5

United States v. Windsor

DOMA seeks to injure the very class New York seeks to protect.

Windsor (2013)

United States v. Windsor1 is, as of this writing, the last of the cases firmly and explicitly grounded on an animus theory. (The 2015 decision striking down state same-sex marriage bans, discussed in Chapter 12, is more equivocal on this point.) In many ways it is also the most paradoxical. It seemed to matter greatly to the Court that the statutory provision Windsor struck down, Section 3 of the Defense of Marriage Act (DOMA),2 had the effect of disrespecting states’ decisions about who could marry whom, yet it disclaimed reliance on federalism. Its focus on that disrespect would seem to buttress, not limit, states’ autonomy to provide or deny marriage rights to same-sex couples, yet after the opinion came down a long string of federal judges cited it when striking down state same-sex marriage bans.

Perhaps most relevant for our current purposes, at one level the Court’s animus conclusion seems straightforward (if nevertheless highly controversial): In finding animus, the Court relied heavily on actual statements of Congress’s purpose and even the statute’s title. But on the other hand, its animus conclusion appears to result from a complex interaction of factors, some of them (most notably, the statute’s disrespect of state marital decisions) not even directly related to equal protection at all. Given these puzzles—and especially in light of Windsor’s impact on same-sex marriage claims beyond the limited confines of federal recognition—it bears giving this case a close look.

Windsor’s Facts

Windsor is an utterly dry case about the amount of estate tax a surviving spouse owes the federal government. But—of course—it is much more than that. As the stories introducing the prior three chapters have made clear, abstract (though important) legal issues often come wrapped in deeply human personal stories. The story of Windsor is no less human for being straightforward: Edie Windsor and Thea Spyer loved each other for over forty years, and when they had an opportunity to get married they did. Windsor was the culmination of Edie’s challenge to the federal government’s refusal to accept New York’s recognition of their marriage.

That refusal was not the first instance of discrimination Edie Windsor faced. At work, she never came out as lesbian during the sixteen years she worked for IBM, even to her co-workers, whom she called “my buddies.”3 In an act of passing that any gay or lesbian person of a certain age (and many younger ones) would recognize, when in 1967 Thea proposed to Edie she gave Edie a diamond pin, rather than a ring—a ring would have raised too many questions.

Of course, Edie’s and Thea’s engagement in 1967 could not have led to a marriage that any government would have recognized—no governmental authority in the world recognized same-sex marriage until the Netherlands, in 2001.4 After 2001, a trickle of foreign nations and American states began to recognize same-sex marriage unions. The supreme court of Edie and Thea’s home state, New York, decided in 2006 that the state constitution did not guarantee same-sex marriage rights. However, in 2008, New York, though a combination of legislative and gubernatorial action, recognized same-sex marriages legally performed elsewhere.

By then, Edie and Thea had married, in Canada. By 2007 Thea had suffered a number of health setbacks; in that year her doctor warned her she had less than a year to live. Wishing to die married to the woman she loved, Thea—confined to a wheelchair owing to multiple sclerosis—and Edie traveled to Toronto, where they were married by Canada’s first openly gay judge. Friends had to lift Thea’s hand up so she could receive her wedding ring. She died in 2009.

In 2009, a federal law, the Defense of Marriage Act, had been on the books for thirteen years. The Defense of Marriage Act was Congress’s reaction to early marriage rights litigation, which bore preliminary fruit in the mid-1990s. The prospect of one state potentially recognizing same-sex marriage had galvanized anti-marriage-equality forces that feared that such legalization would open the door to more widespread recognition. One part of DOMA responded directly to that fear by preventing federal courts from insisting that a state recognize a same-sex marriage validly performed in another state. Another part of DOMA, Section 3, provided that, for federal law purposes, marriage consisted only of a union between one man and one woman.

One of those federal laws was the Internal Revenue Code. When Thea died, the IRS, citing Section 3, disallowed the estate tax exemption Edie would otherwise have enjoyed by virtue of New York’s recognition of her Canadian marriage. That discrimination prompted Edie to sue.

Windsor’s Analysis

As noted in the introduction to this chapter, Windsor’s analysis of DOMA is at once straightforward but also layered and complex. The Court, speaking again through Justice Kennedy (who has written the Court’s four opinions vindicating gay rights claims), began by acknowledging both the normal practice of states enjoying primary authority for defining marriage and the eligibility for marriage, as well as the federal government’s power, when managing its own programs, to override state choices. But despite that latter power, he noticed something unusual about DOMA. In particular, he noted its broad reach, affecting, as he wrote, “over 1,000 federal statutes and the whole realm of federal regulations.” Importantly, the Court then immediately reminded the reader: “And [DOMA’s] operation is directed to a class of persons that the laws of New York, and of 11 other States, have sought to protect.” Even at this early stage of its analysis, then, the Court was painting a picture of an unusual law—unusual in its deviation from the normal federal practice of respecting state marriage decisions and in its “reach and extent,” not just to “over 1,000 federal laws,” but also to a particular class to whom (by 2013) eleven states had decided to make available the dignified status of “married.” Citing his own opinion in Romer, Justice Kennedy then reminded readers that “discriminations of an unusual character especially suggest careful consideration to determine whether they are obnoxious” to the Constitution.5

In the next section of his opinion, where he considered the implications of all this, he pulled no punches. “DOMA,” he said, in the very first sentence of that section, “seeks to injure the very class New York seeks to protect.” Immediately moving to Moreno’s statement that, “at the very least . . . a bare congressional desire to harm a politically unpopular group cannot” justify disparate treatment of that group, Justice Kennedy then choreographed an intricate dance between federalism, equality, and fundamental rights: “The responsibility of the States for the regulation of domestic relations is an important indicator of the substantial societal impact the State’s classifications have in the daily lives and customs of its people.” Based on that observation he concluded that “DOMA’s unusual deviation from the usual tradition of recognizing and accepting state definitions of marriage here operates to deprive same-sex couples of the benefits and responsibilities that come with the federal recognition of their marriages. This is strong evidence of a law having the purpose and effect of disapproval of that class.”

At this point Justice Kennedy then moved on to the direct evidence of congressional intent—speeches and statements on the floor of Congress and the title of the statute. But before getting there, let us pause to inspect the stage Justice Kennedy had set against which to evaluate that evidence. Up to this point in the opinion, he has described DOMA as a law that deviates from the normal federal practice of respecting state law marriage definitions and does so in a way that is both across-the-board (the “over 1,000 federal statutes” it affected) and narrow (focusing solely on same-sex marriage). Moreover, DOMA imposed this unusual, both-broad-and-targeted disability in an area carrying significant dignitary interests—the area of marital rights.

So understood, Justice Kennedy’s description of DOMA shared important features with Amendment 2, the Colorado initiative the Court struck down in the Romer opinion he authored seventeen years earlier. Just like Amendment 2, DOMA denied a targeted group (again gays and lesbians) rights across the board. And just like he had in Romer, Justice Kennedy in Windsor cautioned that such an unusual law merited closer judicial scrutiny. But his description of DOMA suggested that there was even more cause for concern. In Justice Kennedy’s view, DOMA denied a uniquely meaningful right—the right not to have one’s lawful marital status denied by the federal government. Moreover, the fact that marital status was granted by states and then denied by the federal government rendered DOMA, if not a violation of federalism (something Justice Kennedy disclaimed), then at least another reason to be suspicious.

All of these warning signs set the stage for Justice Kennedy’s examination of DOMA’s text and legislative history. That examination “demonstrate[d] that interference with the equal dignity of same-sex marriages, a dignity conferred by the States in the exercise of their sovereign power, was more than an incidental effect of the federal statute. It was its essence.” As if to reinforce the interlocking nature of his federalism, substantive rights, and equality arguments, he concluded, after quoting from the bill’s supporters and the title of the bill itself, that “[t]he arguments put forward by [DOMA’s congressional supporters at the Court] are just as candid about the congressional purpose to influence or interfere with state sovereign choices about who may be married.” The intimately connected nature of those arguments is reflected in his observation that, while New York, in granting same-sex marriage rights, “sought to eliminate inequality,” DOMA “frustrates that objective” by “writ[ing] inequality into the entire United States Code.”

Let us cut through the detail and think about the structure of this argument. After painting a picture of DOMA as even more suspect than Amendment 2, Justice Kennedy, armed with that suspicion, then found his worst fears confirmed by the text of the statute and the rhetoric of its sponsors. In other words, those earlier observations about DOMA’s unusual nature gave Justice Kennedy cause to look carefully for animus. And when he did, he found it.

This analysis makes it more understandable why Justice Kennedy did not emphasize the normal step of considering whether DOMA was rationally connected to any legitimate government purpose. Don’t forget that he took care to note that step in Romer, concluding that Amendment 2 failed the rational relationship test given the extremely bad fit between the state’s justifications and the classification Amendment 2 created. He downplayed that inquiry in Windsor, a fact Justice Scalia noted in his dissent. But understanding the structure of Justice Kennedy’s opinion reveals why that de-emphasis makes sense if one has a particular understanding of animus. If one believes that animus is an affirmative constitutional violation—that is, if a conclusion that a statute is motivated by animus is fatal to the statute’s constitutionality, regardless of its other merits—then it makes sense for a judge to strike a law down as soon as he concludes that animus is in fact the reason behind a statute. No search for a redeeming justification is necessary. Game over.

And this is what Justice Kennedy did—almost. After reciting both DOMA’s text and the rhetoric of its supporters, he then returned to the effect DOMA had on state-recognized same-sex marriages. That recitation led him to conclude that DOMA’s effect was to place such marriages at a second-class level—literally, as valid for state law purposes but not federal law ones. Then, in the penultimate substantive sentence of the opinion, he finally mentioned, if only by implication, the rational basis requirement:

The federal statute is invalid, for no legitimate purpose overcomes the purpose and effect to disparage and to injure those whom the State, by its marriage laws, sought to protect in personhood and dignity.6

So it turns out there was a conclusion about the rational relationship test. Unfortunately for all of us, Justice Kennedy did not explain the process by which he performed that test. Indeed, he did not even identify the legitimate purposes DOMA might be thought to serve—a point Justice Scalia made in his dissent, which did identify some potential legitimate purposes behind DOMA.

The point here is not that Justice Scalia had the better of the argument with regard to the rational relationship test. It is, instead, to acknowledge that even Justice Kennedy implied that he had performed rational relationship review, even after concluding that DOMA was motivated by a “principal purpose . . . to impose inequality.”7

Did he need to make this concession at the very end of his opinion? Was there anything gained by at least nodding in the direction of traditional rational relationship review? Indeed, did he add that last phrase simply in response to Justice Scalia’s dissent (which would have been circulated before the opinions were finalized)? While we will likely not know the answer to this final question until justices’ papers from this case become available, at least we can think about the first two. Viewed one way, his acknowledgment of rational relationship review, even if conclusory and perhaps even an afterthought, nevertheless brought his opinion within the channel cut by the earlier animus opinions, including his own opinion in Romer. All of those earlier opinions—Moreno, Cleburne, and Romer (as well as Justice O’Connor’s separate opinion in Lawrence)—considered whether the challenged action promoted a legitimate government interest. Thanks to his quick rejection of DOMA’s rational basis, so did Windsor. In fact, Justice Kennedy’s very brief application of traditional rational basis review placed Windsor not just within the channel cut by the earlier animus cases but also within the much broader channel of equal protection review more generally.

That brief nod in the direction of traditional rational basis review may also have reflected Justice Kennedy’s desire to avoid confronting an interesting, important, but difficult problem—indeed, a problem on which Part II of this book will focus. By concluding, however quickly, that DOMA did not serve any legitimate government interest (or at least any such interest that “overcame” its disparaging purpose and effect), Justice Kennedy was able to avoid asking the following question: To what extent and by what mechanism does a legitimate motive cleanse a law of the taint of animus?

The Defense of Marriage Act itself provides an illustrative example of this question. In Windsor, Justice Scalia protested that DOMA did in fact survive rational basis scrutiny because it arguably promoted legitimate government interests (which is all that scrutiny requires). For example, he argued that DOMA avoided enmeshing the federal government in the difficult question of which state’s laws to apply to a same-sex couple for federal income tax purposes when a couple married in marriage equality jurisdiction A and then moved to marriage inequality jurisdiction B.8 Accept for the moment Justice Scalia’s argument that DOMA did in fact rationally further this government interest. What would that conclusion mean? Would it mean that DOMA was not really motivated by animus because it featured this second, more benign, set of effects? Or would it mean that DOMA might still be motivated by animus but that its promotion of a legitimate government interest served to cleanse (or, to use Justice Kennedy’s language, “overcome”) that illegitimate motivation?

It may seem like a hair-splitting distinction to ask whether Justice Scalia’s argument, if accepted, rescued DOMA from the allegation of animus or whether that rational justification stood alongside the animus conclusion but sufficed to provide the necessary justification needed for the statute to survive rational basis review. But this distinction is important. It is important as a practical matter because, depending on the approach we take, we may ask different questions, or at least apply different standards when answering them. It is also important theoretically. If we adopt the first of these views (that a benign explanation for the statute cleanses it from the stigma of animus), we can still conclude that statutes without such a cleansing—statutes that we do in fact label as based in animus—are unconstitutional, full stop. But if we adopt the other approach, then we sign up for the proposition that a statute may be labeled as reflecting “animus” but nevertheless may be constitutional.

Again, Justice Kennedy’s cursory rejection of any such rational basis allowed him to avoid these questions. But as observers attempting to make sense of the doctrinal concept of animus, we do not have the luxury of forcing the facts into conceptually simple packages. Windsor, as the last case to date to explicitly rely on the animus idea, leaves us with the question of how to relate allegations of unconstitutional animus to responses that, in fact, the challenged law promotes legitimate government interests and hence is constitutional. In a sense, it forces us to return to the question with which we started this book: What exactly counts as unconstitutional animus, and how do we uncover it? The cases we have examined in Part I provide us with the tools we need to construct a doctrinal structure answering that question. But those cases have not built that structure. That task will fall to us in Part II.

* * *

Part I of this book has identified, explained, and discussed the major modern Supreme Court statements about animus. Those statements have made clear that animus is a real concern of the Court. Nevertheless, that doctrine remains undertheorized. For example, these cases do not explicitly explain how animus can be identified. Moreno appeared to indict the subjective motivations of the congresspersons who reveled in the chance to deny food stamps to “hippies,” and Cleburne rejected as illegitimate the city council’s seeming capitulation to constituent dislike of the people they wished to exclude from the neighborhood. But in Romer, Justice Kennedy notably declined to directly accuse the people of Colorado from acting out of ill will. How necessary is such subjective ill will to a conclusion about animus?

Consider some of the other factors the Court focused on in these cases. In Romer Justice Kennedy emphasized the breadth of the burdens Amendment 2 imposed on gays and lesbians. Similarly, in Windsor he noted DOMA’s applicability to “over 1,000” federal statutes. In both of these cases he characterized the challenged laws as unusual. While these considerations make intuitive sense as warning signs about a law’s fundamental fairness and thus its constitutionality, intuitions by themselves are insufficient. Can we find a doctrinal foundation for these and other commonsense warning signs?

Finally, what result does an animus conclusion yield? One might think that, of all these questions, this one is the easiest: Given how troubling “animus” is as a motivation for government action, one might think that surely, once we find it, that should be the end of the case. But remember what Justice O’Connor said in Lawrence, describing the Court’s animus cases up to that point: “When a law exhibits such a desire to harm a politically unpopular group, we have applied a more searching form of rational basis review to strike down such laws under the Equal Protection Clause.” While the last part of that sentence (“to strike down such laws”) jibes with our intuition, we have to pay attention to the first part of the sentence (“When a law exhibits such a desire . . . we have applied a more searching form of rational basis review”). Is animus really a magic weapon—what one scholar describes as a silver bullet9—that is necessarily fatal to a statute? Or is Justice O’Connor telling us that animus merely triggers “a more searching form of rational basis review”? And what about the very last phrase, which states that that “more searching . . . review” triggers a strike-down? Giving Justice O’Connor credit for accurately describing the Court’s animus jurisprudence, it appears as though we need to investigate more carefully the consequences of a court uncovering animus.

The cases I have described in Part I raise these, and other, questions. They constitute the raw materials and tools constituting the Court’s animus doctrine. But the lack of clear answers to important questions those cases pose means that the Court has not yet used those materials to construct a coherent structure. That’s what Part II sets out to do.