Marriage feels timeless, which is exactly why its history is so disorienting. Almost everything we now consider essential to it (love, choice, equality, even the couple deciding for themselves) is recent. Here is the long arc, painted one era at a time.
Marriage predates writing, so its origins can only be inferred, but by the time law codes appear it is already a thoroughly economic institution. The Code of Hammurabi, from Babylon around the eighteenth century BC, devotes long sections to marriage: bride price, dowry, inheritance, the financial consequences of divorce and adultery. What it does not discuss is affection. Marriage existed to bind families, transfer property, organize labor, and produce legitimate heirs whose claims to inheritance nobody could contest.
The historian Stephanie Coontz, whose book Marriage, a History remains the best single survey of this territory, puts the point sharply: for millennia, marriage was too important an economic and political contract to be left to something as unstable as love. Marrying for love, where the idea appears at all in ancient sources, is often treated as slightly irresponsible, like choosing a business partner by infatuation.
This version of marriage was also, in most of the ancient world, openly unequal. Wives were in varying degrees legal dependents of husbands or fathers. Polygyny was common across many societies for men who could afford it. The couple's own preferences ran a distant second to the interests of their families.
The shape then: a merger of households. Love was optional and slightly suspect; the alliance was the point.
Roman law contributed the idea that would eventually remake everything: consent makes the marriage. Classical Roman marriage in its later, freer form required no ceremony, no priest, and no state registration; it existed because a man and a woman regarded each other as husband and wife, and it could be ended by either party. It remained deeply patriarchal and thoroughly entangled with property, but the legal core was agreement, not ritual.
This mattered enormously later, because medieval church lawyers inherited Roman legal thinking. When they asked what actually creates a marriage, the answer they settled on was the Roman one: the consent of the two people themselves.
The seed of everything modern: agreement, not ceremony, makes the marriage. Every later revolution grows from this root.
Early Christianity was surprisingly lukewarm about marriage. Paul famously conceded that it is better to marry than to burn, but celibacy held the higher spiritual rank for centuries. The church did not initially perform weddings at all; marriage remained a family and civil matter that Christians happened to enter.
Over the medieval period this reversed. Marriage was drawn step by step into the church's jurisdiction: blessed at the church door, then regulated by canon law, then counted among the seven sacraments (a status affirmed through the twelfth and thirteenth centuries and formally defended at the Council of Trent). By the high Middle Ages, church courts across Europe were the venue where marriage disputes were decided.
And here the consent doctrine had a quietly radical consequence. If the free consent of the two individuals is what makes a marriage, then in principle a marriage contracted against the parents' wishes, or across class lines, was still a marriage. Church courts sometimes upheld exactly such unions. The doctrine planted, deep inside a hierarchical institution, the seed of the idea that marriage belongs to the couple. (The machinery the church built to manage the resulting chaos of secret marriages, banns and witnesses and priest interviews, is a story we tell in the companion post.)
The radical seed: if the couple's yes is what makes the marriage, then the couple's yes outranks the family's plans.
Early modern states and churches, Protestant and Catholic alike, spent the sixteenth through eighteenth centuries formalizing marriage: required ceremonies, registers, parental consent rules for minors. England's Hardwicke Marriage Act of 1753 is the emblematic case, ending centuries of ambiguity by requiring a formal church ceremony for a valid marriage. Marriage was becoming what it had never quite been before: a standardized legal status administered by institutions, with paperwork.
Throughout all of this, the economic logic still ruled. Among the propertied, marriage negotiations were openly financial. Among everyone else, a spouse was above all a work partner; a farm or a workshop could not run without one. The household was the unit of survival, and marriage was how you formed one.
The shift: marriage becomes paperwork. Standardized, registered, provable, and administered by institutions rather than families.
The genuinely disruptive idea arrived in the late eighteenth and nineteenth centuries, carried by wage labor, cities, and Enlightenment individualism: people should choose their own spouses, and they should choose for love. Once individuals could earn wages independent of their families' land, parents lost their economic veto. The companionate marriage, the spouse as best friend and chosen partner rather than assigned collaborator, became first thinkable and then expected.
Coontz's central, counterintuitive argument is that this victory of love destabilized marriage permanently. The moment marriage's justification became the quality of the relationship, a bad relationship became a reason to leave. Every subsequent development that alarmed traditionalists (rising divorce, cohabitation, staying single) follows from the love-based premise itself. If you marry for love, you can un-marry for its absence. The nineteenth century made the modern divorce debate inevitable.
The paradox: love wins the argument, and by winning, quietly makes divorce thinkable. The two arrive together and cannot be separated.
The love revolution took another century and a half to reach the law. Under the common-law doctrine of coverture, a married woman's legal identity was absorbed into her husband's: she could not own property, sign contracts, or keep her own wages. The Married Women's Property Acts of the nineteenth century began dismantling this, but the process ran long into the twentieth century. Married women in parts of the West needed a husband's permission to open a bank account within living memory. Marital rape remained legally impossible as a category in much of the United States until the 1970s and 1980s, on the theory that consent was given permanently at the altar.
It is worth sitting with how recent this is. The marriage of equals, two people with the same legal standing inside the relationship, is not the ancient tradition. On the scale of marriage's whole history, it is roughly one lifetime old.
Worth sitting with: the marriage of equals is the youngest version of the institution, not the traditional one.
The second half of the twentieth century completed the shift from marriage as obligation to marriage as choice. No-fault divorce, beginning with California in 1969 and spreading across the West, meant a marriage could end because one person wanted it to end. Reliable contraception separated marriage from childbearing. Women's mass entry into paid work meant marriage was no longer an economic necessity for survival. Cohabitation stopped being scandalous and became the normal first step.
Divorce rates surged in the 1970s and have since declined from their peak, partly because marriage itself became more selective. The sociologist Andrew Cherlin describes the change as marriage moving from cornerstone to capstone: it was once the foundation you laid at the start of adult life, and it is now the monument you place on top of an already-built one. People marry later, after education, career establishment, and usually cohabitation.
The new reality: staying married is a daily choice, not a legal fate. The relationship has to earn its own continuation.
In 2001 the Netherlands became the first country to open marriage to same-sex couples, and the United States followed nationwide in 2015 with Obergefell. After thousands of years in which marriage was defined by property, family alliance, and legal hierarchy, the surviving definition is the simplest one: two people who choose each other and commit.
The settling point: everything else fell away. What remained, and what the law now protects, is the committed couple itself.
Which brings us to now. The psychologist Eli Finkel calls the modern arrangement the all-or-nothing marriage. We ask one person to be what a whole village once was: lover, best friend, co-parent, financial partner, intellectual companion, growth coach. Marriages that meet those expectations are almost certainly the best marriages that have ever existed. Marriages that fall short feel more disappointing than ever, because the bar is so high and because leaving is now possible.
Notice the through-line of the whole story. Every century removed a layer of external scaffolding: the family alliance, the economic necessity, the legal compulsion, the social stigma of leaving. What remains is the relationship itself, load-bearing and unassisted. Marriage did not get weaker; the supports were removed, so the quality of the bond is now the whole structure.
If marriage were still an economic arrangement enforced by family, church, and law, compatibility would be a minor detail; the scaffolding would hold regardless. It is precisely because the modern marriage runs entirely on the quality of the relationship that the patterns two people bring to it (how they fight, how they repair, what they assume about money and children and loyalty) now decide the outcome. History removed every other predictor. What is left is the two of you.
Why this history matters here: the institution has never asked more of couples, and it has never given them better tools to prepare. That is the case for examining the relationship seriously, while the decision is still genuinely open.
A note on sources. The historical arc here follows Stephanie Coontz's Marriage, a History (2005), with the cornerstone-to-capstone framing from Andrew Cherlin and the all-or-nothing analysis from Eli Finkel's The All-or-Nothing Marriage (2017). Specific legal landmarks (Hardwicke Act 1753, California no-fault divorce 1969, Netherlands 2001, Obergefell 2015) are matters of record. This is a synthesis for general readers, not original scholarship; where historians disagree, we have stayed with the mainstream account.