Networks from Genealogies and Linked Censuses: Why do they matter?

Douglas R. White, University of California, Irvine
James Moody, Ohio State University

Networks of kinship and marriage, whether from ethnographic research, linked censuses, or genealogies of sufficiently large populations, provide crucial measures for delimiting the boundaries and intersections of socially cohesive subgroups. New methods are available for this type of boundary delimitation in large networks of all types, including structural endogamy, or cohesion in genealogical networks that occurs through marriage. Membership within common cohesive subgroups, even on a large scale, may provide a delimitation of population subgroups based on structural rather than attribute data. These delimitations may be relevant to the formative processes involved in social class, ethnicity, diffusion, social mobilization, cultural and property transmission, and elite structure. The redundancies of both number of independent paths between members of a cohesive block and the robustness of its connectivity (the minimum number of its members whose removal would disconnect the group) insure that even large cohesive blocks with sparse ties and structural holes where face-to-face ties are missing can exert extensive social influence not only on members, but through interlock with other institutions (e.g., political or institutional officeholding, occupational specializations) and thereby on other populations. Predictive cohesion theory is a new body of theory that spells out and tests the expected consequences of structural cohesion and (in the case of kinship networks) structural endogamy. An important part of the theory relates to those members of a population whose ties are not cohesive, and how lack of cohesion impacts on migration, lack of participation in community activities and organizations, lack of influences, and lower reception of information and potential resources that circulate in cohesive communities.