The contrast between alliance and descent is an idea developed by Louis Dumont to characterize differences between Meyer Fortes and his school of structural functionalists on the one hand and Edmund Leach and his school of structuralists on the other. Fortes argued that descent and descent groups were primary to most social systems based on kinship and could thus explain more about how these systems functioned. Leach, taking off from Levi-Strauss's arguments in The Elementary Structures of Kinship, argued that matrimonial alliances were more fundamental. The two sides drew for the most part on different types of societies and so their explanations were not mutually exclusive. Dumont pointed out that even so the divergence of views turned on different understandings of social structure itself. Fortes argued that descent groups served to create a political and jural order of rights and obligations, which itself was a reflection of a universal principle of kinship amity rooted in the biological family. Levi-Strauss's account of marriage alliances starts with a concept of kinship categories as a break with nature. For Levi-Strauss, the symmetrical and complementary categories of kinship derive from the fact that marriage is an exchange in which people are gifts between groups. This reciprocity among segments of society is what creates the social order, not the supposedly precultural facts of kinship amity and descent identity.

Dumont, Louis. 2006. An Introduction to Two Theories of Social Anthropology: Descent Groups and Marriage Alliance, Robert Parkin, trans. New York: Berghahn Books.

Leaf, Murray J. 2013. “Alliance-Descent Debate.” In Theory in Social and Cultural Anthropology: An Encyclopedia, Richard L. Warms and R. Jon McGee, eds. Thousand Oaks, Calif.: SAGE Publications, Inc. http://dx.doi.org/10.4135/9781452276311.

Lévi-Strauss, Claude. 1969. The Elementary Structures of Kinship, Rodney Needham, trans. Boston: Beacon Press.

Fortes, Meyer. 1953. “The Structure of Unilineal Descent Groups.” American Anthropologist 55 (1): 17–41. doi:10.2307/664462.