Marriage exists, in other words, to solve a problem that arises from sex between men and women but not from sex between partners of the same gender: what to do about its generativity. It has always been the union of a man and a woman (even in polygamous marriages in which a spouse has a marriage with each of two or more persons of the opposite sex) for the same reason that there are two sexes: It takes one of each type in our species to perform the act that produces children. That does not mean that marriage is worthwhile only insofar as it yields children. (The law has never taken that view.) But the institution is oriented toward child-rearing. (The law has taken exactly that view.) What a healthy marriage culture does is encourage adults to arrange their lives so that as many children as possible are raised and nurtured by their biological parents in a common household.
That is also what a sound law of marriage does. Although it is still a radical position without much purchase in public opinion, one increasingly hears the opinion that government should get out of the marriage business: Let individuals make whatever contracts they want, and receive the blessing of whatever church agrees to give it, but confine the government’s role to enforcing contracts. This policy is not so much unwise as it is impossible. The government cannot simply declare itself uninterested in the welfare of children. Nor can it leave it to prearranged contract to determine who will have responsibility for raising children. (It’s not as though people can be expected to work out potential custody arrangements every time they have sex; and any such contracts would look disturbingly like provisions for ownership of a commodity.)
When a marriage involving children breaks down, or a marriage culture weakens, government has to get more involved, not less. Courts may well end up deciding on which days of the month each parent will see a child. We have already gone some distance in separating marriage and state, in a sense: The law no longer ties rights and responsibilities over children to marriage, does little to support a marriage culture, and in some ways subsidizes non-marriage. In consequence government must involve itself more directly in caring for children than it did under the old marriage regime — with worse results.
Thoughtful proponents of same-sex marriage raise three objections to this conception of marriage. The first is that law and society have always let infertile couples marry; why not treat same-sex couples the same way? The question can be tackled philosophically or practically. The philosophical answer boils down to the observation that it is mating that gives marriage its orientation toward children. An infertile couple can mate even if it cannot procreate. Two men or two women literally cannot mate. To put it another way: A child fulfills the marital relationship by revealing what it is, a complete union, including a biological union. A man and a woman who unite biologically may or may not have children depending on factors beyond their control; a same-sex couple cannot thus unite.
The practical problems with using fertility as a criterion for marriage should be obvious. Some couples that believe themselves to be infertile (or even intend not to have children) end up having children. Government could not filter out those marriage applicants who are certain not to be able to have children without extreme intrusiveness. Note that we do not generally expect the eligibility criteria and purposes of marriage to exhibit a rigorous fitness in other respects. This is true about those aspects of marriage about which proponents and opponents of same-sex marriage alike agree. Nobody believes that people should have to persuade the government that they really are capable of a deep emotional union or that they are likely to stick around to take care of an ill partner before getting legally married, because that would be absurd. Nobody would try to devise a test to bar couples with no intention of practicing sexual exclusivity from getting married. It does not follow that marriage is therefore pointless or has nothing to do with monogamy, emotional union, or caregiving. (Those are indeed goods that marriage advances; but if sex did not make children, they would not be a reason to have the institution of marriage.)
The second objection proponents of same-sex marriage raise is that the idea that marriage is importantly linked to procreation is outdated. In our law and culture, the ties between sex, marriage, and child-rearing have been getting weaker thanks to contraception, divorce and remarriage, artificial reproduction, and the rise of single motherhood. Yet those ties still exist. Pregnancy still prompts some couples to get married. People are more likely to ask nosy questions about whether and when children are coming to couples that have gotten married. And we have not at all outgrown the need to channel adult sexual behavior in ways conducive to the well-being of children: The rising percentage of children who are not being raised by their parents, and the negative outcomes associated with this trend, suggest that this need is as urgent as ever. Our culture already lays too much stress on marriage as an emotional union of adults and too little on it as the right environment for children. Same-sex marriage would not only sever the tie between marriage and procreation; it would, at least in our present cultural circumstances, place the law behind the proposition that believing that tie should exist is bigoted.
The third objection is that it is unfair to same-sex couples to tie marriage to procreation, as the traditional conception of marriage does. Harm, if any, to the feelings of same-sex couples is unintentional: Marriage, and its tie to procreation, did not arise as a way of slighting them. (In the tradition we are defending, the conviction that marriage is the union of a man and a woman is logically prior to any judgment about the morality of homosexual relationships.)