Marriage & Kinship curriculum

Every society recognizes some unions and not others, but what the recognized union carries varies: rights over children, over labor, over property, over sexual access, and the duty to support. One society may bundle all of these; another may split them among different relatives. When you meet a new marriage custom, ask which rights it transfers and to whom, rather than whether it "counts" as marriage by your own society's bundle.

Edvard Westermarck, writing in 1891, built his definition around duration: a connection between male and female that lasts past mating and into the rearing of the child. He chose it to include every form then known and to exclude passing liaisons. Later scholars found it too narrow, since it leaves out childless and same-sex unions, but it fixed the question every later definition answers: what makes a union a marriage rather than a pairing?

Anthropologists have argued for a century over whether every society has marriage. The answer turns on the definition: a definition built on shared residence or on a husband who supports the children excludes societies where a woman's brothers, not her partners, hold those roles. Writers who want a universal "marriage" loosen the definition; writers who stress diversity keep it tight. Neither side is wrong about the facts, only about the word.

Many definitions center on what marriage does for a child. A child born to a married woman gets a socially recognized father, who passes on his name and kin and takes on obligations, even if another man is the biological parent. The social father (the pater) and the biological father (the genitor) are different roles, and a society can give them to different men. Westermarck put the father supporting mother and child at the institution's core.

Your parents, siblings, cousins and grandparents are consanguineal kin: you share descent. Your spouse's relatives, and your siblings' spouses, are affinal kin: the tie runs through a marriage. The two kinds carry different duties and different marriage rules: many societies forbid marrying close consanguines but encourage marrying certain affines, such as a brother's widow. English blurs the line with "in-law"; kinship diagrams keep it sharp.

Not all kin are born or married into the family. Adoption makes a stranger a child; godparenthood makes two families co-parents; blood brotherhood makes two men kin by rite. Morgan recorded the Iroquois right of adopting outsiders into a gens, which then treated them as members by descent. Anthropologists call these ties fictive, but the word misleads: the obligations are real, and the people involved rarely think of them as pretend.

Kinship charts use a few symbols: a triangle for a male, a circle for a female, a horizontal line joining siblings, a vertical line from parents to children, and an equals sign for a marriage. Every chart is read from Ego, the person whose relatives are being mapped. Relatives are written as chains of primary terms, so "father's brother's daughter" becomes FBD. Learn the notation and the next chapter's systems become readable at a glance.

Bachofen, McLennan and Morgan each held that marriage grew out of an original state of unregulated mating. Westermarck's book was written largely to test that claim, and he argued that jealousy, paternal care and the pairing seen in many societies made an original promiscuity unlikely. The debate matters less for its answer than for its method: it taught the field to treat present customs as evidence for the past only with great care.

A survival is a custom thought to be left over from an earlier social stage, like a fossil. McLennan read the levirate, in which a widow passes to her husband's brother, as a survival of polyandry. Westermarck replied that the custom fits inheritance of a widow as property just as well, so it proves nothing about polyandry. Rule: a custom can be explained by its present function before it is read as a relic of the past.

English has "aunt" for a parent's sister and a parent's brother's wife alike; other languages split those or merge them differently. Each language's set of kin terms forms a system, and the systems are remarkably constrained given how many are logically possible. Kinbank, a database of more than 210,000 kin terms from 1,229 languages, exists so that these systems can be compared rather than described one at a time.

Lewis Henry Morgan, whose 1871 survey Kinbank has digitized, saw two ultimate forms. In a classificatory system one term covers a whole category: your father's brother's sons are your "brothers". In a descriptive system relatives are specified by combining primary terms: "father's brother's son". Morgan read the classificatory type as older, which later work does not support, but the two-way split is still the first cut anthropologists make.

Theoretical models collapse billions of possible organizations into six key types, each named for a people among whom it was first described. The names are conventional labels for patterns in the parent and cousin generations, not claims about those peoples' present practice. Learn each type by asking two questions. First, does the system separate cousins from siblings? Second, does it treat the father's side and the mother's side alike?

In an Eskimo-type system, mother, father, brother and sister each get a term shared with no one else, and every cousin, on either side, is simply a "cousin". The father's side and the mother's side are treated symmetrically. English and most European languages work this way, which is one reason English speakers find the other types surprising.

Morgan used the Hawaiian system as his simplest case: the only distinctions are generation and sex, so a mother's sister and a father's sister are both "mother", and all cousins are "brother" or "sister". Morgan took this to prove an ancient stage of group marriage, a conclusion the field has abandoned. The pattern itself is real and common, and it is the most frequent destination of change in the Austronesian languages studied phylogenetically.

Parallel cousins are children of a father's brother or mother's sister (a parent's same-sex sibling); cross cousins are children of a father's sister or mother's brother (a parent's opposite-sex sibling). Iroquois-type systems call parallel cousins "siblings" and give cross cousins their own term. This "crossness" is often paired with a rule allowing or preferring cross-cousin marriage, which is why the two were long assumed to evolve together.

In the parental generation, father and father's brother share a term, while mother's brother gets his own; mother and mother's sister share a term, while father's sister gets her own. It "merges" same-sex siblings and "bifurcates" the opposite-sex ones. Kinbank's authors tested the old claim that this pattern co-evolves with cross-cousin marriage in Bantu languages and found no evidence for it: a textbook pairing can fail a statistical test.

In an Omaha-type system, the mother's brother's children are called by terms from the parent generation on the mother's side; in a Crow-type system the mirror image happens on the father's side. The skewing follows the lineage: it treats members of one line as alike regardless of generation. Omaha skewing has long been linked to patrilineal descent and Crow to matrilineal, a link that modern comparative work finds patchy rather than universal.

At the opposite pole from Hawaiian, a Sudanese-type system gives separate terms to father's brother, mother's brother, father's sister and mother's sister, and often to each kind of cousin. Reconstructions of Proto-Uto-Aztecan suggest a system with Hawaiian-style grouping in Ego's own generation and Sudanese-style separate terms for each parent's sibling, which fits none of the six types cleanly: the types are a convenience, not a law.

A 2020 study of Austronesian, Bantu and Uto-Aztecan languages found that terminologies descend vertically along the language phylogeny and show no single direction of change. Of 18 classic theories that link terminology with descent, marriage or residence, tested across 176 societies, only patchy support appeared. Rule: a kin terminology is first a piece of inherited language, and only sometimes a readout of social structure.

Kinbank's authors showed a phonological bias in parent terms worldwide: terms for mother tend to begin with a nasal consonant and an open vowel, while father terms lean on stops like "p" or "t". One explanation is that these are the sounds babies produce first, which adults then hear as names for themselves. The exceptions matter too: in many Australian languages "mama" refers to the father, so the bias is a tendency rather than a universal.

In social anthropology, descent is patrilineal when children belong to the lineage or clan of the father, matrilineal when they belong to the mother's, and bilateral when there is no such corporate group at all. The rule is about membership and its consequences: inheritance, ceremony, marriage eligibility, political standing. Everyone has two parents; descent is the convention that decides which parent's group claims the child.

A patrilineal lineage is a chain of fathers and children. Sisters stay members of their birth lineage for life, but their children belong to their husbands' lineages, so the group renews itself through its men. In many patrilineal societies the lineage manages land, organizes ritual, mediates disputes and regulates whom its members may marry. It is the most common rule in the ethnographic record, but not the only one, and it was never the first.

A matrilineal lineage is a chain of mothers and children. A man belongs to his mother's group, and so do his sister's children; his own children, meanwhile, belong to his wife's group. This is why a man's property and office often go to his sister's son. Engels described exactly this: under "maternal law" a dead man's herd went to his siblings and his sisters' children, never to his own. Matrilineal descent says nothing about who holds power.

Most present-day European societies, and the United States, reckon descent bilaterally. There are no lineages or clans; the household is the largest corporate kin group, and each person's set of relatives, the kindred, is unique to them and overlaps with everyone else's. Unilineal groups are thought to make collective action easier because membership is unambiguous; a kindred has no clear boundary, so it cannot act as a body.

The two unilineal rules are not the only options. Under double descent a person belongs to a patrilineal group for some purposes, such as land, and to a matrilineal group for others, such as movable goods or ritual. Under ambilineal descent membership runs through either parent and a person affiliates with one group, often the one where they live. Both show that descent is a toolkit a society assembles, not a single switch.

The difference is demonstrable genealogy. Lineage members can name the links back to the founder. Clan members share a name and believe they have a common ancestor but cannot trace the links; the founder may be a totemic animal. Morgan described the Iroquois gens, his term for clan, as a body of kin with a common name, a rule against marrying inside it, mutual inheritance, shared burial and the right to elect and depose its chiefs.

Above the clan, kindred clans may be bundled into a phratry with shared ceremonies and mutual obligations; Morgan documented this among the Iroquois, Choctaw and others. When a whole society is split into two halves, each is a moiety, and a member of one must usually marry into the other. A Neolithic cemetery in eastern China, read through ancient DNA, shows two maternal clans burying their dead separately and marrying across the line.

A segmentary system is built from branches of one genealogy. Close branches quarrel with each other but join against a more distant branch, and all join against outsiders. The structure gives a society without a state a way to mobilize for defense and to settle feuds by balancing like against like. The same nesting explains why "brother" in such a society may name a whole segment.

Studies of Neolithic and Bronze Age European cemeteries found high male relatedness and low Y-chromosome diversity and concluded descent was patrilineal. A 2025 simulation study found the data compatible with patrilocal residence, but it cannot show patrilineal descent, since bilateral societies with the same residence rule leave the same genetic signature. Rule: residence leaves a genetic trace; descent is a social rule that may not.

It is tempting to read "patrilineal" as "sons inherit", but descent governs group membership, inheritance governs property, and succession governs office or title. A society can trace descent through mothers and pass land through fathers, or admit daughters to a lineage and exclude them from land. Engels's account of the shift to "paternal law" is a story about inheritance being re-routed, which is why he treated it as the engine of the change.

Residence rules decide which spouse leaves home. In wage economies most couples set up a new household, but in the ethnographic record couples usually settle with or close to the parents of one spouse. The decision affects descent, kinship terminology, inheritance and the form of marriage, because the spouse who stays keeps daily support from kin and the spouse who moves does not.

About 71 percent of the societies in the Ethnographic Atlas are predominantly patrilocal. Some writers reserve "patrilocal" for living in the husband's father's household and use "virilocal" for residence with the husband's kin in a broader sense; many studies merge the two because the difference cannot be seen in older records. Either way the wife is the one who moves, and she lives among her husband's relatives rather than her own.

In a matrilocal arrangement the husband moves and the wife stays among her relatives. The pattern is far less common than patrilocality but occurs on every continent, and a study of Nama and Cederberg communities in South Africa found it persisting across three generations despite the surrounding society's patrilocal norms. "Uxorilocal" is the broader term, residence with the wife's kin rather than specifically her mother's household.

Neolocality is the familiar pattern of modern wage economies, where newlyweds rent or buy a home of their own. In the cross-cultural record it is rare, grouped with ambilocal, multilocal and avunculocal residence in the 18 percent of societies that are neither patrilocal nor matrilocal. Its rarity in traditional societies is a reminder that living apart from kin depends on an economy that lets a couple support itself.

Avunculocality can be treated as a special case of virilocality: the couple lives with a man on the husband's side, but the man is his maternal uncle rather than his father. The logic is matrilineal. A man's heir is his sister's son, so having the nephew move in keeps the lineage's property and its next holder together, while the lineage itself still runs through women.

Ambilocality gives the couple a choice between the two sets of relatives, often made on practical grounds such as land or labor. Multilocality means moving between the households of both sets of parents over time. Hunter-gatherer societies are most often ambi- or multilocal, which gives them a flexibility that settled farmers with fixed fields usually lack.

In a few societies a marriage does not create a shared household. Each spouse continues to live with their own mother's group, and the husband visits. The Mosuo of southwest China are the best-known modern case and are described as duolocal in the ancient-DNA literature on matrilineal communities. Children live with the mother's kin, and the mother's brothers do much of the work a husband does elsewhere.

A phylogenetic comparison of Austronesian societies found that the best model of dwelling size combined residence pattern with sedentism, and that sedentism, not agriculture, was the single best predictor of house size. Once settlement is fixed, matrilocality tends to go with large dwellings that hold related women and their families. The result lets archaeologists infer residence, cautiously, from the floor plans of prehistoric houses.

It is easier to keep a lineage together when its members live together, so patrilineal societies are usually patrilocal and matrilineal ones matrilocal or avunculocal. The correlation is strong enough that simulation studies use the most frequent residence rule for each descent rule. It is not a law: patrilocal societies with bilateral descent are common, which is exactly why ancient DNA cannot read descent from residence.

A ban on unions between parents and children and between siblings is close to universal, and beyond that the rules diverge. Some societies forbid all first cousins; others require marriage to one kind of cousin. Westermarck's chapters on prohibited degrees compile dozens of such rules and show that the boundary of "too close" is drawn by each society, not read off a biological fact.

Westermarck argued that there is no innate aversion to marrying relatives as such; the aversion is to sexual intercourse between people who grew up in very close contact from early youth. Relatives usually fit that description, so the feeling shows up as horror of incest. He supported it with cases like Greenland, where a boy and girl raised in one household were not expected to marry. The proposal is still tested today under his name.

Westermarck's is one explanation among several. Alliance theory holds that the taboo forces families to exchange spouses and so binds a society together. A family-role theory holds that sex between members would wreck the authority and cooperation a household needs. A biological theory points to reduced survival among children of close kin. These are positions, and each explains a part of the record that the others leave untouched.

The group that must be married out of is set by the society. Morgan listed the obligation not to marry within the gens as a defining feature of the Iroquois clan; moiety systems require marriage into the other half; some communities practice local exogamy, marrying outside the settlement. Westermarck noted that where the group is defined by residence, the prohibited partners are those one grew up beside, which is what his theory predicts.

Endogamy and exogamy are not opposites that a society picks between; most societies have both at different levels, requiring spouses to come from outside the lineage yet from within the same ethnic group or faith. Westermarck treated class and caste endogamy as the product of a want of sympathy across lines of rank and noted its decline in modern societies as mixed marriages increase. Religious endogamy remains one of the most widespread forms.

A spousal union between people who share a grandparent is common worldwide and was more so before the demographic transition. Many societies that allow it prefer cross cousins, the children of a parent's opposite-sex sibling, and forbid parallel cousins, who under unilineal descent belong to the same lineage as Ego. A smaller number prefer the father's brother's daughter, a parallel cousin, which keeps property inside the patrilineage.

A study of two populations with high rates of cousin marriage found a fitness cost in reduced child survival in both, alongside a benefit in increased fertility for men among the Dogon and for women among the Ancien Regime nobility. The trade-off helps explain why a practice with known genetic costs persists: in some settings the social and reproductive gains offset them. The finding is about those populations, not a general verdict.

Both customs keep an alliance between two families alive after a death and keep children inside the same kin group. Westermarck described the levirate as widespread and rejected McLennan's reading of it as a relic of polyandry, arguing instead that wives who count as property are just inherited along with everything else. The sororate is the mirror image, and sororal polygyny, marrying two sisters at once, is its living form.

Where groups are ranked, rules about crossing the line usually run in one direction. A system that lets a woman marry a man of higher caste but not lower is hypergamous for women. The pattern interacts with marriage payments, since a family marrying a daughter upward may have to give more with her. Westermarck's chapters on class endogamy treat the one-way exceptions as the first cracks in a closed marriage circle.

Where descent theory asks who belongs to which group, alliance theory asks how groups are tied to each other, and answers that they exchange spouses. A preference for cross-cousin marriage, on this view, is a standing arrangement by which two lineages keep giving each other partners generation after generation. It is a position that reframes the incest taboo as the first rule of exchange: to give a sister away is to gain a brother-in-law.

Prohibited degrees are not only about blood. Many systems bar a man from marrying his stepmother or his wife's close kin, treat adoptive relatives as if born into the family, and extend the ban to godparents and their children. Westermarck explained these as associations of ideas: a relative who is treated like kin comes to feel like kin, so the aversion follows the social role rather than the genealogy.

Monogamy allows one spouse at a time; polygyny allows a man several wives at once; polyandry allows a woman several husbands. "Polygamy" covers both plural forms, though in everyday speech it usually means polygyny. Keep the terms apart: a claim about "polygamy" in the ethnographic record is almost always a claim about polygyny, because polyandry is rare.

In a standard cross-cultural sample, 153 of 186 societies (82 percent) are classified as polygynous, 31 as monogamous and 2 as polyandrous. Westermarck made the complementary point in 1891: although polygyny occurs among most peoples, monogamy is by far the most common form of human marriage, and plural marriage is almost everywhere confined to a minority of men. When you read a statistic, ask whether it counts societies or marriages.

A society that forbids concurrent spouses but allows divorce and remarriage produces sequences of monogamous unions. The children of such sequences grow up with step-parents and half-siblings, as children of polygynous unions do. A 20-year study in rural Tanzania, where both polygyny and serial monogamy are common, treated the two together precisely because they shape children's lives in overlapping ways.

Co-wives who are sisters have grown up together and may share a household; unrelated co-wives are often given separate huts or houses within a compound. Westermarck noted that polygyny is commonly modified in a monogamous direction by giving the first wife a higher position, and that bigamy, two wives, is its most common form. The arrangement of the compound tells you much about how the system handles rivalry.

A 2025 analysis of 84 million census records from Africa, Asia and Oceania found that the association between the prevalence of polygyny and the share of unmarried men is negative or null in almost every country, and that counties of the 1880 American West with Mormon polygyny had fewer unmarried men than other Western counties. Marriage markets are skewed feminine enough, by age gaps and mortality, to sustain some polygyny without excluding men.

Cross-regional studies link polygyny to worse child outcomes but risk the ecological fallacy of reading a regional average as a household effect. A prospective study of 3,693 children in rural Tanzania found no consistent advantage for monogamy in survival, growth or schooling. Where polygynous families are wealthier or co-wives cooperate, the authors note, plural marriage can go with better outcomes; where women lack agency, with worse.

The best-documented polyandry is Himalayan, where a set of brothers marries one woman and their children belong to the household as a whole. Westermarck recorded the Toda of the Nilgiri Hills, among whom a girl who married a man became the wife of all his brothers as they came of age. The cross-cultural study of religious celibacy lists polyandry alongside monogamy and primogeniture as a strategy for not fragmenting heritable land.

Only 2 of 186 societies in the standard sample are classed as polyandrous, and Westermarck wrote that polyandry is always an exception in the human race, usually modified toward monogamy by giving the first husband a chief position. Its rarity is itself a finding: it tends to appear where land is scarce and impartible, or where the sex ratio is skewed, and it coexists with other forms rather than replacing them.

Engels, following the missionary Fison, described Australian societies divided into two classes in which every man of one class was by birth the husband of every woman of the other. Later anthropologists read these as exogamous moieties, rules about whom one may marry, not as actual marriages of whole groups. The episode is a standing lesson in the difference between a category of eligible partners and a spouse.

A Bayesian study of 352 societies found lifelong religious celibacy mainly in monogamous societies with heritable wealth, and more common for whichever sex faces the higher marriage payment. The authors frame polyandry, monogamy, primogeniture and celibacy as alternative ways to stop siblings from fragmenting an estate. Rule: to understand a marriage form, look at what the family is trying to keep together.

George Reynolds was charged under section 5352 of the Revised Statutes, which made it bigamy for a person with a living spouse to marry another in a territory under federal jurisdiction. He asked that the jury be told a marriage entered under what he believed a religious duty should bring a verdict of not guilty. The trial court refused, the jury convicted, and the Supreme Court decided whether the Free Exercise Clause required that instruction.

The Court reasoned that laws exist to govern conduct: they cannot reach mere religious belief and opinion, but they can reach practices. It gave the examples of human sacrifice and of a widow burning herself on her husband's pyre to argue that a civil government may stop a practice even when the actor calls it a duty. The belief-action distinction became a foundation of free-exercise law for the next century.

The Court asserted that polygamy had always been odious among the northern and western nations of Europe and, before the Mormon church, had been almost exclusively a feature of Asiatic and African life. Modern scholarship reads that passage as the racial and religious attitude of its era rather than as a finding. Learning the case means separating its holding, which still governs, from its rhetoric, which does not.

A 2025 study used the full-count 1880 United States census to compare counties of the West with Mormon polygyny to other counties, finding that the share of unmarried men was lower, not higher, in the polygynous counties. The practice ended as a church institution in the 1890s under federal pressure, and groups that continue it today are separate from the mainstream church. It is part of the record of polygyny, not a stereotype of a faith.

S.B. 102 rewrote Utah Code 76-7-101. Purporting to marry someone while knowing one of you is legally married to another is bigamy, now an infraction. Inducing bigamy by false pretenses, threat or coercion is a third-degree felony, and bigamy combined with listed crimes such as abuse or trafficking is a second-degree felony. The change decriminalized a household arrangement without creating a legal plural marriage.

A state can stop prosecuting adults who live as a plural family while still refusing to license or record more than one marriage per person. Recognition would mean deciding how spousal rights divide among several spouses, which no United States jurisdiction has done, and the Respect for Marriage Act defines a marriage as between two individuals. Keep "legal" and "not prosecuted" apart when you read a headline.

Where a state applies different marriage codes to different religious communities, a man's right to a second wife can depend on the community his marriage is registered under. Islamic law permits up to four wives on condition of equal treatment, and some codes add a consent or court requirement. Customary law in parts of Africa recognizes polygynous marriage alongside civil monogamous marriage, so one country can have both.

The census analysis of 30 countries notes that polygyny's prevalence has declined substantially across sub-Saharan Africa while remaining common in some regions, and it works at the subnational level because practice varies so much within one state. The authors suggest that where polygyny is high, pro-marriage norms are also strong, so men there usually marry at higher rates, not lower. National averages hide this.

Polygamy is defined by marriage: a person has more than one spouse, with the rights and duties a marriage carries. Polyamory is defined by relationships: several intimate partnerships at once, each known to and agreed to by everyone involved, with no claim to marital status. Research on consensual non-monogamy belongs to the psychology of relationships; this chapter is about the institution. Confusing the two produces bad arguments on both sides.

Treaty bodies and non-governmental organizations have described plural marriage and related customs as harmful cultural practices, and have urged their abolition. Ethnographers reply that outcomes vary with age, wealth, kinship networks and residence, and that a practice can constrain women in one setting and protect them in another. Both positions appear in the open literature on bride price, and this curriculum presents them as positions.

A matriarchy would be a society governed by women. A matrilineal society traces descent through mothers. A matrilocal society has couples living with the wife's kin. A matrifocal household is organized around a mother, often with no resident husband. Any of the last three can exist with men holding political power, and in the ethnographic record they usually do. Reading "matrilineal" as "matriarchal" is the most common error in this subject.

Claims that matriarchies exist today point to societies that are matrilineal, matrilocal or both. In them women own houses, control lineage property or choose heirs, while councils, chiefships and warfare remain largely male. The absence of a documented matriarchy is a statement about the evidence, not about what is possible, and it leaves the real question open: how much authority do women hold in these systems, and over what?

Engels summarized Bachofen's four propositions: an original promiscuity, descent therefore traced only through mothers, a resulting rule of women, and a transition to monogamy and paternal law driven by religious ideas. Bachofen drew his evidence from Greek myth and drama, reading the Oresteia as the overthrow of mother right. No evidence supports a universal matriarchal stage, but the thesis shaped a century of theory.

Morgan described the Iroquois gens as descended through women and took that to be the original pattern everywhere. Engels argued that as herds and other wealth grew, men sought to pass property to their own children, overthrew maternal descent and inheritance, and that this downfall of maternal law was the historic defeat of the female sex. The sequence is a theory of patriarchy's origin, taught later as one position among several.

Gimbutas read female figurines and settlement patterns as evidence of a woman-centered Old Europe. Critics answered that figurines do not show who held power and that the reading projected a modern wish onto the past. Genetic studies also now support a large movement of people from the steppe into Europe in the third millennium BCE, which her model also predicted. Keep the two claims apart: migration is supported, matriarchy is not.

Fifty-seven genomes from Durotrigian burials revealed an extended kin group centered on a single maternal line, while unrelated burials were predominantly men who had married in. Comparing mitochondrial diversity across six millennia of European sites, the authors found British Iron Age cemeteries stand out for dominant matrilines. Roman writers had remarked on the standing of Celtic women; the genetics give that remark a structure.

At the Fujia site, ancient DNA, isotopes and burial layout show a matrilineal community with high endogamy, organized strictly by maternal clan, farming millet near the coast. The cemeteries were used for about 250 years, implying a stable matrilineal lineage of at least ten generations. The authors compare the pattern to modern multi-clan matrilineal societies, naming the duolocal Mosuo, the matrilocal Lahu and the Tlingit.

The Mosuo appear in the open literature as the living comparison for ancient matrilineal communities and in work on gendered labor, which finds that within matrilocal settings a man is less likely to help on his wife's farm when more women in her household are reproducing. Their system gives women control of house and lineage; it is not a matriarchy, since public authority has historically sat with men. A case study, not a utopia.

In a matrilineal system a man's lineage interests run through his sister's children while his affection and daily life may run through his own. Residence rules are the usual solution: avunculocality brings the nephew to the uncle, duolocality keeps the man in his natal home, and matrilocality moves him to his wife's. Each solution has costs, which is one explanation offered for why matriliny is less common than patriliny.

Interviews covering nearly 3,000 people across three generations found patrilocality the least common residence pattern in these Khoe-San-descendant communities, despite 150 years of colonial expansion, admixture and incorporation into a patrilocal settler society. The authors conclude that integration into the market economy, not colonial norms, is the main driver of change. Cultural traits of residence can be stickier than conquest.

The term describes a household's working center rather than a society's rule. It was coined for Caribbean families where women headed households and men circulated between them, and it applies wherever economic or legal conditions leave mothers as the stable core. A matrifocal household can exist in a patrilineal, bilateral or matrilineal society, which is why it belongs on the list of things "matriarchy" is not.

Morgan used "patriarchal family" narrowly for the Hebrew pastoral household under paternal power. Later sociology used the word for any system in which men hold a disproportionate share of authority and resources. Some scholars define it by law and institutions, others by norms and expectations, others by outcomes such as who owns land. Before judging a claim that a society is or is not patriarchal, find the definition the claim uses.

The open literature on gendered conflict defines gender norms as norms, rules or ideals that regulate conduct and distribute resources and power on the basis of sex. Many are injunctive: they say what one ought to do, not just what most people do. The definition matters for patriarchy because it gives a way to measure it: list the norms, ask what they allocate to whom, and compare across societies rather than arguing from a label.

In Engels's account, men came to hold growing herds while inheritance still passed through the mother's lineage, so a man's own children were disinherited. The desire to change that, he argued, overthrew maternal law, instituted descent through fathers, and produced a monogamous family founded on male supremacy to secure undisputed paternity. The theory ties patriarchy to property; critics note it rests on Morgan's stages, now abandoned.

Smuts applied sexual-conflict theory to humans, arguing that women's ability to resist male coercion is weakest where men control more resources, as after the shift to agriculture, and where patrilocal residence separates women from their kin. She also stressed culture, reading varied gender ideologies as shaped by selection. The collection that cites her reframes "sexual" conflict as "gendered" conflict to keep culture in view.

The evolutionary and the materialist accounts disagree about mechanism but converge on the same two levers. Where wealth is heritable and held by men, and where wives move to their husbands' kin, women bargain from weakness. Where women farm their own land and stay among their relatives, they bargain from strength. The agreement across rival theories makes these two variables the place to start when comparing societies.

Sexual-conflict theory predicts that the scarcer sex gains bargaining power. Applied to humans the prediction has had mixed success, and the open literature argues this is because gender norms regulate who is on the market, how they are valued and who makes reproductive decisions. Norms can also change the sex ratio itself. The lesson for patriarchy is that demography acts through norms, never directly.

The editors of the gendered-conflict collection changed their own terminology to counter what they call stubborn tendencies to essentialize differences between women and men, directing attention to cultural practices, sanctions and social learning. The critique is itself a position: it does not deny that selection acts on humans, but it holds that an account of patriarchy that ignores culture explains too little of the variation.

Because the concept has rival definitions, measurement means choosing indicators and saying so. Studies in the open literature compare women's control of material and educational resources across kinship systems, work hours in matrilocal and patrilocal settings, and reproductive decision-making power. An index is only as good as its indicators, and two indexes can rank the same societies differently. Treat a ranking as an argument, not a fact.

One line of research frames bridewealth as harmful, with experimental evidence linking complete payment to constraints on women's reproductive autonomy. Ethnographers of Ghana answer that within a wider patriarchal context the payment legitimizes a marriage and secures rights for the wife. Both appear in one collection. The point is that "patriarchal" can describe the context and still leave open what a particular custom does inside it.

This curriculum presents Engels's property theory, the sexual-conflict account, the norms-first critique and the measurement approach side by side because the field has not settled them. Each explains part of the variation across societies and each has known weaknesses. A learner's job is to state each position fairly, name what evidence would decide between them, and resist reading any of them as a political conclusion.

The payment moves from the husband's side to the wife's, and in the ethnographic record it is the commonest form of marriage transaction. In Melanesia most marriages involve transfers of traditional items such as pigs or mats and of money. Anthropologists prefer "bridewealth" to "bride price" since the transfer is usually not a purchase: in many societies the bride's family may waive it, and the marriage stays valid.

In patrilineal societies the payment establishes legal paternity and the father's rights over children, who join his lineage, while his kin provide the wife access to land and support. It also binds the two families, which is why relatives on both sides take an interest in keeping the marriage going. Reading it only as a purchase misses most of what it does.

Melanesian ethnography gives a double-edged picture. The payment can trap a woman in an unhappy marriage, since her kin may be asked to repay it on divorce and may press her to stay. It can also give her status and protection, because her relatives, having been paid, can bring claims against a husband who mistreats her. International bodies have called for bans; ethnographers caution that outcomes depend on age, kinship networks and residence.

Westermarck recorded wives obtained by service in Central America, Peru and elsewhere, and marriage "on credit", where a poor man pays cattle by installments and any children born meanwhile belong to the wife's father until redeemed. Bride service is common where wealth is not stored in herds or land and where the wife's family wants hands rather than goods. It often goes with a period of matrilocal residence.

Westermarck called it the marriage portion and listed its mixed purposes: a return gift, the wife's contribution to the joint household, and a settlement for her if the marriage ends by death or divorce. The husband commonly has the use of the portion while the marriage lasts, which makes it hard to say whose property it is. Dowry is associated with monogamy, heritable land and concern for a daughter's rank.

Westermarck traced how a purchase sum paid to the bride's father could be transformed into the morning gift, given to the bride by the groom, and the dotal portion given by her father. In both the wealth ends up with the wife as her security. The Islamic mahr belongs to the same family of payments: it is owed by the husband to the wife, not to her kin, and remains hers.

The marriage contract specifies the mahr. It may be paid at the wedding or partly deferred, and it belongs to the wife herself, not to her family. Because it is owed by the husband to the wife, it is classed with indirect dowry rather than with bride price. Courts in countries with civil family law have treated deferred mahr in various ways on divorce, which is a reminder that a religious payment can acquire a secular legal life.

Westermarck grouped marriage by exchange with marriage by purchase and noted that among the Kurnai of South Australia it was one of several routes, less common than elopement. Direct exchange removes the need for bridewealth and creates a symmetrical alliance between the two groups. Its drawback is rigidity: if one marriage fails, the other is threatened.

Mock capture, the bride's feigned resistance and the groom's party "carrying off" the bride appear across cultures. Nineteenth-century theory read them as survivals of a stage in which wives were taken by force. Westermarck found capture a real but never exclusive way of contracting marriage and warned against reading the rite as proof of a universal stage. The rite is best understood as dramatizing the bride's transfer between families.

A study of 352 societies found religious celibacy mainly in monogamous societies with heritable wealth, with male celibacy more common where bridewealth is high and female celibacy where dowry is high. The authors read monasteries and convents as a response to sibling competition over an estate: a child who does not marry needs no payment and does not split the land. Buddhism and Christianity are the traditions most likely to institutionalize it.

Westermarck described marriage by purchase decaying among civilized peoples: arbitrary presents, a sham sale, then a return gift, until the transfer became symbolic. The sequence is not a law of progress but a pattern seen where market economies, state law and new religions reach a society. Melanesian ethnography records the same process under Christianity, the colonial state and urbanization, with ambiguous results for women.

Westermarck gathered cases from every continent. In some, nothing so important as the disposal of a daughter could be done without the whole village's advice; in others a woman was free to choose, and marriages were made "sometimes with, and sometimes against" the wishes of elders. The question is not whether love exists but who holds the veto, and the answer can differ for sons and daughters, first and later marriages, rich and poor.

At one end, parents select and the couple meets at the wedding; at the other, the couple selects and informs the family. Most real systems sit between: parents propose candidates and the young person may refuse, or the couple chooses and seeks family approval. Consent is the hinge. A marriage the parties agree to, however arranged, is an arranged marriage; one they do not agree to is a forced marriage, treated in the human-rights chapter.

His chapters on sexual selection "influenced by calculation" list a wife chosen for her ability as a laborer, a husband chosen for his ability to protect and provide, a preference for a fertile bride, and the pull of class and caste. These are the criteria a family can evaluate, which is why arranged systems weigh them. Individual attraction is the criterion only the couple can judge, which is why choice systems weigh it.

A betrothal creates obligations before the marriage exists, and in many societies breaking one required compensation. Where it was made for infants, the alliance between families was the point and the children's future consent was assumed. Modern human-rights law treats the betrothal of a child as having no legal effect, a direct response to the practice.

Westermarck devoted chapters to courtship and recorded among the Kurnai that marriages were brought about most often by elopement. Elopement is a safety valve inside arranged systems: it lets a determined couple marry while leaving the rule intact, usually at the price of a later payment or reconciliation. Where it is frequent, "arranged" describes the ideal more than the practice.

In Obergefell the Court wrote that changes such as the decline of arranged marriages and the abandonment of coverture have worked deep transformations in the structure of marriage, and a separate opinion noted that arranged marriages have largely given way to pairings based on romantic love. Historians date the shift in Europe and North America to the eighteenth and nineteenth centuries, when a couple's choice began to outweigh a family's.

Westermarck argued that love and the appreciation of beauty were not peculiar to the civilized mind and found them in the folklore and conduct of peoples with every form of marriage. What varies is whether love is expected to precede the marriage, follow it, or stay outside it. In an arranged system it is often hoped for as a result; in a choice system it is required as a cause.

Westermarck noted the decline of national and class endogamy in modern society and the increase in mixed marriages, while treating religion as a continuing bar to intermarriage. Freedom of choice moves the mechanism from a rule enforced by the family to a pattern produced by where people meet and whom they find familiar. The pattern of who marries whom therefore changes more slowly than the rule does.

Families that arrange marriages tend to do so early, before the young person has formed attachments. Where individuals choose their own spouses, they marry later, after schooling and work. Since 1890 the Census Bureau has published the median age at first marriage in the United States, and the long rise in that number is one of the clearest traces of the shift from family to individual choice.

The Convention on the Elimination of All Forms of Discrimination against Women requires the same right to enter marriage and the same right freely to choose a spouse with free and full consent. That standard leaves arranged marriage lawful where both parties agree and condemns forced marriage where one does not. It is the line that most national marriage laws now draw, however unevenly it is enforced.

The Current Population Survey defines a household as all the people who occupy a housing unit, a house, apartment or group of rooms with separate living quarters and direct access. It includes relatives and any unrelated lodgers, foster children or employees who share the unit. Group quarters such as dormitories and institutions are excluded. The number of households equals the number of householders.

The family definition requires three things at once: a relationship by birth, marriage or adoption; co-residence; and a householder among the members. Related subfamily members count as members of the family. The number of families equals the number of family households, but family members and family household members differ, because the latter include non-relatives in the home.

A family household is one maintained by a householder who is in a family, and it may also contain unrelated people. A nonfamily household is a person alone or a householder with roommates, boarders or an unmarried partner only. The split matters for statistics: a cohabiting couple without children is a nonfamily household, so a rise in cohabitation shows up as a rise in nonfamily households.

The householder is the person who owns or rents the home in their name, or any adult member if there is none. Before 1980 the husband was always classified as the head when a couple lived together. The Bureau dropped "head" because shared responsibilities had made the term increasingly inappropriate, and a married couple may now name either spouse. Every other relationship in the household is recorded relative to that person.

A related subfamily is related to the householder but does not include the householder; an unrelated subfamily, once called a secondary family, is not related to the householder at all. Subfamilies are how the Census counts the stem and joint households that anthropology describes: a three-generation home appears as a family household containing a related subfamily. Related subfamilies are not counted as separate families.

The technical definitions were written for a different era and still read "husband and wife", while the Bureau's published tables now report same-sex married and unmarried couples separately. The "unmarried couple" category, limited to two adults, is the origin of the unmarried-partner statistics that track cohabitation. Read the definition's date before comparing figures across decades.

The definition counts a child as a stepchild only in relation to the householder, so a household in which the householder's spouse is the step-parent and the householder is the child's biological parent is missed. The Bureau says plainly that this undercounts step families. It is a good example of how a counting rule can make a common family form look rarer than it is.

A nuclear household holds a couple and their children. A stem household adds one married child and that child's family, keeping the line in the house across generations. A joint household holds several married siblings and their families together. "Extended" covers any household beyond the nuclear. Engels described the South Slavic zadruga, generations of a father's descendants farming one holding under an elected head, as the classic joint form.

Where co-wives are unrelated they are usually housed separately within a walled or fenced compound, each running her own hearth and raising her own children, while the husband rotates between them. Census categories built on the nuclear household handle such compounds badly, which is one reason comparative family statistics across countries need care.

A couple that is nuclear today may be a stem household when a married child moves in, a nonfamily household when the children leave and a spouse dies, and a subfamily if the survivor moves in with a child. The Census Bureau's own definitions, with their references to children under 18 and to marital status at enumeration, are built to classify a moment. Longitudinal data are needed to see the cycle.

The Bureau's December 2, 2025 release reports that married couples now maintain less than half of all households. The rest are single-person households, single-parent families, unmarried couples and other arrangements. The figure is a product of later marriage, more people living alone, and cohabitation, and it should be read with the household and family definitions above, which decide where each arrangement is counted.

In Windsor the Supreme Court said that the definition and regulation of marriage has historically belonged to the separate states, and that domestic relations are a virtually exclusive province of the states. Questions such as who may marry, the minimum age, the license required and how a marriage ends are state matters. The federal role is to set constitutional floors and decide how federal law treats a marriage.

Obergefell describes the centuries-old doctrine, citing Blackstone, under which a married man and woman were one legal person and that person was the husband: he controlled her property and earnings and she could not sue or contract alone. As women gained legal, political and property rights, the Court says, the law of coverture was abandoned. Married women's property acts in the nineteenth century did much of the dismantling, state by state.

Loving quoted Skinner v. Oklahoma for the proposition that marriage is fundamental to our very existence and survival and described the freedom to marry as one of the vital personal rights essential to the orderly pursuit of happiness. Obergefell reads Loving, Zablocki and Turner as holding that the Due Process Clause protects marriage as a fundamental right. A fundamental right does not mean no regulation; it means regulation must be justified.

Virginia banned only interracial marriages involving white persons, which the Court read as a measure to maintain white supremacy. It found no purpose independent of invidious racial discrimination. The Court held that limiting who may marry by race alone offends the core meaning of the Equal Protection Clause, and also that the laws took the Lovings' liberty without due process of law. Both grounds mattered later.

Wisconsin barred residents owing support for children not in their custody from marrying without a court order, available only on proof the support was paid and the children would not become public charges. Because the right to marry is fundamental and the law significantly interfered with it, the Court examined the state's interests critically and found the means unnecessarily impinged on the right; counseling need not block a wedding.

The Court listed what remains of marriage behind bars: an expression of emotional support and public commitment, a religious act for many, the expectation of a full marriage after release, and a precondition for benefits and property rights. A Missouri rule allowing inmate marriages only for compelling reasons failed even the deferential standard for prison regulations. The case fixed what the law regards as marriage's essential content.

Loving's facts turn on a related rule: Virginia law provided that a marriage solemnized elsewhere would be governed as if solemnized in the state, with cohabitation as evidence of the marriage. A common-law marriage is a real marriage with all its legal effects, which is why one formed in a state that allows it is generally recognized elsewhere. It is not the same as long cohabitation, which creates no marriage.

Edith Windsor married Thea Spyer in Canada; New York recognized the marriage, but DOMA's section 3 barred her from the federal estate-tax exemption for surviving spouses. The Court held that DOMA, with a directive reaching over a thousand federal statutes, was directed at a class of persons New York and eleven other states had sought to protect and was unconstitutional. The decision left each state's own definition untouched.

The Court gave four principles for why marriage is fundamental and found each applies: the right to personal choice in marriage is part of individual autonomy; marriage supports a two-person union unlike any other; it safeguards children and families; and it is a keystone of the social order. It overruled Baker v. Nelson and held the challenged state laws invalid insofar as they excluded same-sex couples from civil marriage on equal terms.

Public Law 117-228 replaced DOMA's 28 U.S.C. 1738C with a rule that no one acting under color of state law may deny full faith and credit to another state's marriage, or any right arising from it, based on the sex, race, ethnicity or national origin of the couple. The Act defines a federal marriage as one between two individuals valid where entered. Section 6 exempts nonprofit religious organizations from any duty to provide wedding services.

Obergefell's majority wrote that changes such as the decline of arranged marriages and the abandonment of coverture worked deep transformations in the structure of marriage, affecting aspects once viewed as essential, and that these new insights strengthened rather than weakened it. The dissents read the same history as a reason for caution. Be able to state both readings: the disagreement is about the meaning of change, not the facts.

Westermarck's chapter on the duration of marriage found the time a marriage lasts varies widely. Some peoples, such as the Andamanese as described by Man, were strictly monogamous with divorce unknown; among many others separation was easy for one or both spouses. He warned against the inference that easy divorce means no marriage: the looseness of the tie is a feature of the institution, not its absence.

Westermarck organized the record by who holds the power: divorce dependent on the husband's decision, a man permitted to divorce only under certain conditions, and marriage dissolved by the wife. Engels made the asymmetry a defining feature of the monogamous family, which could no longer be dissolved at the pleasure of either party and as a rule only by the man. Modern law in most countries gives both spouses the same grounds.

He recorded that among the Chukchi, the Basuto and the Munda Kols repudiated wives and their children were to some extent still supported by former husbands, and that a Burmese woman could demand a divorce if her husband failed to maintain her. The idea that a man is bound to maintain his family survives the marriage. That duty is the ancestor of modern maintenance and child-support orders.

Under a fault regime the grounds are specific wrongs such as adultery, cruelty or desertion, and a spouse who cannot prove one cannot divorce. Under a no-fault regime either spouse may obtain a divorce by asserting irreconcilable differences or a period of separation. The shift moves the state from judging a marriage's morals to recording its end, and it changes bargaining between spouses because neither can withhold consent.

Annulment applies where a requirement of a valid marriage was missing from the start: a party was already married, under age, too closely related, or did not consent. Divorce presupposes the marriage was valid. The difference matters for property, the legitimacy of children, and religious standing. That is also why some faiths that forbid divorce will still grant annulments.

In Jewish law a divorce requires a bill of divorce, the get, delivered by the husband; in Islamic law a husband may pronounce talaq and a wife may seek a judicial divorce; Catholic practice forbids divorce and uses annulment. Where civil and religious law both apply, a person can be divorced in one system and still married in the other. The world_religions curriculum carries each tradition's own rules; this curriculum carries the interaction.

NCHS reports that the collection of detailed marriage and divorce data was suspended beginning in January 1996 for budgetary reasons and because of limits in what states collected. National rates are now calculated from provisional state counts, and the divorce rate covers only the reporting states, 45 and the District of Columbia for 2023. The Census Bureau's surveys fill the gap with self-reported marital history.

A crude comparison of 672,502 divorces to 2,041,926 marriages in 2023 gives about a third, and earlier decades gave figures near one half. But the people divorcing in a year married in many earlier years, so the ratio is not a probability that a marriage will fail. The honest measure follows a marriage cohort over time, which is what the National Survey of Family Growth was built to do.

Melanesian ethnography records that customary divorce brings in both families and that the husband's kin may demand repayment from the bride's kin, who then pressure her to stay. The effect is strongest where the payment was large. The same institution that gives a wife her relatives' protection within a marriage can make leaving it depend on their willingness to pay.

NCHS reports 2,041,926 marriages in 2023 and a rate of 6.1 per 1,000 population. The denominator includes children and the already married, so a population with few young adults will show a low rate even if nearly everyone marries. The rate is useful for trends within one country and poor for comparing countries with different age structures. Rates refined by age and marital status solve that problem but need more data.

NCHS reports 672,502 divorces in 2023 from 45 reporting states and the District of Columbia, a rate of 2.4 per 1,000. Because several states do not report, the national count is incomplete and the rate is computed over the reporting population. Trend tables run from 2000 to 2024 by state. When a headline gives a national divorce figure, check whether it includes every state.

NCHS describes its own data as records-based and the Census Bureau's as survey-based. Records give exact counts of marriages and divorces but little about the people; surveys give marital histories, ages, education and cohabitation but depend on recall and sampling. The National Survey of Family Growth sits in between, a survey run by NCHS designed to measure cohabitation, marriage, divorce and remarriage directly.

The National Survey of Family Growth estimates the probability of first marriage to an opposite-sex spouse before given ages. For 2017 to 2019 the figures are 2 percent of women and 1 percent of men by 18, 34 and 24 percent by 25, 56 and 50 percent by 30, and 76 and 75 percent by 40. The gap between women and men narrows with age because men marry later. These are the numbers behind the phrase "later marriage".

The survey distinguishes ever married from currently married and splits the latter into first marriages and second or later marriages: 33.6 percent of women were in a first marriage and 5.3 percent in a later one. The ever-married share is lower than older generations' because the sample includes people in their teens and twenties who have not yet married. Age ranges decide what a percentage means.

The survey reports current cohabitation at 13.9 percent of women and 12.0 percent of men, and ever cohabited at 58.9 and 54.3 percent. Cohabitation has become a common stage before or instead of marriage, and because a cohabiting couple without children is a nonfamily household in Census terms, its rise reshapes household statistics as well as marriage statistics.

The Bureau's historical marital status tables report the median age at first marriage for men and women from 1890 onward, alongside marital status by sex. The median is the point at which half of the people who will ever marry have already done so, and it rises when people delay marriage for education and work. Read it together with the survey probabilities above: a rising median with a stable eventual share means postponement, not abandonment.

A person is "married, spouse present" if the spouse is a household member, even if temporarily away. "Separated" includes legal separations and people living apart because of marital discord. "Other married, spouse absent" covers spouses living apart for work or other reasons. Because separated people are still married, a count of the married overstates intact marriages unless the subcategories are shown.

The survey counted 5.3 percent of women and 6.7 percent of men aged 15 to 49 as married for the second time or more in 2017 to 2019. The Census step family definition counts a stepchild only relative to the householder, which the Bureau says undercounts step families. Remarriage is one of the places where the family a person lives in and the family the statistics see diverge most.

A December 2, 2025 release states that fewer than half of households in 2025 were married couples. The figure combines several trends measured elsewhere in this chapter: later first marriage, more people living alone, cohabitation counted as nonfamily households, and longer lives that leave widowed people in one-person households. It is a statement about households, not about the share of people who ever marry.

The definition is by age, not by whether the marriage was arranged or consented to, and it treats formal marriages and informal unions alike. The 640 million estimate is cited in the open literature as the global stock of women and girls who married as children. The practice is recognized as a form of gender-based violence and as an indicator of gender inequality, with consequences for health, education and autonomy.

A study of 12.9 million person-years of survey data found child marriage present in all 14 countries examined, with the lowest proportion in the Philippines at 20.9 percent and the highest in Bangladesh at 76.9 percent. Such ranges show that national law and national prevalence can diverge sharply, and that regional and rural-urban differences within a country are often as large as differences between countries.

Reports from aid organizations had documented child marriages following cyclones and floods, driven by lost livelihoods, displacement and interrupted schooling. The systematic test found little change in national child-marriage rates in the three years after a cyclone, with a small increase in Mozambique and hints among urban populations in two other countries. The lesson is that poverty, schooling and norms, not single shocks, set the rate.

Human-rights law does not condemn arranged marriage as such. It requires that each party enter marriage only with free and full consent, and it treats a marriage contracted under duress, fraud or incapacity as forced. A child is presumed unable to give the consent required, which is why child marriage and forced marriage are addressed together even though an adult arranged marriage is lawful.

The article lists equal rights to enter marriage, to choose a spouse with free and full consent, and the same rights and responsibilities during marriage and at its dissolution, including over children and property, and the right to decide freely on the number and spacing of children. Its final paragraph requires states to set a minimum age and to make registration compulsory. It is the treaty text most often cited against child marriage.

A statute can set 18 as the marriage age and yet allow a younger marriage with parental consent, a judge's order, or a pregnancy. Campaigns to end child marriage therefore focus on closing exceptions rather than on the headline age. In the United States each state sets the age and the exceptions, and they have been changing, so a state-by-state list only captures one moment and needs a date.

The Convention pairs its minimum-age requirement with a requirement that marriage registration be compulsory. Registration creates a record that a registrar can refuse to make when a party is under age, and it gives a wife proof of her marriage when she needs maintenance, inheritance or custody. Where customary or religious marriages go unregistered, the state's minimum age has no point of enforcement.

Treaty bodies and non-governmental organizations have called bride price a harmful cultural practice and demanded that it be banned or made unenforceable, and have treated child marriage as gender-based violence. Ethnographers accept the harm findings for child marriage while arguing that bride price's effects depend on context. A learner should distinguish a practice that denies a child's consent from one whose effects on adults are contested.

Schneider claimed that the comparative study of kinship took the Western idea of relatedness through biological ties as universal and then found it everywhere by definition. His call to abandon kinship studies did not end them; instead the field had to ask what counts as relatedness in each society's own view. The 2026 special issue on "kinship trouble" marks its fortieth anniversary as a turning point rather than an ending.

Students and co-editors of Schneider, at a conference on feminism and kinship theory, advocated treating gender and kinship together rather than as separate fields. The argument is that categories like wife, mother and sister are gender categories as much as kin categories, and that rules about descent and marriage are also rules about what women and men may do. It is more constructive than Schneider's, because it says what to study instead.

Descent theorists, mostly British, built their models on lineages and clans; alliance theorists, following Levi-Strauss, held that culture rather than biology binds social groups and that marriage exchange is what makes a society a whole. The "kinship trouble" review treats the prioritizing of either descent or alliance as a mistake and proposes gender and the house as alternative starting points.

The definition deliberately covers birth, marriage, adoption, shared food, co-residence and ritual alike, because in many societies any of these can make kin. It also leaves room for the finding, noted in the review, that relatedness may extend beyond humans in some cosmologies. Its critics say it is too broad to compare societies; its defenders say that breadth is the point, since the narrow definition was the Western one.

Archaeogenetics can show who was buried beside their genetic relatives and who married in; it cannot show who was counted as kin. The review calls for integrating biological and social approaches and for ethical collaboration with descendant communities. Its evidence includes informal adoption of enslaved children, where co-residence made kin, as a case of "kinning" that genes would miss.

A child may have a genetic mother, a woman who carried the pregnancy and a mother who raises them, and the law of each country decides which one is the legal parent. Adoption has always done this for social parenthood; reproductive technology adds gestational parenthood to the list. These cases are the modern form of the old distinction between the social father and the biological one, and they show that distinction was never only about the past.

The review highlights the diversity and complexity of kinning practices, from adoption to shared residence. Families of choice, in which unrelated people take on the functions of kin, belong here as a kinship category; the attraction_connection curriculum treats their role in wellbeing. The concept matters because it makes kinship a process rather than a status, which is what Schneider's critique and Sahlins's definition both pointed toward.

The "kinship trouble" program names ethical collaboration as one of its three themes and frames kinship studies as acts of care, engaging academic institutions, museums and communities. Ancient DNA from a cemetery makes claims about people whose descendants may be living nearby and who may reckon kin in ways the genetic model does not capture. Method and ethics are the same problem here.