I am opposed to gay marriage, but I am also opposed to straight marriage. I also oppose bans on gay marriage. But read the whole thing, because it is more compli­cated than that.

Marriage is a package deal. Break up the package, and the answers become clear. Marriage combines a constel­lation of legal oblig­a­tions in contract (concerning property, intestacy, finances, parenting, medical decision-​​making, &c.) with a set of priv­i­leges* conferred by the state under various mandates and enti­tlement schemes.

If we get rid of all the ille­git­imate, welfare-​​state junk asso­ciated with marriage, all we are left with is a complex, legit­imate two-​​party† contract. As it stands now in most states, I see no legit­imate aspect of marriage that is unavailable to same-​​sex couples.‡ Absent a poorly drafted ban, same-​​sex couples can form a personal union prac­ti­cally identical to the legit­imate aspects of state-​​sponsored marriage.

There are some excep­tions, due to badly written same-​​sex marriage bans in various states. Key Midwestern Swing State passed a consti­tu­tional amendment back in 2004, purporting to ban same sex marriage. In fact, it does more. Its first sentence defines marriage as a union between one man and one woman, and states that this is the only sort of marriage the state will create or recognize. The second sentence of the KMWSS ban reads:

This state and its political subdi­vi­sions shall not create or recognize a legal status for rela­tion­ships of unmarried indi­viduals that intends to approx­imate the design, qualities, signif­i­cance or effect of marriage.

Emphasis mine.

This is not a common way to draft a same-​​sex marriage ban. It is a very sloppy way to do so, and it has the effect, inter­preted in accord with the meaning of the law, of voiding private contracts between unmarried persons intended to approx­imate some aspect of marriage. In states, like KMWSS, that adopt this type of ban, my previous statement about the legit­imate aspects of marriage being already available to same-​​sex couples does not apply. My point is that, in the absence of either a ban or autho­rization, same-​​sex couples can get “married” using their estab­lished contract rights.

I am opposed to bans on same-​​sex marriage that might impair private rights of contract. And the U.S. Constitution, inter­preted objec­tively, agrees with me when it says, in Art. I, § 10, cl. 1, “No State shall . . . pass any . . . Law impairing the Obligation of Contracts . . . .” But I also see no need to enact legis­lation autho­rizing same-​​sex marriage, because every legit­imate legal aspect of marriage is already available to same-​​sex couples in contract.

If we didn’t have a welfare state, there would be no problem. But the welfare state creates a situation where indi­vidual rights are forced into conflict with one another. The laws that create the ille­git­imate benefits of marriage are ille­git­imate because they violate the rights of indi­viduals. They are either wealth redis­tri­b­ution benefits, which forcibly seize property from some people and give it to other people; or they are contractual impair­ments, which force private parties to include or exclude certain provi­sions from their contracts with other private parties.

The question of whether to allow same-​​sex couples to marry is therefore a question of which right is more important to us. We are forced to weigh property and contract rights against the right of equal protection under the law. If we allow same-​​sex couples to marry, we expand the welfare state. If we do not allow same-​​sex couples to marry, we are treating a group of people differ­ently under the law.

I find it dangerous to argue that because we give ill-​​gotten benefits to one group of people, we must give them to all people in the name of equal protection, all other rights be damned. I would find it morally repugnant to ask my government to give me special benefits that it should not have the power to give. Because marriage is a package deal, an argument in favor of gay marriage is an argument in favor of expanding the welfare state. Therefore, currently, and in the absence of any ban, I do not advocate for the enactment of legis­lation giving equal rights of marriage to more people. I oppose such legis­lation, while noisily advo­cating the elim­i­nation of the ille­git­imate aspects of marriage, so that the law might even­tually be equally applied to all.

The problem with this situation is that there is no polit­i­cally right answer.** One could, instead, argue in favor of gay marriage while advo­cating the rollback of the welfare state. I think this is the more dangerous route, because it is polit­i­cally so much easier to expand the scope of the welfare state than it is to reduce it. I think that more harm will be done to indi­vidual rights by taking this route.

* We can further break up the set of priv­i­leges into a group of wealth-​​redistribution enti­tle­ments (e.g., social security benefits; joint taxation) and private sector mandates (e.g., spousal insurance mandates) if we want, but since both are ille­git­imate functions for a proper government, we can ignore this distinction for now.

† Many of the contractual oblig­a­tions asso­ciated with marriage are not merely histor­i­cally limited to two parties, but are neces­sarily so limited. Take medical decision-​​making: one cannot by contract grant to two different people equal, absolute medical decision-​​making power in the event of inca­pacity. A rule for deciding conflicts would make the power unequal between the two and non-​​absolute as to each one.

‡ Incidentally, I have not yet found a compelling argument in favor of government involvement in the formation of a marriage contract, beyond the normal consid­er­a­tions of compe­tency and whatnot. And I am unable to conceive of any reason why a government should insist on the use of its particular form in drafting that contract. While I have no problem with a legal requirement that a marriage contract be notarized or otherwise formally witnessed, I can find no justi­fi­cation for the current system, where parties do not draft the contract, have no oppor­tunity to read the contract, and, in fact, are unable to do so, because the marriage contract is imputed by common law and statutes, not reduced to a legally operative contractual document. This lack of an objective reference for the terms of the agreement leads to all sorts of awful incon­sis­tency in inter­preting the actual meaning of the marriage contract.

** I think there is an ethically right answer, but only if you do not drop the full context. Legitimate rights never conflict, but what do you do when you are forbidden, as many political agents are, from treating an ille­git­imate right as ille­git­imate? Political agents are often bound to analyze legal questions by dropping the context of the under­lying injustice of the welfare state. I will discuss situ­a­tions like this in more detail, along with a theory on how such agents should act in these situ­a­tions, in a forth­coming post.

Legally, the two relevant cases are Romer v. Evans, 517 U.S. 620 (1996), and Lawrence v. Texas, 539 U.S. 558 (2003), and neither is as helpful for gay marriage as some people hope.

Romer might be used to strike down state bans, but will not consti­tu­tionally force the states to authorize gay marriages. Romer struck down Colorado’s “Amendment 2,” passed by refer­endum in 1992, prohibiting the state government from enacting any civil rights legis­lation applying to non-​​heterosexuals. The Court struck down the amendment on the grounds that the amendment did not seek a legit­imate govern­mental interest. The amendment, the Court deter­mined, was designed to exclude a group of people from the legislative process. Justice Kennedy, who wrote the opinion, is a big fan of the First Amendment, and may have been making an argument that the amendment violated the First Amendment’s guarantee of the right to petition the government for redress of griev­ances. The case is unhelpful for three reasons. First, it declines to apply strict scrutiny, setting persuasive, but not conclusive precedent that strict scrutiny does not apply to laws discrim­i­nating against homo­sexuals. Second, the purpose of the amendment was to exclude a group of people from the legislative process. The purposes of gay marriage bans are various, and generally less clear. “Protecting the sanctity of marriage” is one asserted purpose that the current Court is not likely to find ille­git­imate. There are numerous others that the Court would, rightly or wrongly, find legit­imate. Finally, the case strikes down a ban on pro-​​gay civil rights legis­lation; it does not order Colorado or any other state to enact pro-​​gay civil rights legis­lation. So it is not at all likely to be exten­sible to force states to allow gay marriage.

Lawrence strikes down sodomy laws. It does so on Due Process grounds, not on Equal Protection grounds. The Court (opinion again by Kennedy) overruled Bowers v. Hardwick, 478 U.S. 186 (1986), which had held that prohi­bi­tions against homo­sexual sodomy, based on moral oppro­brium, were acceptable. Kennedy and the Lawrence Court reverse that decision and rule that moral oppro­brium alone is insuf­fi­cient to state a legit­imate state interest. The Court doesn’t exactly get to the Due Process strict scrutiny analysis because a legit­imate state interest is required by both the strict scrutiny and rational basis tests. Therefore, the Lawrence Court never decides the applic­a­bility of strict scrutiny to laws discrim­i­nating against homo­sexuals. Like in Romer, this creates a persuasive, but not conclusive precedent suggesting that strict scrutiny does not apply, this time in the Due Process arena.

In order to apply strict scrutiny in the Equal Protection arena, the Court has to determine either that the rights of a “suspect class”†† are impli­cated, or that a funda­mental right is at stake. The only Justice on the Lawrence Court to suggest applying Equal Protection analysis was Justice O’Connor, and she garnered only her own vote in favor of that approach. Now, she’s gone. And moreover, she did not suggest that homo­sexuals might be a suspect class.

The upshot is that no Supreme Court precedent suggests that homo­sexuals are a suspect class, subjecting discrim­i­natory laws against them to strict scrutiny. The precedent shows that only the rational basis standard has been applied. All that is needed to pass rational basis is a legit­imate state interest and a rational relation of the law to accom­plishing that interest. Most difficult for gay rights advocates, the burden of showing that the interest is ille­git­imate or that the law has no rational relation to the interest is on the chal­lenger. It is not the legal duty of the state to assert or conclu­sively establish the legit­imacy of the interest or the rational relation of the law to achieving it.

Because Equal Protection jurispru­dence grew up in the context of a mixed government, it is crafted to accom­modate the welfare state. The welfare state requires unequal appli­cation of the laws. For example, a law prohibiting distri­b­ution of milk in plastic, non-​​refillable containers, but not in paper, non-​​refillable containers, passes Equal Protection analysis, in part because the aims of the welfare state are accepted as legit­imate state interests.

Bans on gay marriage may fail under an Equal Protection analysis, but the failure of a state to permit gay marriage is not likely to offend the Court’s current Equal Protection jurispru­dence. It is not likely that the current Court will decide that homo­sexuals are a “suspect class” deserving of the special protec­tions of strict scrutiny. It may be right, in an objective sense, that failure to provide equal marriage rights to homo­sexuals, or providing “separate but equal” marriage rights, may “deny to [homo­sexual persons] the equal protection of the laws,” but such arguments are non-​​starters in our current, conflicted legal system.

†† I find “suspect class” analysis highly . . . suspect. Because the Court does not have a rational, objective concept of funda­mental rights, it resorts to aggre­gating the rights of particular groups and treats some groups as more deserving of protection than others. Although, to its credit, the Court, once finding a suspect class, converts the class into a clas­si­fi­cation; e.g., if the suspect class is blacks, the suspect clas­si­fi­cation becomes race, and the Court will apply strict scrutiny to laws discrim­i­nating on the basis of the suspect clas­si­fi­cation, not just against the suspect class. I think suspect class analysis should diminish, and that funda­mental right analysis should expand. In the case of gay marriage, I do not find the situation urgent enough to justify making an argument to expand bad law (i.e., that homo­sexuals are a suspect class). If the threat to indi­vidual rights were dire enough to constitute an emergency, I might think differently.

PS: I am entirely willing to be corrected or convinced to the contrary.