1. This article deals only with the Han Chinese inhabitants of Taiwan. When they began migrating to the island in the 17th century they found a small population of aborigines, whom they gradually absorbed or pushed into the mountains or remote areas of eastern Taiwan. Today the island's approximately 250,000 aborigines are being rapidly sinicized. For a useful history of Taiwan's native inhabitants seeWang, I-shou, “Cultural contact and the migration of Taiwan's Aborigines: a historical perspective,” in Knapp, Ronald G. (ed.), China's Island Frontier: Studies in the Historical Geography of Taiwan (Honolulu: University Press of Hawaii, 1980), pp. 31–43Google Scholar;

2. Chiang, Tao-chang, “Walled cities and towns in Taiwan,” in Knapp, Ronald G. (ed.), China's Island Frontier, pp. 117–42Google Scholar;

3. My discussion of the Qing elite is based in part on Johanna Menzel Meskill's excellent study,A Chinese Pioneer Family: The Lins of Wu-feng, Taiwan, 1729–1895 (Princeton: Princeton University Press, 1979)Google Scholar;

4. This rough estimate is based on Barclay's figures of the urban population in 1920, and extrapolated back to 1905 under the assumption that the growth rate of all five towns between 1905 and 1920 is the same as that of the three towns for which data are available between 1915 and 1920. SeeBarclay, George W., Colonial Development and Population in Taiwan (Princeton: Princeton University Press, 1954Google Scholar; reissued in Port Washington, New York: Kennikat Press, 1972), esp. pp. 13 and 116.

5. Ho, Samuel P. S., Economic Development of Taiwan, 1860–1970 (New Haven: Yale University Press, 1978), pp. 313–14Google Scholar; 319. These figures overstate the share of the Taiwanese population in cities because they include Japanese, who were largely urban.

6. Calculated fromBarclay, , Colonial Development and Population in Taiwan, p. 113Google Scholar;

7. Barclay, , Colonial Development and Population in Taiwan, pp 84–98Google Scholar;

8. Thomas B. Gold, “Colonial origins of Taiwanese capitalism,” in Edwin A. Winckler and Susan M. Greenhalgh (eds.), Authoritarianism and Dependency in East Asia: Comparing Taiwan (forthcoming); andTsurumi, E. Patricia, Japanese Colonial Education in Taiwan, 1895–1945 (Cambridge: Harvard University Press, 1977)CrossRefGoogle Scholar;

9. Based on 1982 year-end population. Housing and Urban Development Department, Council for Economic Planning and Development,Urban and Regional Development Statistics, Republic of China, 1983 (Taipei: Council for Economic Planning and Development, 10 1983)Google Scholar;

10. Tsai, H. H., “Urban growth and the change of spatial structure in Taiwan,” Paper presented at Conference on Urban Growth and Economic Development in the Pacific Region, Institute of Economics, Academia Sinica, Taipei, 9–1101 1984Google Scholar; See also Liu, Paul K. C., “Factors and policies contributing to urbanization and labor mobility in Taiwan,” Industry of Free China (05 1983), pp. 1–20Google Scholar;

11. Tsai shows that between 1960 and 1980 natural increase accounted for 6–10% of the urban population growth, migration accounted for 28–78%, while upgrading of new urban areas accounted for 36–65%. SeeTsai, , “Urban growth and the change of spatial structure in Taiwan”, pp. 9–13Google Scholar;

12. Chang, Ming-cheng, “Economic adjustment of migrants in Taiwan: recent findings and policy implications”, Paper presented at Conference on Urban Growth and Economic Development in the Pacific Region, Institute of Economics, Academia Sinica, Taipei, 9–11 01 1984, p. 6Google Scholar;

13. Ministry of Interior, 1982 Taiwan-Fukien Demographic Fact Book, Republic of China (Taipei: Ministry of Interior, 12 1983), pp. 966–83Google Scholar; These figures are based on the household registers, and thus are likely to understate the amount of migration.

14. From the vantage point of China, where close interdependence between city and countryside has historically been the norm, general theories of urbanism as a distinct form of life appear as ethnocentric constructs that have little bearing on the Chinese experience. A prime example of such theories isWirth, Louis, “Urbanism as a way of life”, American Journal of Sociology, Vol. 44 (1938), pp. 1–24CrossRefGoogle Scholar;

15. Some of these groups are described inGallin, Bernard and Gallin, Rita S., “The integration of village migrants in Taipei”, in Skinner, G. William and Elvin, Mark (eds.), The Chinese City Between Two Worlds (Stanford: Stanford University Press, 1974), pp. 331–58Google Scholar; David Schak, “Normative compliance in a small-scale society: a Taiwanese beggar leader”, undated ms.; Wen, Chung-i, “Studies of social change in Wanhua: an urban community in Taipei. II. Groups and power structure” (in Chinese), Bulletin of the Institute of Ethnology, Academia Sinica, No. 39 (1975), pp. 19–56Google Scholar; and Yin, Alexander Chien-chung, “Voluntary associations and rural-urban migration”, in Ahern, Emily Martin and Gates, Hill (eds.), The Anthropology of Taiwanese Society (Stanford: Stanford University Press, 1981), pp. 319–40Google Scholar;

16. Grichting, Wolfgang L., The Value System in Taiwan, 1970: A Preliminary Report (privately published in Taipei, 1971), p. 350Google Scholar;

17. In the five major cities, about 78% of the employed population works in the private sector (calculated from Ministry of Interior, 1982 Taiwan-Fukien Demographic Fact Book, pp. 352–55)Google Scholar;

18. Susan Greenhalgh, “Microsocial processes of income distribution on Taiwan”, in Winckler and Greenhalgh (eds.), Authoritarianism and Dependency in East Asia: Comparing Taiwan.

19. For families with tangible property division is a discrete event or series of events that entail a partition of the family assets.“De facto division” occurs in families without property. In such families the marriage and physical dispersal of sons effectively dissolve the original unit and establish new families.

20. Cases I am most familiar with include the three-way break-up of the Hsiao Family Group (which includes the well-known Ta Ming Chemical Fiber Corporation) in 1965, and the split of the vast Kung Hsueh She Group (makers of educational materials, importers of Yamaha motorcycles and musical instruments, etc.) in 1973.

21. These terms were first introduced byCohen, Myron L. in House United, House Divided: The Chinese Family in Taiwan (New York: Columbia University Press, 1976)Google Scholar; My definition of the primary household is slightly different from Cohen's. For him the primary household is that containing the majority of family members.

22. Greenhalgh, Susan, “Income units: the ethnographic alternative to standardization”, in Ben-Porath, Yoram (ed.), Income Distribution and the Family (New York: Population Council, 1982Google Scholar; Supplement to Population and Development Review, Vol. 8), esp. p. 79.

23. Greenhalgh, , “Income units”, esp. p. 78Google Scholar; This process of family dispersal has led some sociologists to conclude that societal modernization has created “modern” nuclear families in Taiwan's cities. Blindfolded by the assumptions of modernization theory, these scholars paid inadequate attention to the social and economic links between related households. Available data suggest that, had they fully measured these links, they would probably have reached the opposite conclusion, that modernization has created (modified) traditional joint families in urban Taiwan (this is elaborated below). While some elements of modernization theory are supported on Taiwan (e.g. the fertility decline), these and other changes may be better explained by a “revisionist modernization” perspective such as that developed by Tamara Hareven, Louise A. Tilly, Joan W. Scott and others.

24. SeeMeskill, , A Chinese Pioneer Family, p. 24Google Scholar; The description of Chinese as profit-.’ minded petty entrepreneurs is Elvin's. See Elvin, Mark, The Pattern of the Chinese Past: A Social and Economic Interpretation (Stanford: Stanford University Press, 1973), p. 167Google Scholar;

25. The distrust and suspicion with which people outside the family are viewed is discussed bySilin, Robert H., Leadership and Values: The Organisation of Large Scale Taiwanese Enterprises (Cambridge: Harvard University Press, 1976), esp. pp. 45–48CrossRefGoogle Scholar; and Stites, Richard West, Small-Scale Industry in Yingge, Taiwan, Ph.D Dissertation, Department of Anthropology, University of Washington (1982), esp. pp. 136–52Google Scholar;

26. SeeDeGlopper, Donald R., “Doing business in Lukang”, in Willmott, W. E. (ed.), Economic Organization in Chinese Society (Stanford: Stanford University Press, 1972), pp. 297–326Google Scholar; Mark, Lindy Li, Taiwanese Lineage Enterprises: A Study of Familial Entrepreneur-ship, Ph.D. Dissertation, Department of Anthropology, University of California, Berkeley (1972)Google Scholar; and Stites, Small-Scale Industry in Yingge, Taiwan. On economic relationships in other Chinese societies, see for example Freedman, Maurice, Chinese Family and Marriage in Singapore (London: Her Majesty's Stationery Office, 1957), esp. pp. 87–90Google Scholar;

27. Liu, , “Factors and policies contributing to urbanization and labor mobility in Taiwan”, p. 19Google Scholar; See also Huang, Nora Chiang, “The migration of rural women to Taipei”, in Fawcett, James T., Khoo, Siew-ean and Smith, Peter C. (eds.), Women in the Cities of Asia: Migration and Urban Adaptation (Boulder, Colorado: Westview, 1984), pp. 247–68Google Scholar;

28. Huang, “The migration of rural women to Taipei.”

29. Simon, Denis Fred, Taiwan, Technology Transfer, and Transnationalism: The Political Management of Dependency, Ph.D. Dissertation, Department of Political Science, University of Calfornia, Berkeley (1980), esp. pp. 510–12Google Scholar;

30. This discussion is confined to differences between Han Chinese of mainland and Taiwan origin.

31. These estimates are based onKuznets, Simon, “Growth and structural shifts”, in Galenson, Walter (ed.), Economic Growth and Structural Change in Taiwan: The Postwar Experience of the Republic of China (Ithaca: Cornell University Press, 1979), esp. pp. 27–38Google Scholar;

32. SeeGates, Hill, “Ethnicity and social class”, in Ahern, and Gates, (eds.), The Anthropology of Taiwanese Society, pp. 241–81Google Scholar;

33. A recent survey of 8,000 individuals suggests that about one-quarter of the people have a member of their immediate families married to someone of the other group, while another quarter have a more distant relative whose spouse belongs to the other ethnic group. SeeFar Eastern Economic Review, Vol. 119 (13 01 1983), p. 19Google Scholar;

34. This survey was conducted by the Bureaus of Budget, Accounting and Statistics of Taiwan Provincial Government and Taipei City Government..The information discussed below was coded from the first page of the original questionnaire and computer analysed. My sample of 4,700 households was a 50% random sample of the original sample, which was selected by stratified random sampling methods, using a sampling fraction of 0.3%. Methods of data collection are detailed inDirectorate-General of Budget, Accounting and Statistics, Report on the Survey of Personal Income Distribution in Taiwan District, Republic of China, 1976 (Taipei: Directorate-General of Budget, Accounting and Statistics, 1977)Google Scholar; A convenient overview of these methods is provided by Fei, John C. H., Ranis, Gustav and Kuo, Shirley W. Y. in their book, Growth With Equity: The Taiwan Case (New York: Oxford University Press, 1979)Google Scholar;

35. The definitions of urban levels are based onLiu, Paul K. C., “Effect of population policies on land use and regional development: a simultaneous model for Taiwan”, in National Science Council, Sino-American Workshop on Land Use Planning, Taipei, 4–12 January 1978 (Taipei: National Science Council, 1978), pp. 415–43Google Scholar;

36. These are described inGates, , “Ethnicity and social class”, p. 277Google Scholar;

37. This conclusion is based on an analysis of a sample of 100 of Taiwan's largest business groups. The data come from China Credit Information Service, Ltd.,Business Groups in Taiwan, 1978 (in Chinese) (Taipei: China Credit Information Service, 1978)Google Scholar; Of 80 groups founded by Taiwanese, 66% were family firms and 15% were organized by families and friends. Of 20 groups founded by mainlanders, 40% were organized as family firms, and 15% were owned by family and friends.

38. See supra, n. 37.

39. This follows from the differences in class position observed above (Taiwanese are more likely to be own-account workers and employers). It is also supported by data on house ownership presented by Grichting. These data indicate that 73% of Taiwanese (Minnan and Hakka, N = 1, 538) and only 34% of mainlanders (N = 337) own their houses. SeeGrichting, , The Value System in Taiwan, p. 81Google Scholar;

40. Here I deal only with classes as analytic entities (classes in themselves). I do not discuss classes as class-conscious political actors (classes for themselves).

41. Computed fromMinistry of Interior, 1982 Taiwan-Fukien Demographic Fact Book, pp. 352–55Google Scholar; The five largest cities are, in order, Taipei, Kaohsiung, Taichung, Tainan and Keelung. In the United States in 1980, 2% of the labour force were employers, 85% were employees, 7% were self-employed. An additional 6% were small employers. See Wright, Erik Olin, Costello, Cynthia, Hachen, David and Sprague, Joey, “The American class Structure”, American Sociological Review, Vol. 47, No. 6,(1982), pp. 709–726CrossRefGoogle Scholar;

42. Gates' portrait of the class make-up of the whole society is similar in key respects to the one I draw for the cities. See her article,“Dependency and the part-time proletariat in Taiwan”, Modern China, Vol. 5, No. 3 (1979), pp. 381–407CrossRefGoogle Scholar;

43. Methods of data collection and analysis are detailed in my thesis,Demographic Differentiation and the Distribution of Income: The Taiwan Case, Ph.D. Dissertation, Department of Anthropology, Columbia University (1982)Google Scholar;

44. This does not mean that they all see all parts of the ladder at once; those near the bottom generally see only the rungs just above and below the one they are on.

45. A nuclear family contains a married couple and any unmarried children; a stem family consists of a married couple, any unmarried children, and one married child (usually a son) with his spouse and any children; a joint family includes two or more married children (usually sons) with their spouses and any children, their parents, if alive, and any unmarried siblings. Families in any of these configurations may be “broken” -i.e. one member of a t defining couple is absent – and still retain their original structure. There is no claim that these families are representative of all urban families. In fact, they are probably slightly larger and more complex than average because they were selected so as to include at least one member born between 1915 and 1935.

46. The China News, 23 July 1979.

47. Taiwan's low level of interclass inequality presents a serious challenge for dependency and world-systems theories, which hold that integration into the capitalist world economy exacerbates inequalities between classes. A particularly lucid statement of this position isRubinson, Richard, “The world economy and the distribution of income within states: a cross-national study”, American Sociological Review, Vol. 41 (1976), pp. 638–59CrossRefGoogle Scholar; Why such theories fail to explain the Taiwan situation is discussed by Barrett, Richard E. and Whyte, Martin King, “Dependency theory and Taiwan: analysis of a deviant case, ”American Journal of Sociology, Vol. 87, No. 5 (1982), pp. 1062–89CrossRefGoogle Scholar; and Susan Greenhalgh, “Macrolevel processes of income distribution on Taiwan”, in Winckler and Greenhalgh (eds.), Authoritarianism and Dependency in East Asia: Comparing Taiwan.

48. Mote, F. W., “The transformation of Nanking, 1350–1400”, in Skinner, G. William (ed.), The City in Late Imperial China (Stanford: Stanford University Press, 1977), pp. 103Google Scholar; also 101–119. See also Skinner, G. William, “Introduction: urban and rural in Chinese: society”, in Skinner, G. William (ed.), The City in Late Imperial China, pp. 253–74Google Scholar;

49. Urban-based families are families whose primary households are located in the city of Taipei or the town of Tanshui. Rural-based families are those whose primary households are ‘located in the agricultural community of Sanchih. The differences we find in this sample may be smaller than they would be in other rural-urban comparisons because the rural: community of Sanchih is relatively close (c. 40 km) to Taipei. On the other hand, Taipei is the wealthiest city on the island; using it in our comparison tends to exaggerate urban-rural differences. In any event, all three sites are in the north, which is the richest region on the island. Urban-rural comparisons that crossed regional boundaries would undoubtedly show larger differences in social and economic organization.

50. A classic statement of this argument isSkinner's, G. William, “Mobility strategies in late imperial China: a regional systems analysis”, in Smith, Carol A. (ed.), Regional Systems (New York: Academic Press, 1976), Vol. 1, pp. 327–64Google Scholar;

51. Ties created through marriage are “supra-family” networks because they include people such as married-out daughters and families of daughters-in-law who are not technically part of the family unit. Available evidence suggests that such ties have become increasingly important over the past decades. This is elaborated below. Some data on marriage distances are provided bySando, Ruth Ann E., The Meaning of Development for Rural Areas: Depopulation in a Taiwanese Farming Community, Ph.D. Dissertation, Department of Anthropology, University of Hawaii (1981), pp. 142–50Google Scholar;

52. Chang, , “Economic adjustment of migrants in Taiwan”, p. 6Google Scholar;

53. These estimates are based on my data from Sanchih and Huang's from Sanlin in central Taiwan. SeeHuang, Shu-min, Agricultural Degradation: Changing Community Systems in Rural Taiwan (Washington, D.C.: University Press of America, 1981), p. 161Google Scholar; Data on the number of individuals living outside the village can be found in Gallin, Bernard and Gallin, Rita S., “Socioeconomic life in rural Taiwan: twenty years of development and change”, Modern China, Vol. 8 (1982), pp. 205–46CrossRefGoogle Scholar; Sando, The Meaning of Development for Rural Areas; Wang, Sung-hsing and Apthorpe, Raymond, Rice Farming in Taiwan: Three Village Studies (Taipei: Insitute of Ethnology, Academia Sinica, 1974), p. 102Google Scholar;

54. In 1982, 94% of the registered students abroad were in theU.S. Ministry of Education, Educational Statistics of the Republic of China, 1983 (Taipei: Ministry of Education), p. 52Google Scholar;

55. Boyan, Douglas R. (ed.) Open Doors: 1982/83, Report on International Educational Exchange (New York: Institute of International Education, 1981)Google Scholar;

56. Ministry of Education, Educational Statistics of the Republic of China, 1983, pp. 48–49Google Scholar;

57. Between the early 1950s and early 1970s the total fertility rate declined from 6,708 to 3,427; by 1982 it had fallen to 2,320. This rate indicates the number of children a cohort of 1,000 women would give birth to through the full span of their reproductive life, assuming complete survival and age-specific birth rates of the year to which the rate applies. Thus, these figures suggest that the average number of births per woman fell from 6.7 in the early 1950s to 2.3 in 1982. SeeKuznets, , “Growth and structural shifts”, pp. 116–17Google Scholar; and Ministry of Interior, 1982 Taiwan-Fukien Demographic Fact Book, pp. 786–87Google Scholar;

58. Strauch, Judith, “Women in rural-urban circulation networks: implications for social structural change”, in Fawcett, James T., Khoo, Siew-ean and Smith, Peter C. (eds.), Women in the Cities of Asia: Migration and Urban Adaptation, pp. 60–77Google Scholar;

59. ‘See especiallyGallin, Bernard, Hsin Hsing, Taiwan: A Chinese Village in Change (Berkeley: University of California Press, 1966)Google Scholar; Gallin, Bernard and Gallin, Rita S., “The integration of village migrants in Taipei”, in Elvin, and Skinner, (eds.), The Chinese City Between Two Worlds esp. pp. 351–56Google Scholar; and Gallin and Gallin, “Socioeconomic life in rural Taiwan.”

60. With fewer family members to pool their resources, entrepreneurs will need assistance from outside the family, presumably from the government, to obtain the necessary loans and advanced technology.