Now the shoe is on the other foot, and it is time for the court to strike down a federal statute in order to advance a liberal policy goal rather than a conservative policy goal. Justice Antonin Scalia’s paean to the democratic process in his dissent sounds a little hollow, coming in the wake of his votes to strike down affirmative action programs and Section 4 of the Voting Rights Act — both of them the result of the democratic process, as much as DOMA was. Meanwhile, none of the liberals pipe in to explain how to reconcile the outcome of this case with the concerns about democracy that they expressed in dissenting opinions in the other cases. (Ruth Bader Ginsburg, in Shelby County, Ala.: “That determination of the body empowered to enforce the Civil War Amendments ‘by appropriate legislation’ merits this Court’s utmost respect.”)
But this is a trite point, and never mind. The problem faced by opponents of the Defense of Marriage Act is that there was no clear constitutional hook for striking it down. The Equal Protection Clause does not seem to apply because gay people (unlike, say, African-Americans) have not been regarded as politically weak enough to be a “suspect class,” justifying heightened review. That means that only a rational basis is necessary to uphold DOMA and a rational basis is easy to find (uniformity, efficiency, blah, blah, blah). The Due Process Clause does not seem to apply because that clause protects only rights that are rooted in history and tradition, and the right of same-sex marriage, however compelling a moral issue it may seem today, is not such a right. Federalism says that (under ill-defined conditions) the U.S. government cannot trump state law, especially in an area like family law, but in fact there are plenty of federal laws that regulate marriage, at least along the margins.
In a remarkable opinion for the majority, Justice Anthony Kennedy manages to hit on all these theories without really endorsing any of them. After a long peroration on the importance of federalism, he disclaims it as a basis for the opinion — for the reason, I suspect, that if the unconstitutionality of DOMA is based on federalist principles, this case will not be a precedent that can be used to strike down state laws that refuse to recognize same-sex marriage in the future. He then vaguely invokes both due process and equal protection, without explaining how he overcomes the limits on those doctrines. He does not, for example, declare homosexuals a suspect class, nor does he (or could he) claim that same-sex marriage has roots in history and tradition. In the end, he seems to hold DOMA unconstitutional because he is convinced that the purpose of the statute was to stigmatize gay people, and indeed there is some precedent for the idea that statutes based on animus are unconstitutional. But he does not provide a very convincing account of the motives of the legislators. Isn’t it possible to oppose same-sex marriage without hating gay people?
Kennedy’s opinion reminds me of two cases, both heavily criticized by constitutional scholars (for their reasoning, not necessarily their result). The first is 1965’s Griswold v. Connecticut, where the Supreme Court struck down a statute that prohibited the use of contraceptives, based on what Justice William O. Douglas called “emanations” from and “penumbras” of other constitutional provisions that said nothing about sex or contraception but did endorse other liberty interests. The second is 1990’s Employment Division v. Smith, where Justice Scalia, in the course of putting limits on the free exercise of religion doctrine of the first amendment, explained away some old cases on the grounds that they involved “hybrid rights”: The statutes in question were struck down not because they violated the religion clause per se, but because they burdened both the practice of religion and other constitutionally protected activities.
So there is this vague idea that certain constitutional interests standing alone may not invalidate statutes, but may suffice when combined together. Something like this idea might ultimately be the basis of Kennedy’s opinion. Gay people do not form a suspect class, but they almost do. Same-sex marriage is not a longstanding tradition, but same-sex relationships are. Federalism principles are not broken but they are eroded. Put together three almost violations, and you have a real violation.
But I think trying to find a jurisprudential explanation for this opinion, like the opinions in Fisher v. University of Texas and Shelby County v. Holder, is a fool’s errand. Same-sex marriage is advancing while affirmative action is receding because that’s what the relevant majorities of the justices care about.
Eric Posner, a professor at the University of Chicago Law School, is a co-author of “The Executive Unbound: After the Madisonian Republic and Climate Change Justice.” Reach him on Twitter at @EricAPosner.