The romantic mind imagines a wedding as two people joining. The anthropological mind sees two kin groups joining and using the couple as the joint. For most of human history, this second view was the dominant one — held not just by anthropologists looking back but by the participants themselves. A marriage made two families into affines: people now bound by mutual obligation, claim, and interest. The couple at the center of the ceremony was the medium through which the alliance was struck. They were important, but they were not the principal stakeholders.
Lévi-Strauss built his theory of kinship on this insight. Marriage, he argued, is the elementary structure through which human groups exchange women (in his framing, controversially) and create the lattice of alliance that turns isolated bands into wider societies. The incest taboo, near-universal across cultures, forces marriage outside the immediate family and thereby forces alliance. You cannot keep your daughters; you must give them away, and in giving them you receive sisters and daughters from other groups. The web of marriages becomes the web of peace.
The history of European royal and aristocratic marriage is a long demonstration of marriage-as-alliance in concentrated form. Habsburg Spain married Habsburg Austria; English princesses married French kings; the dynasties of Europe knit themselves into a single extended family precisely so that war between them would carry the weight of civil war. The famous Habsburg motto — bella gerant alii, tu felix Austria nube — others wage war, you happy Austria marry — captured the principle exactly. Marriage was foreign policy. The bride was an ambassador with permanent posting. Her children were the treaty's enforcement mechanism. When she failed to produce children, the treaty failed; when she produced too many, succession disputes could undo the alliance she had been sent to create.
Below the aristocracy, the same logic ran at smaller scale. Peasant families married into neighboring peasant families to consolidate land, share labor, and build the web of mutual obligation that made village life sustainable. Merchant families married into other merchant families to cement trade partnerships, share capital, and extend networks across cities. Artisan families married into the families of allied trades. The marriage choices of any given generation were the alliance strategy of the kin group, executed person by person. Coontz documents the practical seriousness with which families across class lines treated marriage as a strategic decision affecting the whole house, not just the couple.
In societies with clear lineage structures, the alliance function is built directly into the kinship terminology. Many cultures distinguish cross-cousins (children of a parent's opposite-sex sibling) from parallel cousins (children of a parent's same-sex sibling), and prescribe or prefer marriage between cross-cousins as a way to maintain alliance across generations. Sister exchange, where two men marry each other's sisters, doubles the alliance and shares the risk. Generalized exchange systems, where group A gives women to group B, B to C, and C to A, create circular alliance structures that bind whole regions. Lévi-Strauss called these elementary structures; their elegance is striking once you stop reading marriage as a private romantic act.
The Christian Church's long campaign against close-kin marriage in medieval Europe — banning marriage within seven degrees of consanguinity at one point, later relaxed to four — had the effect of forcing European families to make alliances with more distant kin, which broke up powerful clan structures and made the Church and the developing state the principal alternative organizers of social life. Goody argued that this was, in part, the point: a Church that controlled marriage rules could weaken the corporate kin groups that competed with it for inheritance and loyalty. Whatever the intent, the effect was real. European kinship structures became unusually shallow and bilateral compared to most of the world, and the European nuclear household became correspondingly more central.
The alliance function did not disappear with the rise of romantic marriage. It became less central, less explicit, and less openly discussed, but it persists. Families still meet before weddings, still evaluate the in-laws, still judge whether the marriage will be good for the wider house. Class endogamy — marriage within social class — remains powerful in every modern society, and the mechanisms that produce it (educational sorting, residential sorting, social network sorting) generate alliance patterns nearly as predictable as the formal arrangements of earlier centuries. Two doctors marry; two professors marry; two factory workers marry. The alliance is implicit but it is there.
What changes in the modern era is the locus of decision. Where families once chose, individuals now choose. The choice is still constrained by class, network, and inherited disposition, but the formal authority has shifted. The romantic lens treats this as liberation; the institutional lens notes that the alliance function is now being performed by markets and networks rather than by elders, with results that are less coordinated and less protective of long-term kin interests. Both readings catch something true. The alliance is older than the romance, and when romances end the alliances often persist in muted form — divorced parents who still see each other at grandchildren's weddings, in-laws who remain friends across a broken bond, family networks that absorb the shocks the couple could not contain.
To marry is still to ally families, whether the couple notices or not.