For roughly a century — from Lewis Henry Morgan's Systems of Consanguinity and Affinity of the Human Family in 1871 to David Schneider's A Critique of the Study of Kinship in 1984 — kinship was anthropology's central technical problem. Departments hired on it. Dissertations turned on it. Jokes inside the discipline assumed it as common reference. And then the field, in the space of about fifteen years, walked away.
Both halves of that story are interesting.
Why kinship looked like the master key
Morgan's insight was deceptively simple. Different societies use different terminologies to refer to relatives — the same English word cousin may, in another society, be split into four distinct categories, with elaborate marriage and avoidance rules attached to each. Morgan compiled a worldwide tabulation of these terminological systems and noticed that they sorted into a finite number of types. There were, on his count, six basic patterns into which the kinship terminologies of every documented society fell.
If terminologies sorted into types, the argument went, the underlying social structures must too. To know how a society's kinship terminology worked was to know, with surprising specificity, how marriage was contracted, who could and could not be a sexual partner, how property descended, how households formed and dissolved, how political authority was inherited. Kinship was the load-bearing skeleton on which everything else hung.
The British social anthropologists ran with this for sixty years. Radcliffe-Brown, Evans-Pritchard, Meyer Fortes, Edmund Leach — the major monographs of the mid-century are all, at their core, kinship monographs. The Nuer (1940) is a kinship study. African Political Systems (1940) is a comparative study of kinship-grounded polities. The technical apparatus became dauntingly elaborate: descent groups, lineage segments, prescriptive versus preferential marriage, classificatory versus descriptive terminologies, ego-centered versus society-centered diagrams.
The Lévi-Strauss leap
Claude Lévi-Strauss, working in Paris from a base in Brazilian fieldwork during the 1930s, took the kinship problem in a direction that detached it from the British tradition almost entirely. His argument, in The Elementary Structures of Kinship (1949), was that kinship is fundamentally about the exchange of women between groups. The incest taboo — the universal prohibition on marrying close kin — forces every social group to obtain its women from elsewhere, which forces every social group into a relationship of exchange with at least one other group. Marriage is, in his framing, a kind of communication: women circulate the way words circulate, and the rules that govern who-marries-whom are a grammar in the linguistic sense.
That move — from social institutions to underlying structures analogous to linguistic structures — became the founding gesture of structuralism. The argument extended outward from kinship to myth (The Raw and the Cooked, 1964), to totemism, to food and ritual, eventually to almost any cultural phenomenon you cared to apply it to. By the 1960s, structuralism in this sense had become a major intellectual movement well outside anthropology, with effects in literary theory (Barthes), psychoanalysis (Lacan), Marxism (Althusser), and historiography (the early Foucault).
Why the project broke
Schneider's A Critique of the Study of Kinship (1984) is, oddly, the document that ended the project from within the discipline. Schneider, an American who had built his career on kinship work in Yap, argued — devastatingly, in a slim volume — that the entire kinship apparatus rested on a Euro-American folk theory that biological reproduction is the natural ground of social relations. Anthropologists had imported, without examining, the European intuition that "kinship" picks out a real, biologically grounded, cross-culturally comparable thing. But the assumption was a folk theory, not a scientific finding. There was no good reason to think the Yapese category of tabinau mapped onto English "kin" in any deep way; there was every reason to think the systematic appearance of mappability was an artifact of the analytic apparatus the discipline had built around the assumption.
The critique was uncomfortable enough that the field largely accepted it and changed the subject. Within a decade, kinship had moved from the central technical problem of anthropology to a specialized topic. The American departments turned toward political economy, post-colonial critique, science studies, the body, gender. The terminological diagrams disappeared from textbooks.
What was left
The strange thing is what survived the dismantling. Specific kinship findings — that bridewealth and dowry tend to occur in different kinds of agricultural systems, that matrilineal societies cluster around certain forms of subsistence, that systematic preferential cross-cousin marriage produces over generations a recognizable pattern of inter-group ties — those have not been refuted. They are simply no longer the discipline's organizing problem. They sit in handbooks now, available, used.
What also survived is the style of attention the kinship project enforced: the willingness to take seriously, and to diagram, what your interlocutors actually said about who their relatives were, against what the ethnographer's intuition wanted them to say. That style migrated outward into other ethnographic problems and is, arguably, the most durable methodological legacy of the project that produced it.
What stayed with me
That an entire generation of anthropologists patiently mapped kinship terminologies in two thousand societies, that the resulting archive is a real intellectual achievement, and that the discipline then, on the strength of one well-argued critique, decided collectively to move on. Disciplines, like crops, can be overgrown. The interesting question is not whether kinship will return — small parts of it have, under different names — but whether the next overgrown field will be recognized as quickly.