MARRIAGE
The freedom to marry has long been recognized as one of the vital personal rights essential to the orderly pursuit of happiness by free men . . . Marriage is one of the basic civil rights of man, fundamental to our very existence and survival.
—LOVING V. VIRGINIA1
Modern marriage has lost its meaning—consequently one abolishes it.
—FRIEDRICH NIETZSCHE2
Is there a constitutional right to marry? On several occasions, the Supreme Court has said so.3 But the very idea of a “right to marry” presents two sets of puzzles. The first involves the content of the right: what it provides for those who are entitled to it. The second set of puzzles involves the scope of the right to marry: the kinds of relationships that can claim it.
Begin with the first puzzle. As an official matter, marriage is no more and no less than a government-run licensing system. Why should governments license marriages? Some people are skeptical of the official institution of marriage and argue that it should be abolished—not by forbidding private arrangements, religious or otherwise, but by eliminating the special status that governments confer, including a unique set of legal benefits and burdens. Most people strongly favor this particular licensing system, and the doubts are not widely held, but they have cast in sharp relief the vast range of possibilities, with respect to human relationships, that the simple notions of “married” and “single” tend to erase.
But are states under a constitutional obligation to recognize an official institution called marriage? Does the Constitution really mean that the government must provide tax, Social Security, and other benefits to those who are lawfully married? Or does it mean, much more modestly, that states may not forbid religious and other organizations from performing “marriage ceremonies” and allowing those who go through such ceremonies to claim that status as a matter of personal choice?
And what relationships are included within the right to marry? People do not have a right to marry their dog, their cell phone, their house, their refrigerator, Afghanistan, July 21, a power plant, or a rose petal. At most, people have a right to marry people. But the Supreme Court cannot possibly have meant to suggest that “people” have a general right to marry “people”; it did not mean to say that, under the Constitution, any “person” has a right to marry a thousand other people, or five, or even two. We might conclude that the Court is saying (at most) that one person has a right to marry one other person. But if there is a right to marry, what is the basis for this particular limitation on the right? Does the Constitution require states to recognize same-sex marriages?
My initial suggestion is that the right to marry is best understood as an analogue to the right to vote. In both cases, states are under no obligation to create the relevant institutions; but once those institutions are created, the Constitution imposes large barriers to government efforts to deny people access to them. But what is the institution of marriage? It seems to have two characteristics: the expressive legitimacy that comes from the public institution of marriage, and the panoply of material benefits—both economic and noneconomic—that the marital relationship confers.
The right to marry, then, is a right of access to the expressive and material benefits that the state affords with the institution of marriage. Unless a compelling justification can be found, no one can be denied access to those benefits. This understanding of the right to marry suggests that so long as the official institution of marriage exists, the right to marry entitles people not to any particular set of expressive and material benefits but to exactly that panoply of benefits that the relevant state offers.
But what is the scope of that right? At a minimum, the right includes relationships between one man and one woman—the right, of one adult man and one adult woman, to enter into the marital relationship, with whatever expressive and material incidents the state affords, unless the relevant restriction is supported by compelling justifications.
This minimal understanding is fully consistent with the Supreme Court’s decisions. It can also claim support in a tradition-centered approach to constitutional interpretation, one that attempts to root an understanding of liberty in longstanding practices. Under the minimal understanding, the bans on same-sex, bigamous, and polygamous marriages are legitimate, simply because such marriages do not involve one man and one woman.
The principal problem with the minimal understanding is that it seems to draw arbitrary lines. Why should the scope of the right be limited in that way? Compare a far-from-minimal understanding of the right to marry: a right, of two or more adults, to enter into the marital relationship, with its expressive and material incidents, unless the relevant restriction is supported by compelling justifications. This approach would essentially convert marriage from a closed licensing system into an open-ended one, allowing people to enter into marital agreements as they see fit. Such an approach might tailor its economic incidents to the particulars of the relationship—refusing, for example, to accord economic benefits when there is no sufficient reason for them.
From the constitutional point of view, the problem with this understanding is that it depends on a broad and unanchored understanding of “liberty,” one that endangers restrictions on marriage that are at least time-honored (such as the ban on incestuous or polygamous marriages). One task is to produce an understanding of the scope of the right that lacks the arbitrariness of the minimal understanding without being unnecessarily broad.
I shall try to provide that understanding here. To get ahead of the story, I believe that there are no reasonable grounds for prohibiting same-sex marriages. That prohibition is a form of unacceptable discrimination. But this conclusion raises several puzzles about the nature of the marriage right and the role of the federal courts in a democratic society.
MARRIAGE AND THE SUPREME COURT
The constitutional right to marriage has deep roots. In 1888 the US Supreme Court described marriage as “the most important relation in life”4 and called it “the foundation of the family and of society, without which there would be neither civilization nor progress.” In Meyer v. State of Nebraska,5 decided in 1923, the Court said that the Constitution protected the right “to marry, establish a home, and bring up children.” In Skinner v. State of Oklahoma ex rel. Williamson (1942),6 striking down a compulsory sterilization law, the Court described marriage as “fundamental to the very existence and survival of the race.” Griswold v. Connecticut7 (1965) held that states could not ban married couples from using contraceptives. The Court emphasized that it was dealing with “a right of privacy older than the Bill of Rights—older than our political parties, older than our school system. Marriage is a coming together for better or for worse, hopefully enduring, and intimate to the degree of being sacred.” (So wrote Justice William O. Douglas, who was married four times.)
None of these cases, however, involved the right to marry as such. In its modern form, the right to marry is principally a product of three cases. The initial decision, in 1967, was Loving v. Virginia,8 in which the Court struck down a ban on interracial marriage. Most of the Court’s opinion spoke in terms of the equal-protection clause, seeing that ban as a form of racial discrimination. The Court could easily have stopped there. But in a separate and broader ruling, it also held that the ban was inconsistent with “the freedom to marry,” which “has long been recognized as one of the vital personal rights essential to the orderly pursuit of happiness by free men.” It added that “[m]arriage is one of the ‘basic civil rights of man,’ fundamental to our very existence and survival.”
The second decision was Zablocki v. Redhail (1978).9 There the Court invoked the equal protection clause to strike down a Wisconsin law forbidding people facing child support obligations to remarry unless they obtained a judicial determination that they had met those obligations and that their children were not likely to become public charges. The Court proclaimed that “the right to marry is of fundamental importance for all individuals,” adding that “the decision to marry has been placed on the same level of importance as decisions relating to procreation, childbirth, child rearing, and family relationships.”
The Court did say that it would uphold “reasonable regulations that do not significantly interfere with decisions to enter into the marital relationship.” But it announced that any “direct and substantial” interference with the right to marry would be scrutinized carefully. In a concurring opinion, Justice John Paul Stevens underlined the point, urging that the Constitution would cast serious doubt on any “classification which determines who may lawfully enter into the marriage relationship.” This suggestion appears to be extremely broad, applying to incestuous and same-sex marriages, among others—though it is most doubtful that Justice Stevens meant to suggest that states must recognize those marriages. In Zablocki, the Court concluded that the restriction could not be justified, for it was an unnecessarily intrusive means of ensuring compliance with child support obligations.
In Turner v. Safley (1987),10 the Court followed and extended Zablocki, striking down a prison regulation that prohibited inmates from marrying unless there were “compelling reasons” to allow them to do so. “Compelling reasons” were understood to include pregnancy or the birth of an illegitimate child. In fact, the Court went beyond its previous decisions to spell out some of the foundations of the right to marry. It said that marriages, by inmates as by others, “are expressions of emotional support and public commitment,” and emphasized that these are “important and significant aspects of the marital relationship.” The Court added that marriages are often recognized as having spiritual significance, and that the “marital status often is a prerequisite for” a number of material benefits, including property rights, government benefits, and less tangible advantages.
WHAT IS THE RIGHT TO MARRY?
What, exactly, does it mean to say that there is a right to marry? Are states obliged to provide the institution of marriage? This is a more difficult question than it might seem. To answer it, we need to be clear on what that institution entails.
Imagine that a state abolishes the institution of marriage in the following sense: it says that it will not recognize anything called “marriage,” nor will it provide marriage licenses in any form. It will not legitimate particular relationships through declaring them to be “marriages.” Nor will it confer special benefits, economic and otherwise, on people who deem themselves to be “married.” But the state will not forbid such private arrangements as people choose. Above all, it will allow private persons to organize their personal relationships as they see fit, subject to limitations stemming from the criminal law (such as the ban on sexual relations with children).
Religious ceremonies, constituting relationships that the parties may call “marriage,” would not be abolished. If the parties follow the proper formalities for making contracts, their agreements would be enforceable under the ordinary terms of contract law. But as a matter of law, and apart from these points, there would be no such thing as “marriage” as an official matter of state licensing.
Does the “right to marry” mean that the abolition of official marriage would be unconstitutional? Under the Supreme Court’s decisions, this would not be an implausible conclusion. Perhaps the Court is best read as having recognized the existence, within Anglo-American law, of the institution of marriage as one that the government recognizes and safeguards. But if there is a right to an official institution of marriage, what must the state do or provide? The initial question is what marriage actually entails. We need to distinguish here between material benefits on the one hand and expressive ones on the other.
Many material benefits, economic and noneconomic, accompany the marital relationship. Of course, state laws vary, but these benefits fall into six major categories.11
1. Tax Benefits (and Burdens). While a great deal of public attention is paid to the “marriage penalty,” the tax system rewards many couples when they marry—at least if one spouse earns a great deal more than the other. There is a marriage “bonus” for couples in traditional relationships, in which the man is the breadwinner and the woman stays at home. The marriage penalty can be significant if the spouses both earn substantial incomes. Married couples can also file joint returns. Members of such couples are allowed to transfer property to each other without being subject to gain-loss valuation.
2. Entitlements. While the Obama administration took numerous steps to put unmarried same-sex couples on a plane of equality with married opposite-sex couples, federal law continues to benefit married couples through a number of entitlement programs. Under the Family and Medical Leave Act, for example, employers must allow unpaid leave to workers who seek to care for a spouse; they need not do so for “partners.” Veterans’ programs provide some economic benefits (involving medical care, housing, and educational assistance) to the spouses, but not the partners, of veterans. Those who are married to federal employees can also claim certain benefits unavailable to those who are unmarried. Under state law, the entitlement to consortium protects spouses; the status of members of unmarried couples is unclear. With respect to same-sex couples, the law is evolving rapidly, but in general, married couples have significant advantages under entitlement programs.
3. Inheritance and Other Death Benefits. The surviving member of a married couple obtains a number of benefits at the time of a spouse’s death. The rules favor wives and husbands for those who die intestate, or without a will, and many states forbid people to refuse to leave money to the person to whom they are married. Under the Uniform Probate Code, those who die intestate give much of their estate to their spouse, even if they had children. In wrongful death actions, spouses automatically qualify for benefits.
4. Ownership Benefits. Under both state and federal laws, spouses may well have automatic ownership rights that nonspouses lack. In community property states, people are given automatic rights to their spouses’ holdings, and they cannot contract around the legal rules. Even in states that do not follow community property rules, states may presume joint ownership of property acquired after marriage and before legal separation.
5. Surrogate Decision Making. Members of married couples are given the right to make surrogate decisions of various sorts in the event of their spouse’s incapacitation. More generally, a spouse might be appointed formal guardian, entitled to make decisions about care, residence, and money, as well as about particular medical options.
6. Evidentiary Privileges. Federal courts, and a number of state courts, recognize marital privileges, including a right to keep marital communications confidential and to exclude adverse spousal testimony.
This is a large set of benefits. Some states do make them available to same-sex couples, but most do not. While there are sharp political constraints on any effort to rethink such benefits, the state is not constitutionally required to provide them. Suppose that Illinois altered a host of laws to place married people closer to, on the same plane as, or even below unmarried couples or single people. It defies belief to suggest that the alteration would be an unconstitutional violation of the right to marry.
But if the right to marry does not require economic benefits, what, exactly, does it entail? Perhaps it operates as a kind of precondition for certain familial rights that are, broadly speaking, associational in character—for example, the right to visit and make choices for a loved one in case of incapacitating illness. Under existing law, marriage may or may not be a literal precondition for these rights, but it usually makes their exercise significantly easier. Does the Constitution require states to provide the institution of marriage so as to recognize these rights? Suppose that a state abolishes official marriage and denies members of committed relationships the right to make decisions on behalf of their incapacitated partners in the hospital. The Constitution may well forbid states to take such (absurd) steps. But this point does not suggest that states must provide the institution of marriage as such. If official marriage were abolished, the Constitution might grant people a right to some of those benefits in any case—without requiring the institution of marriage.
But is that institution required for other reasons? Recall that in Turner, the Court stressed that marriages are “expressions of emotional support and public commitment.” If a state says that people are married, then they are, in fact, married, and not only for purposes of financial and other benefits. They are married in the sense that the relationship is taken, by everyone who knows about it, to have a special quality as an official and public matter. In short, marriage has an important function in signaling the nature of the relationship. Quite apart from material benefits, the official institution of marriage entails an important form of public legitimation and endorsement. Perhaps states must provide the institution for that reason.
Consider two people who announce to the world that they are married, or who seem to act as married couples do, or both, but with one wrinkle: under state law, they are not, in fact, married. (Suppose, too, that they have not been married through any formal ceremony, religious or otherwise.) To be sure, people can become “engaged” simply by announcing that fact. But to be married as a matter of formal law, they have to go through certain official procedures. Does the right to marry mean that the Constitution requires states to make those procedures available? Perhaps it does.
But perhaps not. Constitutional rights usually do not require affirmative provision by the state. The right to choose abortion, for example, forbids the state to impose undue burdens on the right to choose, but it does not require the state to fund abortion. For most privacy rights, the Constitution requires government noninterference. It does not require government to provide money, institutional arrangements, or anything else.
Of course, this analogy is not decisive. We could imagine a ruling to the effect that the legal institution of marriage is so time-honored, and so important to family formation, that states must make it available. But this argument might end up falling on deaf ears in any situation in which official marriage really was abolished. To see why, note the difficulty, at the present time, of even imagining a state that has actually abolished the legal institution of marriage. This is difficult to imagine simply because most people cherish that institution. If a state did abolish marriage, it would be because most people, or most influential people, had come to believe that the official institution should be abolished. In those circumstances, the idea that there is a right to that institution would be difficult to accept. Whatever the content of the right to marry, it is not at all clear that it requires states to maintain an official licensing scheme for recognizing and legitimating marriage.
It follows that the “right to marry” entails an individual right of access to the official institution of marriage so long as the state is providing that institution. The best analogy is the right to vote. As the Constitution is now understood, states are not required to provide elections for state offices. But when elections are held, the right to vote qualifies as fundamental, and state laws that deprive people of that right will be carefully scrutinized and generally struck down. For both the right to marry and the right to vote, there is an unanswered question whether and when the state is required to create the practice in the first instance. But so long as the practice exists, the state must make it available to everyone.
LOVE AND MONEY
I have said that as a matter of state law, the institution of marriage contains both material and expressive features. And the discussion thus far should be enough to suggest that when the Supreme Court says there is a right to marry, it is establishing a right to both features. Suppose that a white person and an African-American seek to marry, and that a state says that such relationships can be formed as civil unions but not as marriages. Suppose, too, that civil unions have all the legal benefits of marriage. It is clear that under Loving, the state cannot deny interracial couples the right to marry merely by insisting that it is providing such couples with the right to the material incidents of marriage. (So, too, a state cannot say that unless people have met their child support obligations, they may enter into civil unions but not marriage.)
Just as a state could not grant the material benefits of marriage while denying the expressive benefits, it could not provide expressive benefits while denying the material benefits. Certainly a state could not say that mixed-race couples can marry but without receiving the material advantages that normally flow from marriage. The Court’s opinions seem to mean that for those who enjoy it, the right to marry conveys a right of access to the expressive and the material benefits of marriage, so long as the institution of marriage exists.
THE MINIMAL RIGHT TO MARRY
But what is the scope of this right? Begin with a minimal understanding. By deeming the right to marry fundamental, the Court did not mean to suggest that it would question any law that departed from the traditional idea that a marriage is between (one) woman and (one) man. It meant to say only that when a man and a woman seek to marry, the state must have exceedingly good reasons for putting significant barriers in their path. The minimal understanding of the right to marry is that without very good reason, states may not deny an adult man and an adult woman access to the institution of marriage. This rationale would not question bans on same-sex marriages, polygamous marriages, or marriages between people and cats.
But the minimal understanding does have two serious problems. First, it might turn out to be less minimal than it appears, for one simple reason: it raises serious questions about bans on incestuous marriages. If this problem is deemed serious, two options are available. The right might be described more narrowly still: without very good reason, states may not deny an adult man and an adult woman access to the institution of marriage unless the marriage runs afoul of longstanding views about who may enter into a marital relationship. This understanding of the right to marry also fits with the Court’s decisions, and it would insure that the Court would uphold any restriction that is not novel—a benefit for those who believe in a cautious judicial role in this domain. Alternatively, bans on incestuous marriages might be permitted only if they can be compellingly justified—and struck down if they cannot be. For example, an uncle might be prohibited from marrying his niece simply because of the risk of coercion and psychological harm. But a ban on marriage between cousins might well be struck down.
The second problem with the minimal understanding is much more formidable. It draws lines that seem arbitrary in principle. Why, exactly, should the right to marry be limited in this way? Why should the state be required to give a strong justification for any law that forbids marriage between one woman and one man—but not have to justify any other law that impinges on marital choices?
A possible answer would be that marriage is a legal status, with a scope that depends on nothing more than convention. On this view, the Court has not (until recently) been willing to revisit the question of what marriage is; it has worked entirely within the convention on that count. But are conventions really determinative? Suppose that a white person and an African-American seek to marry, and a state responds (in, say, 1961) that they cannot marry because marriage is defined as, or is conventionally understood as, a legal relationship between people of the same race. The mere fact that there is a convention to this effect cannot be decisive. The convention is constitutionally unacceptable. There are countless conventions, and their legal validity depends on whether they conform to the Constitution. Their status as conventions cannot resolve that question.
TRADITIONALISM
Apart from convention, what might be said in favor of the minimal understanding of the right to marry? One answer would endorse constitutional traditionalism. A decision to root constitutional rights in traditions might be the best way of reducing judicial mistakes and judicial burdens (in economic terminology, the costs of errors and the costs of decisions). Let us see how this argument might be spelled out.
Some members of the Supreme Court have been drawn to constitutional traditionalism simply as a way of disciplining themselves—of limiting judicial discretion. If judges follow traditions, they will not have to ask hard questions about basic values, and if judges are not good at answering such questions, traditionalism will have considerable appeal. An initial objection to this approach is that traditions are not self-defining; they do not come prepackaged for easy identification. It is tempting to object that constitutional traditionalism is a charade, in which the key value judgment—how should the tradition be defined?—ends up doing all the work. But the objection is overstated. We should be able to agree, for example, that in the United States, there is no tradition of respect for incestuous marriages or marriages that involve more than two people.
Many people believe that the discipline imposed by tradition is far from arbitrary. Suppose that with the great British statesman and theorist Edmund Burke,12 we believe that simply because traditions represent the judgment of countless people over a long period of time, they are likely to be wise. If so, traditions have some of the advantages of free markets, reflecting as they do the assessments of many rather than few. This is not to say that longstanding practices are always justified. They might reflect prejudice or ignorance rather than wisdom. But perhaps practices are likely to be longstanding only if they make some sense and promote important goals. If so, there should be a presumption in their favor.
In any case, the question is a comparative one. If we believe that judges are prone to error, an effort to root constitutional understandings in traditions might well be better than any alternative. And even if we believe that judicial decisions have some advantages, we might agree that in the face of doubt, democratic judgments, especially in a federal system, deserve a measure of respect, in part because self-government is one of the rights to which people are entitled.
For all of these reasons, constitutional traditionalism is far from irrational or arbitrary, even if it produces bad results in particular cases. And if we are constitutional traditionalists, we might insist that if there is a right to marry, it includes only the time-honored form: one man and one woman.
RATIONALITY, ARBITRARINESS, INVIDIOUSNESS
To clarify the issue, imagine a much broader understanding of the scope of the right to marry: two or more people have a right of access to the marital relationship, with its expressive and material features, unless any restriction is supported by compelling justifications. And with the benefit of a little science fiction, it should not be so difficult for us to imagine a parallel world—unfamiliar to be sure, but perhaps not so unrecognizably different from our own—in which this understanding were accepted. In such a world, people could deem themselves married and receive the appropriate license from the state, so long as force and fraud were not involved. Should this understanding be accepted? What would be the implications?
Even under the broad understanding, prohibitions on nonconsensual marriages are plainly legitimate; so, too, are prohibitions on marriages involving people who are underage, as well as incestuous marriages (except perhaps if they involve cousins). There are only two differences between the minimal and the broad understanding. First, the latter eliminates the limitation to two people. Second, the latter eliminates the ban on same-sex marriages.
Put aside the fact that no federal judges seem even a little bit interested in ruling that states must recognize polygamous marriages. Is their skepticism justified? An affirmative answer might insist that many of the material benefits of marriage make sense only for couples, and no sense at all for groups of three or more. To come to terms with this justification, we would have to go through those benefits one by one. Perhaps states could compellingly justify the decision to withhold some—but not all—of the material incidents of marriage from polygamous relationships. But even if so, the broad understanding might nonetheless require states to recognize certain relationships as “marital” for expressive reasons.
But these points do not capture the real reason that polygamous relationships cannot be deemed “marriages.” After all, most states not only refuse to recognize them as such but also impose criminal penalties on the people involved. The reason is that states believe that polygamous relationships cause individual and social harm—and, in any case, are immoral. Under the broad understanding, however, states must produce compelling justifications for prohibiting polygamous marriages. Should they be required to do so?
I do not believe so. A central reason involves the limited role of the judiciary in a democracy. The broad understanding would put courts in a position for which they are extremely ill-suited. It would require them to assume the exceptionally difficult task of policing the adequacy of official justifications for refusing to recognize marriages involving more than two people. Those justifications, which involve complex questions of both policy and morality, are best evaluated democratically, not judicially. If this objection is convincing, then we can imagine an intermediate understanding of the scope of the right to marry, one that puts the question of same-sex marriage in sharp relief: two adults have a right of access to the marital relationship, with its expressive and material incidents, unless the restriction is supported by compelling justifications.
I believe that this is the right principle, and that it is not possible to make a convincing defense of the prohibition on same-sex marriage. The principle is especially appealing in light of the fact that discrimination on the basis of sexual orientation has long been rooted in hostility and prejudice, and should therefore be carefully scrutinized by courts. What sorts of social harms would follow from recognizing marriages between people of the same sex? Some people argue that refusing to recognize same-sex marriage is a way of protecting the marital institution itself. But aside from purely semantic arguments, this is very puzzling. How do same-sex marriages threaten the institution of marriage? Extending the right to enter into the institution of marriage would hardly threaten traditional marriages—unless it were thought that significant numbers of heterosexuals would forgo traditional marriages if same-sex marriages were permitted (a difficult causal argument, to say the least). Or perhaps same-sex marriages would harm children. But what evidence supports that speculation?
Some people believe that the state can legitimately deny marriage to same-sex unions for expressive reasons. They do not want to “endorse” such unions or suggest that they are appropriate or legitimate, or have a standing similar to that of traditional marriage. But what is their justification? Why should states refuse to endorse such unions?
For those who believe, as I do, that the ban on same-sex marriages cannot be plausibly justified, the only questions involve the limited role of the judiciary in a democratic society. In 2013, the Supreme Court struck down the Defense of Marriage Act as discriminatory, but it declined to rule on bans on same-sex marriage as such. There have been legitimate prudential reasons for federal courts to hesitate before requiring states to recognize same-sex marriage, not least because that issue is under intense discussion at the local, state, and national levels. Committed minimalists (see chapter 10) would want judges to leave the question undecided, not because they are ambivalent about the underlying principle but because they believe in a modest role for the federal judiciary.
To be sure, the minimalist position raises questions and objections of its own. Should courts really hesitate to vindicate constitutional rights? When? Won’t there come a time when courts should strike down unjustified discrimination? Why isn’t that time now? This is not the place to answer these questions. The only point is that if judges are to hesitate, it is because of the claims of judicial humility, not because there is a good argument, in principle, for forbidding same-sex marriage.
RIGHTS AND DEMOCRACY
My goal has been to make progress on two questions. The first is the content of the right to marry. The second is the scope of that right.
I have suggested that like the right to vote, the right to marry is one of equal access to a publicly administered institution. It is a right of access to the expressive and material benefits that official marriage provides.
I have also identified competing understandings of the scope of the right to marry. The minimal understanding recognizes the right of access, by any couple consisting of one adult man and one adult woman, to the expressive and material benefits of marriage, so long as the institution of marriage exists. The chief advantage of the minimal understanding is that it promises to minimize judicial discretion and to rely on practices that, simply because they are time-honored, have a claim to social respect. Its chief disadvantage is arbitrariness. The disadvantage is serious. In principle, there is no convincing constitutional defense of the ban on same-sex marriage.
What about the marital institution itself? Like most people, I believe that it should be continued, if only because it provides the basis for a kind of precommitment strategy that is beneficial to adults and children alike. But this conclusion should not obscure a central point: official marriage is unambiguously a form of government intervention, and its future form should be a matter not of following supposed dictates of any kind, but of our own free choices.