildings.

26. Each mosque on Taiwan has its own board of directors, called tung-shih, and a separate board of directors, called li-shih and chien-shih, for the organizations associated with the mosque. While at each mosque they are formally two separate boards, in fact certain key individuals dominate both.Gallin observed in Hsin Hsing that faction leaders “are usually wealthy men with extensive local contacts and influence, which place them in positions of power from which they are able to dispense political favours. It is not unusual to find in this group some wealthy, educated and prestigious men who were former politically influential village or tsu (lineage) leaders. While many of this group have withdrawn from active political life, others continue to wield important political influence in the area primarily by virtue of their positions in the new factional system.” This description also fits quite well the Muslim faction leaders. See Bernard Gallin, “Political factionalism and its impact on Chinese village social organization in Taiwan,” in Marc Swarz et al. (eds), Local-Level Politics (Chicago: Aldine, 1968), p. 389.

27. F. G. Bailey, “Decisions by consensus in councils and committees: with special reference to village and local government in India,” in M. Banton (ed.), Political Systems and the Distribution of Power (London: Tavistock, 1965), pp. 10–12.

28. “Ideology” is to be understood here in its broader sense, namely, as any belief system linked to and legitimizing the interests of the group that subscribes to it.

29. British anthropologist Joan Vincent cautions that this is by no means uniquely characteristic of Chinese political culture; her field research in Northern Ireland revealed the same attitude there (personal communication). Perhaps the distinction might be expressed more accurately as an American tendency to look for solutions and a European and Chinese (and perhaps general tendency of others) to live with problems.

30. See Solomon, Mao's Revolution, pp. 116–20. Solomon's interviewees estimated it only about 30 per cent probable that co-operation could ever be achieved between peers who had come into open conflict. A comparable sample of Americans, however, thought it more likely than not (55 per cent probable) that co-operative relations would be re-established.

31. Ralph Nicholas, “Factions: a comparative analysis,” in M. Banton (ed.), Political Systems and the Distribution of Power, p. 27. See also Harold Lasswell, “Faction,” in E. Seligman (ed.), Encyclopedia of the Social Sciences (New York: Macmillan, 1931), pp. 49–51.

32. For first-hand observations on factionalism that has persisted over 50 years, see Lawrence Crissman, “Each for his own: Taiwanese political response to KMT local administration,” unpublished paper prepared for the London–Cornell Project for East and Southeast Asian Studies (June 1969), and “The development of local Taiwanese political factions,” unpublished paper prepared for the American Anthropological Association Annual Meeting, Washington D.C., 17–21 November 1976.

33. Not all persons who frequent either mosque or are members of either formal organization are members of the corresponding faction. While a faction does exist within each mosque-cum-organization, most Taipei Muslims find the factional conflict among their leaders uncomfortably discrediting to their collective face and try to avoid direct involvement.

34. Robert Scalapino and Junnosuke Masumi similarly found that Japanese factional coalitions employ the tactic of organizing themselves into “clubs” or “societies.” The dominant coalition is generally known as the “Main Current” group and it is opposed by the “anti-Main Current” group, they explain. Each faction is organized as a club, and each takes a name. See their Parties and Politics in Contemporary Japan (Berkeley: University of California Press, 1962), pp. 59, 8

35. Gallin and Crissman similarly observed that members of central Taiwan factions also refer to themselves by their respective leaders' names. See Gallin, “Political factionalism,” p. 389, and Crissman, “Development of local Taiwanese factions,” p. 5.

36. Certain PRC officials have expressed to me similar scorn towards past strategies of KMT opponents. We have seen that critics of Ma's faction accuse Ma of chicanery in setting up a sham structure to secure a claim to legitimate power (“T'a kao-le nei-ke chia-ti hui …”). In the same vein, certain PRC officials have criticized the KMT of sham tactics in holding nationwide elections and establishing the ROC National Assembly upon which to base its claim to being the sole legitimate government of China (“T'a-men kao-le nei-ke chia-ti hui-i …”).

37. See Nicholas, “Factions,” p. 29.40. Bailey, Stratagems and Spoils, p. 52.

38. Nor should this be interpreted as precluding a basic ideology shared by both competing factions. As Nathan emphasized, “factions operate within a broad ideological consensus while exaggerating the small differences that remain among them.” See Nathan, “A factionalism model,” p. 47.

39. Bailey, Stratagems and Spoils, p. 52. Gallin also discovered that factions “do not seem to have any dedication to special principles or objectives, but … have always been organized around particular personalities.” Political factions, he concludes, “are not founded on any particular set of ideals or objectives other than the control of the leadership of the area; in this sense, there is no real distinction between them. Their membership is recruited primarily on the basis of interpersonal relations …” (my emphasis). Gallin, “Political factionalism,” pp. 384, 397.

40. See Nicholas, “Factions,” pp. 28, 45.

41. Nicholas, “Factions,” pp. 26–27.

42. Andrew Nathan, “‘Connections’ in Chinese politics: political recruitment and Kuan-hsi in late Ch'ing and early Republican China,” unpublished paper prepared for the American Historical Association Annual Meeting, New Orleans, 27–30 December 1972. See also Bruce Jacobs, Local Politics in Rural Taiwan: A Field Study of Kuan-hsi, Face, and Faction in Matsu Township (Ph.D. dissertation, Columbia University, 1975).

43. Nathan, “Connections,” p. 18.

44. Faction recruitment in Hsin Hsing was based, according to Gallin, on such things as allegiance to an employer, past favours rendered by faction leaders, promises of future aid, particular local incidents, and ch'in-ch'i as well as lineage relationships. See Gallin, “Political factionalism,” p. 397. See also John Young, “The terminology and structure of interpersonal relations,” in Business and Sentiment in a Chinese Market Town (Taipei: Orient Cultural Service, 1974); and Bernard Gallin and Rita Gallin, “Sociopolitical power and sworn brother groups in Chinese society: a Taiwanese case,” in R. Fogelson and R. Adams (eds), The Anthropology of Power (New York: Academic Press, 1977).

45. Nathan, “A factionalism model,” p. 49.

46. The criterion of disruption can be used to distinguish two kinds of dispute: (1) normal dispute which, while members of the group may regard it as detrimental, remains relatively non-disruptive; and (2) factionalist dispute, which involves major organizational changes. The latter can be further distinguished as pervasive factionalism or schismatic factionalism. See Alan Beals and Bernard Siegel, Divisiveness and Social Conflict: An Anthropological Approach (Stanford: Stanford University Press, 1966), pp. 20–25.

47. For theoretical perspectives, see Max Gluckman, “Gossip and scandal,” Current Anthropology, Vol. 4, No. 3 (June 1963), pp. 307–16; F. G. Bailey, Gifts and Poison: The Politics of Reputation (New York: Schocken Books, 1971); and Tamotsu Shibutani, Improvised News: A Sociological Study of Rumor (Indianapolis: Bobbs-Merrill, 1966).

48. Solomon also found in his Taiwan research that there “tends to be meager interaction between various factions out of fear of provoking conflict.” See Solomon, Mao's Revolution, p. 149.

49. For discussion of imperial, ROC and PRC government involvement in Muslim affairs, see my “Pig and policy: maintenance of boundaries between Han and Muslim Chinese,” in B. E. Griessman (ed.), Minorities: A Text with Readings in Intergroup Relations (Hinsdale, Illinois: Dryden Press, 1975), pp. 136–45.

50. See Pillsbury, Cohesion and Cleavage, pp. 144–49, and “No pigs for the ancestors: pigs, mothers, and filial piety among ‘Taiwanese Muslims’,” prepared for the Conference on Chinese Religion (University of California at Riverside, 20 April 1974).

51. The mosque was built in 1960 in response to the international political situation. With religious expression curtailed on the mainland, the ROC was eager to prove itself the protector of religious freedom for the Chinese people. Partly by demonstrating how well Muslims were treated on Taiwan, it succeeded in winning the favour of numerous Arab and other countries with Muslim populations, many of which initially opposed Communism on the grounds that Communism opposed religion. Given Taiwan's successful courting of these Muslim countries, both the government and many Muslims felt increasingly embarrassed when the time came for foreign dignitaries to visit the East Harmony Road mosque, actually only a small converted Japanese-style house. Pai Ch'ung-hsi and Foreign Minister George Yeh are credited for negotiating the large Arabic-style mosque to which visiting Muslim heads of state could be proudly taken. Construction was financed by local Muslims, overseas Chinese Muslims, King Hussein of Jordan, King Faisal of Saudi Arabia, Shah Pahlevi of Iran, the government of Iraq, and the KMT.

52. Instructive comparisons can be made to the entertaining of officials by clan temple associations in Taipei. See Morton Fried, “Some political aspects of clanship in a modern Chinese city,” in M. Swartz et al. (eds), Political Anthropology (Chicago: Aldine, 1966), pp. 285–300.

53. Since 1955, Muslim pilgrimage and good-will delegations have also gone from the PRC to Mecca and Islamic centres en route. Peking began sending delegations as early as 1952, but the first three were “stopped on their way by imperialist interference.” See New China News Agency, 11 December 1956.

54. Many of the Chungli Muslims are retired KMT soldiers from Yunnan who came to Taiwan via Burma and Thailand where some of their families still live.58. The desire of Chinese “common people” to avoid involvement with the central government and supra-local authorities is well documented. On the Chinese reluctance to even talk about these authorities, or to discuss politics in general, see Solomon, Mao's Revolution, pp. 141–50; and Sheldon Appleton, “Silent students and the future of Taiwan,” Pacific Affairs, Vol. 43, No. 2, pp. 227–39.

55. To see this relationship called into play when a major Taiwan corporation threatened to evict the Kaohsiung Muslims from their mosque, see Chung-kuo Hui-chiao Hsieh-hui: “Chiao-wu yü Hui-wu” (“Religious matters and association matters”), Chung-kuo Hui-chiao (The Islam in China), Vol. 139 (16 July 1970), pp. 1–2; “Ho Hui-hsieh Ts'ai-t'uan Fa-jen Tung-shih-hui Ch'eng-li” (“Congratulations on the establishment of the board of directors of the juridical person of the Chinese Muslim Association”), Chung-kuo Hui-chiao (The Islam in China), Vol. 139 (16 July 1970), p. 2; and “Lun Tai-ts'ai yü Yüan-wang” (“On malicious speculation and false accusations”), Chung-kuo Hui-chiao (The Islam in China), Vol. 140 (16 September 1970), p. 1. Credit for preventing the eviction is given to two Muslims who are ROC National Assemblymen and utilized that connection to appeal for mercy to the corporation's two top executives, also National Assemblymen. The negotiations are summarized in Pillsbury, Cohesion and Cleavage, pp. 170–71.

56. See Raphael Israeli, Muslims in China: A Study in Cultural Confrontation (London: Curzon Press, 1976); and, for an account by an eminent PRC Muslim scholar, Pai Shou-i, Hui-min Ch'i-i (The Rebellions of the Hui People), 4 vols. (Shanghai: Shen-chou Kuo-kuang She, 1952). As recently as 1958, Muslims engaged in factional conflict in north-west China were still being attacked for New Sect activities. See my forthcoming chapter, “Ningsia,” in Edwin Winckler (ed.), A Provincial Handbook of China (Stanford: Stanford University Press, forthcoming).

57. This practice was particularly entrenched in north-west China where it came under fierce attack in the early 1950s when the ahungs were branded as feudal elitist exploiters living off the masses. See Pai Shou-i, Hui-hui Min-tsu ti Hsin Sheng (A New Life for the Hui-hui Nationality) (Shanghai: Tung-fang Shu-she, 1951), pp. 85–90.

58. The tendency to label a factional opponent a deviant is discussed in Bailey, Stratagems and Spoils, p. 53. Ma's willingness to “walk the grave” definitely does attract Muslims to his mosque. Traditionally, only unfilial Muslims would neglect their deceased by failing to have Quranic verses recited on their behalf. Commemorating ancestors in this manner appears equally important for Chinese Muslims as maintaining an ancestral altar has been for other Chinese. The form is different but the content seems quite similar. In response to criticism, Ma insists his reciting the Quran this way has nothing to do with ancestor worship or earning money but merely permits Muslim common people to fulfil obligations of filial piety which not only Chinese tradition but also Islam enjoins upon them.Carrying out the duties of Islam means that leaders of both factions pray daily, fast during Ramadan, and have made the pilgrimage to Mecca. They also rigidly abstain from pork. All this is a considerable feat in China and represents a zealousness that Saudi officials resident on Taiwan admit is rare even in the Middle East.

59. For comparative perspective, see David French, “Ambiguity and irrelevancy in factional conflict,” in M. Sherif (ed.), Intergroup Relations and Leadership (New York: Wiley, 1962); and Lewis, “Medicine and politics,” pp. 415–26.

60. This and the other factional stratagems employed on Taiwan appear virtually identical to those used back on the mainland. In Peking stands the famous Oxen Street (“West”) Mosque (to which foreign Muslims in the PRC currently go for prayer); nearby was a smaller “East Mosque.” The relationship between the two is said to have been characterized for about 20 years by three major contradictions (mao-tun). One was consistent lack of agreement on when to begin and end Ramadan. A second was the Old and New Sects' conflict and the presence of two power groups – the old generation and the “more educated” new generation. See Sun Sheng-wu, “Ts'un-ch'eng-chai Sui-pi: Pei-p'ing Niuchieh Li-pai-ssu Mien-mien Kuan” (“Miscellaneous notes from the ‘sincerity preservation den’: a comprehensive view of Peiping's Oxen Street Mosque”), Chung-kuo Hui-chiao (The Chinese Islam), Vol. 137 (16 March 1970), p. 16, and Vol. 138 (16 May 1970), p. 15. Muslims from Shantung report their communities were also characterized by the same “contradictions.”

61. See, for example, Solon Kimball and Marion Pearsall, “Event analysis as an approach to community study,” Social Forces, Vol. 34, No. 1, pp. 58–63; and Harold Garfinkel, “Common sense knowledge of social structures: the documentary method of interpretation,” in J. Scher (ed.), Theories of the Mind (New York: Free Press, 1962), pp. 689–712.

62. Clifford Geertz, “Epilogue: on proving paradigms,” in The Social History of an Indonesian Town (Cambridge: M.I.T. Press, 1965), p. 153.

63. Victor Turner, Schism and Continuity in an African Society (Manchester: University Press, 1957), p. 93.

64. For use of this model in analysis of other factionalist episodes, see Ronald Frankenberg, “British community studies: problems of synthesis,” in M. Banton (ed.), The Social Anthropology of Complex Societies, A.S.A. Monograph No. 4 (London: Tavistock, 1966); Max Gluckman, Analysis of a Social Situation in Modern Zululand (Manchester: Rhodes Livingstone Institute, reprint 1958); and Beals and Siegel, Divisiveness and Social Conflict, pp. vii–x.

65. See Ch'en Shao-hsiao (pseud.), Hsi-pei Chun-fa Chi (Record of the Northwest Warlords) (Hong Kong: Chih-ch'eng Ch'u-pan-she, 1969), especially pp. 31–36.

66. Among Chinese, according to Francis Hsu, the length of time the coffin is kept at home indicates the strength of esteem of the living for the dead. Only poor families, he explains, have the coffin removed from the house within three days after death. See Francis L. K. Hsu, Under the Ancestors' Shadow: Kinship, Personality, and Social Mobility in Village China (Garden City, N.J.: Doubleday, 1967), pp. 156–57.

67. Distinctly more so than in the west, postponement of discussion when conflicting opinions have arisen seems modus operandi for the Taiwan Muslim elite councils – and perhaps in all Chinese elite politics. Chinese in general tend to place comparatively great emphasis on careful planning and forethought for achieving control and predictability. “San ts'e erh hsing” advises the Chinese proverb – “Think over three times, then act.” Note also the Confucian precept “In all things success depends on previous preparation, and without such there a sure to be failure.” See Weakland, “The organization of action,” p. 365.

68. See discussion of the use of irrelevant arguments in factional conflict in French, “Ambiguity and irrelevancy,” pp. 415–26.

69. Such councils, Bailey has stated, consider a majority vote much inferior to manimity. Elite councillors, he explains, have a strong incentive to present a front of consensus and keep their ranks closed in the face of their public. See his “Decisions by consensus,” p. 10.

70. Central Daily News, 22 May 1967. Muslims say mediators played important roles in their communities on the mainland – just as in most Chinese communities – in resolving conflict and in keeping conflict that did exist from going beyond the community. Among Taiwan Muslims, the role of ahungs and mosque elders as mediators of conflict is steadily diminishing. The breakdown in mediation among Taiwanese is discussed in Bernard Gallin, “Conflict resolution in changing Chinese society: a Taiwanese study” in Swartz et al., Political Anthropology, pp. 265–74.

71. See Kao Wen-yüan, Hui-chiao ping-li (Islamic Funeral Ritual) (Taipei: I-ssu-lan Hsiao Ts'ung Shu No. 4, 1969). The author is the brother of Kao Hao-jan who was engaged in United Front work among Muslims in the PRC until he left the Chinese mainland in 1957 and came, via Mecca, to Taiwan. (See The Imam's Story, Hong Kong: Green Pagoda Press, 1960.)

72. See Pillsbury, Cohesion and Cleavage, pp. 201–210.

73. See Bailey, Stratagems and Spoils, pp. 132–36.

74. Cf. Bailey, Stratagems and Spoils, p. 139.

75. A Muslim colonel and instructor of military history likened their enmity to that between Mao Tse-tung and Chiang Kai-shek, between whom an outside “umpire,” the United States' General George Marshall in 1946, also unsuccessfully attempted to negotiate reconciliation.

76. Stratagems differ somewhat when factional spoils include electoral office and there is not the same felt need to present a face of harmony. A prime example is the local Taiwanese elections in which, while my research was in progress, nearly all votes were purchased in one way or another.

77. Gallin's study of Taiwanese factionalism similarly concludes that factions existed for no other reason than the control of leadership in the arena. See his “Political factionalism,” p. 397.

78. Likewise, G. William Skinner in his studies of Chinese in Bangkok concluded that the spoils of the contest for leadership were leadership itself, along with its accompanying prestige. A reason why so many hundreds of associations exist is to provide hundreds of leadership positions (personal communication).

79. Here, as in Chinese elite politics, succession has been problematic. For most of the years since Pai's death the Federation has had no chairman but only a non-threatening, partially deaf “old gentleman” as “acting chairman” since the faction members could not decide which of them should occupy this pre-eminent position.

80. There are intriguing parallels in the eruption of factional conflict over funerary issues after the deaths of both Pai and of Chou and Mao in the PRC. In the Taiwan Muslim conflict, factional issues related to funerary details include not only the two discussed above (Pai's funeral and “walking the grave”) but several others as well, such as how to commemorate a deceased leader. This suggests that, in addition to the problematic question of succession, deep-rooted traditional Chinese concerns make potentially volatile any issue related to the death and “after life” of important individuals.

81. This is clearly a factor in contemporary Chinese politics. For discussion of the priority, however, of personality versus intellectual issues in Chinese policy debates, see Michel Oksenberg and Steven Goldstein, “The Chinese political spectrum,” Problems of Communism, Vol. 23, No. 2 (1974), pp. 1–13; Michael Pillsbury, “How ‘useful’ a model?” Problems of Communism, Vol. 24, No. 1 (1975), pp. 72–73; and Oksenberg and Goldstein, “A Rejoinder,” Problems of Communism, Vol. 24, No. 1 (1975), pp. 75–77.