On the
constitutionality of the Defense of Marriage Act and its effect on
immigration law:
There are two principal parts to the DOMA:
§ 2(a), which provides that
"No State, territory, or possession of the United States, or Indian tribe, shall be required to give effect to any public act, record, or judicial proceeding of any other State, territory, possession, or tribe respecting a relationship between persons of the same sex that is treated as a marriage under the laws of such other State, territory, possession, or tribe, or a right or claim arising from such relationship."
and
§ 3(a), which provides that:
"In determining the meaning of any Act of Congress, or of any ruling, regulation, or interpretation of the various administrative bureaus and agencies of the United States, the word 'marriage' means only a legal union between one man and one woman as husband and wife, and the word 'spouse' refers only to a person of the opposite sex who is a husband or a wife."
Out of those, the section that seems to me to have the clearest chances at being invalidated by a court is
§ 2(a), which
arguably runs afoul of the
U.S. Constitution's "
Full Faith and Credit Clause." (FFC)
Art. IV § 1 U.S. Constitution. The Full Faith and Credit Clause basically requires that administrative and judicial acts handed down in one state be given "Full
Faith and Credit" (i.e.
recognition) in every other state. The Full Faith and Credit Clause was included in the Constitution in order to break down the barriers to interstate travel that existed prior to the Constitution's
enactment, when each state had its own
currency, etc. A marriage licence or the document establishing a
civil union under
Vermont law (or any other form of same-sex union that may be allowed in any other state) is clearly a
document evidencing an administrative or judicial act. As such, a civil union clearly would constitute a "public act[], record[], [or] judicial proceeding[]" of one state within the meaning of the FFC.
The next question is whether the DOMA falls within the authorisation of
Congress to "by general
laws
prescribe the manner in which such acts, records, and proceedings shall be proved, and the effect thereof."
Art. IV §1 cl. 2. Most proponents of the DOMA argue that it does constitute a regulation of the "manner" in which records, etc. "shall be proved, and the
effect thereof." Clearly, a
statute that says that one state can disregard the "acts, records, and proceedings" of another state on a particular subject does not regulate the "manner" in which they can be proved. "No
manner" is not a regulation of manner.
A tougher question is whether Congress can permissibly regulate "the effect" of a marriage licence in one state by simply declaring that another state may disregard it entirely if it so chooses. It's a lot easier to say that
declaring a marriage licence to be without any
binding effect on another state is "prescribing the effect" of the licence on another state than it is to say that "no recognition" is a regulation of the manner of recognition.
This requires an examination of purpose. As noted above, a purpose of the FFC Clause was to ensure that citizens would be able to travel freely throughout the US without worrying that their settled rights would be left
at the state line. The purpose of ensuring
freedom of movement of persons and goods throughout the states is evidenced elsewhere in the Constitution,
as well. It's obvious that freedom of interstate travel and
migration is burdened by the DOMA. There are plenty of cases of people who have moved from Vermont to some other state, where, not only was their civil union a
nullity, but the state to which they'd moved would not even allow it to be
dissolved because it was considered never to have existed in the first place.
Most cases surrounding freedom of
interstate travel have involved state laws (e.g. state
welfare laws,
residency requirements for voting, etc.). The cases involving state statutes, such as
Saenz v. Roe, 526 U.S. 489 (1999) (different levels of welfare
benefits based on duration of residency in state), have generally required at least a "
good-faith basis" for treating long-time state residents and recent arrivals differently, similar to a rational basis for
Equal Protection purposes. This test gives the states a
lot of leeway, but isn't a
rubber stamp.
Assuming that
Saenz and its companion cases would be applied to the DOMA, the US would have to show that there was some "good-faith" basis for burdening interstate travel. Those trying to show a "rational basis" for discriminating based on sexual orientation haven't been doing well lately. E.g.
Romer v. Evans, 517 US 620 (1996),
Lawrence v. Texas, ___ U.S. ___ (2003). The
government would probably have to try to come up with something other than "
public morals," or "allowing states to enforce their voters' own moral choices," etc. to justify the burden on interstate travel and
commerce.
However, even if
§ 2(a) were to be held
unconstitutional, that wouldn't really have any effect on federal immigration law, since
§ 2(a) deals with
what the states can do. The DOMA is structured so that, even if all 50 states were to decide to recognise gay marriages, thus rendering
§2(a) meaningless, this would still have no effect on the
federal government,
because
"In determining the meaning of any Act of Congress, or of any ruling, regulation, or interpretation of the various administrative bureaus and agencies of the United States, the word 'marriage' means only a legal union between one man and one woman as husband and wife, and the word 'spouse' refers only to a person of the opposite sex who is a husband or a wife"
DOMA § 3(a).
So, for federal purposes, even if
same-sex marriage were legal in all 50 states, the federal government would still be bound by
§ 3(a) not to give any federal recognition to the marriages for the purposes of interpreting or administering federal law. This would include
binational same-sex marriages.
However, if the
PPIA, which would mandate equity in immigration for same-sex couples, were passed, it would supersede
§ 3(a) and grant
recognition of same-sex marriages for immigration purposes. Simply
repealing § 3(a) might have a similar effect.
Then, "marriage" and
"spouse" in federal laws would be interpreted as they were prior to the
enactment of the DOMA, based on state law.
An argument could be made that Congress has no power to say what marriage is, since this is a traditional province of state law; however, I don't think it would succeed. Congress isn't saying who can get married, but rather
whose marriage counts as a marriage under federal law.
An argument based on Equal Protection would have better chances of succeeding.
§ 3(a) clearly singles out a
class, namely people in same-sex marriages, who, absent some really extraordinary circumstances, will be
gay or
lesbian. The purpose of the statute, based on its clear language, is to
single out GLBT couples for treatment that is worse than that afforded any other couple. A
state-sanctioned marriage of a straight couple will be
recognised without a
hassle by the federal government, but a state-sanctioned marriage of a same-sex couple is categorically barred from similar treatment.
§ 3(a) singles out legally married same-sex couples for a
"
disability" not suffered by other couples who are similarly situated. The
onus would be on the federal government to advance a "
rational basis" for the
discrimination. This would be the same
minefield presented by
§ 2(a)'s
burden on interstate travel and migration. The most popular argument ("
family values" et al.) has essentially
been foreclosed by
Romer v. Evans, and, indirectly, by
Lawrence v. Texas.
However, even if the entire DOMA is invalidated as described above, I still don't think that would, in itself, have any effect on the immigration status
of binational same-sex couples. Getting rid of the DOMA would return things to the way they were pre-DOMA, and same-sex couples did not get any immigration recognition under federal law at that time. However, since the
passage of DOMA, at least one
state has allowed for "civil unions," which are similar to marriage. Without the constraints of § 3(a) of the DOMA, a partner in a civil union might be considered a "spouse" for immigration purposes. Moreover,
BCIS might be able, in that case, to adopt
interpretive rules stating that a civil union partner is considered to be a spouse for the purposes of the
Immigration and Nationality Act.
As we have recently been reminded by
Lawrence,
substantive due process is very much alive; thus, substantive due process provides another avenue for attack:
The following arguments are possible:
(1) The statute unduly burdens the fundamental
liberty interest in marriage. It's been held that marriage is a fundamental liberty interest, but this would be a long shot, since it assumes that a single state could create (and make binding upon every other state and the Federal government) a "fundamental liberty interest" in marriage for people who traditionally had no such fundamental interest simply by creating a new
legal status ("civil
union") or expanding an old one.
Lawrence definitely opened the door, but I
don't think it's open quite wide enough for this sort of challenge.
(2) The statute unduly burdens the right to interstate travel. This, I think, is a little closer to hitting the target than argument (1), simply because it doesn't rest on creating a completely new "fundamental right" (as
much as I would enjoy seeing
Scalia rant in
dissent if it actually worked).
One obvious problem would be the way the challenge is framed: Is it the statute that is burdening interstate travel or the state's refusal of recognition pursuant to the statute? Either one would require a court to
reach the issue of the statute's constitutionality, but it seems to me that the second framing has a better chance of making it to a judgment on the
merits.
There are some pretty crusty old cases in very different contexts that say that the right to regulate interstate commerce doesn't include the right to
interdict it altogether. For example, the
Supreme Court once invalidated an act of Congress that prohibited trafficking in goods made by child labour in interstate commerce . Obviously, things have changed substantially since those
holdings, but
Griswold v. Connecticut, 381 U.S. 479 (1965), and
Roe v. Wade, 410 U.S. 113 (1973), which are very much alive after
Lawrence, were
both based partially on
Lochner v. New York, 198 U.S. 45 (1905), which is similarly archaic. So,
obviously, it's possible to resurrect bits and pieces of the old substantive due process cases without re-entering the
dark ages. The question is whether
there's anyone on the Court who would do it.