In its Windsor decision last June, the Supreme Court identified an unconstitutional burden upon federalism, namely, “two contradictory marriage regimes within the same state.” The “contradiction” arose at the intersection of the federal Defense of Marriage Act (DOMA), which defined “marriage” as the union of man and woman, and some states’ exercise of their “historic and essential authority to define the marital relation” (the Court’s phrase) to include same-sex couples. At ground level, Edith Windsor was considered “married” to her late lesbian partner under New York law, but was unmarried for federal estate tax purposes.

The Court solved this apparent problem by requiring the federal government to “defer to state-law policy decisions with respect to domestic relations.” So, since Windsor was deemed to be a surviving “spouse” under New York law, she was deemed to be a surviving “spouse” under federal tax law, too. The central declared goal of Windsor was just this vertical uniformity: a legally married New York resident is married for all local, state, and federal government purposes.

But the Court’s solution raised an obvious question, one that the Justices ignored entirely: which state? What if Edith Windsor and her New York “spouse” had moved to the Poconos? Or to Palm Beach? Or to any one of the thirty-two states where marriage is rightly defined as the union of man and woman? The obvious question is whether the federal government must “defer” to the laws of a couples’ resident state—their “domicile”—or to the laws of the state where their marriage was celebrated. (Windsor and her partner were originally “married” in Canada, and New York law chose to treat them as such even before the state permitted same-sex marriage locally. The Court’s decision of the case nonetheless raises the question just asked.)

It is surely accurate to say that Windsor created a vacuum when it threw out Section 3 of DOMA, which stipulated that the term “marriage” in all federal laws referred to the union of a man and a woman. But we can now say more perspicuously that Windsor created an acute choice-of-law question. Do the laws of the state of domicile or celebration prevail? Congress has done nothing to answer this conspicuous question.

Start your day with Public Discourse

Sign up and get our daily essays sent straight to your inbox.

That is, until yesterday, when Texas Representative Randy Weber introduced HR 3829, the “State Marriage Defense Act.” Its operative sentence provides that, in determining the meaning of any federal law as it pertains to “individuals domiciled in a state,” “the term ‘marriage’ shall not include any relationship which that State . . . does not recognize as a marriage.” [Emphasis mine.]

There are many good reasons to support this bill. The most important is that it does what can be done after Windsor to promote genuine marriage in those states that continue to recognize and legally protect it.

Another important reason swings free of one’s views about the truth concerning marriage. Representative Weber hit upon this reason when he wrote (in his “Dear Colleague” letter introducing the Act) that it would stop the Obama administration’s “unlawful” practice since Windsor.

Which “practice” is that? And how is it “unlawful”?

Since Windsor, the administration has followed the marriage definition of a couple’s domicile state in a few cases—but only, as far as I can tell, where a program-specific directive from Congress leaves it no choice. The Copyright Act, for example, states plainly that when an author dies, ownership rights shift to the “surviving spouse” according to the “domicile” state’s law. The Social Security Administration is likewise bound, as is the Department of Veterans’ Affairs, to follow the marriage laws of the state in which a couple is domiciled.

This partial practice shows that “domicile” is a familiar and usable concept. But the administration has nonetheless announced that it will ignore the Veterans’ law. And most federal agencies have decided to treat a couple as “married” if they participated in any legally recognized marriage ceremony, anywhere in the United States. The IRS, the Defense Department, the Departments of State and Education—among others—have adopted this state-of-celebration practice.

These agencies have no inherent legal authority to define marriage. Neither does the President or his Attorney General, so long as Congress has exercised its paramount authority to do so. The State Marriage Defense Act will thus restore proper legal order to the scene and correct the administration’s unlawful practice.

The administration’s practice no doubt stems from a sincere conviction that laws defining marriage as a union of man and woman are unjust. But it defies Windsor, which condemned DOMA for trying to “put a thumb on the scales and influence a state’s decision as to how to shape its own marriage laws.” The Obama administration is actually re-instituting “two contradictory marriage regimes within the same state.” A person married to someone of the same sex in New York or Massachusetts would be “married” when she files her 1040 with Uncle Sam, but would be “single” for  state tax purposes if she (or they) moved to Indiana. It seems, too, that the administration is putting into place the premise of a further Court decision, one that would finally sweep same-sex “marriage” into every corner of the country.

Some might object to the State Marriage Defense Act by saying that it is unfair for a same-sex couple in federal employ, or perhaps in the military, to be transferred from one state to another and thereby to come “unmarried.” But this is true now, as any such couple transferred from New York to Indiana or Texas could attest, since they would be unmarried for all state purposes in their new domicile. This objection is more cogent when it is limited to the more significant federal benefits, should they too be attenuated or lost in the transfer. But this is the situation that the Windsor Court created, and so is more a criticism of that case than it is of the Act. Besides, this prospect would be one factor for both the government employer as well as the same-sex couple to consider when a move is in the offing.

Others might object that part of each state’s relevant “sovereignty” consists of having its marriages recognized by other states. This objector would say that a state’s “historic and essential authority to define the marital relation” entails opposition to the Act, because the Act would decrease interstate respect for some states’ marriages—namely, those between same-sex partners.

But the Windsor Court, with eyes wide open, carved out an exception to this general rule of interstate comity when it comes to any state’s same-sex “marriages.” The majority opinion unequivocally affirmed continuing state authority to define marriage as the union of man and woman, in a case that presented a question about federal effects of state laws recognizing same-sex marriages.

Besides, the states’ “sovereignty” over marriage is, principally, each state’s authority to regulate marriage within its own borders, for the common good of the people who live there. The precise extent to which any state’s marriages travel to other states is a comparatively minor matter. This is especially true now, when marriage’s migratory prerogatives implicate not sundry tangible incidents of marriage (government payments, for example) but its fundamental meaning.

In other words: the state authority that Windsor praised so highly refers mainly to how marriage law contributes to sustaining a common public sense of what marriage is, what its responsibilities and satisfactions are, and how people can benefit by participating in that special form of life.

So, the objection about celebratory travels fails. But it is an illuminating failure, in that it shows why it is time to say goodbye to one “historic” norm about marriage in our federal system. The Utah federal court that recently overturned its marriage laws relied in part upon a time-tested constitutional rule of interstate comity. This notion is, basically, “married here, married everywhere.” Married in Ohio (as were my wife and I) meant married in New York, Illinois, Indiana, and, presumably, in any other state to which we might yet move. And it meant married everywhere for both state law and federal law purposes.

Windsor paid no attention to this enduring norm, which established an important, but limited, national uniformity in marriage without an overweening federal command. But the presence of same-sex marriage in eighteen states already—and polygamy on the horizon—means that the sun has set on this longstanding norm. Indeed, Windsor implicitly recognized as much.

The “married here, married everywhere” rule arose precisely from the states’ common recognition of the same basic reality. The states could and did differ considerably in defining ancillary features of marriage and its legal consequences. But they were united in understanding it to be the exclusive union of man and woman for life.

This common embrace of the truth about marriage persisted at least up to World War II. While it endured, national laws that predicated benefits upon marriage did not intrude upon state sovereignty. Sister states did not impose upon one another by expecting their marriages to be recognized across the country. The subject of the “here, everywhere” norm was, in other words, a common property (if you will) of each state, as well as the federal government.

That world has passed away.