Andrew Sullivan
Andrew Sullivan; drawing by David Levine


For most Americans today, acquiring a spouse is the most important task in the years after high school. You are supposed to find someone you love and who loves you, to get to know each other and confirm that you are compatible, and then move into shared quarters and take up a common life. Ideally, you and your spouse then stay together, in the formula of the Book of Common Prayer, “so long as ye both shall live.”

If you want to start a family, your spouse is the person with whom you will have and raise children. This is one reason sex is important to marriage. But you may not want to raise children and (while this may distress your parents) that is your right. Sex still remains important. It is a natural and pleasurable expression of marital love; which is to say that marital love, at least as most of us now conceive of it, is largely erotic.

But your marital relationship will, of course, affect almost everything you do; it will be central—especially if it goes well—to the meaning of your whole life. As you grow old and look back, you will be looking back on a life lived together. Browning’s Rabbi Ben Ezra is so often quoted because he captures this promise so exactly:

Grow old along with me!
The best is yet to be,
The last of life, for which the first was made…

It is a vision expressed with a more prosaic eloquence by Genora Dancel, one of the lesbian partners in the suit against the limitation of marriage to heterosexual couples now pending before the Supreme Court of Hawaii: “I want to be able to say at the end of my life that I had loved somebody really well for a long time.”

This image of marriage belongs to a narrative of romance that has deep roots in Western culture, but it became a possibility for large numbers of men and women only with the emergence of a sizable middle class. Its ascendancy reflects the triumph of bourgeois romance, in which, like the horse and carriage of Sammy Cahn’s lyric, love provides the motive power and marriage the natural vehicle.

This narrative—with its emphasis on lifelong monogamy—is, for the many who believe in it, an ideal, not something they can be sure of sustaining. Half the marriages in America now end in divorce. Casual observation (reinforced by TV soaps and sociological surveys) confirms that one of the reasons for this is that the commitment to exclusivity—especially on the part of husbands—is difficult to maintain. But this ideal is not only at odds with reality; it is wildly at odds with most traditional conceptions of marriage.

First of all, in earlier Western societies—as in most societies everywhere else—marriage was the creation of a relationship not between individuals but between families, a fact still signified by the role of the father in “giving away” the bride in many forms of marriage rite. (There is a resonant Hebrew word for the relationship between the parents of one spouse and those of the other: they are mehatanim.) When our sisters and brothers or our children marry, we say that we have acquired “in-laws”; there is a name for this relationship or something like it in most languages.

These bonds were once important because they created socially underwritten obligations. Insofar as modern married couples recognize obligations to in-laws, by contrast, they are likely to see them either as expressions of love for their partner, or as growing out of the particular relationships that they develop with the family of a spouse. One asks a partner to do something for “my father”—stress on the “my”—not for “your father-in-law”; one says, “Come on, you know my mother always liked you,” not “You’re not behaving like a good son-in-law.”

A second difference is that marriage once had substantial ritual meaning. In the Middle Ages in Europe, it was a sacrament, one of the seven holiest rites of the church (along with the Eucharist, Baptism, Confirmation, Penance, Ordination, and Extreme Unction). According to the Book of Common Prayer, the marriage of man and wife signifies the mystical union between Christ and his Church. Even today, for devout Catholics and Jews, obtaining a “real” divorce requires appeal to religious authority.

Many people still marry in churches and synagogues, under the supervision of priests, pastors, and rabbis (who are authorized by the state to create—en passant, so to say—the civil bond as they solemnize the religious one). For many young couples a wedding still requires organ music and hymns as well as white dresses; and many rabbis and priests still insist on explaining the sacramental meaning of the wedding. But the solemnity of the occasion is a reflection, for most people, of the significance of the commitment to a life of love.


Central to these religious understandings—this is a third difference from the modern picture of marriage—is the begetting and raising of children. For Orthodox Jews, having and raising children within Jewish tradition is, where possible, a religious obligation; for Catholics, sex within marriage is permissible only if nothing “unnatural” is done to prevent conception; the different Protestant traditions and forms of service almost always mention marriage as the setting for bearing and raising children. The same Book of Common Prayer informs the congregation: “First, it was ordained for the procreation of children, to be brought up in the fear and nurture of the Lord.”

It is true that the Jewish tradition, early on, saw marital sexuality as an expression of the conjugal relation-ship and not merely as procreative; Luther and Calvin also grasped this idea. For Catholics, such an understanding has (notoriously) emerged only in more recent years. But most of the religious traditions we inherit here in America—and since much of our moral thinking about sex is religious, this means most of the moral traditions—have had a hard time accepting sex outside marriage. (Bishop William Waitland, the Anglican Bishop of Eau Claire, Wisconsin, said recently, apropos his church’s attitude to homosexuality, “there’s not a single support in Scripture for adultery, fornication, or incest,” reflecting his church’s historic view that all sexual relations between two unmarried people must fall under at least one of these unwholesome labels.)1

With the triumph of the narrative of bourgeois romance, these three elements—the alliance between families, the sacramental meaning, and the centrality of procreation—have lost their force for many even here in America, the Western country with the highest proportion of its population still nominally committed to religion. The end of a marriage strikes many as a terrible thing: but this is neither because it is the betrayal of a union between families nor because it is the breaking of a sacred bond. And a marriage that is childless by choice is not ipso facto a failure. Conversely, for many couples children enrich a marriage, but they are not its central meaning; and the success of a marriage is now celebrated, in the end, neither as the bonding of families, nor as the fulfillment of a sacred covenant, but as the loving confluence of two lives.

The ideal of the marital relationship I have been describing can be embedded, for heterosexual couples, in a vast web of laws. To invoke these laws in some states, you must go through a formal ceremony; in others—the common-law marriage states—you can end up legally entangled even if you deliberately and explicitly refuse both ceremony and legal license.2 But however marriage gets established, its legal consequences are enormous.

In The Case for Same-Sex Marriage, William Eskridge (co-counsel in Dean v. District of Columbia, a recent federal case about gay marriage) lists a host of rights that both members of married couples have in Washington, D.C., which, he says, is “representative of other jurisdictions.”3 They include: rights to support and alimony; the ability to act for an incapacitated spouse; the guarantee of benefits in insurance; visitation rights in hospital and in prison; “an evidentiary privilege for marital communications”; rights to inherit property; and various other survivor’s benefits. While others may adopt children in some jurisdictions, married couples have the right—subject to some conditions—to do so everywhere. Mr. Eskridge lists the “modest ‘marriage penalty”‘ in income taxes for two-income families among the disadvantages of marriage, but being treated as a single household for tax purposes is often an advantage: an unemployed spouse can be claimed as a dependent, and no gift taxes have to be paid on transfers between spouses.

These legal rights (and their correlative obligations) are justified by three fundamentally distinct sorts of consideration.

The “evidentiary privilege for marital communications” reflects the distinctive obligations of the marital relationship. I can no more be asked to provide evidence that will convict my spouse than I can be forced to testify against myself. These are the rights of marital relationship—among the most important of which is the exclusion of the government from the regulation of our (consensual) sexual life.

A second set of rights—rights of trusteeship—reflects the recognition that I am my spouse’s natural representative because I know him, his character, and his desires, because I care for him and have his interests at heart. It is as if each spouse acquires in marriage the rights (and correlative duties) to act for each other that their parents once had.

The final class of rights derives from a recognition of the household as an economic unit, where transfers between members do not count as public economic events. This reality is captured in the prayer-book formula: “with all my worldly goods I thee endow.” Such are the rights of economic identity.


Many of the rights of trusteeship that married people automatically have can be created by special legal arrangements: you can give your partner a permanent power-of-attorney and explicitly instruct medical authorities to recognize him as your representative if you become incapacitated. Some of the rights of economic identity can be recreated, too: you can write wills; you can create contractual obligations of support to come into effect if you cease to live together. (In some cities couples can acquire some rights through registering as “domestic partners.”)

In short, with a good lawyer—and at less cost than the average wedding—you can cobble together a bundle of legal rights of economic identity and trusteeship similar to those of married couples. Outside of marriage, however, some of the rights of economic identity will remain unavailable: you cannot claim the tax advantages (or disadvantages) of married people; you cannot benefit from the other’s Social Security or veteran’s benefits; what rights to insurance you have depend on your employers. And, so far as I can see, the rights of marital relationship—including the right to sexual privacy that the Supreme Court, in its decision in Bowers v. Hardwick, explicitly denied to homosexuals—cannot be acquired by any sort of legal agreement.

A final, more general benefit flows from the state’s recognition of heterosexual marriage: in picking out married couples for special grace and favor, the state expresses support for their commitment and respect for their relationship. Dr. Johnson once rightly insisted (with his distinctive mixture of hyperbole and good sense) on the difficulty of the marital commitment and on the importance of the social rules and attitudes that recognize and maintain it:

It is so far from being natural for a man and woman to live in a state of marriage, that we find all the motives which they have for remaining in that connection, and the restraints which civilized society imposes to prevent separation, are hardly sufficient to keep them together.4

Denying that recognition is rightly seen, both by those who favor and those who oppose gay marriage, not only as a refusal to support such marriages, but also as an expression of indifference—even hostility—toward gay relationships. It is clear that for many people (whether or not they want to marry) this standing insult, as they regard it, is the most important issue of all.

A married couple can gain a great deal, then, from legal recognition. They gain it whether their relationship is deep or shallow, faithful or faithless. Any currently unmarried man and woman can wander into a municipal office in their home town and get this full panoply of benefits without even the most perfunctory investigation. (Since prisoners—even those serving life sentences without parole—have the constitutional right to marry, you can get a marriage license even if you can’t wander in off the street.) On the other hand, no two people, however deeply loving their relationship, however sanctified by religious rite or celebrated by their families and friends, can acquire these benefits through the laws of marriage if they are of the same sex.


I have ignored so far the fact that bourgeois romance and the traditional vision of marriage had something in common: neither was a union of equals. The Book of Common Prayer had Mrs. Smith promising to obey Mr. Smith; this was not a promise that Mr. Smith was expected to reciprocate. And she had the major responsibility for raising the children.

Facts like these are crucial in interpreting matrimonial law; they require us to recognize how central the experience of middle-class patriarchy is to the legal meaning of marriage. The practice of allowing a wife access to health and Social Security benefits associated with her husband’s job and his income, for example, emerged when the wives of the middle-class men who had the most valuable forms of insurance and pension provisions did not have incomes of their own. The tax law’s treatment of a couple as a single unit thus at first reflected not the fungibility of two incomes but a shared dependence on one. And alimony was conceived for women whose unfaithful husbands moved on, because being a middle-class wife (and, especially, a mother) meant you didn’t do paid work.

Much of the economic structure of marriage still reflects the old inequality of husband and wife. Access to health care should be something that every adult has for him or her-self, not something that you get as someone else’s dependent. Women with adequate incomes and careers of their own would not have to be dependent on alimony. In sum, some of what straight people—and especially wives—get through marriage, they ought to be able to get without it.

Reflection on the current structure of marriage, therefore, suggests plenty of scope for rational reform. It is clear, however, that such reforms are not about to take place.

The obvious interim solution would be to allow gay couples to get married. Unfortunately, this proposal faces what must in a democracy count as a serious impediment: as many as two thirds of the population are opposed to it.5 The extreme unpopularity of gay marriage is, no doubt, why, in this election year, Bob Dole is cosponsoring a bill to deny federal recognition to gay marriages, even though no state has allowed any to take place, and even though marriage has always been largely a matter of state law, and Mr. Dole is from the party of state’s rights. It is equally clear that this is why Mr. Clinton, whose first major policy fiasco in the White House was his attempt to allow gay men and women to serve openly in the military, long ago sought to preempt gay marriage as an election issue, declaring himself opposed to it in 1992, and now has declared himself willing to sign the bill Dole has cosponsored. It follows—especially in a democracy—that those who wish to see gay marriages legally recognized must offer arguments against this opposition; and such arguments can proceed in two ways.

One strategy begins from first principles. It would examine why state recognition of heterosexual marriage is a wise and just policy, and show that extending it to homosexuals would be a just and wise continuation of that policy, meeting some (or all) of the same ends—and perhaps some additional ones. Another strategy would be to show that the recognition of gay marriage flows from principles embodied in mainstream American political practice, principles that are important to most of us; and that anyone who genuinely supports those principles must, in good conscience, accept that extension.

The first strategy faces an obvious difficulty: from what first principles should we start? In a modern pluralist society such as ours, there’s likely to be disagreement about such matters. If we are to persuade a plurality of our fellow citizens—as Andrew Sullivan aims to do in his elegant Virtually Normal—we are therefore likely to have to make the argument (as he does) in several ways.

Mr. Sullivan’s argument for gay marriage is part of a wider defense of what he calls “the politics of homosexuality.” His aim, in his short book, is to explore how “we as a society” should “deal with the small minority of us which is homosexual.” His answer is clear enough. State action in relation to homosexuals, Mr. Sullivan thinks, should be governed by a simple principle: “that all public (as opposed to private) discrimination against homosexuals be ended and that every right and responsibility that heterosexuals enjoy as public citizens be extended to those who grow and find themselves emotionally different. And that is all.”

Mr. Sullivan’s approach is to examine critically four current positions in the politics of homosexuality, and, in so doing, to develop the case for his own. As I say, any politics that stands a chance of getting the support of an American majority must address the claims of more than one ideology, more than one group of principles. Mr. Sullivan believes that the four positions he addresses—he calls them “prohibitionist,” “liberationist,” “conservative,” and “liberal”—represent “the most prominent views at war in our society, the four essential choices we are being asked to make.” If he can show that the people who are opposed to full civic equality for homosexuals have beliefs that are inconsistent with the founding ideas of each of these basic positions, they can be asked to accept his proposal not for his reasons but for theirs.

Prohibitionism—in the form that Mr. Sullivan addresses it—is, essentially, the official position of his own Roman Catholic Church. It holds that homosexuality is a “crime against nature,” because we are all naturally heterosexual and the “heterosexual union and its social affirmation” are “at the core of the possibility of civilization.” The politics that follows is one of curing or punishing homosexuals, and deterring others who might “stray into the homosexual milieu.”

Mr. Sullivan is keen to insist that this is not necessarily a bigoted view, though he recognizes that some who profess it are merely using it “to prettify a prejudice.” Indeed, he does not insist that accepting the Bible’s few explicit pronouncements on sex between men (there are no clear Biblical references to sex between women) is consistent with regarding sex between gay men as permissible.6 But that is because the debate within his Church about these matters is not fundamentally based on scripture. Rather it is rooted in the natural-law philosophy of Saint Thomas Aquinas.

Sex, in Aquinas’s view, is fundamentally for procreation; homosexual acts (along with masturbation and non-procreative heterosexual intercourse) are therefore “unnatural.” Engaging in sex for any other purpose than procreation is wrong because there is only one proper end of sexual activity: making babies. Catholic teaching on homosexuality has advanced beyond Saint Thomas’s. Since 1975, the Church has held that some people are naturally homosexual, so that they have, in their individual natures, homosexual desires. But homosexual acts continue to be sinful. Mr. Sullivan draws attention to the core paradox here: “In other words, something in nature always and everywhere violated a vital part of the nature of human beings; something essentially blameless was always and everywhere blameworthy if acted upon.”

Mr. Sullivan’s argument with this position is elegant and persuasive; but one wonders how central it can be to contemporary American political debate. After all, as he points out in closing, the natural-law argument should lead prohibitionists to argue for heterosexual sodomy laws and legal proscriptions against masturbation. They should also be opposed to the marriage of post-menopausal women and all others who know they are infertile. A prohibitionist who does not want to do all these things is treating homosexuality—which he says he opposes as non-procreative sex—differently from other forms of non-procreative sex. He does not have a consistent principled position. He is, in short, “prettifying a prejudice.”

Mr. Sullivan has much to say about the militant homosexuals he calls “liberationists.” For them, he writes, “homosexuality as a defining condition does not properly exist because it is a construct of human thought,…generated in human consciousness by the powerful to control and define the powerless.” With this analysis goes a prescription. “For the liberationists, the full end of human fruition is to be free of all social constructs, to be liberated from the condition of homosexuality into a fully chosen form of identity.”

The source of the idea of liberationism, Mr. Sullivan claims, was the late Michel Foucault. “In many ways,” he tells us, Foucault “is to liberationism what Aquinas is to prohibitionism.” In my judgment, however, Mr. Sullivan is a far less reliable interpreter of Foucault than he is of Aquinas. Fortunately for our purposes this does not matter much, since the politics of liberationism he attacks has little connection with his misreadings of Foucault.

Mr. Sullivan admits as much himself. In discussing “outing” as an example of the politics of liberationism, he allows that “to be sure, there was no direct link between this tactic and the philosophical structure I have just been describing…. It is doubtful whether many recent gay activists have ever heard of Foucault, let alone read him. But indirectly, ‘outing’ follows the logic of liberationist [i.e., Foucauldean] politics.” This sort of argument gives polemic a bad name. How can someone who asserts that homosexuality “does not properly exist,” as Foucault is alleged to have done, be the intellectual ancestor of a practice that presupposes that “homosexual” is what the outed person “really” is; especially when those who created the practice were largely unaware of Foucault (and would have been hostile, had they known of him, to his claim that the identity they were celebrating was itself merely another form of prison)?

The real enemy here is not Foucault but what is now called “queer” politics. It is exemplified in the explicit theatrical performances of ACT UP, showing, for example, the suffering of people with AIDS; the bared breasts of the Lesbian Avengers; the “kiss-in” at the local mall. Mr. Sullivan objects to all of this activity on grounds that it mistakes culture for politics. In a free society, he argues, “cultural redeployment…is always subject to a cultural response…. The techniques of ACT UP lend legitimacy to the techniques of Operation Rescue”—the group that violently attacks abortion clinics.

“Culture”—in the sense of deep-rooted attitudes—obviously makes a difference; and there are cultural battles to be won. Theater—of the sort that ACT UP sponsored in its battle to speed up the approval of drugs to treat AIDS by the FDA—has been successful as politics. All this Mr. Sullivan concedes. What he opposes is the notion that this is the only way to act. He resents the view of some activists that politics—conceived of, in Mr. Sullivan’s humane manner, as reasonable conversation about the formal political institutions of the state—is inevitably pointless. He argues that any approach to liberation that ignores the power and politics of the state (whether or not it is inspired by a Foucauldean conviction that the state must always be a source of constraint) will simply fail to

achieve actual results, to end persecution of homosexuals in the military, to allow gay parents to keep their children, to provide basic education about homosexuality in high schools, to prevent murderers of homosexuals from getting lenient treatment….

Mr. Sullivan quotes Michael Walzer’s telling characterization of Foucault, turning it against such militant groups as Queer Nation, which use rhetoric dismissing the entire society as hopelessly sick: “Angrily he rattles the bars of the iron cage. But he has no plans or projects for turning the cage into something more like a human home.” At this point, the argument is utterly convincing. But what convinces us has little to do with the critique of Foucault, and much to do with Sullivan’s sensible acknowledgment of practical politics.

Liberationists, on Mr. Sullivan’s account, are skeptical of marriage, because they see it as yet another of the traps of power. The liberationist agrees with the old comic’s line: “Marriage is a fine institution, if you want to live in an institution.” Mr. Sullivan’s reply is that there is no reason why we cannot, through politics, turn it into “something more like a human home.”

The people Mr. Sullivan calls “conservatives”—his terminology here can be confusing—accept the basic liberal assumptions of the American constitutional order. For them, homosexuality is an essentially private matter; it is not the government’s business. But the public acceptance of homosexuality is another thing entirely; and this conservatives reject because it would offend “the identity…of married heterosexuals and so make it harder for them to practice marriage as it should be practiced. It devalues the social meaning of sex and undermines the very basis of familial life.” What conservatives endorse is a combination of private tolerance and state-endorsed public disapproval.

The conservative, then, is opposed to gay marriage not, like the prohibitionist, because gay sex is immoral; not, like the liberationist, because marriage is yet another form of imprisonment; but because the recognition of gay marriage will undermine straight marriage, which is the central institution of American society. This is a surprisingly widespread conviction, and Mr. Sullivan provides good answers to each of its arguments.

The entire notion that public acceptance and respect for gay relationships would undermine straight ones is very hard to make sense of. What, after all, are the effects of the current public disapproval of homosexuals? First, many people who are strongly inclined toward homosexual love enter into unhappy heterosexual marriages. Second, many young people have the agonies of adolescence intensified by the discovery of gay desire. Third, insofar as social recognition is part of what sustains relationships, gay couples are less likely to sustain them; and, if one accepts the narrative of bourgeois romance, many of them will therefore live less fulfilled lives. Fourth, some who have the choice—because they are bisexual or able to shift their desires—will choose straight marriage over a stigmatized gay relationship.

Of these results, only the last could be thought by any decent person to be a good thing: and only then if that person thought that it was intrinsically better to live a heterosexual life than a homosexual one. But even someone who thought this would also have to hold that it was appropriate for the state to place the burden of suffering on homosexuals in order to tilt the scales in the right direction for the odd bisexual.

The conservative arguments assume, bizarrely, that straight marriage is so intrinsically unappealing that enormous burdens must be placed on homosexuals in order to keep it attractive. After all, if you really want to encourage straight people to stay married, why not toughen the divorce laws and increase the tax advantages? Taking it out on homosexuals is not merely mean; it is ineffectual.

Mr. Sullivan identifies, I think, the key difficulty for conservatism: and it is, again, an objection from the world of practical politics. More and more homosexual men and women are refusing to live in the closet that conservatism prescribes for them. As a result, the combination of private toleration and public stigma that is the essence of the conservative position is no longer a practical possibility.

In this climate, conservatives who wish to maintain public disapproval of homosexuality…will be increasingly forced to resort to crude moral arguments, or publicly express ugly contempt for gay people or for gay practices. They will find this disagreeable in the extreme, especially since they are required by the exigencies of the moment to cooperate, and even join forces with, the prohibitionists, whose illiberalism they disdain.

Mr. Sullivan suggests that there is a different strategy that better suits conservativism’s fundamental respect for liberal notions of privacy and its concern for sustaining the family; and that, of course, is to endorse gay marriage. In so doing conservatives could, if they wanted, try to recruit gay couples for the purposes of the campaign to restore the family, combining private tolerance with its more natural partner, which is a public respect.

Mr. Sullivan’s final argument is with what he calls liberalism; but here he has to go carefully. For surely his own position “that every right and responsibility that heterosexuals enjoy as public citizens be extended” to gay people is a liberal one. In fact, however, Mr. Sullivan wants to argue something different: that a modern liberal will be inclined to grant to homosexuals the affirmative action and anti-discrimination protections that are currently available to another stigmatized group, namely African Americans. Since he has, to put it mildly, doubts about affirmative action for blacks, he is also against affirmative action for homosexuals.

To put the matter this way is to imply (wrongly) that political creeds can be identified with clear principles clearly articulated with respect to one another and embodied in settled understandings of the proper conduct of politics. Mr. Sullivan points out that fewer and fewer people are willing to call themselves “liberals.” But he fails to see those who still do, like the many more who did so before the 1980s, are not defined by a single, articulate political creed. They are heirs to the liberalism of the American Founding Fathers, a liberalism that is expressed in the Bill of Rights, to the extent that they believe in the importance of freedom of expression, association, and religion, and in civic equality. But they are heirs also to the liberalism of the New Deal in their conviction that the state has a part to play in guaranteeing the basic welfare of all citizens and in protecting employees from exploitation by powerful corporations.

Mr. Sullivan’s disapproving account of the genesis of the modern liberal position begins in the nineteenth century with John Stuart Mill and Benjamin Constant, and continues by way of the civil rights movement and affirmative action, without, it is fair to say, so much as a nod toward the New Deal.7 As history goes, this is eccentric. To begin with, although Mr. Sullivan wants to portray Mill’s idea of liberty as the previously accepted standard of liberalism, he has to concede that no nation or party—certainly not the United States or the Republican or Democratic parties—ever conducted itself according to Mill’s postulates assigning equal liberties to all.8 Furthermore, freedom of expression and association have, in any case, expanded in the United States since the nineteenth century, as has freedom of religion; they have not, as Mr. Sullivan seems to imply, declined. What has declined is freedom of contract, for instance the freedom to contract with labor at below minimum wage or minimum age. And what accounted for that was the recognition by New Deal liberals, especially in the 1930s, that allowing a few people who ran banks and other large corporations to do as they wished with their property could lead to the poverty of the many and to the kind of social crisis created by the Great Depression.

Mr. Sullivan, on the other hand, wants to persuade us that it was dealing with race that forced Americans to abandon Mill’s dream. He objects, for example, that liberals now want “to deny others…complete freedom of contract and…complete freedom of expression, in order to protect a specific minority,” i.e., blacks. These illiberal horrors are achieved by such means as “antidiscrimination laws in employment and housing, and forced integration—busing—in school areas, to achieve what free human beings refused to achieve on their own.”

It would not be difficult to show the inaccuracy of Mr. Sullivan’s account of American political history or to challenge his distorted descriptions of the current practices of anti-discrimination and affirmative action. But to do so is not necessary. Someone who sees the end of “complete freedom of contract” as the result of legislation aimed to protect black Americans, or who does not notice that the institutions forcibly integrated by busing were public institutions, is not a safe guide through this difficult territory. (Remember that Mr. Sullivan wants forcibly to integrate another public institution—the armed forces—in the name of avoiding discrimination by the state.)

What Mr. Sullivan essentially argues is that the modern laws intended to mitigate the consequences of social prejudice ignore the boundary between public and private that is the key to a proper liberalism. I agree that the distinction between what is public and what is private is critical to understanding liberalism. But Mr. Sullivan doesn’t seem to understand its significance. The private, for liberalism, is indeed the sphere into which the state should not intrude: but that is just a definition of privacy. The question that must be decided is where the private lies. If a man beats his wife at home, he is, in a sense, acting in a private relationship in a private place: but this is, for the liberal, a public act, in the sense that, as an assault, it is the government’s business. When the Supreme Court respects the privacy of the marriage bed, it does so—rightly—only to the extent that what happens there is consensual. By contrast, loud and excited discussions of politics in a public park between citizens are, for the liberal, protected from the control of the state; they are, in the relevant sense, private.

Mr. Sullivan writes as if there are, for the purposes of political morality, only two forms of human interaction: those people have with the government and those with one another. He calls the former “public,” the latter “private.” This puts my relationship with the officers of the corporation that employs me, or with the agent of the real estate company that manages my apartment block, on the same basis as a conversation over drinks in my sitting room with family and friends. Mr. Sullivan’s view ignores, so it seems to me, the fact that between the domestic and the civic sits the economy and the economic relations people have with one another. A liberalism that wants to pull the government out of the economy altogether—which appears to be Mr. Sullivan’s aim—is simply not a serious candidate for a politics of a modern society.9

Roughly speaking, the liberal case for anti-discrimination laws is that, in the modern world, an autonomous and dignified social life requires access to work and housing; that, if racism makes these unequally available to black people, it is, therefore, a proper aim of a state concerned with respecting the autonomy of individuals to rectify such unequal availability. It is true that this limits the freedom of action of corporate persons (who are, in my view, of no moral interest save as they have an impact on real persons) and of some individuals; but this limitation is, on the whole, not a heavy burden on the dignity and the autonomy of those real persons whose actions it restricts. Property rights need not be regarded by a sensible liberalism as having the weight that should, for example, be accorded to freedom of association and expression.

There seem to me to be plausible parallel arguments for anti-discrimination laws in housing, employment, and public accommodations for gay people; and so limiting gay people’s political rights—as Mr. Sullivan aims to do—to equal treatment by the state strikes me as quite unwarranted.


Mr. Sullivan’s arguments lead to an obvious question: namely, why should the state recognize any marriages at all? If, after all, the problem is that the state is “becoming intimately involved in the details of private life” (his emphasis), then surely nothing could be more objectionable than the laws of marriage. Marriage, for Sullivan, is a paradigm of a private relationship. Mr. Sullivan’s skepticism about modern liberalism should lead him to argue not for extending state recognition of marriage to gay people but against state recognition of marriage at all.

Mr. Sullivan’s approach reveals a fundamental difficulty about arguing from first principles—such as the absolute exclusion of the state from private life—to reform our political and social institutions. Current arrangements reflect a history of compromise in response to a great variety of often incompatible forces and ideas; they are, in other words, the result of a history of politics. To demand reform on the basis of principles is to require new policies to meet an argumentative burden that existing ones could not meet. We are more likely to succeed if we proceed in the second way I mentioned earlier: we can begin with principles that are now embodied in everyday American political practice, and argue that those principles require us to make existing institutions open equally to straight and to gay citizens. That is the strategy adopted in William N. Eskridge, Jr.’s book The Case for Same-Sex Marriage.

William Eskridge combines argument with anecdotal accounts of the lives of lesbian and gay couples. He tells us of happy-sounding gay families, like Susan Silber and Dana Naparsteck—the former “a successful lawyer” in “one of the leading law firms in Takoma Park, Maryland,” the latter a therapist “with raven hair and a ready smile”—and their thirteen-year-old daughter, Danielle Rachel Naparsteck Silber. Danielle knows her biological father, Chris, a gay man she calls Papa, and his former partner, Art; and being “a brilliantly alert child,” she knows that she has “more people to explain things to her and to help her, because she has ‘two mommies and two daddies.”‘

Mr. Eskridge also reminds us of the suffering caused by the present law. He tells the story of Kristen, who was raised by Joan Pearlman, her biological mother, and Janet Ratcliffe. After Joan died, she was taken away from Janet by Joan’s parents even though they knew their late daughter wished Janet to have custody. This story has a happy ending—a judge was eventually persuaded that the child’s wish to be with her mother’s lover should be respected. But the pain of both parent and child in the interim could have been avoided if Janet and Joan had been able to marry.

These tales can be affecting, even if they sometimes read like an awkward yoking of People magazine and the Harvard Law Review. And, so it seems to me, while Mr. Sullivan’s is the more elegant essay, Mr. Eskridge’s work is the more convincing, in large measure because he does not face the difficulties I have tried to identify in arguing from first principles.

The most powerful arguments in Mr. Eskridge’s book address the constitutional basis for the recognition of gay marriage. There are, as Mr. Eskridge sees the matter, two legal bases for challenging the current bans on the recognition of gay couples. One is a “substantive due process” claim grounded in the due process clause of the Fourteenth Amendment to the Constitution. The Supreme Court has held that this clause protects those fundamental rights that are, in Justice Cardozo’s words, “implicit in the concept of ordered liberty,” those rights which belong, in Justice Harlan’s formulation, “to the citizens of all free governments.” This is one of the sources of the privacy rights that underlie the decision in Griswold v. Connecticut striking down laws prohibiting the use of contraceptives within marriage. This case is crucially important for issues involving gays and marriage because it recognizes the importance of protecting marital privacy in relation to a law that is aimed precisely against limiting procreation. Constitutional protections for marital privacy therefore do not derive from the state’s recognition of the importance to marriage of being “geared towards reproduction.”

A second line of constitutional argument for gay marriage makes a claim grounded in the Fourteenth Amendment’s equal protection clause. The Supreme Court has held that state limitations on marriage that place “a direct legal obstacle in the path of persons desiring to get married” are in violation of the equal protection clause of the Constitution unless the policy is “supported by sufficiently important state interests and is closely tailored to effectuate only those interests.”10 Arguments about what are “compelling state interests,” in this sense, are hard to summarize; they proceed by exploring the principles and the decisions in earlier cases about marriage and about equality. In the end, Mr. Eskridge makes a persuasive case that the Supreme Court could strike down laws prohibiting gay marriage on the grounds that they involve gender discrimination—a woman is not permitted to marry a woman, a man may not marry a man.

Some will object that this is preposterous: the current law treats men and women equally in requiring both to marry someone of the other gender. But, by that line of reasoning, as Mr. Eskridge points out, we could defend anti-miscegenation laws: for all these do is require both whites and blacks to marry within their “races.” The Supreme Court has already rejected this line of argument in Loving v. Virginia. They did so because it was clear that the law was rooted in anti-black prejudice.

The hostility that animates anti-gay discrimination when it comes to marriage is not, it is true, against women: it is not, in that sense, gender discrimination. But it is an equally irrational prejudice aimed at gay people; and there is a constitutional requirement that discrimination on the basis of gender be reasonable. Both the equal protection clause of the Constitution and the due process clause provide the basis, therefore, for a claim that if gay marriage is to be disallowed, there must be a compelling state interest for doing so.

In the case now before the Hawaii Supreme Court, the state has offered a tripartite defense of its refusal to recognize the relationship of Ninia Baehr and Genora Dancel. It claims, first, that restricting gay marriage serves the state’s interest in fostering procreation; second, that Hawaii has a compelling reason to avoid expressing approval of non-heterosexual sex; and third—rather comically—that the resulting rush of gay people to Hawaii would distort job and housing markets there.

The constitutional protections for contraception (in the Griswold case) and for abortion (in Roe v. Wade) suggest that privacy rights take precedence over the state’s declared interest in procreation in at least some cases. But it is difficult, in any case, to see exactly what factual assumptions Hawaii’s first argument requires. There are three classes of persons who might procreate who are each differently affected by current law: heterosexuals, bisexuals, and homosexuals. It is hard to see why straight people should stop having children because gay people get the right to marry; gay people are surely more likely to have or raise children once their relationships are given recognition. So that leaves bisexuals. The absence of gay marriage, it could be argued, drives bisexuals into straight marriages, which are procreative, and away from gay ones, which are not. But is the state of Hawaii or any other state dependent for the replacement of its population on the difference between the numbers of children who would be born this way and the numbers who would be raised by gay couples? It is not an argument I would want to have to make before a reasonable judge.

But there is no guarantee that the judges will be reasonable. The record of the Supreme Court, for example, in thinking about homosexuality—a record that reaches its recent nadir in Bowers v. Hardwick, a decision later regretted by Supreme Court Justice Powell, whose vote gave it a bare majority—has been erratic at best. (In that case, the court held that homosexuals had no right to private consensual sex.) On May 20 the Court finally handed gay rights advocates a modest victory, striking down, by six to three, in Romer v. Evans, Amendment 2 to the Colorado Constitution, which had aimed to deny to gay people the possibility of protection by anti-discrimination laws. But Justice Scalia, for the dissenting minority, recalled to his colleagues the decision in Bowers (which, though “not even mentioned in the Court’s opinion,” was, he said, “the case most relevant here”) and thus reminded us all how far it is from this decision to a constitutional recognition of any positive gay rights.11

Still, it is worth asking, finally, what would happen if the constitutional arguments for the recognition of gay marriage prevailed in court. First, some gay couples would get some entitlements they badly want: lesbian and gay parents could gain equal recognition; some inequities in tax and insurance law would be put right; gay people could insist on their rights to have a part in determining the medical treatments of their incompetent spouses.

Second, just as the abolition of capital punishment in Britain led public opinion away from support of the practice, and school desegregation helped lead some to greater racial understanding, so, no doubt, in the unlikely event of the Supreme Court’s declaring that the relationships of gay people deserved the protection of the Constitution, the opinion of the odd bigot might be shifted. But these effects would not bring on a revolution in attitudes. Nearly thirty years ago, in Loving v. Virginia, the Supreme Court declared interracial couples entitled to the Constitution’s protection; but we live in a country where there is still widespread hostility to interracial marriage.

American hostility to homosexuals is irrational and leads to much cruelty. But in a liberal society the Constitution’s role—the law’s role—in dealing with unreason and unkindness is circumscribed by the rights of citizens to think and say what they will, to associate with whom they will, to worship before whatever idols beguile them. While Mr. Sullivan misinterprets liberalism’s history and practice in America, he is right, in the end, to stress this fundamental truth.

May 23, 1996

This Issue

June 20, 1996