At his New York Times blog, Ross Douthat has been doing a yeoman’s work, making me almost regret my critique of his essay on gay marriage by offering a patient, sophisticated case for preserving the “ideal” of heterosexual marriage.

Specifically, I was pleased to see him affirm my point that the legal affects the culture in addition to reflecting it, a point often lost on my peers. One of my favorite moments is when he turns the civil rights narrative on his opponents to prove the point:

Second, I think that most of Greenwald’s examples of cultural norms that aren’t legally enforced actually tend to back up my belief that law and culture are inextricably bound up, rather than his case that they needn’t be. A stigma on racism, for instance, would hopefully exist even in a libertarian paradise, but it draws a great deal of its potency from the fact the American government has spent the last 40 years actively campaigning against racist conduct and racist thought, using every means at its disposal short of banning speech outright. The state forbids people from discriminating based on race in their private business dealings. It forbids them from instituting policies that have a “disparate impact” on racial minorities. It allows and encourages reverse discrimination in various settings, the better to remedy racism’s earlier effects. It promulgates public school curricula that paint racism as the original sin of the United States. It has even created a special legal category that punishes crimes committed with racist intentions more severely than identical crimes committed with non-racial motivations. In these and other arenas, there isn’t a bright line between the legal campaign against racism and the cultural stigma attached to racist beliefs; indeed, there isn’t a line at all.

Douthat suggests that as gay marriage is legalized, the stigma against those who have old-fashioned views of marriage will increase and the line between culture and law will continue to blur. That’s near the center of the social conservative anxiety over gay marriage, despite the Constitutional protections that are in place to preserve the “freedom of religion.” Like freedom of speech, the ability to practice our religion doesn’t seem to be absolute when the “harm” of other individuals is in question.

But the heart of Douthat’s case—so far—is his description of heterosexual relationships as “thick,” which I joked over at Mere Orthodoxy is a term philosophers use when they have nothing else to say. The argument is that the gender differences and procreative impulses add an additional layer of complexity to heterosexual relationships that distinguishes them from homosexual relationships, which inevitably have to be characterized by “love and commitment”—and nothing more.

On this account, the slippery-slope argument isn’t a social argument, but what I would call an “in-principle” argument. There is nothing that stops the “logic of gay marriage [from being] extended to encompass all kinds of relationships that we definitely don’t want to call marriages.”

That’s a more interesting way of framing the debate over marriage, in that it refuses to bow to the pressure to identify a single good as determinative of heterosexual relationships and instead identifies a inextricably linked cluster of attributes that demarcates them as unique.

But in terms of the state’s interest, Douthat seems to come down where social conservatives always have: procreation.

And the fact that this interplay determines how and when and whether the vast majority of new human beings come into the world is what makes it possible to argue — not necessarily convincingly, but at least plausibly! — that both state and society have a stronger interest in the mating rituals of heterosexuals than in those of gays and lesbians.

There’s a key point lurking here which he doesn’t make, but is worth teasing out: the complex relationship of procreation, gender differences, and reproductive impulses that is heterosexual marriage exists “pre-politically.” It is a relationship that only needs a man and a woman to exist, without any need for the surrounding society or its institutions, even while they might be interested in it.

The same, I suggest, cannot be said for gay marriage. It’s tempting to respond to Douthat by suggesting that the possibility of gay adoption resolves the “thickness” objection by introducing children into the equation. But that ignores the structural difference between gay unions and heterosexual unions. The former necessarily depend upon some third party for the introduction of children, either through adoption, artificial insemination, or some other method. They cannot be a child-bearing couple, and to be a child-rearing couple they have to get someone from outside the relationship involved. So while the adoption of children may make gay unions more “complex,” and hence more similar to heterosexual unions on that score alone, it is a complexity that is not intrinsic to the relationship itself, but that lives between the gay couple and the world.

There is, though, a corollary to this: gay families demand a higher level of social involvement than heterosexual unions, and by extension, political involvement. Within a legal order, adoptive parents are parents by virtue of the state’s recognition of them as such, rather than because they were the procreators of the child and hence naturally responsible for its well-being. Their obligations to their children exist by choice, a choice that exists within the structure of the state and its guidelines. In that sense, same-sex marriage necessarily depends upon the state to support and buttress homosexual couples in their pursuit of having children.

But let’s take another step back. Abortion proponents have suggested that the state’s interest in children is only in their rearing, and not in the fact that they are conceived. In that sense, the predominate position on the question is one of choice—”children” become the sort of things that are valid only when chosen, rather than having their own value by virtue of the fact that a mother conceived them. What constitutes the relationship between parent and child is not the biological fact that they are his progenitors, but the commitment, love, and acceptance that is expressed in the decision by the parents to keep the child.

Because the state (at a federal level) seems to have adopted this view of children and procreation (as measured primarily by Roe), the argument that procreation is enough to justify the state’s exclusive interest in heterosexual marriages has little weight. It certainly has little cultural weight, where the notion that children are chosen is so deeply ingrained that people thought Juno was a pro-life film. For our society, the state’s interest is in relationships that choose children, heterosexual or otherwise, not those that have or can have them. Hence, we have no ground to stand on to distinguish between heterosexual couples that have children naturally and homosexual couples that adopt them.

In that sense, it’s hard to separate the logic of the pro-choice crowd from the pro-gay-marriage crowd. Their view of children is fundamentally the same. What this means for those who want to be pro-life and pro-gay-marriage (even politically) I’m still working out. But I suspect there’s a deep dissonance at some level that inevitably leads to the downplaying of pro-life commitments.

To return to Douthat, though, I’ll simply note that the one thing he hasn’t yet clarified is his claim that the death of the marriage ideal might make the legal recognition of homosexual marriages “morally necessary.” Perhaps I’m pushing the worry too far, but it seems like the state has moral obligations not only to protect the rights of the people it governs, but to protect only those rights that the people actually have.

If heterosexual marriage is a different sort of thing than homosexual unions—as Ross has persistently and elegantly argued—then I’m curious to hear what legal and moral deliberation leads him to the surprising conclusion that those differences should not just be ignored by the law, but that the law would eventually be wrong to ignore them.

Cross-posted at Mere Orthodoxy.