There is a first time for everything. This past weekend, it was the first time a Supreme Court justice performed a gay wedding. It was Justice Ruth Bader Ginsburg, one of the five justices who voted in United States v. Windsor to invalidate the federal Defense of Marriage Act (DOMA) on constitutional grounds. There is no legal connection between these two events: The District of Columbia, where the wedding took place, legalized same-sex marriage of its own accord, and the ruling in Windsor stops the federal government from refusing to recognize valid same-sex marriages. But the solemnization of a gay wedding in a very public place—the Kennedy Center in Washington, DC, of which one of the spouses is the current president—by a Supreme Court Justice signifies the normalization of same-sex marriage and the inevitable march towards full acceptance of it.
Although Windsor did not directly facilitate the Kennedy Center wedding, it did change the meaning of that new marriage. Because of Windsor, the couple will have a marriage that is recognized by both the District of Columbia and the federal government—not the “skim-milk marriage” that Justice Ginsburg decried in the oral argument in the Windsor case. The Windsor ruling will protect other gay married couples as well. But the full effects of Windsor are still unfolding. One question relating to the ruling is whether the federal government must recognize all validly celebrated same-sex marriages, or only those recognized by a couple’s home state. A second question is whether Windsor can be used to force states to recognize same-sex marriage from sister states, even though the ruling did not directly implicate this issue. Both of these questions implicate the law of marriage recognition, and the unusual form that it has taken in the same-sex marriage controversy.
The Ruling in United States v. Windsor
Writing for a 5-4 majority, Justice Kennedy authored the majority opinion in Windsor, which held that Section Three of the federal Defense of Marriage Act (DOMA) violated the equal protection principles embodied in the Fifth Amendment’s Due Process Clause. (A full description of the case is available here.)
DOMA was enacted at the beginning of the anti-same-sex-marriage frenzy in the mid-1990s, as Congress and the states worried about the potential impact of the legalization of same-sex marriage in a single state:Would every jurisdiction have to recognize same-sex marriages that were legally valid somewhere? DOMA was Congress’s attempt to make sure that that wouldn’t happen. It provided, in Section Two, that states could not be forced, as a matter of full faith and credit, to give effect to same-sex marriages from sister states and, in Section Three, that the definition of marriage for federal-law purposes is a union between a man and a woman.
The challenge in Windsor involved only Section Three. It arose out of an IRS ruling, which held that a decedent’s estate was subject to more than $350,000 in federal estate taxes because, although transfers to a spouse at death are usually tax-free, the federal government could not recognize the decedent’s lesbian marriage.
The marriage in Windsor was not the only one adversely affected by Section Three. Almost every same-sex married couple faced some consequence of the federal government’s refusing to recognize their union. There are over 1000 federal laws that turn in some way on marital status, and, in most instances, federal law defers to state law on marital status. In other words, if two people are married under New York law, they are married under federal law for Social Security, income and estate tax, and myriad other purposes. But DOMA created an exception to this scheme just for same-sex marriages.
It was this deviation from the norm that ultimately led to the invalidation of DOMA. While the federal government is not obligated to defer to state definitions of marital status, its refusal to do so with respect to only one type of marriage raises suspicions that Congress acted out of animus or hostility to gays and lesbians, rather than out of some legitimate government need. And in a prior equal-protection case, Romer v. Evans (1993), the Court had held that even under the lowest level of scrutiny, a law motivated by animus against a disfavored group could not survive.
Justice Kennedy’s opinion acknowledged the developing social norms about gay rights and relationships, which had gone from unheard of to fundamental, at least in some jurisdictions. It also considered the traditional regulation of marriage. Although “by history and tradition” marriage has been “treated as being within the authority and realm of the separate states,” Kennedy wrote, Congress has the authority to “make determinations that bear on marital rights and privileges” when acting “in the exercise of its own proper authority.” Congress thus can, for example, refuse to grant citizenship rights to the non-citizen spouse in a sham marriage (one that is entered into solely for purposes of procuring immigration rights) even if the marriage would be valid for state-law purposes. Congress can also make its own determinations about marriage, if it chooses to, when doling out Social Security benefits, or impose special protections on spouses under pension plans regulated by ERISA. But, as noted above, Congress mostly has not chosen to exercise this prerogative, choosing instead to defer to the state’s own laws of marriage.
This history of deference made DOMA all the more suspicious. It was not a targeted exclusion like the INS rule on sham marriages; it had “far greater reach” and was “applicable to over 1,000 federal statutes and the whole realm of federal regulations.” Moreover, it was not a generally-applicable rule. It was targeted at a single class of people, which has historically suffered discrimination. DOMA’s fatal flaw was its rejection of “the long-established precept that the incidents, benefits, and obligations of marriage are uniform for all married couples within each State, though they may vary, subject to constitutional guarantees, from one State to the next.” The “unusual character” of the discrimination created a so-called Romer problem, one that was exacerbated by the fact that Congress was seeking to penalize a group that states like New York (the home state of the couple in Windsor) specifically sought to protect. The Court in Romer wrote that the guarantee of equality “must at the very least mean that a bare congressional desire to harm a politically unpopular group cannot justify disparate treatment of that group.” The more unusual a discriminatory law is, the more likely it is to be the product of animus. DOMA had to go.
Implementing Windsor
Although the Court’s ruling made clear that Section Three of DOMA could no longer be enforced, questions arose immediately about the fallout. The biggest question was whether the federal government would recognize all same-sex marriages, or only those in which the couple’s home state recognized their marriage. (Thirteen states and the District of Columbia allow gay couples to marry; thirty-seven do not.) Windsor seemed to leave open both possibilities, since it was the practice of not deferring to the home state (just for same-sex marriages) that seemed to raise the Romer problem. It was also not clear, after Windsor, whether the federal government would treat civil unions or domestic partnerships as marriages if state law provides for such equivalency.
The federal government, as it turns out, is inconsistent in its marriage-recognition rules. In some instances, it relies on a place-of-residence rule, which means that you are considered married for federal-law purposes if your home state treats you as married. In others, it relies on a place-of-celebration rule, which means that you are considered married for federal-law purposes if your marriage was valid where celebrated, regardless of which state you call home. In either case, the federal government is deferring to the state law of marriage, rather than providing its own definition, but the question is, Which state? For couples who live in a marriage-equality state, either rule will produce the same outcome. But couples who live in a state that does not allow same-sex marriage will benefit from a place-of-celebration rule, but not from a place-of-residence rule.
The Social Security Administration, for example, applies a place-of-residence rule when determining whether an individual can apply for benefits based a spouse’s earning records. This means that the members of a couple who live and marry in Massachusetts will be entitled to Social Security benefits that a couple who lives in Texas, but marries in Maryland, will be denied. But it also appears to allow civil-union partners to claim such benefits because it defers to state inheritance law to determine eligibility (and all civil union laws allow partners to inherit as if they are married).
As of August 2013, the IRS, however, will apply a place-of-celebration rule to same-sex married couples. Regardless of the laws of their home states, such couples will be allowed (and forced) to file either joint federal tax returns or married-filing-separately tax returns. The new rule, just announced by the Department of the Treasury, allows couples to file amended tax returns for past years if doing so would yield a refund, but does not require amended returns for those who would pay more. (Married couples pay the so-called “marriage penalty” if their earnings are relatively equal to one another.) The IRS also made clear, however, that it will not treat couples in civil unions or other marriage-equivalent statuses as married.
For couples who live in one of the thirty-seven non-marriage-equality states, the new approach may require them to shuffle some paper in order to comply with both state and federal tax laws. Because state returns are often based on federal returns, these couples may have to first prepare and file a real return—reflecting their joint taxpayer status—and then prepare (and not file) a dummy return as a single taxpayer in order to get the correct adjusted gross income amount necessary for transfer to the state return. (Under DOMA, gay married couples in marriage-equality states had to do the reverse in order to produce a single-taxpayer federal form and a joint-taxpayer state form.)
In a separate move, the Department of Health and Human Services said that it would also apply a place-of-celebration rule for purposes of Medicare benefits. There aren’t as many marital-status implications for Medicare, but the rule could affect things like the right to live in a nursing home where one’s spouse lives.
As reflected in the Medicare and IRS announcements, the Obama Administration has come out strongly in favor of same-sex marriage recognition. It started on this path years ago, when the Department of Justice announced that it would no longer defend Section Three in litigation. It has followed through on its commitment in the post-Windsor months, pushing federal agencies to act swiftly to bestow federal benefits on gay married couples.
DOMA’s Section Two
A second set of questions also relates to interstate marriage recognition for same-sex marriages. As noted above, Section Two of DOMA purports to give states the right to refuse recognition to marriages from sister states. In DOMA’s wake, more than 40 states enacted statutory provisions or constitutional amendments that adopted categorical rules of non-recognition that applied only to same-sex marriages.
Under the traditional rules of interstate marriage recognition, states had the right to decide whether to recognize marriages from other states that the states themselves would not allow. As a general matter, states followed the place-of-celebration rule, which meant that marriages that were valid where they were celebrated were valid everywhere (and marriages that were void where celebrated were void everywhere). States sometimes applied two exceptions to this rule, one for marriages deemed to violate natural law and one for marriages that could not be recognized because there was some positive law (a statute) preventing extraterritorial recognition.
Marriages were never understood to command full faith and credit. Rather, states recognized each other’s marriages as an exercise of comity, or respect, for one another’s laws. For the most part, states leaned towards recognition, given the consequences for couples’ mobility and stability of having marital status change at state borders. The decision whether to recognize any particular marriage, however, was left largely to courts.
The mini-DOMAs passed between the mid-1990s and late 2000s deviated from this historical approach, singling out a certain type of marriage for a categorical rule of non-recognition. Could Windsor be used to challenge the state-law non-recognition rules?
Windsor has no direct relevance to state marriage laws; it ruled only on the constitutionality of the federal DOMA. But its reasoning is highly relevant here. State laws are subject to constitutional challenges that are rooted in both equal protection and due process principles. (The proposition that state marriage laws must conform to federal constitutional standards has been clear since the Court struck down Virginia’s ban on interracial marriage in 1967 in Loving v. Virginia.) Thus, Romer can be used to challenge state laws that target a disfavored group for adverse treatment, just as it was used in Windsor to challenge a federal law that did that. And Windsor makes clear that even if a governmental body is not compelled to recognize a particular type of marriage, its decision not to do so is suspect if it deviates from longstanding historical precedent. Mini-DOMAs that target same-sex marriages for non-recognition are as vulnerable as DOMA on this theory.
Moreover, when the rules of interstate marriage recognition were developed, federal constitutional norms of equal protection and due process were neither as broad nor as clear as they are now. Thus, these rules developed in a world in which states had more latitude to deny civil rights based on state prerogative or policy. Those same rules today—giving states the power to pick and choose among out-of-state marriages for recognition purposes—may simply not be valid in some cases.
A pending case in federal district court in Ohio raises some of these questions. In Obergefell v. Kasich, which my fellow Justia columnist David Kemp has written about in detail here, the spouse of a terminally-ill man sued to ensure that Ohio would issue his husband a death certificate that reflected his marriage. The couple flew to Maryland to marry, given Ohio’s ban on same-sex marriage. In granting a temporary restraining order (TRO) against the state of Ohio, the judge ruled that, under the principles of Windsor and Romer, the state cannot refuse recognition to same-sex marriages validly celebrated elsewhere. Important to the ruling was the fact that Ohio has a long history of recognizing prohibited marriages, even those where the couple fled Ohio in order to evade its restrictive laws and then returned home to demand recognition. That raised a troubling question: If Ohio bent over backwards to recognize all marriages, why does it refuse to do the same for gay marriages?
The Obergefell case is only the first of many challenges that will follow Windsor. It may well be, as Justice Scalia predicted in his angry dissent in Windsor, that even without ruling that state bans on same-sex marriage are unconstitutional (which the Court could have done, but did not do, in Perry v. Schwarzenegger), the Court may have done them in indirectly.