Response to left and right-wing critics of the sex discrimination argument against laws banning same-sex marriage [updated with rejoinders to Christopher Green and Andrew Hyman]

Thanks to its appearance in the Supreme Court oral argument as a result of a question by Chief Justice John Roberts, the sex discrimination argument against laws banning same-sex marriage has attracted the attention of critics on both the left and right.
As it turns out, many of these criticisms was anticipated and addressed in the amicus brief on the sex discrimination argument I coauthored with Northwestern Professor Andrew Koppelman and in our recent USA Today op ed summarizing our argument (the brief may have inspired Roberts' question).

I. Response to Critics on the Right.

On the right, Roger Clegg and Ed Whelan have put up posts criticizing the argument at National Review. Legal scholars Christopher Green and Andrew Hyman have done so at the Originalism Blog. Clegg argues that laws banning same-sex marriage don't discriminate on the basis of sex in the same way as laws banning interracial marriage discriminate on the basis of race, because the former are not motivated by sexism in the same ways as the latter are motivated by racism:

If blacks cannot marry whites and vice versa, that's clearly a racist law, aimed at blacks. If men can't marry men and women can't marry women, is it seriously suggested that this law is sexist? Of course it isn't; it's a statement against same-sex marriage all right, but it's not a statement against men or women. Anti-miscegenation laws were about racial discrimination; laws against same-sex marriage deny nothing because of sex, but because of sexual orientation.

Green similarly argues that "racial segregation typically connotes racial insult," whereas gender-based restrictions on marriage do not similarly "insult" people on the basis of sex. Hyman raises much the same argument, as well.

There are two problems with this argument, First as we discuss at length in our brief, the Supreme Court has always held that laws that restrict rights based on sex qualify as sex discrimination even if they are not motivated by hostility towards men or women as such. Indeed, Craig v. Boren (1976), the case where the Supreme Court first ruled that gender classifications are subject to heightened judicial scrutiny involved an Oklahoma law that forbade 18-20 year old men, but not 18-20 year old women to buy 3.2% beer. It is highly unlikely that the Oklahoma state legislature was trying to make "a statement against men or women." But the law was invalidated anyway.

The Supreme Court has also long held that racial classifications are suspect regardless of whether they are motivated by racism. Laws banning interracial marriage would surely be unconstitutional even if they were not motivated by racism. For example, a state could enact such a law out of a desire to promote cultural diversity. Interracial marriage facilitates cultural assimilation of minority groups, and thereby tends to reduce the distinctiveness of their cultures. Clegg himself is a leading advocate of the argument that the use of racial preferences in affirmative action programs is unconstitutional, even though these programs are rarely motivated by racism or hostility to whites and Asian-Americans as a class.

Second, it is not true that laws banning same-sex marriage lack any connection to sexism. As explained in our brief and op ed, such laws are often in part motivated by rigid sex-role conceptions about the abilities of men and women and the roles they play in family and society.

Finally, it is simply not true that "laws against same-sex marriage deny nothing because of sex." To the contrary, the whole point of these laws is to restrict marital options based on the sex of the participants involved. As Chief Justice Roberts put it in the oral argument, "If Sue loves Joe and Tom loves Joe, Sue can marry Joe but Tom cannot." Tom is denied the right to marry Joe solely based on his gender. As the Chief Justice suggests, that is "a straightforward case of sexual discrimination."

Whelan raises a different kind of objection from the right. While most of the other critics seek to minimize the extent to which laws restricting marriage to opposite-sex couples focus on gender, Whelan embraces and justifies that focus:

The sexual complementarity of male and female inheres in what marriage has perennially been understood to be and in the core reason for its existence. (As Roberts stated at oral argument, "The fundamental core of the institution [of marriage] is the opposite-sex relationship.") It is true that laws that define marriage as the union of a man and a woman draw lines based on sex, not on sexual orientation. That is a powerful argument against the claim that state marriage laws discriminate on the basis of sexual orientation. But it is not a good argument for the claim that such laws "discriminate" on the basis of sex. To label as "discrimination" the line-drawing that is intrinsic to marriage (as it's been understood for millennia) is to engage in word games that, as Sherif Girgis explains, would confound the reasons why discriminatory classifications do receive heightened scrutiny.

I think it is actually Whelan who is engaging in "word games" here, not the defenders of the sex discrimination argument. The fact that sex discrimination is built into a long-standing definition of marriage that has existed for "millenia" does not make it any less discriminatory. Marriage was also long understood—also for millenia, in fact—to require domination of the female spouse by the male one, and that both participants must have the same religion. Yet laws imposing male domination within marriage still constitute sex discrimination, and laws banning interfaith marriage still qualify as discrimination on the basis of religion.

If "sexual complementarity" is understood to mean ability to engage in biological reproduction, then it is not now nor has it ever been an essential attribute of marriage. To the contrary, people who are sterile or unable to engage in biological reproduction because of age have always been permitted to marry. Such cases are far from marginal exceptions. As William Saletan notes, some 15% of new heterosexual marriages involve women 45 years old or older, and thereby probably unable to engage in biological reproduction. For millenia, marriage has served the purposes of promoting child-raising through adoption, sexual fidelity, economic security, and other objectives that it is now clear do not require "sexual complementarity" of a kind that only exists in opposite-sex relationships.

Whelan also relies on Sherif Girgis' criticisms of the sex discrimination argument. I addressed Girgis' position in detail here and here.

II. Addressing a Critique from the Left.

Liberal legal scholar Ari Ezra Waldman is also very critical of the sex discrimination argument:

It's central failing is why even conservatives like the Chief Justice may be inclined to adopt it. Conservatives are not "there" yet when it comes to homosexuality, in general, and antigay discrimination, in particular. For various reasons—religion, morality, Burkean conservatism—they are uncomfortable setting precedent that would prohibit individuals from discriminating against gays….

And therein lies the problem with using the sex discrimination argument to win this battle. It could lose us the war, or, at best, set us back years.

There are countless examples of discrimination faced by gay persons that do not also constitute sex discrimination, but the many forms of employment discrimination are paradigmatic. In 31 states, you can be fired simply for being gay. That's not sex discrimination: it's not as if the fired employee would have been saved had he or she had different sexual organs.

The sex discrimination is not only bad for the long term legal health of our community. It also ignores the fact that bans on gays marrying is in fact sexual orientation discrimination! That cannot be ignored. If you ban an entire group of people, coherent based on their sexual orientation identity, from participating in an institution for the sole reason of their sexual orientation identity, that is not only sexual orientation discrimination, but it is also odious under the Fourteenth Amendment's guarantee of equal protection to all persons.

Notice that Waldman does not offer any reason why the sex discrimination argument is legally wrong. He merely believes that a victory for same-sex marriage on that basis would not have the same beneficial effects as one that relies on the theory that laws banning same-sex marriage discriminate based on sexual orientation. Even if his critique is completely correct, it does not offer any constitutional reason why the Court should reject the sex discrimination theory.

Like Roger Clegg, Waldman believes that "sexual orientation discrimination" is the real issue here, and gender is somehow tangential. But, as Koppelman and I explain in our brief, the two forms of discrimination are in fact closely linked. Here's the relevant passage from our op ed, which summarizes the longer argument laid out in the brief:

Some defenders of same-sex marriage fear that the gender discrimination argument ignores what they see as the true motivation for the laws they oppose: prejudice against gay men and lesbians. But hostility to gays and lesbians and sexism are often closely linked. At least until recently, most Americans learned no later than high school that one of the nastier social sanctions one will suffer for deviating from traditional gender roles is the imputation of homosexuality. Much anti-homosexual prejudice is closely linked to gays' and lesbians' supposed deviation from conventional gender norms. While gay men are often stigmatized for being "effeminate," lesbians are stereotyped as too masculine.

More generally, the restriction of marriage to opposite-sex couples was historically in large part based on an assumption that marriage requires adherence to traditional sex roles that can only be performed by opposite-sex couples, because a woman could not fulfill a man's role in the family and vice versa. As we discuss in the brief, such assumptions persist today in some of the arguments raised by defenders of laws banning same-sex marriage in the current litigation.

Ultimately, laws banning same-sex marriage discriminate on the basis of both sex and sexual orientation; the two are not mutually exclusive. As Prof. Stephen Clark, one of the signers of our brief puts it, "because homosexuality is itself nothing but a derivative construct defined by reference to sex, there is no such thing as an instance of 'sexual orientation discrimination' that does not turn… on consideration of the target's sex. Sexual orientation is simply not a concept that has any meaning independent of sex."

The two forms of discrimination at issue in this case are inextricably connected. But the sex discrimination argument has a much stronger basis in Supreme Court precedent and in the history and original meaning of the Fourteenth Amendment. Among other things, sex discrimination is built into the very text of laws banning same-sex marriage, which explicitly restrict who can marry whom based on their gender. By contrast, at least in a formal legal sense, they do not allocate rights based on sexual orientation. People are allowed to enter into opposite-sex marriages and forbidden to enter into same-sex ones, regardless of whether they happen to be gays or lesbians or not. Sexual orientation discrimination is present only in the intentions behind many of these laws, which often include a significant element of homophobia.

It is true, of course, that a Supreme Court victory for same-sex marriage would not lead to enactment of laws banning sexual orientation discrimination in employment in those states which don't yet have them. But, when it comes to the private sector, neither would a decision relying on the sexual orientation theory. Both decisions would only cover state government discrimination, because the Fourteenth Amendment's Equal Protection Clause generally applies only to state action, not private action. When it comes to discrimination in the public sector, a Supreme Court decision invalidating laws banning same-sex marriage would also likely cut against sexual orientation discrimination in hiring for government jobs. After all, if the government refuses to hire Joe because he is attracted to other men, they are rejecting him for an attribute they would not object to if Joe were a woman.

Although I think both the left and right-wing critiques of the sex discrimination argument are flawed, I am grateful for the attention renewed attention to the issue from thoughtful commentators on both sides of the political spectrum. Part of my purpose in coauthoring the brief and writing other pieces focused on the subject was to attract greater attention to this somewhat neglected aspect of the debate over same-sex marriage.

UPDATE: I have made a few minor revisions to this post, such as adding a point made by Prof. Stephen Clark, one of the signers of our brief.

UPDATE #2: Green argues in his post that laws banning same-sex marriage may not be unconstitutional because they are "symmetrical" in the sense that they burden both men and women equally. But he also recognizes that that argument is vulnerable to the objection that laws banning interracial marriage are similarly symmetrical, in the sense that they burden both blacks and whites in the same way. Thus, he ultimately emphasizes that the real difference between the two is that the anti-miscegenation laws constitute a form of "racial insult," whereas laws banning same-sex marriage are not motivated by sexism. That distinction, in turn, is vulnerable to the objections I note above. The symmetry argument is covered in greater detail in our brief, and in this post.

UPDATE #3: Christopher Green responds in an update to his post:

I'm not sure [Somin] appreciates the simplicity of Bursch's proposal that intermediate scrutiny does not apply if a statute does not "involve[] treating classes of men and women differently." That's not an inquiry into motive, as Somin suggests, but merely a proposal that symmetric gender distinctions deserve categorically less scrutiny than gender exclusions. My rationale for lessened scrutiny was the typical lack of a connotation of inferiority in usual cases of segregation, such as in bathrooms or athletics, but I offered this as a reason to adopt Bursch's idea, not a suggestion that we make ill motive an additional threshold requirement before applying intermediate scrutiny.

I do appreciate the simplicity of the idea. I just think its simplicity fails to save it from being obviously wrong. For reasons noted above and in my earlier critique of Bursch, symmetry can't justify lowered scrutiny of sex discrimination, especially when the discrimination in question is in part motivated by sex-role stereotyping.

Green also emphasizes the rarity of the type of law struck down in Craig v. Boren, noting that by 1976 few states still had sex-discriminatory drinking laws. However, nothing in the majority opinion turns on how common the law at issue was. Rather, Craig establishes a general rule that "classifications by gender must serve important governmental objectives and must be substantially related to achievement of those objectives." It in no way distinguishes between "classifications by gender" that are adopted by many states and those that exist in only a few or only one. Obviously, the justices knew that many such classifications still existed in 1976. They could have limited their reasoning to relatively unusual laws like Oklahoma's restriction on the sale of beer, but chose not to do so. This is even more clear in later Supreme Court sex discrimination cases, most of which are discussed in our brief.

Justice John Paul Stevens' concurring opinion in Craig does note that one reason why the Oklahoma law is "objectionable" was "because it is a mere remnant of the now almost universally rejected tradition of discriminating against males in this age bracket." But that is just one of several reasons he lists for striking it down. He also found it "objectionable because it is based on an accident of birth" and that passage of his opinion makes clear this is true of all laws based on gender classifications. Ultimately, the purpose of heightened scrutiny—and of the Fourteenth Amendment more generally—is not to strike down only unusual discriminatory laws but to uproot more widespread traditions of state-enforced discrimination.

UPDATE #4: In a further update, Green clarifies that his argument is that "Heightened scrutiny in this context only makes sense if gender segregation in general is suspicious, and the lack of suspiciousness to most reasonable observers of of gender-segregated bathrooms and athletics shows that they are not." Given the extensive historic use of gender classifications as a tool of oppression, I think it is pretty clear that such classifications (whether used to segregate or not) are suspicious. They are particularly so when, as in the case of marriage law, they are in part motivated by sex-role stereotypes. Some such classifications can still pass heightened scrutiny. Sex-segregated bathrooms and sports teams are good examples. But the fact that there are a few such cases (and there are very few, if any, besides the two Green mentions) does not undermine the case for applying heightened scrutiny to sex classifications generally. One can equally argue that racial and ethnic classifications shouldn't get heightened scrutiny because we can find a few cases where their use is both well-intentioned and benign in its effects.

UPDATE #5: Andrew Hyman responds to this post here:

I didn't mention Craig v. Boren in my original blog post because the legislation struck down in that case so obviously discriminated against men; it reduced the ability of men to buy beer, but not the ability of women to buy beer. The statutes at issue in Obergefell are nothing like it, because they do not discriminate against either gender as a class. The Oklahoma law did not explicitly denounce men or insult men, but still it was against men, and so I don't think Professor Somin really has a point there.

Second, Professor Somin claims that laws banning interracial marriage would be unconstitutional even if motivated by a sincere desire for cultural diversity instead of by racism. Well, sure, marriage as defined for millennia has been deemed a fundamental right, and so strict scrutiny would apply according to the Court's precedents; therefore the law would inevitably be struck down. The strict scrutiny would not be because of nonexistent discrimination against a suspect class, but rather because of the fundamental right being violated, and the person being discriminated against as an individual. Even if one insists that statutes be 100% colorblind (a notion rejected by the Supreme Court in Parents Involved v. Seattle), still Chris Green cogently argued at this blog that gender and race should not automatically be treated the same in that regard.

Hyman's first point relies on the symmetry argument, which I have already extensively addressed above, and in previous posts. On his point regarding interracial marriage, it may be true that laws banning interracial marriage would be unconstitutional even if enacted for benign purposes because they violate a "fundamental right" protected by the Due Process Clause. But they would also be unconstitutional because they engage in unconstitutional discrimination on the basis of race. Certainly, a symmetrical form of racial discrimination that did not implicate a fundamental right and was not motivated by racism would still be subject to heightened scrutiny. For example, a requirement that blacks and whites attend segregated public schools or use segregated public services would be unconstitutional even if it were adopted for nonracist reasons and even if the schools and other services in question were all of exactly equal quality.

Finally, Hyman objects to my categorizing his criticism of the sex discrimination argument as "right wing." My reference to "right wing" critics was simply to denote those whose political views are generally associated with the conservative side of the political spectrum, which I think is true of Hyman and the others so described. I apologize if I have misunderstood his views.

UPDATE #6: Green has made a brief further rejoinder, at the same link as before. His main point is that various types of gender classifications, particular sex-segregated bathrooms and sports teams, but also requirements that both men and women be represented on some types of committees, do not deserve heightened scrutiny because they seem benign. It may indeed be the case that these gender classifications are benign. However, the long history of sexism and gender-based oppression suggests that heightened scrutiny is needed to separate the relatively few benign gender classifications from the many that are not. Moreover, even well-intentioned gender classifications—like well-intentioned racial classifications—can still sometimes be harmful and oppressive.

Green also asks why segregated bathrooms would pass intermediate scrutiny. The answer is that, at least in many circumstances, they may be the only way to protect the important state interest in individual privacy. By contrast, there is no comparable state interest that can be achieved only by excluding same-sex couples from marriage.