by Michael Dorf
In my latest Verdict column, I discuss last week's same-sex marriage (SSM) rulings: Judge Feldman's district court opinion upholding Louisiana's SSM ban; and Judge Posner's 7th Circuit opinion invalidating the Indiana and Wisconsin SSM bans. My column praises Posner's opinion overall but especially for his injection of a new argument into the growing genre of opinions striking down state SSM bans. As he foreshadowed during the oral argument, Posner performs a kind of judicial jujitsu. He turns the state claim that it is concerned about the consequences of accidental heterosexual procreation against the state: If the state were really concerned about those consequences, he says, it would permit same-sex couples to marry, because they are raising thousands of adopted children who were born as a result of such accidental heterosexual procreation.
That's a very nice move, but as I note in the column, it is potentially vulnerable to a rejoinder by the state: Rational basis scrutiny permits the state to tackle a problem one step at a time; here the state is tackling the problem for the heterosexual couples who keep their offspring; it need not also tackle other aspects of the problem. I say in the column that this is not a persuasive rejoinder but I do not elaborate at length why. Consider this blog post an extended footnote to the column.
The state rejoinder can only get off the ground if laws that facially disadvantage people based on sexual orientation are subject to mere rational basis scrutiny, rather than to heightened scrutiny. For the last 18+ years (since Romer v. Evans) the Supreme Court has found it unnecessary to decide what level of scrutiny applies to such laws, invalidating laws disadvantaging gays and lesbians even under rational basis scrutiny. There is a robust academic literature that argues about whether those decisions apply conventional rational basis scrutiny or a kind of covert heightened scrutiny. When I teach the relevant line of cases, I call this, tongue in cheek, "extra crispy rational basis scrutiny." (Occasionally students, to my horror, write that phrase on an exam as though it were an actual legal term of art.)
Meanwhile, precedents in some states and circuits say that sexual orientation is not a suspect or semi-suspect classification, so that judges in these states and circuits take the position that conventional rational basis scrutiny is the correct standard. I think those precedents are wrong for two reasons. First, sexual orientation is, or should be understood as, a form of sex discrimination, both literally and substantively. It is literally a form of sex discrimination because, in the case of SSM, whether a marriage license will issue depends on the sex of the applicants. It is substantively a form of sex discrimination because prejudice against sexual minorities relies on impermissible stereotypes about sex roles--the very evil that is at the core of what makes sex discrimination constitutionally objectionable.
Second, even if one thinks that sexual orientation should not be considered sex discrimination, standing alone it bears the hallmarks of a suspect (or semi-suspect) classification. Judge Posner summed up the reasons matter-of-factly and pithily when he wrote that "homosexuals are among the most stigmatized, misunderstood, and discriminated-against minorities in the history of the world ...." Only an unprincipled commitment to the notion that the Supreme Court is done recognizing suspect classifications--a kind of arbitrary numerus clausus--can support the view that sexual orientation is not a suspect classification.
But let us suppose for the sake of argument that the SSM bans were subject only to rational basis scrutiny. Even so, the one-step-at-a-time defense would fail as a response to Judge Posner's argument about adoption. To see why, one needs to examine exactly what the state is saying in its accidental procreation argument. The state is not saying that preventing same-sex couples from marrying in some way prevents accidental procreation by unmarried heterosexuals. And for apparently good reason, because that would be both illogical and, as Judge Posner shows, counter-factual. There is absolutely no evidence that permitting same-sex couples to marry in any way discourages opposite-sex couples from marrying.
Now maybe . . . just maybe . . . the state could argue that permitting same-sex couples to marry does discourage opposite-sex couples from marrying. Sure, there's no evidence for this supposed phenomenon, but traditional rational basis scrutiny does not require evidence. As expressed in the most forgiving formulation, if a court can conceive of a state of facts in which the state's law would further the state's goals, then the law satisfies rational basis scrutiny. Can we conceive of such a world? Sure, but the problem is that the only rational way of conceiving such a world is itself constitutionally problematic.
What would be the mechanism by which permitting same-sex couples to marry discourages opposite-sex couples from marrying? In a word, homophobia. Some number of opposite-sex couples who would otherwise marry would see that marriage ain't what it used to be now that those disgusting people can git married, so they would decide instead just to shack up. Then, when their fornicatin' led to babies, they would be less likely to stay together than if they had married. Presto! Permitting same-sex marriage would have undermined traditional marriage and thus contributed to the problem of accidental procreation by unmarried heterosexuals.
Let me emphasize that, to their credit, Indiana and Wisconsin did not make this argument. And to the credit of heterosexuals in Indiana and Wisconsin (and so far as I am aware, the rest of the country), homophobia is not so extreme that it actually leads to the foregoing hypothetical scenario. But again, traditional rational basis scrutiny does not require actual facts, only those that can be "conceived" (pun foreseen but not intended). Suppose that some other state were to advance the hypothetical scenario as an ostensible rational basis for a SSM ban. Surely a judge could conceive the hypothetical facts that I just conceived, right?
Yes, but whatever else the extra-crispy cases mean, at a minimum, they stand for the proposition that animus cannot be the basis for legislation, and in the imagined scenario the heterosexuals who are discouraged from marrying are discouraged because of animus: they are so disgusted by the prospect of SSM that they forgo marriage themselves. To be sure, the state would say that it doesn't share the animus; it is merely taking account of it for the benefit of the children. But that option is foreclosed by Palmore v. Sidoti, which held that, even in determining the best interest of actual children in a custody case, a judge could not in effect incorporate private bias. As the Court said there: "Private biases may be outside the reach of the law, but the law cannot, directly or indirectly, give them effect." Palmore involved race discrimination, but nothing about its logic would make it inapplicable to other constitutionally forbidden motives, such as the general prohibition on animus.
Accordingly, Indiana and Wisconsin were right to place no reliance on the imagined scenario in which straight people don't marry because they are demoralized by gay people marrying. It doesn't happen; it is an offensive claim; and even if the state were to advance the argument, it would fail under the logic of the animus cases and Palmore.
So what is the argument that the states advance regarding accidental reproduction? Their argument is not that that they need to ban SSM to mitigate the harm of accidental reproduction by unmarried heterosexuals. Their claim is simply that they do not need to extend marriage to same-sex couples because same-sex couples do not accidentally procreate.
At this point I want to make one very, very, very simple observation: The laws being challenged in the SSM litigation do not simply fail to extend marriage to same-sex couples. They ban SSM. So even if mere rational basis scrutiny should apply to such laws (and it shouldn't), the argument that there is a rational basis for failing to extend the institution of marriage does not explain why it is rational to ban SSM, rather than merely to fail to extend it.
Here we can use a variant on a hypothetical example offered by Judge Posner to illustrate the point. A state issues driver's licenses, initially simply for the purpose of ensuring that people operating motor vehicles can do so safely. If that is all that a driver's license does, then the state has a rational basis for failing to issue "non-driver's licenses" to people who don't drive. But over time, the driver's license begins to play multiple other functions. Private firms like banks, airlines, and bars demand to see a driver's license as identification; the government itself does the same in various contexts; perhaps even hospitals demand to see a driver's license as a condition of visiting a sick relative. Eventually, the driver's license comes to be a kind of universal ID card. Now imagine that non-drivers (including people who are physically unable to drive for various reasons such as blindness) organize to seek non-driver's licenses. If at that point the state passes a law or a state constitutional amendment forbidding the issuance of non-driver's licenses, the state can no longer rely on the original reason for issuing licenses only to drivers. The original distinction--between people who might endanger the public by operating motor vehicles and those who don't--has no bearing on all of the collateral benefits conferred by having a govt ID which are sought by the non-drivers. So even though the state may have originally had a rational basis for failing to issue non-driver's licenses, it lacks such a rational basis for banning them once the collateral benefits are clear.
Likewise with respect to SSM. (Indeed, even more so, because in the foregoing scenario, the non-drivers have another, albeit more expensive, option: they can obtain federal passports. But there's no way to get the state benefits of marriage by getting "federally married.") Perhaps it's true that at some point in the past state recognition of marriage for opposite-sex couples was rooted in concerns about accidental procreation. Even so, everybody knows that the SSM bans were not enacted because of a judgment that same-sex couples don't need protection against accidental procreation. My research in a news database reveals not a single reference to the phrase "accidental procreation" in connection with same-sex marriage before 2006, with nearly all of the references coming in the last couple of years.
The best that can be said for the accidental procreation argument is that it is not as ugly as some of the anti-gay arguments that have been raised in the past to defend anti-gay laws. (See, e.g., here.) Nonetheless, the sheer ridiculousness of the accidental procreation argument reveals it to be an obvious pretext, so that any judge or Justice who accepts it will likely be ridiculed, if not reviled as a bigot, in the not-too-distant future.