Pete Marovich/Zuma Press/Corbis

Pro-same sex marriage activist Ryan Toney demonstrating outside the Supreme Court during arguments on the constitutionality of the Defense of Marriage Act, March 2013

The Equal Protection Clause of the Constitution, adopted three years after the Civil War as part of the Fourteenth Amendment, was designed principally to protect emancipated slaves from states that would deny their newly gained status as free and equal citizens. Yet in the Supreme Court this term, African-Americans were the big losers, while the winners were groups that the clause’s authors would not have imagined they were protecting at all: (1) southern states with a documented history of race discrimination; (2) whites challenging state efforts to help blacks by means of affirmative action; and (3) same-sex married couples.

This paradoxical departure from the Equal Protection Clause’s “original” purpose is not in itself a ground for criticism. Like much else in the Constitution, it was written in broad, general terms for the ages. It was not narrowly limited to the specific problem to which it was addressed at the time. It guarantees “equal protection of the laws” to all, not just to blacks. Through extended political struggle and evolving judicial interpretation, the scope of its protection has grown, embracing, among others, women and foreign nationals. So the fact that the Court, on the last day of its term, recognized the equal dignity of same-sex married couples should be seen as simply another important step in the inexorable development of our nation’s constitutional commitment to equality.

The Court’s concurrent denial of important protections to African-Americans, however, is of grave concern—not simply because it parts company with the Fourteenth Amendment’s original intent, but more fundamentally because it underscores that a majority of the Court believes that the time for legal measures designed to redress the nation’s long history of subordinating black citizens is past, and that the way forward is through mandated color-blindness. As Chief Justice John Roberts put it in 2007, invalidating a program designed to integrate public schools, “The way to stop discrimination on the basis of race is to stop discriminating on the basis of race.” For five of the Court’s justices, it’s as simple as that.

Only such a worldview could impose on states the same burden of justification when they seek to help African-Americans as when they seek to oppress them, as the Court has done in its decisions on affirmative action. And only such a worldview could lead the Court to elevate the “equal sovereignty” of states, a principle not even mentioned in the Constitution, over the voting rights of minority citizens—as the Court did this term in striking down the heart of the 1965 Voting Rights Act.* By imposing this view on the nation, the Court has not only refused to interpret the Equal Protection Clause as mandating measures to ameliorate racial inequality, but has used the clause to obstruct the states and Congress from doing so.


First, the good news. In a historic, eloquent, if at times cryptic opinion, the Supreme Court took its first important step toward recognizing the constitutional right to same-sex marriage. On the last day of its 2012–2013 term, in United States v. Windsor, the Court ruled unconstitutional a provision of Congress’s 1996 Defense of Marriage Act (DOMA) that selectively denied many federal benefits to same-sex couples whose marriages were recognized by state law. Justice Anthony Kennedy, joined by the Court’s four liberal justices, said that when the states chose to extend marriage to same-sex couples, they “conferred upon them a dignity and status of immense import.” When Congress, by contrast, denied those same couples a raft of federal benefits, Kennedy wrote, it “impose[d] a disadvantage, a separate status, and so a stigma upon all who enter into same-sex marriages.”

Kennedy’s decision was carefully limited to the question of Congress’s authority to restrict marriage to opposite-sex couples; he insisted that he was not deciding the much larger question of whether states can restrict marriage along those lines. And by a different five-justice majority, the Court in Hollingsworth v. Perry the same day declined to decide the issue of state authority to restrict marriage. Chief Justice John Roberts, joined in an unlikely majority by Justices Antonin Scalia, Stephen Breyer, Ruth Bader Ginsburg, and Elena Kagan, concluded that the citizens who sought to appeal a district court ruling invalidating California’s anti–gay marriage Proposition 8 lacked standing to bring the appeal. By dismissing the appeal and leaving the district court decision intact, the Court effectively legalized same-sex marriage in California, but without making any pronouncements about how it would decide the matter on its own merits.

Together, these decisions are a consummate act of judicial statesmanship. They extend federal benefits to all same-sex married couples in states that recognize gay marriage, expand the number of states recognizing gay marriage to thirteen, yet leave open for the time being the ultimate issue of state power to limit marriage to the union of a man and woman. The Court took a significant step toward recognition of the equality rights of gays and lesbians. But by not imposing same-sex marriage on the three quarters of the states whose laws still forbid it, the Court has allowed the issue to develop further through the political process—where its trajectory is all but inevitable.


The case against Congress’s attempt to define marriage through DOMA was always easier to argue than the case against a state marriage restriction, because marriage is largely the province of the states. The federal government has historically deferred to state definitions of marriage to determine who should receive marriage benefits. Congress broke from that tradition when it enacted DOMA, after a Hawaii court decision threatened to recognize same-sex marriage for the first time. DOMA displaced all state definitions in favor of a federal law that restricted marriage to unions between a man and a woman.

The fact that DOMA was an “unusual deviation from the usual tradition” of respecting state law, Kennedy reasoned, was “strong evidence of a law having the purpose and effect of disapproval of that class,” i.e., the class of same-sex married couples. Looking further into the law’s legislative history, as well as the arguments advanced in its defense in Windsor itself, Kennedy found that “interference with the equal dignity of same-sex marriages” was not just the law’s effect, but its very “essence.” Laws designed to demean or stigmatize a group have long been treated, for that reason, as a violation of equal protection of the law, and Kennedy appeared to conclude that DOMA was of the same sort.

I say “appeared,” however, because Justice Kennedy was not always clear about which of two constitutional principles he was relying on. He suggested that DOMA denied equal protection because it created a double standard, treating marriages recognized by states differently depending on whether they were unions of opposite-sex or same-sex couples. But he also said that DOMA was a violation of due process, suggesting that the right of same-sex couples to marry, at least when recognized by the states, is a fundamental “liberty” that cannot be infringed without a substantial justification. Thus, he wrote, “while the Fifth Amendment [due process clause] itself withdraws from Government the power to degrade or demean in the way this law does, the equal protection guarantee…makes that Fifth Amendment right all the more specific and all the better understood and preserved.”

Justice Kennedy said nothing, however, about one of the most hotly debated questions in the case—namely, what standard of scrutiny should apply to laws that discriminate on the basis of sexual orientation? Much as the Court did when it first invalidated laws discriminating on the basis of sex, the Court in Windsor avoided specifying a standard of review, but seemed silently to apply a healthy measure of skepticism to the distinction.

In his dissent, Justice Scalia vehemently objected to the majority’s lack of clarity about whether it was relying on equal protection or due process, and what standard of review it was employing. And he criticized Kennedy for failing to address more forthrightly the justifications the government asserted for DOMA. The law’s defenders said it served a desire to have a uniform standard of marriage for purposes of federal law, and to avoid difficult questions that might arise when different state marriage laws conflict.

Kennedy’s decision would have been stronger had he addressed these arguments head-on, and it should not have been difficult to do so. The federal government has long tolerated differences in the way states define marriage (such as varying age limits for when one can marry, and different limits on marriage between cousins), so why demand uniformity only with respect to the gender identity of the married couples? And the desire to avoid difficult “choice of law” questions between conflicting state laws, the kinds of questions courts routinely confront and decide, can hardly merit relegating an entire group of people to second-class status, as Scalia seemed to suggest it should.

Moreover, a clear discussion of the appropriate standard of review for laws that treat people differently on the basis of sexual orientation would have offered useful guidance to lower courts. Equal protection doctrine has long identified certain types of distinctions as inherently more “suspect” than others. Thus, laws that classify people on the basis of race or gender call for stringent scrutiny, while laws that classify on many other bases, such as age, merit only deferential review. The levels of scrutiny guide the courts in how suspicious they should be of particular types of laws. The argument for “heightened scrutiny” of laws discriminating on the basis of sexual orientation is very strong, in view of the long history of discrimination against gays and lesbians, and the irrelevance of sexual orientation to most official decisions. Indeed, the Obama administration, which declined to defend DOMA, argued forcefully in its brief that heightened scrutiny was appropriate for these reasons.


So as a matter of pure judicial craft, Justice Kennedy could have been more clear and definite. But judging is not only a matter of craft. Justice Kennedy sought to walk a fine line—invalidating a federal law limiting marriage to opposite-sex couples without simultaneously resolving the question of whether similar state laws are constitutional. Had he pronounced that all classifications based on sexual orientation are suspect and subject to heightened scrutiny, or had he held that the due process clause affirmatively protects the right of two committed adults to marry, regardless of sex, he would have, in effect, decided the broader issue. By focusing instead on the federal government’s unusual intrusion into a prerogative of state law, and its dramatic imposition of a double standard in states that recognize same-sex marriage, he wisely limited his decision to the federal law at issue. By allowing the transition to full recognition of same-sex marriage to take place gradually, Kennedy’s opinion avoids the backlash that a federal mandate to recognize same-sex marriage in every state might have triggered.

Ironically, Justice Kennedy did not join the majority opinion in Hollingsworth v. Perry, the California case challenging Proposition 8—the ban on gay marriage in California that was adopted in 2008 and found unconstitutional both by a district court and by the US Court of Appeals for the Ninth Circuit. That case squarely presented the broader question of state authority to restrict marriage to unions between a man and a woman. While the majority dismissed the appeal by Proposition 8 proponents on the narrow ground that they lacked “standing” to appeal the district court’s decision, Justice Kennedy dissented, joined by three other justices, and would have decided that case on the merits.

Had he attracted a fifth vote for that view, it would have rendered all the care he took in the Windsor decision moot, because the Court would have had to address directly the broad question of state authority to ban same-sex marriage. The statesmanship of the two decisions, then, comes not from any one justice, but from the collective action of the Court as a whole. There’s a reason we have nine justices.

As for the future, Chief Justice Roberts wrote a separate dissent in Windsor for the sole purpose of underscoring that Kennedy’s majority opinion did not resolve the issue of state authority to restrict marriage. But notwithstanding its carefully limited reasoning, it is difficult to read Justice Kennedy’s opinion in Windsor without sensing, as Justice Scalia warned in his angry dissenting opinion, that the “other shoe” will drop eventually, and the Court will ultimately recognize that denying same-sex couples the right to marry violates the Constitution in the same way that denying mixed-race couples the right to marry does. (Significantly, in his opinion for the Court, Justice Kennedy cited Loving v. Virginia, the Court’s 1967 decision on interracial marriage, for the proposition that the state’s otherwise broad authority to define marriage is subject to constitutional limits.)

That would be the right outcome, both morally and constitutionally. The justifications for denying a “dignity and status of immense import” to gay and lesbian couples are unpersuasive, whether at the federal or the state level. Same-sex couples, like opposite-sex couples, can and do make long-term commitments to live together as a family unit. They can and do raise children, pay taxes, care for dependent family members, and the like. The real reason so many states have denied marital status to same-sex couples is a sense, founded on tradition and moral disapproval, that these couples do not deserve it, simply because they are of the same sex. But the Court has already held, in Lawrence v. Texas, its 2003 decision invalidating state sodomy laws, that tradition and moral disapproval are insufficient grounds for differential treatment of gay and straight couples.

We will have to wait for recognition of a full-fledged constitutional right of same-sex couples to marry, but it is hard to imagine that it will not come.


Now, the bad news. African-Americans, the original beneficiaries of the Equal Protection Clause, fared poorly this term. Most disastrously, in Shelby County v. Holder, the Court struck down the Voting Rights Act’s requirement that states with histories of discriminatory voting practices must obtain advance approval, or “preclearance,” from the federal government before changing their voting rules. This mechanism, which requires the states to show that their rule changes will not dilute minority voting rights, has been the statute’s single most effective provision. It provides an efficient administrative tool to prevent discriminatory practices from taking effect.

By a 5–4 vote, the Court’s conservatives found that the formula used to select which states are subject to the preclearance process, first developed when the law was enacted in 1965, was outdated. Times have changed, Chief Justice Roberts wrote, and Congress’s effort to protect minority voters no longer justifies this intrusion on states’ rights. But as Justice Ginsburg pointed out in dissent, application of the preclearance requirement is in fact determined not just by the initial formula (tied to historic discriminatory voting practices), but also by “bail-out” and “bail-in” provisions that effectively update the law’s coverage on an ongoing basis.

Those provisions allow states to free themselves from the obligations if they abide by the law for ten years, and empower courts to impose preclearance obligations as a remedy on new states and localities that have violated the law. Thus, Congress started with the states with the worst records, allowed those who avoid problems to opt out, and allowed new transgressors to be added. The Court had upheld this scheme four times previously. Moreover, Congress found in 2006, after many months of hearings and 15,000 pages of evidence, that discrimination remained a problem in the covered jurisdictions.

What was really going on in Shelby County? The Court’s five conservatives resent the Voting Rights Act because, by requiring states to ensure that their voting rule changes don’t harm minority voters, it effectively requires them to take account of race. The Court’s conservatives, committed to color-blindness, find this practice objectionable—even though it is designed to help, not harm, minorities. But that ideological objection should not carry any weight. The Fifteenth Amendment, which guarantees the right to vote, also vests in Congress the authority to protect voting rights by “appropriate legislation.” The Court has previously said that as long as Congress adopts rational means for doing so, the Court should defer to its judgment.

That deference is now gone, it appears, in the interest of an abstract commitment to color-blindness that is in fact blind to the reality that race continues to play a pervasive role in voting regulation. Times have changed, but a strong and legitimate concern about racial gerrymandering remains. The crude barriers to voting that states once erected out of pure racism have been eliminated, largely because of the act. But because of the persistence of “racially polarized voting,” in which racial groups predictably divide in whom they support, politicians are often tempted to deny or dilute the votes of those groups that favor their challengers. The Republican efforts to suppress voting through “voter-identification” laws in 2012 illustrate the point. Race matters, and simply saying it shouldn’t won’t solve the problem.

The same color-blind approach frames the Court’s skeptical attitude toward affirmative action. In Fisher v. University of Texas, Abigail Fisher, a white student denied admission, argued that the university’s modest consideration of race, as a “factor of a factor of a factor,” in assessing applicants violated her right to equal treatment. The Court has long ruled that affirmative action plans must satisfy “strict scrutiny,” the same standard it applies to laws that discriminate against blacks. But it has also acknowledged that the educational benefits of diversity may under some circumstances permit admissions officers to consider race, as one of many factors, to ensure a diverse student body.

When the Court agreed to decide Fisher, observers feared it might end affirmative action. The practice’s survival in 2003, the last time the Court addressed it, had hinged on Justice Sandra Day O’Connor’s vote, and she has been replace by Justice Samuel Alito, no friend of affirmative action. The case was clearly the subject of a lengthy internal fight; the Court took longer to decide the Fisher case than any case this term. The final result, however, was a compromise, joined by seven of the Court’s eight justices. (Justice Kagan recused herself; only Justice Ginsburg dissented.) The Court essentially maintained the status quo. It vacated the lower court’s decision upholding the Texas plan, but only because the lower court did not apply sufficiently rigorous review. Specifically, the Court reaffirmed that universities may not take race into account unless they can show that no reasonably available race-neutral alternatives could achieve sufficient diversity.

This ruling will change admissions policies only if universities are currently using race even where there are available race-neutral alternatives that could work just as well. That seems highly unlikely. No one likes to consider race; if universities already do so only as a last resort, their plans should be sustained. Moreover, where a race-neutral option would work, it is certainly preferable. Some have suggested, for example, in view of the large numbers of poor black and Latino students, that considering socioeconomic class in admission might have the effect of achieving significant racial diversity. But the Court in Fisher also reaffirmed that courts should defer to universities on the diversity they need, as courts are ill-equipped to second-guess such academic judgments. As a result, universities should still have substantial leeway in using race where necessary to achieve diversity.

In both the voting rights and affirmative action settings, the Court has in essence turned the Equal Protection Clause on its head. It has interpreted the clause, designed to assist newly freed slaves, as an affirmative obstacle to federal and state officials’ efforts to improve the position of African-Americans and other disadvantaged minorities through race-conscious measures.

In the end, then, the Court took an important step forward on equality for married gays and lesbians, a disturbing step backward on voting rights, and largely marched in place on affirmative action. A majority of the justices seem poised to recognize the rights of gays and lesbians to marry even in states that do not afford such recognition; but they will evidently wait until more progress has been made politically on that issue before taking it on directly. As for racial equality, however, the Court remains steadfastly blindered in its commitment to color-blindness. As a result, most of its constitutional race discrimination decisions in recent years have ruled for white plaintiffs, or states, challenging laws designed to protect the initial intended beneficiaries of the Equal Protection Clause.

Our ongoing constitutional commitment to equality requires us to recognize that groups previously excluded from protection—such as women and gays and lesbians—should be included. But it also requires a more nuanced understanding of how pervasive and insidious certain forms of discrimination remain. This term, as in prior years, the Court proved much better at extending the scope of equal protection to new groups than at providing, or even permitting the political branches to provide, a meaningful measure of equality to those for whom the clause was written in the first place.