The End of a Bad Era: Congress Repeals the Defense of Marriage Act | Joanna L. Grossman | Verdict

More than a quarter century after p،ing it, Congress has repealed the Defense of Marriage Act (DOMA). This law was p،ed in haste in 2003 as anti-same-،-marriage sentiments ،ned momentum in the United States. Through a new law, the Respect for Marriage Act, Congress has provided a legislative backstop for marriage rights in case the Supreme Court decides to junk them as t،ughtlessly as it has done with abortion rights.

The Respect for Marriage Act closes an ، chapter in our history, in which Congress and a Democratic President allowed themselves to be enlisted to intervene in an unprecedented manner in state marriage law in the name of bigotry. And alt،ugh DOMA itself became irrelevant many years ago due to two Supreme Court rulings (discussed below), this new law will prevent it from rearing its ، head in the future s،uld the cons،utional landscape suddenly change. It is also symbolically important for Congress to erase from its books a law that singled out gay and ، couples for adverse treatment for no reason other than animus a،nst the LGBT community.

In this column, I’ll explain the history of the Defense of Marriage Act, its odd place in the national controversy over same-، marriage, and the changes effectuated by the Respect for Marriage Act.

The Origins of the Defense of Marriage Act

Congress enacted DOMA in 1996, after very brief deliberation and hearings. DOMA p،ed both ،uses of Congress by a wide margin—342-67 in the House and 85-14 in the Senate. And somewhat surprisingly, a Democratic President, Bill Clinton, signed it swiftly into law.

DOMA did two things. Section Two of the Act purported to give states the right to refuse recognition to same-، marriages that have been cele،ted in other states. Section Three provided that, for any federal-law purpose, the word “marriage” means only a legal union between one man and one woman, and a “spouse” refers only to someone of the opposite ،. It did these things at a time when same-، couples could not legally marry anywhere in the United States—and indeed would not be able to for another eight years.

DOMA was a reactionary response to the first real possibility of legal same-، marriage in the United States. In 1996, Hawaii was on the verge of legalizing same-، marriage after the state’s highest court had ruled, in Baehr v. Lewin (2003), that banning same-، marriage is a form of ، discrimination—for men were allowed to marry women, but women were not allowed to marry women—that merited the highest level of judicial scrutiny. It was widely predicted that the State of Hawaii would lose the case on remand because it would be unable to s،w that it had a compelling reason for engaging in this type of discrimination and that the state’s existing ban on same-، marriage would be held uncons،utional under the Hawaii cons،ution.

Hawaii loomed large in the growing national controversy over same-، marriage. Prior to 1996, there basically was no law regarding same-، marriage anywhere in the United States. No state explicitly allowed it, but very few expressly banned it either. Most state marriage laws were silent on the gender of the parties. Several lawsuits were brought in the 1970s to test the proposition that silence on gender meant that same-، couples already had the right to marry—and that a contrary interpretation of the marriage licensing statutes would violate the federal or state cons،utions on due process or equal protection grounds. The courts w، heard these cases ruled unanimously that silence meant exclusion, as the legislative intent from decades or even centuries in the past could not be understood to em،ce a type of marriage never contemplated; and, just as unanimously, they refused to seriously engage with the possibility that exclusion of same-، couples from the right to marriage presented any kind of a cons،utional problem.

During the 1980s, the gay rights movement backed off the quest for marriage rights, focusing on other issues and also seeking the enactment of local domestic partner،p ordinances that carried more limited rights than marriage but made some inroads into the traditional system of non-recognition. But in the 1990s, a second round of court challenges to state marriage laws emerged, including the one in Hawaii. These challenges were filed in a variety of different states and were carefully constructed to raise only state cons،utional questions in order to prevent removal of the cases to federal court as well as a ،entially adverse ruling from the U.S. Supreme Court that would stop the marriage-rights movement in its tracks. Unlike the first round of challenges, the second round was brought a،nst a backdrop of increasing support for gay rights and the firm establishment in other contexts of a cons،utional right to marry. These challenges were thus taken more seriously on both sides.

The litigation in Hawaii ، the central question into the national s،light: S،uld same-، couples be allowed to marry? And it also raised a related question: If Hawaii were to allow same-، marriage, would that effectively mean that such marriages would be ، upon every state? Both opponents and proponents of same-، marriage ،umed that if Hawaii legalized same-، marriage, then couples from all over the country would marry there and then seek recognition of their marriages at ،me. Evan Wolfson, a strong proponent, argued that many same-، couples “in and out of Hawaii” would take advantage of such a “landmark victory”; and the “great majority of t،se w، travel to Hawaii to marry will return to their ،mes in the rest of the country expecting full legal nationwide recognition of their marriage unions.” Senator Trent Lott, an ardent opponent, argued that Hawaii could be left alone if “such a decision affected only Hawaii.” But it wouldn’t, he argued: “[A] court decision . . . would raise threatening possibilities in other States.”

Enter DOMA. Congress decided to “defend” traditional marriage by allowing states to avoid compelled recognition of same-، marriages cele،ted in other states; and by allowing the federal government to ignore same-، marriages cele،ted in any American state, or anywhere abroad.

DOMA and Interstate Recognition of Same-Sex Marriages

For several years after DOMA was p،ed, it had no practical effect. Hawaii did not legalize marriage by same-، couples until 2013—almost two decades after the court ruling that spurred DOMA and its state-level counterparts. While Baehr was pending on remand, Hawaii voters amended the state cons،ution by referendum to give the legislature the power to ban same-، marriage, which it did (and then reversed itself in 2013). So DOMA, at that time in history, was just background noise—“defending” marriage a،nst an attack that didn’t exist.

DOMA became at least tangentially relevant in the United States in 2001, when the first same-، marriages were legally cele،ted in foreign jurisdictions like Ca،a. Normally, a marriage in the foreign country would be recognized in any American state as long as it was valid where cele،ted. But despite the ،ential for a conflict with DOMA, there were no court cases at this stage challenging any state’s refusal—or the federal government’s refusal—to give effect to same-، marriages legally cele،ted in a foreign jurisdiction.

DOMA only became relevant—and problematic—when the first same-، marriages were cele،ted in the U.S. By virtue of a 2003 ruling of the state’s highest court, in Goodridge v. Department of Public Health, M،achusetts began issuing marriage licenses to same-، couples in May 2004. But even with the legalization of same-، marriage in M،achusetts, Section Two of DOMA had no meaning. It ostensibly granted states the right to refuse recognition of each other’s same-، marriages. As a technical matter, this provision DOMA amended the federal Full Faith and Credit Act (28 U.S.C. § 1738) to provide that states need not grant “full faith and credit” to same-، marriages. In response, four-fifths of the states p،ed so-called mini-statutes or cons،utional amendments banning both the cele،tion and recognition of same-، marriages.

This was all somewhat of a ruse, ،wever. In more than two centuries of cases, courts had never understood full faith and credit to compel interstate marriage recognition. Rather, the “exacting” obligations of full faith and credit had been reserved for final judgments in judicial proceedings—including divorce. Marriage is not the ،uct of a court judgment; it is merely the application of a state law, which requires only that other states meet “certain minimum requirements” of full faith and credit. Under the general conflict of laws principles, states can prefer their own law—including their law denying same-، marriage—over the competing c،ice of another state—one allowing same-، marriage—as long as the c،ice is “neither arbitrary nor fundamentally unfair.”

The law of interstate marriage recognition—which is completely independent from the law of marriage cele،tion—had always left room for a state to refuse recognition of marriages to which it strenuously objected. By c،ice rather than necessity, all states follow some version of the “place of cele،tion” rule when deciding whether to give effect to a marriage that it would not have permitted to be cele،ted. Under this rule, a marriage is valid everywhere if valid where cele،ted. It reflects the principle of comity, or respect for sister states, as well as the practical need for certainty and portability of marital status.

This general pro-recognition rule is subject to two standard exceptions for marriages that violate “natural law” (reserved for the denial of recognition to closely ،uous or bigamous marriages) or violate positive law (such as a statute expressly denying extraterritorial recognition for a certain type of marriage). Historically, some states that prohibited interracial marriage refused to recognize such marriage even if they had been validly cele،ted elsewhere under one of the two exceptions; other states gave effect to interracial marriages despite not allowing them to be cele،ted. As many law professors explained in the congressional hearings on DOMA, Section Two of DOMA did not grant the states any right they did not already possess. And, indeed, a، the ،dreds of cases about same-، marriage litigated between 1996 and 2015, not a single one involved the application of this provision. And once the Supreme Court held in Obergefell v. Hodges (2015) that it is uncons،utional to deny same-، couples the right to marry, the question of interstate recognition became irrelevant (for now, see below) as well.

DOMA and Federal-Law Recognition of Same-Sex Marriages

Section Three of DOMA became relevant, ،wever, as states began to legalize same-، marriage. This provision said that regardless of their validity under state law, same-، marriages could not be recognized for any federal law purpose. Because of the nature of the interaction between federal and state law where marital status is relevant, this imposed a variety of completely senseless bureaucratic h،les on same-، couples as well as significant substantive hard،p in some cases.

The basic problem stemmed from the fact that alt،ugh Congress routinely p،es laws for which marital status is relevant, they rarely provide their own definition of a valid marriage. Rather, the standard approach has been for federal laws to rely on state determinations of marital status (as well as other family law questions like whether a legal parent-child relation،p exists between two people). Thus, if a person is seeking spousal benefits from the Social Security Administration (a federal benefit), en،lement would turn on marital status at the state level. This is because marriages are licensed and regulated at the state level, and people rely on certainty about their status. Imagine the silliness of a system in which a married person was simultaneously considered married and single by different governmental agencies. That is exactly what happened to gay and ، couples w، legally married in the United States. They were married under state law, but unmarried under federal law. This led to both predictable and surprising headaches.

This provision also deprived legally married same-، couples of important rights, such as Social Security benefits, estate tax exemptions, marriage-based immigration rights, and so on. It also imposed a serious stigma on couples w، were told that their marriages just did not count on the federal level—for no reason. Section Three represented an unusual power grab by the federal government, as well as an ironic rebuke of federalism for the conservative sponsors w، championed the bill, and w، would champion federalism—styled as “states’ rights”—in virtually any other context.

Cons،utional Protection for the Right of Same-Sex Couples to Marry

The unprecedented nature of DOMA ultimately led to its cons،utional downfall. In 2013, the Supreme Court held in United States v. Windsor that the federal-law provision ran afoul of the equal protection principles guaranteed by the Fifth Amendment (parallel to the Equal Protection Clause that applies to the states through the Fourteenth Amendment). By the time the Court heard this case, thirteen states permitted same-، couples to marry (and thirty percent of the American population lived in one of t،se states). Public opinion had also ،fted dramatically by that point, with more than half of survey respondents supporting marriage equality. Moreover, the Obama administration had ceased defending the cons،utionality of the law, for reasons laid out in the “Holder Memo.”

Windsor arose when the widow of a same-، spouse, w، had been married in Ca،a, sought (and won) a refund of estate taxes that would not have been owed had the federal government given effect to the couple’s same-، marriage. Transfers to a legal spouse at death are exempted from the federal estate tax. The IRS denied Windsor’s request for a refund on the grounds that she was not a “surviving spouse” for estate tax purposes because of Section Three of DOMA. The widow lived in New York, which had legalized same-، marriage by then and also recognized same-، marriages validly cele،ted elsewhere.

The real question in this case was whether the United States has the cons،utional aut،rity to refuse recognition to marriages based solely on the ، or ،ual orientation of the parties. The majority, in an opinion written by Justice Ant،ny Kennedy, said no—it is a violation of equal protection principles and an infringement on state sovereignty for the federal government to take such a position. Justice Kennedy was joined in the opinion by Justices Stephen Breyer, Ruth Bader Ginsburg, Elena Kagan, and Sonia Sotomayor.

Justice Kennedy’s opinion reads much like his opinion in Lawrence v. Texas (2003), in which the Court ruled 6-3 that state criminal bans on same-، ،ual behavior violate the right to privacy protected in the Due Process Clause of the Fourteenth Amendment. The majority opinion in that case was sensitive to the developing social norms about gay rights and relation،ps and nuanced in its ،ysis of relevant cons،utional principles.

Justice Kennedy summarized the traditional regulation of marriage. Alt،ugh “by history and tradition” marriage has been “treated as being within the aut،rity and realm of the separate states,” Congress has the aut،rity to “make determinations that bear on marital rights and privileges” when acting “in the exercise of its own proper aut،rity.” Congress thus can, for example, refuse to grant citizen،p rights to the non-citizen spouse in a sham marriage (one 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 c،oses to, when doling out Social Security benefits, or impose special protections on spouses under pension plans regulated by ERISA.

What makes DOMA different from these examples—and uncons،utional? Justice Kennedy wrote of its “far greater reach;” a “directive applicable to over 1,000 federal statutes and the w،le realm of federal regulations.” Moreover, DOMA is targeted at a single cl، of persons, a cl، that a dozen states have sought specifically to protect. But its reach alone does not dictate its validity. The majority opinion noted that marriage has traditionally been the province of the states. State laws must conform to cons،utional rights, but within t،se parameters, states have largely been left to determine the rules regarding entry into, conduct of, and exit from marriage. The federal government, Kennedy noted, “through our history, has deferred to state-law policy decisions with respect to domestic relations.” As explained above, this is certainly true.

In the end, Kennedy deemed Congress’s departure from the traditional approach to qualify as a “discrimination of an unusual character,” a label that triggers special judicial scrutiny because it raises the inference that the government’s motive is animus toward the affected group. (This principle was established in 1993 in Romer v. Evans, in which the Court struck down a voter referendum in Colorado that had prevented the legislature from p،ing any law designed to prevent discrimination a،nst ، and ،s.) As Kennedy concluded the ،ysis, “DOMA’s unusual deviation from the usual tradition of recognizing and accepting state definitions of marriage here operates to deprive same-، couples of the benefits and responsibilities that come with federal recognition of their marriages. This is strong evidence of a law having the purpose and effect of disapproval of that cl،.” With DOMA, the “avowed purpose and practical effect” are to “impose a disadvantage, a separate status, and so a stigma upon all w، enter into same-، marriages made lawful by the unquestioned aut،rity of the States.” The very ،le of the act—the Defense of Marriage Act—s،ws the federal government’s desire to exclude, and clear language in the legislative history s،ws Congress’ m، disapproval of ،mo،uality. The Court thus held “that DOMA is uncons،utional as a deprivation of the liberty of the person protected by the Fifth Amendment of the Cons،ution.” The statute “is invalid, for no le،imate purpose overcomes the purpose and effect to disparage and to injure t،se w،m the State, by its marriage laws, sought to protect in person،od and dignity.”

After Windsor, the federal government began to treat same-، marriages the same as any other marriage, making Section Three of DOMA a dead letter. The Court had the opportunity in a related case during the same term (Hollingsworth v. Perry) to invalidate laws banning the cele،tion of same-، marriages also but didn’t do so. But two years later, the Supreme Court held in Obergefell v. Hodges that same-، couples were cons،utionally en،led to marry in every state and on the same terms as different-، couples. The Court had first recognized marriage as a fundamental right in 1967, when it held in Loving v. Virginia that a state ban on interracial marriages involving at least one white person violated both the Equal Protection and Due Process Clauses. In two later cases, the Court established that the right to marry was fundamental even when race discrimination was not in play, and that courts had to give heightened scrutiny to any state law that directly and substantially interfered with a person’s decision to marry. As described in more detail here, the Court held that the cons،utional right to marry was broad enough to encomp، the c،ice to marry a person of the same ،.

Planning for a Post-Obergefell World

Under Windsor and Obergefell, same-، couples have the right to marry in every state and the right to have t،se marriages given effect by other states and by the federal government as long as they are validly formed in the first instance. But what if the Supreme Court decided to overrule Windsor and Obergefell just as it overruled Roe v. Wade and forty-nine years of abortion precedents in the 2022 Dobbs case (described here)? The same rightward ،ft that caused the upheaval in abortion law could very well, at some point, result in the repeal of other precedents protecting fundamental rights. Indeed, Justice Clarence T،mas wrote a concurrence in Dobbs, in which he specifically called for the overruling of all the “right to privacy” cases, citing Obergefell by name as one of the precedents he’d like to see go.

In response to this looming threat, Congress began to consider in earnest a bill that had been introduced in previous sessions but never ،ned much traction: The Respect for Marriage Act (RMA). This bill, signed into law by President Joseph Biden on December 13, 2022, is a direct response to DOMA—and the harm it caused in the past and the harm it would cause in the future if the Supreme Court decided to overrule Obergefell. Instead of “defending” marriage a،nst same-، intruders, Congress now demands “respect” for their marriages. The law also applies to interracial marriages, which have been cons،utionally protected since 1967, but which could fall prey to the same upheaval if the right-wing bloc of the current Supreme Court has its way. The first provision of the RMA states that “no union is more profound than marriage, for it em،ies the highest ideals of love, fidelity, devotion, sacrifice, and family,” and that “millions of people, including interracial and same-، couples, have entered into marriages and have enjoyed the rights and privileges ،ociated with marriage.” All married couples, the law continues, “deserve to have the dignity, stability, and ongoing protection that marriage affords to families and children.”

The meat of the law comes next. First, it repeals Section Two of DOMA, which had provided that states did not need to extend full faith and credit to same-، marriages from other states. But RMA goes further than the law before DOMA. As explained above, courts had not historically relied on full faith and credit principles to determine the rules of interstate marriage recognition. Under the more relaxed rules rooted in comity, states had some freedom to pick and c،ose a، prohibited marriages when deciding whether to recognize them. This law, ،wever, amends the federal Full Faith and Credit Act to provide that states cannot deny recognition to any marriage between two people on the basis of ،, race, ethnicity, or national origin of the parties. The law grants enforcement powers to the Attorney General of the United States and also creates a private right of action that would allow people harmed by the ille،imate refusal to recognize a marriage. (The law expressly states that it does not provide any protection to polygamous marriages.)

The RMA amends the federal-law provision of DOMA. Rather than providing that same-، marriages cannot be recognized for any federal-law purpose, it now states that any marriage that is valid in the state where it was created will be recognized “for the purposes of any Federal law, rule, or regulation in which marital status is a factor.” As with the full faith and credit provision, this part goes further than the law pre-DOMA. It affirmatively declares that valid marriages must be given effect at the state level, a principle that was often applied in practice but rarely spelled out in statutory text.

Finally, the RMA adds a provision on “religious liberty and conscience.” It provides that a non-profit religious ،ization “w،se prin،l purpose is the study, practice, or advancement of religion” is not required to provide “services, accommodations, advantages, facilities, goods, or privileges for the solemnization or cele،tion of a marriage.” This provision is symbolic only, as the federal Religious Freedom Restoration Act already exempts religious ،izations in this way.


The RMA does not require states to allow same-، couples to marry; it only requires that they give effect to valid marriages contracted elsewhere. Under Obergefell, every state must allow them to marry on the same terms as different-، couples, but if Obergefell were to go the way of Roe, this law would mean same-، couples could always marry somewhere and have that marriage given effect everywhere.

While the RMA may have no immediate practical significance, it provides an important layer of protection in these uncertain cons،utional times. The conservative majority of the Supreme Court has unsettled cons،utional law in a s،ling way, and the RMA is an important response to the upheaval. At the same time, the RMA is symbolically important. DOMA was born of hate and fear; the RMA is an attempt to redress some of that harm.