When the marriage vow was drafted, "till death do us part" meant, on average, about fifteen years. A peasant couple in fourteenth-century Europe could expect their union to end with one of them in a churchyard before either turned forty. The vow was a small promise, manageable, witnessed by the certainty of early death. Industrial sanitation, antibiotics, and vaccination doubled the expected duration of marriage in the twentieth century. The longevity revolution now underway, biological aging treated as a malleable phenomenon rather than a fixed sentence, may double it again. A couple marrying at thirty in 2040 could plausibly celebrate their seventieth anniversary at one hundred. The vow they speak will be, in functional terms, a different vow from the one their grandparents spoke. We have not yet decided what it means.
The collective implications are larger than the individual ones. A society in which half of its marriages last seventy years is structurally different from one in which the median marriage lasts twenty-five and ends in widowhood or divorce around year forty. Inheritance patterns extend. Sibling rivalries between adult children in their seventies persist while parents in their hundreds still hold property. The cousin networks of nonagenarians become the practical unit of family. The labor of late-life caregiving, currently borne by adult daughters in their fifties, is borne instead by spouses in their nineties. Real estate markets, which currently rely on intergenerational transfers in the donor's seventies, lock up. Pension systems, designed for thirty years of contribution and fifteen of withdrawal, invert.
Inside the marriage itself, the question is whether the institution can absorb the new duration. The honest answer is that we do not know, because no human institution has ever been asked to hold the same two people in voluntary partnership for seven decades. Andrew Cherlin's data show that the median modern marriage already strains at twenty-five years; the second half of a long marriage is statistically more fragile than the first, with divorce rates rising in the 65+ cohort even as overall rates decline. Add forty more years and the strain may snap the form, or it may, with the right cultural scaffolding, find a new equilibrium.
There are three plausible futures. In the first, marriage becomes serial by default: the seventy-year monogamy is replaced by a sequence of three or four twenty-year unions, each appropriate to a phase of the extended life. In the second, marriage bifurcates: a small minority achieves the seventy-year companionate ideal while the majority cycles through shorter arrangements, with the long marriage acquiring the prestige a fifty-year marriage carries today. In the third, the institution itself is renegotiated: the vow is rewritten to include scheduled renewals every fifteen or twenty years, with the option to dissolve, renew, or reconfigure. This third future is currently the least visible and probably the most likely, because it is what couples are already doing informally under the cover of the old vocabulary.
The biological substrate matters here. Andrew Steele and the broader longevity research community have spent a decade arguing that aging is a tractable disease, that senescent cell clearance, partial reprogramming, and metabolic intervention can extend not just lifespan but healthspan, the years lived in functional vigor. If this program succeeds even partially, the seventy-year marriage will be conducted between two people who remain biologically active, sexually interested, and cognitively present into their nineties. This is a fundamentally different prospect from a long marriage conducted in shared decline. The current cultural model of the long late marriage, two affectionate elders holding hands in a garden, is replaced by something we have no template for: two vital adults negotiating a seventh decade of erotic, intellectual, and economic partnership.
The collective scaffolding required is substantial and largely absent. Legal regimes treat marriage as a binary state; we will need gradients. Tax codes assume one or two retirements per couple; we will need to accommodate three or four sequential careers. Housing stock is built for nuclear families with twenty-year occupancy; we will need flexibility for unions that move through multiple phases. Religious institutions, the primary cultural authors of the marriage script, will need to revise their texts or watch the texts go unused. None of this is happening at the speed of the underlying biological change.
The Fifth Law, revision, lands here with unusual force. The marriage vow is one of the oldest continuously used pieces of cultural technology. It has been revised before, the abolition of coverture, the legalization of divorce, the extension to same-sex couples, but each revision has lagged the underlying change by decades. The longevity revision is bigger than any of these, because it is not a change in who can marry or under what terms; it is a change in the duration that any marriage must, if successful, sustain. We can revise the vow consciously, or we can let it be revised by the gap between the words and the lives they no longer describe.