On the Holes in Chinese Genealogies

Vol. 8, No. 2 Late Imperial ChinaDecember 1987 ON THE HOLES IN CHINESE GENEALOGIES Stevan Harrell* The Problem The social history of Europe from the sixteenth to the twentieth centuries has been greatly enriched by the incorporation of the detailed demographic history of small communities. As students of Chinese history begin to move from and exclusive focus on large-scale generalizations to a finer-grained examination of local phenomena, we naturally hope to learn from our Europeanist colleagues and to emulate, where possible, their successful efforts at local history, including local population history. We, too, would like to be able to reconstruct families from sources resembling the English and French parish registers, for example. But it is not to be. With the exception of certain local and specialized populations,1 there exist very few population registers suitable for Cambridge-style family reconstruction. If we are to study population process in a wide variety of localities over a considerable time span, we seem to have only one kind of source in any abundance: the lineage or clan genealogy. There exist in the world's libraries thousands of these genealogies, ranging in size from a few hand-scribbled pages to a small shelf of weighty tomes, covering most of the coast provinces and a few in the interior, containing useful data, in most cases, from at least the early Ming, and sometimes from as early as the Song.2 Beginning in about 1970, a few scholars have begun to extract I am indebted to a large number of people for various kinds of help with this project. Arthur Wolf, Sophie Sa, and Ted Telford all provided access to valuable data. The Chester Fritz endowment of the China Program of the Henry M. Jackson School of International Studies and the Graduate School Research Fund, both of the University of Washington, generously helped fund the project. Able research assistance was provided by Liou Hsiao-hsiun, Hsieh Shih-chung, and Huchun. A preliminary version was presented at the Workshop on Qing Population, sponsored by the Joint Committee on Chinese Studies of the American Council of Learned Societies and the Social Science Research Council, and held at the California Institute of Technology in August, 1985. I am grateful to the Committee and to the Workshop organizers, James Lee and William R. Lavely, for providing me with this opportunity . 1 Lee and Eng 1984; Lee and Gjerde 1986. 2 See Telford 1986 for systematic tabulation of the data available in a sample of about 900 of these genealogies, mostly printed ones from coastal provinces of the mainland; see Telford, Thatcher, and Yang 1987 for a comprehensive listing. 53 54Stevan Harreil data from these genealogies to try to answer various kinds of populationrelated and kinship-related questions. Liu Ts'ui-jung3 has attempted to derive fertility and mortality schedules by a variety of ingenious methods. Fei and Liu4 have used genealogical population data to construct a model of the rise and decline of lineages. In a previous paper5 I have attempted to use genealogical data to model the internal demographic differentiation within lineages. Other scholars, including Telford, Hazelton, Dennerline, Marmé, and others in the US, as well as Ju Deyuan, Guo Songyi, and others in China, are now engaged in more ambitious projects, attempting to tabulate data for large numbers of genealogies in a local area, often to try to say something about population trends in that area at a scale intermediate between the local lineage and the large-scale studies of traditional demographic history. A conference was held in January 1987 on the topic of lineage demography in late traditional China and Korea, at which a larger number of scholars (still a relative handful, given the esoteric nature of the field) compared their results for various areas, mostly in the Lower Yangzi Macroregion. Genealogies are the best sources we have for comparative local historical demography in Ming and Qing China, but they are problematical from the standpoint of demographic research. Genealogies were not compiled for the convenience of future historical demographers, or indeed even for keeping track of the population trends of the lineages or clans themselves. Rather they were (and are, at least where they continue...