Ross Douthat warns that same-sex marriage can indeed weaken the institution of marriage for everyone, especially in further removing us from the ideal of marriage as a procreative union. Kevin Drum responds by saying that was never what marriage was about. Drum’s wrong, and Douthat might be right—he notes that as same-sex marriage has became increasingly accepted, “the marriage rate has been falling faster, the out-of-wedlock birthrate has been rising faster, and the substitution of cohabitation for marriage has markedly increased.”

Homosexuals are not the cause of this, as Douthat notes—“Underlying these trends is a steady shift in values”—but it’s possible that gay marriage will contribute to the redefinition of the marriage ideal as one that doesn’t necessarily include children. Damon Linker provides some context for that ongoing redefinition. The relationship between marriage and childbearing has severed by stages, and the key battles were lost by the long before today. When exactly?

Between five and six decades ago, to be precise. That’s when the birth control pill—first made available to consumers for the treatment of menstrual disorders in 1957 and approved by the FDA for contraceptive use three years later—began to transform sexual relationships, and hence marriage, in the United States. Once pregnancy was decoupled from intercourse, pre-marital sex became far more common, which removed one powerful incentive to marry young (or marry at all). It likewise became far more common for newlyweds to give themselves an extended childless honeymoon (with some couples choosing never to have kids).

The Supreme Court’s 1965 Griswold v. Connecticut decision, which held that married couples could purchase prophylactics—and discovered broader rights to privacy in the Constitution’s “penumbras” and “emanations”—gave this development the force of law. Linker argues that same-sex marriage is an effect, not the cause, of marriage’s redefinition.

Against this backdrop, the case against same-sex marriage might still have had a fighting chance if opponents consistently kept to procreation as a defining purpose of marriage. The public and the state do have an interest in the continuation of the citizenry, after all. (Although at this point the nation’s governing class seems happy to substitute immigration for sexual reproduction.) But Charles Cooper, defending Proposition 8 before the Supreme Court, surrendered on this front:

JUSTICE KAGAN: Well, Mr. Cooper, suppose a state said that, “Because we think that the focus of marriage really should be on procreation, we are not going to give marriage licenses anymore to any couple where both people are over the age of 55.” Would that be constitutional?

COOPER: No, Your Honor, it would not be constitutional.

Lawyers make tactical errors, but even the theoretical defenders of procreative marriage have yielded. Linker paraphrases the natural-law argument of Princeton’s Robert P. George:

Whether or not a particular male-female couple can produce a child is irrelevant. In cases of infertility due to medical defect or advanced age on the part of one or both members of the marriage, the union falls short of reaching its goal but remains oriented to that goal nonetheless. (The union would produce a child if the bodies of both members were functioning as they should.)

Within a natural-law tradition this is a strong argument: what matters is the abstract capacity to reproduce, not actual reproduction. (That may be a concession already, but it certainly doesn’t concede everything.) Outside of a natural-law tradition, though, there isn’t an obvious reason why this is the correct kind of abstraction. That is, why should it be more valid to say, “if this man and woman were not past child bearing age, they could bear children,” than to say, “if these two women were not both women, they could bear children”? Because natural law does, in fact, have a tradition, it’s plain that one of these scenarios is formally correct and the other is not. But the Supreme Court doesn’t follow traditional natural law, so it has no reason for preferring one impossibility over another.

The strong argument for procreation as essential to marriage would have to say that yes, the state can choose not to grant marriage licenses to people who categorically cannot conceive, whether they are old, of the same-sex, or physically incapable. There would then have to be some rationale for applying this legitimate prohibition in some cases (against same-sex couples) but not in others (nobody is seriously suggesting denying marriage licenses to AARP members).

There is, of course, a good reason why even Robert George and Charles Cooper don’t argue that heterosexuals who cannot bear children may be denied access to the institution of marriage: because marriage is not only about children, and in fact the West has a long history of balancing priorities between marriage-as-about-children (proles) and marriage-as-about-love-between-two-people (fides)—until now, specifically a man and a woman. Carle Zimmerman’s Family and Civilization, oft-cited here, sketches that history. Although the relative weights of proles and fides have shifted over time, both have been definitive components of the ideal until now.

One weakness of the traditionalist argument has been its failure to adhere strictly enough to proles. But the failure to give fides its due has perhaps contributed at least as much to the rout. Can anything other than marriage, if homosexuals are to be excluded from that, accommodate fides? Would any alternative be acceptable to others who cannot bear their own children—“domestic partnerships” for the aged? The question answers itself.