In addition to the due process argument, the New York gay-marriage plaintiffs claimed that denying them marriage violated the state constitution's guarantee of equal protection. (Here, New York generally follows federal precedent, offering no more protection than would be offered under the federal constitution.) The court confronted two arguments for heightened scrutiny: first, that denying same-sex couples marriage is a form of sex discrimination, to which the court should apply intermediate scrutiny; and second, that their exclusion from marriage is a form of sexual orientation discrimination, to which the court should apply some form of heightened scrutiny.

The first claim, that excluding gay couples from marriage is a form of sex discrimination, has been a staple of gay-rights litigation and especially of academic theorizing for a couple of decades. It's fair to say that sex-discrimination arguments have been the dominant mode of legal-academic writing supporting gay-rights claims for a generation now. One of the primary theorists of this argument has been Andrew Koppelman, a Northwestern University Law School professor and friend of mine, who has been both tireless and remarkably articulate in defense of it.

The argument operates at two levels: a formal one and a deeper sociological one. The formal argument is that much anti-gay legislation, like a same-sex marriage exclusion, that appears to be merely "anti-gay" is actually sex discrimination on its face. After all, Jack can marry Jill, but he can't marry Adam. The exclusion turns on Jack's biological sex, since Jack could marry Adam if he (Jack) were female. Thus, Jack is being denied a government benefit because of his sex. It does not matter, according to this argument, that males and females are equally denied the right to marry someone of the same sex. In Loving v. Virginia, the sex-discrimination theory goes, the Supreme Court easily dispatched the argument that an antimiscegenation law that applied to both blacks and whites was not race discrimination.

Further, the sex-discrimination argument holds, a same-sex marriage exclusion operates sociologically to reinforce traditional notions about the proper gender roles of men and women; women must be "wives" to men, men "husbands" to women, and so on. It is a goal of sex-discrimination jurisprudence to root out legislative classifications grounded in traditional gender roles.

The New York court addressed only the formal part of the sex-discrimination argument, rejecting it in a single paragraph:

By limiting marriage to opposite-sex couples, New York is not engaging in sex discrimination. The limitation does not put men and women in separate classes, and give one class a benefit not given to the other. Women and men are treated alike -- they are permitted to marry people of the opposite sex, but not people of their own sex. This is not the kind of sham equality that the Supreme Court confronted in Loving; the statute there, prohibiting black and white people from marrying each other, was in substance anti-black legislation. Plaintiffs do not argue here that the legislation they challenge is designed to subordinate either men to women or women to men as a class. (pp. 13-14)

The New York court's conclusion here is very much in line with the vast majority of decisions at the state and federal levels that have confronted the sex-discrimination argument in one form or another. Indeed, even courts favorable to gay-rights claims have either explicitly rejected the sex-discrimination argument (as did the Vermont and Massachusetts high courts in their marriage cases) or have ignored it (as did the Supreme Court in Lawrence). In some two decades it has succeeded just once in a high-profile case, Baehr v. Lewin, in which the Hawaii Supreme Court held that the denial of marriage to same-sex couples was sex discrimination. Overall, its won-loss record in litigation has been abysmal.

But should it be? I have never been persuaded by the sex-discrimination argument. At the formal level, the argument over-reads Loving, a single case from a different and special area of equal-protection jurisprudence (race). Classifications based on race are far more likely to be based simply on racism than classifications based on sex are likely to be based simply on sexism. We have separate bathrooms for men and women; separate men's and women's sports teams at public universities; men and women are treated differently by the military; different standards for presuming parentage depending on sex; different penalties and standards for statutory rape depending on sex, and so on. I doubt these sex classifications are unconstitutional, or that they would get anything more than rational-basis scrutiny. Even if we applied intermediate scrutiny to them, they'd probably pass that more forgiving and flexible test.

Further, I understand Loving to have concluded that antimiscegenation laws, however "equal" they were in form, to have one basic and overriding purpose that deeply offends the heart of the 14th Amendment: the maintenance of "White Supremacy." While I agree that the limitation of marriage to opposite-sex couples both reflects and reinforces traditional gender roles, I think it is a crabbed view of traditional marriage to say that that is all it reflects or to say that maintaining a gender hierarchy is even its dominant cultural function nowadays. Many, many legal academics would strongly disagree with my view.

Especially troubling for the formal sex-discrimination argument in the marriage context, I think, are cases in which the Supreme Court has suggested that legislative classifications based on biological differences between men and women get only rational-basis scrutiny (see, e.g., Geduldig). The argument of the state in the New York case, and the one the court ultimately accepted, is that the limitation of marriage to opposite-sex couples (the legislative classification) is rooted in a biological fact -- that most opposite-sex couples may procreate while no same-sex couple can. That's not a good policy reason to exclude gay couples from marriage, in my view, but it may be a good constitutional basis for subjecting their exclusion to rational-basis review despite what looks facially like a sex classification.

Next, the New York plaintiffs argued that their exclusion from marriage was a form of sexual orientation discrimination that should be subjected to heightened scrutiny. Under the federal and state precedents, this claim is even more dubious than the sex-discrimination argument. Federal courts do not apply heightened scrutiny to sexual orientation discrimination. Neither do state courts. Nevertheless, the Supreme Court has never squarely addressed the arguments for heightened scrutiny of sexual orientation discrimination (no, not even in Romer). It is certainly open to New York to go its own way on this question.

To its credit, the New York court at least acknowledged that the gay-marriage exclusion is a form of discrimination based on "sexual preference," implicitly rejecting the glib claim that the exclusion doesn't discriminate against gays because they can still marry someone of the opposite sex. But the court rejected the argument for heightened scrutiny by saying that, while heightened scrutiny might apply to some forms of sexual orientation discrimination, it would not apply to "legislation governing marriage and family relationships," (pp. 14-15) since homosexual "preference" doesn't produce children, a fact "relevant" to state interests in marriage.

This, I think, is the weakest part of the opinion. The usual equal protection approach is categorical, not context-specific. The courts usually ask whether classifications aimed at a group are in general suspect, and then apply the appropriate scrutiny to a particular classification depending on the answer to that question. They do not usually, as the New York court did here, look at each law and decide whether the state has interests "relevant" to the classification and then apply only rational-basis scrutiny wherever the state has such "relevant" interests. For example, a court would not apply rational-basis scrutiny to a decision by a state prison system that has interests "relevant" to race in separating black from white prisoners. The court would apply strict scrutiny to such a classification because that's the level of scrutiny the court applies to all race classifications. (There are cases, however, involving alienage and sex classifications where the level of scrutiny is more context-specific.)

While the Supreme Court has never fully laid out the guidelines for the application of heightened scrutiny, it has suggested some criteria, like the political powerlessness of the group, a history of invidious discrimination against the group, and so on. The New York court acknowledges a history of "serious injustice in the treatment of homosexuals, a wrong that has been widely recognized only in the relatively recent past" (p. 9). But the court fails to cite the criteria for the application of heightened scrutiny or the decisions that employ them, much less analyze the issue. The dissent, by contrast, contains a nice primer on some of the relevant points (Kaye dissent at pp. 12-16). Whether the dissent is correct or not is beside the point. The majority failed even to engage the issue.