The Volokh Conspiracy

Before I go, I wanted to leave you with a few thoughts about these legal issues that go beyond the issue of DOMA and same-sex marriage.

As you may have gleaned, it turns out that marriage isn’t the only area where federal law draws upon or relies upon a pre-existing concept under state law. As has been mentioned in passing in my posts and in the comments, bankruptcy and tax law frequently draw upon state property law. (Before you tax and bankruptcy experts complain, yes, I know that federal law doesn’t always follow exactly what the state does, but in many cases the variations in state law are incorporated.) That’s also true in criminal law (where prior state convictions trigger increased punishments for later federal crimes), in federal enclaves (where the Assimilative Crimes Act incorporates the criminal law of the surrounding state), and much more. Henry Hart called this kind of federal law “interstitial.”

In some of areas of “interstitial” law, the choice of law problems are pretty well worked out. It’s pretty obvious to everybody that your prior state convictions are determined by the state where you were convicted, and so on. But in others, the choice-of-law problem is subject to the same kinds of ambiguities as marital law is, so my article provides some insight for resolving those areas too. (For example, my article shows why the Klaxon rule, which some courts have used, shouldn’t be extended to those areas.)

I think I have another normative point, too. Sometimes when interstitial law is invoked in a particular context, it strikes people as weird or counter-intuitive. Shouldn’t the scope of a federal right or the meaning of a federal law be uniform? I hope that our exploration of interstitial law convinces you that the answer is “not necessarily.” In many areas that are politically controversial, legally technical, or both, federal law has drawn much of its effect from state law.

At least two cases pending at the Supreme Court this term involve a dispute over interstitial law — the social security case of Astrue v. Capato (thanks, Nick, for calling it to my attention), and the habeas case of Gonzalez v. Thaler. In Capato, the federal government champions interstitial law — it wants state law to determine when a child conceived after his father’s death (by artificial insemination, of course) can inherit — in Gonzalez, it’s the habeas petitioner — he thinks state law makes his federal habeas petition timely. I’m not sure about the correct ultimate outcome in either case, but I think it’s important for the Court to see that there’s nothing anomalous or weird about relying on state law in such cases.

Thanks to Eugene and his co-conspirators for letting me crash here for the week, and thanks to the commenters, many of whom were insightful or informative. If you want to hear more from me, you can read my articles on SSRN (stay tuned for my blockbuster essay on the Jurisdiction and Venue Clarification Act), and you can follow me on Twitter (though it’s really not my medium). I’m sure I’ll see you all around the blogosphere again someday — maybe soon, maybe in a year or two.

Amidst all of my many posts about what federal courts should do if DOMA is struck down, I thought I should say a little bit about Congress.

First — and this may be obvious to some readers but perhaps not to all — Congress could of course fix the choice of law problem by providing a statutory rule. Indeed, one of the few virtues of DOMA is that it is just such a rule. And in one of the most recent proposed bills that would repeal DOMA, Congress has also proposed replacing it with a choice of law rule.

I think this is a great idea. If DOMA goes away, Congress should exercise its power to replace it with some legitimate choice of law rule (obviously, if DOMA is unconstitutional, the replacement rule couldn’t be identical to DOMA, but there are a lot of other possibilities).

A congressional choice of law solution would have more legitimacy than a common-law solution, and could be more far-reaching, considering a broader range of policy considerations. (It could even extend federal marital benefits to civil unions, as discussed in my exchange with JHW.) Indeed, the opportunity to provide a replacement rule is one reason that Congress ought to repeal DOMA rather than just waiting around for the courts to have their way with it.

As for what rule Congress should adopt, I think the rule proposed in S.598 is a pretty good one. That rule is:

For the purposes of any Federal law in which marital status is a factor, an individual shall be considered married if that individual’s marriage is valid in the State where the marriage was entered into or, in the case of a marriage entered into outside any State, if the marriage is valid in the place where entered into and the marriage could have been entered into in a State.

For domestic marriages, the bill makes the marriage valid if it was valid where the couple got married. (Note, by the way, that with very rare exception, a marriage is almost always valid where the couple got married, because the couple usually chooses to get married someplace where their marriage will be valid.) For international marriages, there is the extra requirement that it must be a marriage that is valid in at least one state — no polygamy, no brother-sister marriages, no marrying nine-year-olds.

This is something of a “maximalist” approach to the choice of law problem, because in practice it will make almost all same-sex marriages valid. (Or at least it is “maximalist” within the realm of the traditional choice of law options.) So I think it would be problematic for courts to impose a solution like this without Congressional authorization. But it would also maximize stability and predictability for same-sex couples, at least with respect to federal law. As with the common-law solution, there’s plenty of room for argument about what rule Congress should adopt, but it would be nice if it stepped in with a rule.

However, I am pessimistic about this actually happening. Scholars have been calling for Congress to fix various choice of law problems for decades, and it rarely happens. While the DOMA repeal bill is a sign that at least somebody in Congress has thought about this issue, I don’t think anybody should be holding their breath for that bill, or any other statutory solution. That’s why I spent so much time blogging about the courts.

So I’ve explained why I reject two of the three possible approaches that federal courts take to the marital choice of law problem. The one that’s left over is the so-called federal common law approach, where courts craft a common-law doctrine for deciding which state’s marriage law is relevant to a federal statute.

You might ask, wasn’t federal common law forbidden by the Court’s decision in Erie v. Tompkins? No, not really. Here is how the Court explained it just last term in AEP v. Connecticut:

“There is no federal general common law,” Erie R. Co. v. Tompkins famously recognized. In the wake of Erie, however, a keener understanding developed. See generally Friendly, In Praise of Erie — And of the New Federal Common Law, 39 N.Y.U. L. Rev. 383 (1964). Erie “le[ft] to the states what ought be left to them,” id., at 405, and thus required “federal courts [to] follow state decisions on matters of substantive law appropriately cognizable by the states,” id., at 422. Erie also sparked “the emergence of a federal decisional law in areas of national concern.” Id., at 405. The “new” federal common law addresses “subjects within national legislative power where Congress has so directed” or where the basic scheme of the Constitution so demands.

I elaborate on this more in the paper, but basically it should be pretty clear that this kind of choice-of-law rule — choosing the state law on which a federal statute has chosen to rely — is the kind of “area of national concern” where a federal common law rule is okay.

That leaves us with the question: well, what rule? It may be a little anticlimactic to say so, but I don’t actually think it is that important what the rule is. The important thing is that there be a rule, so that federal law have a consistent test for deciding whether a couple is married. (As others have noted, whether a couple is married will still depend on state law, but at least the federal law will have a single approach for evaluating it.)

The rule I advocate in the paper is that the courts should look at the choice-of-law rules of the couple’s domicile. So, if a same-sex couple lives in Massachusetts or Maryland or New Mexico, they are married for purposes of federal law. If they live in Texas, they are not.

Basically, I think the federal government’s goal here should be to leave the smallest footprint possible, and I think the federal government does that best by treating couples the same way at home as their state does. (Obviously there are arguments that federal law should guarantee a constitutional right to same-sex marriage, but if so, that should be done directly, through interpretation of the 14th Amendment, not in a roundabout fashion by manipulating choice of law.)

Another important consideration is that there are at least a couple of statutes that do provide marital choice of law rules (for social security and for some veterans benefits) and they both look to the law of the domicile. These statutes were trumped, for same-sex marriages, by DOMA, but they will presumably spring back into effect if DOMA is gone. They affect just a tiny piece of the federal law of marriage recognition, but it would be best for a federal common law rule to be consistent with them, so that marriage can get a unified treatment throughout federal law.

So there you have it. My experience has been that a lot of people have different intuitions about what the right federal common law rule should be, and that those intuitions tie very closely to whether they like or don’t like the right to same-sex marriage. But if I’ve convinced you that there should be a rule, and that courts have the power to create a rule, that’s the important part.

I have a little bit more to say about this– about both Congress’s role, and about a few other areas of law. I’ll try to post about them tonight or this weekend, before my magic Volokh credentials turn back into a pumpkin, and I turn back into a regular, non-blogging fellow.

Before I get to the last phase of my stint here, I thought it would be helpful for me to address one recurring theme in the comments. John D’s comment this morning is representative: “how are opposite-sex married couples treated in conflict-of-laws situations? We’ll take that, then.”

The important thing to recognize is that there is no established answer for how opposite-sex married couples are treated! All of these posts about different approaches aren’t hypothetical, they’re all real doctrines already applied to the conflicts problem outside of the same-sex context.

I think this is non-intuitive for many people — lawyers and non-lawyers — because they have a sense that legal uncertainty is pretty commonplace, and people usually manage to just muddle through. Even if doctrine is uncertain on the margins or theoretically incoherent, most of the time it just doesn’t matter much. But choice-of-law doctrine is unusually uncertain, and unusually incoherent, even compared to other legal doctrines!

Also, there are two practical reasons that these uncertainties haven’t made much of a difference, most of the time, to straight couples.

First, the uncertainties really bubble to the surface only when some states so strongly oppose a type of marriage that they exercise their traditional prerogative to refuse to recognize that marriage when it is consummated out of state. That hasn’t happened very often.

Second, the times when it has happened have simply not featured the same numbers as same-sex marriages do. In the last census, more than 130,000 same-sex couples described themselves as married! (Interracial marriage might have featured sizable numbers, I’m not sure, and if so, I’m not sure why the problem didn’t come up as much as you would expect in that context. I wonder if it had to do with the smaller size and scope of federal regulation.)

Anyway, I agree that, without DOMA, the conflicts rules for opposite-sex marriage will be applicable to same-sex marriages. But the point is that there is no single conflicts rule for opposite-sex marriages, and the same-sex marriage controversy will probably force us to resolve the old conflicts problem at long last.

So earlier this morning (yesterday, Eastern Time, which apparently governs the Volokh Conspiracy), I explained why the federal government does indeed need to look in large part to state law when it tries to figure out whether a couple is married. In this post we’ll discuss another alternative solution with a goofy case name: “Klaxon.”

To recap, Klaxon is a general rule for solving choice of law problems in federal court when there is no federal law involved. And the solution Klaxon provides depends on the physical location of the federal trial court where the litigation takes place. It says that the federal trial court should imagine itself as a state court in the same state, and then do whatever that state’s courts would do.

Commenter “JHW” asked, in the previous post mentioning Klaxon:

Is there a practical difference, in the marriage recognition context, between the Klaxon rule and the “federal common law” rule that the law of the domicile is controlling?

There would be, presumably, in a diversity case, if for example the spouses lived in different states. But for a couple living in the same state seeking federal recognition, it seems that applying a state’s choice-of-law rules and applying a state’s marriage recognition rules amounts to applying precisely the same set of rules. Perhaps because I’m nearly totally ignorant of this area of law, the way you’ve framed this confuses me a bit.

There is a big difference: a Klaxon rule will often pick a very different state than the parties’ domicile, because Klaxon depends entirely on what state the lawsuit is in, and federal lawsuits can often be filed in one of several different districts. (There is a legal doctrine called venue that determines what district a lawsuit can be filed in.) So a couple might live in Texas, but wish to challenge an agency action in the District of Columbia or someplace totally different.

Now, as I noted earlier, this Klaxon rule controls in federal lawsuits that are brought in diversity jurisdiction, i.e. when no federal law is involved. And several courts (though not the Supreme Court) have suggested that the rule should be extended to federal cases too. I don’t think that makes much sense.

Here’s the most important reason why. Federal law is frequently enforced and administered by the executive branch. That’s a huge amount of what administrative agencies do all day, and we expect agencies to apply federal law in a huge range of cases that will never go to court, or will only go to court years later, after the agency has hopefully gotten things pretty much right.

But the Klaxon rule and the executive branch don’t mix very well. Under the Klaxon rule, you only know what law to apply once there has been a federal lawsuit, because only then will you know in what state the district court is located. So the executive branch just has to guess where the litigation will be brought. Moreover, since the plaintiff often has a choice of several different venues, he can strategically select a venue that will produce a choice of law rule that will contradict what the executive branch did.

This isn’t a problem in the diversity jurisdiction context, because the executive branch isn’t very involved in enforcing state law. But executive branch enforcement is a big part of federal law which results in federal question jurisdiction. The fact that Klaxon doesn’t mesh with executive branch enforcement is a very important reason not to extend Klaxon to federal question cases.

By process of elimination, that means that federal courts will have to craft their own choice of law rule as a matter of so-called federal common law. Tomorrow, we’ll talk about what that rule should be.

Yesterday I outlined the three different ways that courts have dealt with the concept of marriage in federal statutes. I ultimately hope to convince you that they should be using the so-called “common law” approach. But my argument is essentially one of process of elimination. So in this post I’ll explain why the “Borax” method of defining marriage without reference to state law is bad.

First of all, let me acknowledge a point made repeatedly by Andy Grewal in the comments. It is true, in a technical but important sense, that the meaning of the word “marriage” in a federal statute is a question of federal law. Congress gets to decide what it’s referring to when it uses that word, and so the “choice of law” question that I’ve been grappling with has a hidden premise, and the premise is that Congress did indeed intend the word to draw upon the state law concept of marriage.

Nonetheless, I think this premise is right. The Supreme Court has said that it is right, holding in Desylva v. Ballentine that the word “children” in the Copyright Act should be defined by reference to “the ready-made body of state law” because “there is no federal law of domestic relations, which is primarily a matter of state concern.” (The Court also went on to look at the law of a particular state, not to try to derive some sort of generic consensus state law.)

This approach makes sense too. When people think of marriage, as a legal matter, they think about the institution that is created and defined by state law. It doesn’t really make sense to ask for a court to ask whether a couple is “married” except in the state-law sense. This brings us back to the point that federal officials (territories, enclaves, and other exceptions aside) don’t issue marriage licenses.

Now, let me acknowledge a second Grewal point, which is that even if what I say is mostly right, surely there is some theoretical state law that is so crazy that the courts wouldn’t rely on it. (Commenters like to invoke the old saw about calling a “tail” a “leg” here– imagine a state that for some mysterious reason used the word “marriage” to describe all contractual relationships of any kind.) Indeed, in DeSylva the court said that a “State would [not] be entitled to use the word ‘children’ in a way entirely strange to those familiar with its ordinary usage,” (but, it added, “at least to the extent that there are permissible variations in the ordinary concept of ‘children,’ we deem state law controlling.”).

This is an accurate caveat. But the whole premise of a world where DOMA is invalidated or repealed is that it is at least “permissible” for the federal government to use the word marriage to include state-sanctioned same-sex marriages. So we are clearly inside the zone where we should “deem state law controlling,” meaning that we do indeed need to figure out which state’s law to so deem.

I’ve noticed that the Borax approach has intuitive appeal for a lot of readers. But I want to encourage you to think about how hard it would be to make it work in this context. Imagine that Congress repeals DOMA, without replacing it with any other interpretive rule. A same-sex couple says that they are married, and tries to file their taxes jointly, or one of them tries to claim a spousal employee benefit. Is the same-sex couple married, and what information would you need to decide? I just don’t think there’s a workable answer that doesn’t include “their marriage license” (or equivalent).

Before we dive deeper into the federal cases I blogged about this morning, I thought it would be helpful to clarify a point featured in several comments. Namely, isn’t the Full Faith and Credit Clause involved here somewhere? The answer is: not very much, and certainly not enough to make the problem go away.

The Constitution provides that “Full faith and credit shall be given in each state to the public acts, records, and judicial proceedings of every other state. And the Congress may by general laws prescribe the manner in which such acts, records, and proceedings shall be proved, and the effect thereof.” The clause has some application in this area, but it doesn’t force states to recognize marriages from other states for several reasons.

First of all, even though the Clause mentions “public acts,” it has not been interpreted to apply to statutes (like a law stating which marriages are valid) or marriage licenses with the same force it has for judicial proceedings. Relatedly, states have traditionally refused to give effect to another state’s statute if they think doing so is against their “public policy,” and this has been thought to be okay under the Full Faith and Credit Clause.

Moreover, even if the clause did normally require states to give effect to foreign legislation (or if the parties somehow get a judicial judgment based on their same-sex marriage), the clause delegates substantial power to Congress to decide what “effect” those acts, records, and proceedings have. And Congress has explicitly provided (in that other section of DOMA that I said wasn’t very important) that “No State . . . shall be required to give effect to any public act, record, or judicial proceeding of any other State . . . respecting a relationship between persons of the same sex that is treated as a marriage under the laws of such other State . . . or a right or claim arising from such relationship.” (For lots of fascinating history demonstrating Congress’s power under the Full Faith and Credit Clause, I recommend Steve Sachs’s Full Faith and Credit in the Early Congress.)

Now, some of these points are controversial. There are scholars (like my own law school’s dean, Larry Kramer) who argue that Section 2 of DOMA is unconstitutional and that the Full Faith and Credit Clause requires greater interstate recognition of marriage. Steve Sanders has made a similar argument under the Due Process Clause. But that’s not the state of judicial doctrine today, so the interstate disagreement about same-sex marriage is still something that courts have to deal with unless and until that doctrine is radically changed.

To sum up: Yes, there are some federal rules about interstate recognition of marriages. But those rules give states enough leeway that there’s still a great deal of state disagreement, which is all that matters for purposes of my argument. If you want to know more about this, there is a ton of recent scholarship on Full Faith and Credit, some of it cited in the paper.

So states have different rules about who can get married, and states also have different rules about which state’s law to look to when a same-sex couple moves around the country. Now it’s time to answer the question some of you have been patiently waiting with: what does the federal government normally do about this? As I’ve noted, outside of the DOMA context, the federal government usually looks to state law, and surely there must be other areas where state law differs. So what do the feds do?

The answer is ... it’s an even bigger mess! Notice that as a matter of legal theory and doctrine the question we’re asking here isn’t necessarily marriage-specific. Rather, it can be asked more generally as: “when federal law relies upon state law, what’s the system for choosing which state’s law?” In the twentieth century, federal courts have adopted at least three different approaches to this problem.

One approach is based off of a choice-of-law classic called Klaxon v. Stentor Electric Manufacturing (I’ve always loved the name of this case, and it may be the real reason I got into conflicts in the first place). Klaxon is the rule that the courts apply to the similar problem in diversity cases. It says that you apply the same choice-of-law rule as would be applied by a state court in the state where the district court is physically located. If you sue in the federal district of New Jersey, you apply New Jersey choice-of-law principles; if you sue in the southern district of New York, you apply New York choice-of-law principles. It has nothing to do with where any of the actions in the case took place — it’s just about where the plaintiff decides to bring the lawsuit.

As I said, this is the law in so-called diversity jurisdiction, which features cases that are brought entirely under state law, but which nonetheless get to federal court. It’s designed to make the federal court imitate the equivalent state court as closely as possible. Some courts have also applied this rule to cases brought under federal law, which would include the marital choice-of-law situation we’re talking about here.

A second approach is for federal courts to make up their own choice-of-law rule as a matter of so-called “federal common law.” The good news is that this gets you away from the weird formalism of the Klaxon rule. The bad news is that now the federal court has to come up with a rule. And within this approach, federal courts have come up with all sorts of different rules for marriage.

For example, in Tatum v. Tatum, an old Ninth Circuit case about life insurance for a federal employee named Erwin Tatum (who had a somewhat exciting romantic life, for a postal worker), the court summarized the state of disagreement about the life insurance statute as follows (with citations omitted):

Under similar conditions, the adjudicated National Service Life Insurance Act cases have produced differing conclusions. One point of unanimity is that state law will govern. But there exists little agreement as to what state law is applicable. Some would view the law of the place of marriage as controlling; others would be guided by the law of the domicile of the parties, either at the time of the marriage, or when the alleged claim accrues; another court has declined to resolve the matter.

This was over fifty years ago, and things haven’t gotten much clearer! Some courts follow their own ad hoc rules, some look to one of the Restatements. So this approach is really a welter of sub-approaches.

Finally, the third approach is to sort of give up on using state law after all. This is the least common solution (hence the point that marriage is usually a creature of state law) but it’s a thread that some courts return to. Commenter “Andy Grewal” summarizes this attitude pretty well:

That is, if a federal statute generally refers to marriage under state law, but SSM is not what Congress had in mind when it used the term “marriage,” then SSM should not qualify under the federal statute whether recognized in a the couple’s home state or not. On the flip side, if Congress intended to reach marriages of any sort when it (e.g.) refers to a married couple in the tax code, the fact that the residence state does not recognize a same-sex couple’s marriage should not disqualify the couple from the federal statute.

So in at least two cases, federal courts have done something like this. One is an old case called Borax, where the Second Circuit held that a couple was married for purposes of the tax code even though the man’s prior divorce was invalid under state law. Judge Friendly (one of Orin’s favorite judges) dissented. Another is Adams v. Howerton, a Ninth Circuit case that actually deals with same-sex marriage. But the case is from 1982, so it was easy for the court to say that the “ordinary, contemporary, common meaning” of the word “marriage” meant only opposite-sex marriages. That would be a much more debatable claim today!

So federal courts are all over the place. The Supreme Court has never resolved this issue and individual circuits waffle back and forth between different approaches. Discovering these sets of federal cases, and learning how little people knew about them, inspired me to write my paper. In some more posts, I’ll explain which approach I think should be used.

Here are a few follow-up questions to my first post this morning.

“Burt Likko” says: Prop. 8 is worded: “...only marriage between a man and a woman is valid or recognized in California.” I have a hard time getting past the plain meaning rule to reach a result that an Iowan or Canadian same-sex marriage license would be “valid or recognized” in California.

Answer: Maybe. But note that in Strauss v. Horton, (starting on p. 128) the California Supreme Court has already made an exception to the text of that provision, holding that same-sex-marriages that preceded the enactment of Prop 8 are indeed valid and recognized in California. The court relied on background principles of anti-retroactivity, as well as respect for vested rights (and the state’s due process provision). And Maryland recognizes out-of-state same-sex marriages despite a law that provides that “only a marriage between a man and a woman is valid in this State.”

Would a marriage validly celebrated in another state — especially by a resident of that other state, though that is not the only scenario — be treated as a “vested right” that Prop 8 left intact? I do not know, but I am not as sure as the commenter is that the answer is obvious.

“Henry Clay” asks: I’ve been looking forward to this post because I’ve been wondering what examples there are of states not recognizing marriages from other states apart from the obvious examples of SSM and Jim Crow. Are you really going to just wave your hand at that (“sometimes they don’t”) or is there another post on the way?

Answer: If you really want to dig into the historical examples here, you’ll have to leave the blogs. The short answer is that most of the examples are indeed from same-sex marriage and the interracial marriage context. Basically, most states claim a right to refuse to recognize a marriage only in extreme circumstances, known as the “public policy exception,” and this exception doesn’t get invoked that often.

For further reading about the rarity (outside of same-sex marriage) of the “public policy exception,” see Steve Sanders’s very engaging paper, The Constitutional Right to (Keep Your) Same-sex Marriage.” For further reading about the interracial marriage precedents (and others), see Andrew Koppelman’s excellent book, Same Sex, Different States (the only e-book I’ve ever purchased!). To see what other complications I’m trimming out for blogging purposes, see my paper.

“Rocket Scientist” says: I take some small issue with the claim that same-sex marriages would be recognized in New Mexico. The opinion is an attorney general opinion not really a court ruling, so there may still have to be litigation in order to resolve the question.

Answer: A good point! While I’m not aware of a court in any of the states I mentioned that has bucked the ruling of the state Attorney General, those issues are still up for grabs in litigation, so the state of affairs is even more uncertain and confusing than I made it sound.

This brings me to my ultimate point — which is that this is a mess. (There is a saying you only need to know two things about conflict-of-laws: that there is no area where 20th-century academics have had a greater impact on the law, and that there is no area of the law that is more screwed up.) And we haven’t even gotten to the worst part of the conflicts problem yet, which will come when we bring in the federal government. So stay tuned; it gets worse before it gets better.

So we’ve established that in a world without DOMA, state law is the natural place to look to figure out whether a same-sex couple is married. Here is where we meet our first conflict-of-laws problem.

You surely know that some states allow same-sex couples to marry, and others do not. So what happens when a couple lives in a no-same-sex-marriage state, but gets married out of state? Or lives and marries in a same-sex-marriage state, but then moves to a no-same-sex-marriage state? Well, it depends.

In Maryland (or Rhode Island, or New Mexico), the couple will still be treated as married, even though they couldn’t have gotten married in that state. In many other states, they won’t be. Texas is an extreme example: a same-sex couple moved there from Massachusetts and wasn’t even given access to divorce court, because Texas figured they were never married in the first place.

And in other states, the answer is still up in the air. Remember the controversy about District Judge Vaughn Walker’s same-sex relationship, and possible marriage plans? Steven Gillers suggested that if Walker had any interest in getting married, he could just travel to Iowa or another same-sex-marriage state. Ed Whelan’s response was that such a marriage should not be recognized in California. But the strange thing is that California law isn’t really clear on this point, so even after looking into it, I don’t know for sure how the California courts would apply Prop 8 to an out-of-state same-sex marriage.

Now, in a sense, this is not a new problem for the states. People have been getting married and moving around for a very long time. But there’s never been widespread agreement about the solution to that problem. Most of the time states defer to the state where the marriage was “celebrated,” but sometimes they don’t.

So when you ask whether a same-sex couple is married under state law, the answer depends on which state you ask.

A number of commenters have asked about the relationship between law and marriage. Some, for example, have taken issue with the statement in my paper’s abstract that “marriage is primarily a creature of state law.” So I thought it was worth explaining a little but more about the relationship between state law and marriage.

Obviously, marriage is partly a private, non-governmental act. You can “marry” in the eyes of your religion or your community without ever marrying in the eyes of your state. Or you can have two separate marriage ceremonies — one religious and another one designed to satisfy the state’s requirements — which is what I did when I got married. In that sense, marriage is not just something the government makes up.

But the government also has a lot of rules which depend on whether or not you are married — being married affects your taxes, your health benefits as a federal employee, rights under an ERISA plan, child custody, your right not to testify, and so on. So even if you think the private, non-governmental part of marriage is more important than the government-recognized ceremony, the government still needs a way to figure out who is married and who isn’t. It generally uses state marriage ceremonies (or common-law marriage, now mostly of historical interest) to do so. Of course, you could also try to get rid of all laws that treat married couples differently from unmarried couples, but I no longer think that’s wise, and it’s certainly not going to happen any time soon.

So when I say that “marriage is primarily a creature of state law,” I really mean: “when the law deals with marriage, it’s mostly state law, not federal law, that determines marital validity.” You get a marriage license from the state where you’re getting married, not the federal government, and the state determines who can officiate, how old you have to be, whether you have adequately terminated any previous marriages you had, and so on. In that sense marital status is somewhat like property: federal law frequently turns on whether you have it, but it often uses state law to figure that out.

Finally, for purposes of this paper it doesn’t actually matter whether the federal government has the constitutional power choose to create its own independent marriage regime — with its own federal officiants, federal marriage licenses, etc. The main point is that it hasn’t created one. But I’ll also add that I think it would be impractical and unwise for the federal government to try to create such an independent marriage regime.

Thanks to Eugene and the conspiracy for having me here. In my first post I thought I’d explain what the Defense of Marriage Act does and the circumstances under which it may be held unconstitutional. We can get into the choice-of-law stuff later. (The act, by the way, is known as “DOMA,” which gives rise to lots of great puns, from Andrew Koppelman’s Dumb and DOMA, to the title a colleague suggested for my paper: Beyond ThunderDOMA.)

Section Three of DOMA defines marriage for purposes of the thousand-some federal statutes that deal with marriage. It says that for purposes of federal law, “the word ‘marriage’ means only a legal union between one man and one woman as husband and wife.” This is an exception to the usual federal practice, which is to look at state law to decide whether a couple is married. (Another part of the statute deals with states’ refusing to give “full faith and credit” to marriages from other states, but for present purposes, it is mostly unimportant.)

DOMA has become very controversial. A number of trial courts (and at least one appellate judge, acting in his administrative capacity) have held it unconstitutional. Last February, the Obama Administration joined in, and began arguing to the courts that it is unconstitutional.

On what basis? Here is where it gets interesting. The courts and the challengers have not really argued that states must recognize same-sex marriages (as the district court in Perry v. Schwarzenegger held). Instead, they’ve argued that even if states can make their own decision about same-sex marriage, the federal government is constitutionally required to respect the state’s choice.

Doctrinally, these challenges turn on levels of scrutiny and government interests, but the closest analogy might actually be the Supreme Court’s confusing opinion in Romer v. Evans. In Romer, the Court struck down a state constitutional amendment that repealed local and state laws that protected against sexual-orientation discrimination, and thereby “impose[d] a special disability upon [gay people] alone.” In almost every other respect, the federal government looks to state law to decide whether a couple is married — age, consanguinity, officiant, previous divorce, etc. — except for same-sex marriages, on which, “alone,” DOMA “imposes a special disability.”

Anyway, you may or may not buy that constitutional argument, but a lot of courts are buying it and the President is selling it. And if that argument wins out, federal law will have to rely on state law to determine whether a same-sex couple is married. (The same thing happens if DOMA is repealed.) But relying on state law, it turns out, is much more complicated than it seems. (That’s where the choice-of-law stuff comes in. We’ll get to it today or tomorrow.)