1 The major collections of essays in comparative household history simply juxtapose a variety of studies of household structure in different societies. Laslett, Peter and Wall, Richard, eds., Household and family in past time (Cambridge, 1972)CrossRefGoogle Scholar; Wall, Richard, ed., Family forms in historic Europe (Cambridge, 1983)CrossRefGoogle Scholar; McC, RobertNetting, , Wilk, Richard R. and Arnould, Eric J., Households: Comparative and historical studies of the domestic group (Berkeley, 1984). Very few of the essays explicitly contrast different populations. The only prominent exceptions are:Google Scholar; Hajnal, John, ‘Two kinds of pre-industrial household formation’Google Scholar; Laslett, Peter, ‘Family and household as work group and kin group: Areas of traditional Europe compared’Google Scholar; Wall, Richard, ‘Introduction’, in Wall, ed., Family forms, 65–107, 513–64, 1–64.Google Scholar

2 The Chinese households were peasants who lived in a number of villages in Shenyang County surrounding the provincial capital of Liaoning (formerly Shengjing) Province. Most of them were well-off. Our data derive from eleven Eight Banner registers dated 1774, 1780, 1786, 1792, 1798, 1801, 1804, 1810, 1813, 1816 and 1819. These registers record for each person their names, age, occupation, family relationship, lineage relationship, birth date, physical health, recent demographic events and much more. They survive today in the Number Three Historical Archives in the Peoples Republic of China and are available on microfilm at the California Institute of Technology.

3 The Norwegian households were derived from individual level censuses of Leikanger and Balestrand counties conducted in 1801 and 1865. The bulk of the population were members of ‘middle class’ landed households. Sustained population growth, however, between 1801 and 1865 in conjunction with strict unigeniture inheritance led to widespread landlessness. The result was increasing emigration from these two counties to the United States.

4 The American households come from the 1880 and 1900 United States federal census for Filmore and Renville Counties, Minnesota. Previous censuses either do not indicate headship or relationship to the head. The 1890 manuscript census schedules were destroyed by fire. The population consists largely of ‘middle class’ farmers primarily of north-western European stock including Scandinavians, Germans, and Irish.

5 Manuel de demographie historique (Paris, 1967).Google Scholar

6 Laslett, and Wall, , eds., Household and family.Google Scholar

7 ‘Comparing household structure over time and between cultures’, Comparative Studies in Society and History 16 (1974), 73–111.CrossRefGoogle Scholar

8 To date, our work has been mainly latitudinal. We have begun our longitudinal analyses which will appear in a separate article on comparative household cycles.

9 George, Murdock, Ethnographic atlas (Pittsburgh, 1967).Google Scholar

10 The classic text is Le Play, P. G. F., Les ouvriers Européens: l'organisation des families (Paris, 1855).Google Scholar

For an elaborate discussion of the stem household structure, see Mitterauer, M. and Sieder, R., The European family (Oxford, 1980).Google Scholar

11 For the purposes of these and other comparisons we include the conjugal units of servants and other workers in our definition of the household. Joint households often included kin as household workers. Stem households sometimes listed kin as servants and boarders. The result substantially modifies the distribution of household types.

12 Many people have proposed variant schemes, usually, however, for specific societies: Chudacoff, Howard, ‘Newlyweds and family extension: First stages of the family cycle in Providence, Rhode Island’, in Hareven, Tamara, ed., Transitions: The family and the life course in historical perspective (New York, 1978), 179–205Google Scholar; Hammel, Eugene, ‘Household structure in fourteenth-century Macedonia’, Journal of Family History 5 (1980), 242–73CrossRefGoogle Scholar; Kolenda, Pauline, ‘Region, caste and family structure: A comparative study of the Indian joint family’, in Singer, Milton and Cohen, Bernard, eds., Structure and change in Indian society (Chicago, 1968), 339–96Google Scholar

Schmidtbauer, P., ‘Households and household forms of Viennese Jews in 1857’, Journal of Family History 5 (1980), 375–89CrossRefGoogle Scholar; Smith, Elizabeth, ‘Family structure and complexity’, Journal of Comparative Family Studies 8 (1977), 299–310Google Scholar; Smith, Robert, ‘The domestic cycle in selected commoner families in urban Japan, 1757–1858’, Journal of Family History 3 (1978), 219–35CrossRefGoogle Scholar; Wolf, Arthur, ‘Family life and the life cycle in rural China’, in Netting et al., Households, 279–98. Very few of these particularistic models are suitable for wide application. None of them have achieved the popularity of the Henry-Laslett-Hammel classification scheme.Google Scholar

13 Goody, Jack, The developmental cycle in domestic groups (Cambridge, 1958)Google Scholar; Berkner, Lutz, ‘The stem family and the development cycle of the peasant household’, American Historical Review 77 (1972), 398–418CrossRefGoogle Scholar and ‘The use and misuse of census data for the historical analysis of family structure’, Journal of Interdisciplinary History 5 (1975), 721–38.CrossRefGoogle Scholar

14 Hareven, , ed., Transitions.Google Scholar

15 Gaunt, David, ‘The property and kin relations of retired farmers in northern and central Europe’, in Wall, , ed., Family forms 249–80.Google Scholar

16 Smith, Daniel Scott, ‘Life course, norms, and family system of old Americans in 1900’, Journal of Family History 4 (1979), 285–98.CrossRefGoogle Scholar

17 Wheaton, Robert, ‘Family and kinship in western Europe: The problem of the joint family household’, Journal of Interdisciplinary History 5 (1975), 601–28.CrossRefGoogle Scholar

18 Wolf, Arthur and Huang, Chieh-Shan, Marriage and adoption in China (Stanford, 1982).Google Scholar

19 Kessinger, Thomas, Vilayatpur, 1848–1968: Social and economic change in a north Indian village (Berkeley, 1974)Google Scholar; Palli, H., ‘Estonian households in the seventeenth and eighteenth centuries’, in Wall, , ed., Family forms 207–16Google Scholar; Plakans, A., ‘Serf-emancipation and the changing structure of rural domestic groups in the Russian Baltic provinces: Linden Estate, 1797–1858’, in Netting et al., Households, 245–78Google Scholar; Duben, Alan, ‘Turkish families and households in historical perspective’, Journal of Family History 10 (1985), 75–97.CrossRefGoogle Scholar

20 ‘The perennial multiple family household, Mishino, Russia, 1782–1858’, Journal of Family History 10 (1982), 5–26Google Scholar; ‘A large family: The peasant's greatest wealth: Serf households in Mishino, Russia, 1814–1858’, in Wall, , ed., Family forms, 105–52.Google Scholar

21 In defence of Hammel and Laslett, they are cognisant of the asymmetry of headship designation between joint and extended households. They contend that they would make appropriate changes to their scheme if they had universal knowledge of who indeed was household head ‘instead of being in the position of having to assume that the first named was always the head’. Our experience in working with three diverse populations is that headship is often clearly delineated within the document. We concur that any scheme based on headship faces difficulty in those cases where headship is assumed. But we contend that such an instance would adversely affect most schemes, including the Henry–Laslett–Hammel system itself.

22 The conjugal unit is identical in most instances with the conjugal unit as defined according to the Laslett–Hammel system. There are, however, two exceptions in that widowed offspring, without offspring of their own form part of the conjugal unit of the parent as do grandchildren residing with a grandparent in the absence of the intervening generation (see Figure 5 D, E). Under the Laslett–Hammel system any offspring included in the parental conjugal unit must be unmarried while the grandparent–grandchild combination constitutes one of the variants of the non-family household (see Laslett and Wall, eds., Household andfamily, 29, 4 1–2.)

23 Under the Laslett–Hammel system the conjugal unit is formed by the youngest generations present; see Laslett and Wall, Household and family, 29.