In his latest reply to our argument in the Harvard Journal of Law and Public Policy that marriage is the conjugal union of husband and wife, Kenji Yoshino presents a truncated and distorted version of our view. Nevertheless, his answers to our challenges to him and others who demand the redefinition of civil marriage force Yoshino into awkward moral and political positions—including one that seems directly at odds with a stance he has prominently taken.

Yoshino’s Flawed Defense

Our first reply challenged Yoshino to explain his own view of marriage, such that two men or two women could form what is truly a marital relationship. Yoshino: “I thought the answer would be intuitive: I want . . . marriage to widen to permit same-sex couples to enter it.” Translation: Yoshino wants marriage to be whatever it must be such that two men or two women could truly marry. But this is to dodge the crucial—and ultimately unavoidable—question: what is marriage? We had hoped Yoshino would offer what we had offered: a holistic defense of a view of marriage that accounts for marital norms that he wishes to retain, assuming that there are some (e.g., monogamy and sexual exclusivity).

But Yoshino sees his ad hoc, results-oriented approach as a virtue of his view because, he says, proponents of “trans-historical . . . definitions of marriage have often been time’s fools.” Since “we do not stand at the end of history today,” Yoshino thinks that “only time will reveal” what the moral ideals of “liberty, equality, and justice” require of our marriage law.

Start your day with Public Discourse

Sign up and get our daily essays sent straight to your inbox.

We reject this idea of history as a quasi-divine judge. We doubt that Yoshino himself believes that each generation is necessarily more enlightened than the previous one. Such a belief would play into the conservative caricature of progressivism’s alleged faith in the inevitability of moral progress. In any event, it is demonstrably false. Nor can the passage of time as such reveal new principles of justice or equality. History tells us what has happened, not what should happen. Though it might help us predict a policy’s effects on certain human goods, it cannot give us principles for evaluating those effects, or for determining the structure of those goods. But what we sought from Yoshino was his view of the normative structure (the defining norms) of the human good of marriage.

Finally, liberty, equality, and justice forbid imposing arbitrary norms. But the question of whether any norm (complementarity, permanence of commitment, monogamy) is essential to marriage and its public purposes, or irrelevant and therefore arbitrary, cannot be answered without a holistic view of the human good of marriage and the point of marriage policy. So to know what justice requires, Yoshino must first address the question that he resolutely refuses to answer, and to which no mere succession of historical events gives any hints: what is marriage?

All of this is clear from Yoshino’s only positive statement of what he thinks marriage does require: though he expects future demands for the recognition of multiple-partner unions, Yoshino would “currently . . . distinguish polygamous marriage primarily on the intuitive ground that one can give one’s full self to only one other person.” So Yoshino thinks that marriage requires a comprehensive union, and that this requires monogamy. We agree, as our article says explicitly. But history (in which monogamy is both observed and flouted) could not have yielded that conclusion. Only reasoning about what the human good requires of our natural and public institutions—moral and political philosophy—could. We have offered our reasons, according to which truly comprehensive union involves bodily union—and thus coitus, and thus sexual complementarity. Yoshino, who disagrees, refuses to make a counterproposal.

If all of this undermines even the one criterion for legal recognition (monogamy) that Yoshino embraces here, perhaps that is because he already rejected it elsewhere. In a 2006 statement entitled “Beyond Same-Sex Marriage,” some 300 self-described “lesbian, gay, bisexual, and transgender… and allied activists, scholars, educators, writers, artists, lawyers” and others declared their support for “legal recognition for a wide range of relationships, households and families,” including (among others) “committed, loving households in which there is more than one conjugal partner.” Kenji Yoshino was one of the signers.

Yoshino evidently thought then that fairness requires legally recognizing polyamorous sexual partnerships. Did he forget having publicly endorsed that idea? Did he change his mind about it? Would he square what he says now with what he said then by distinguishing legal recognition of multiple “conjugal partner” relationships from marital legal recognition, reserving the term “marriage” for partnerships of exactly two? (If so, on what basis would he withhold the term from multiple-partner households?) Or did his reply to us soften his position, and profess agnosticism about what justice requires of our law, because showing his cards would (a) cost his position broad support, and (b) vindicate our argument that eliminating sexual complementarity removes any principled grounds for monogamy?

If there are no principled grounds for marital norms, then it must be unjust to fail to recognize any relationships that are just as socially valuable as those that we do recognize. But in that case, as we argued against Koppelman, the reasons to exclude polyamorous unions grow thin indeed:

The social costs of recognizing polyamorous partnerships might include, say, increased administrative burdens for the state. But the benefits would presumably include spousal privileges, inheritance and hospital visitation rights, and in general more practical assistance to, and social acceptance of, the relationships that Americans in an estimated 500,000 polyamorous households find most personally desirable. The stigma against such people and their children would be weaker. They would feel less pressure to hide their romantic inclinations and lifestyle choices. Their economic situation could well be improved.

If it’s an unsettled question whether justice requires recognizing such unions, can we really put off answering it until the possible victims of injustice have clamored long enough? Should we really join Yoshino’s rejection of wholesale thinking about marriage, and remain content to “test such intuitions [about polyamory] if and when such debates become live national controversies”?

So much for the implausibility of what Yoshino (perhaps looking backward as well as forward) labels his “current” position: we can also show that his criticisms of our view miss their mark.

Yoshino’s Flawed Critique

Yoshino imputes to us what he labels “the common procreation argument” about marriage, which he thinks cannot account for the validity or value of marriages that do not produce children. But we denied that actual procreation was necessary for marriage, and defended as philosophically sound the historic law of marriage that has long regarded infertility as no impediment to matrimony. For marriage is no mere means to procreation, but valuable in itself. That is perfectly consistent with holding, as we do, that the distinctive contours of marriage are what they are in significant part because it is the kind of union that would be naturally fulfilled by having and rearing children together.

After all, any serious account must explain how marriage differs from other types of community—and make sense of the evident fact that the idea of marriage would never have been conceived if human beings did not reproduce sexually. The view that we defend and that our legal tradition long enshrined does both: Marriage, valuable in itself, is the kind of commitment inherently oriented to the bearing and rearing of children; it is naturally fulfilled by procreation. This orientation is related to the fact that marriage is uniquely embodied in the kind of act that is fulfilled by procreation: coitus. By coitus alone, a man and woman can be related much as the organs of a single individual are related—as parts coordinating together toward a biological good of the whole. So marriage is consummated in an act that creates in this sense a bodily union—an extension of two people’s union of hearts and minds along their bodily dimension, thus making marriage a uniquely comprehensive interpersonal union. (By contrast, friendships in general are unions of hearts and minds alone, and so are characteristically embodied in conversations and joint pursuits.) Finally, in view of its comprehensiveness and its orientation to children’s needs, only marriage inherently requires of its would-be participants pledges of permanence, exclusivity, and monogamy. (By contrast, friendships do not require a promise of permanence and are often enhanced, not betrayed, by openness to new members.)

Every single sentence about marriage in the previous paragraph applies equally to any man and woman who have made and consummated their marital commitment, regardless of fertility. After all, each such sentence is just as true of a couple on their wedding night as it is after the birth of a third child. By contrast, not one of these same sentences applies to two men, two women, partnerships of three or more, or by-design temporary or open unions. If Yoshino thinks that we offer no “principled ground” for the distinctions we make, perhaps that is because his inapt label for our view (“common procreation”) has clipped and obscured it.

Nor do we salvage the validity of childless marriages at the price of denigrating their value, as Yoshino also charges. That an orientation to procreation distinguishes marriage from other unions does not mean that procreation must be the most important aspect of a marriage, much less its sole point. Comprehensive union itself—of mind, heart and body; permanent and exclusive—is of great inherent value, and distinct from the value of general friendships (unions of hearts and minds), however deep and fulfilling in their own right. Hence infertile spouses realize an important value distinguishable in significant ways from that of other friendships.

Moreover, in agreeing that marriage is a comprehensive union of persons but denying that it includes true bodily union, Yoshino must be reducing the person to a center of consciousness and emotion, which just uses a body as an extrinsic (and thus subpersonal) instrument for achieving satisfactions or other goals. For reasons we and others have articulated in various writings, we believe that this is a serious philosophical error, one at the heart of much contemporary confusion about the meaning of sex and marriage. In truth, our bodies are integral aspects of us as human persons, so that no interpersonal union is comprehensive if it leaves out bodily union.

Note, too, that legally recognizing infertile opposite-sex unions does nothing to undermine opposite-sex parenting as a public ideal. Now Yoshino denies that opposite-sex parenting is ideal even as a rule. He points to the values of “liberty, equality, and justice.” But in light of these, do we not owe it to children to ensure that they are, wherever reasonably possible, reared by the mother and father who conceived them—that our policies privilege this arrangement as a norm?

In this connection, Yoshino mistakenly claims that we deny that adoptive parents are the real parents of the children entrusted to their loving care. The sentence in our article that Yoshino sees as “denigrating” adoption was not about, and did not mention, adoption. As its context makes clear, its (perhaps inartfully phrased) point was that every child’s having two biological parents is related to the ground for monogamy: for only two-person unions (specifically, those capable of coital consummation) can be of the procreative type—and only unions of the procreative type can be marriages. We further argued that each child would ideally know and love, and be known and loved by, her biological parents, in the security provided by their marital commitment and love for each other. We never denied that the best available approximation of this through adoption is at times necessary, and in such cases laudable. Those who adopt should enjoy the same rights, and be held to the same responsibilities, as non-adoptive parents, legally and socially. They are real parents. That as such does not make them part of a procreative type of community, or therefore capable of marrying—witness, say, an orphan being adopted by his single aunt or two cohabiting bachelor uncles.

Of course, the normative issue of which arrangements our policies should privilege as generally ideal for procreation cannot be resolved by descriptive social-scientific studies alone, but such studies would contribute importantly relevant information. What we need, however, are studies that meet the acknowledged gold standard of social-scientific research, by drawing on large, random, and representative samples observed longitudinally. But so far, none of the studies comparing children reared by same-sex couples to children reared by their married biological parents has these features (for reasons acknowledged in this literature review by a sociologist and Jonathan Rauch, a gay civil marriage proponent). Yet Yoshino treats the social science as settled.

The designers of currently available studies of same-sex parenting outcomes cannot be blamed for the unavailability to date of large, random, representative samples to track over time. But every parenting arrangement that has been examined in high-quality studies has consistently been shown less effective than parenting by married biological parents: this is true of single- and step-parenting as well as parenting by cohabiting couples. Studies also suggest that mothers and fathers foster—and their respective absences impede—children’s healthy development in different ways. It would therefore be surprising if same-sex and opposite-sex parenting were equally effective. But let the methodologically strong studies be done, and the chips fall where they may.

Finally, Yoshino implies that, despite our argument that the structure of marriage depends partly on its orientation to procreation, we advocate a view that crudely draws a circle around all opposite-sex couples (and only them). But we have consistently argued that sexual complementarity, while necessary, is not sufficient to make a marriage. For example, if a man and woman do not sincerely pledge monogamy and sexual exclusivity—norms that are (like sexual complementarity itself) connected to marriage’s orientation to procreation—then their partnership is not, as a moral matter, a true marriage; nor is a marriage complete unless it is sealed in the generative act. But none of this is—as Yoshino further suggests—a strike against our view. For it is evident that nothing in any principled way distinguishes even opposite-sex bonds without these features from the wide and varied spectrum of non-marital friendships or partnerships. (Of course, the law may ordinarily have reasons not to inquire into such things as the sincerity of spouses’ marital pledges. Similarly, even advocates of defining civil marriage as a loving romantic commitment would not want the state inquiring into a couple’s level of affection or intimacy before granting marriage licenses. And as lawyers and legal scholars know, none of this concern about invasive inquiries is unique to marriage law.)

In “What is Marriage?” we defended a coherent answer to the title question. We showed how that answer makes sense not only of the requirement of sexual complementarity, but also of other marital norms (pledges of monogamy, exclusivity, and permanence)—and of the historic legal practice under which marriages could be annulled or dissolved for non-consummation but not for infertility. We also challenged proponents of redefining civil marriage to defend marital norms (like monogamy) embraced by most on both sides. Professor Yoshino insists that whatever marriage is, it does not require sexual complementarity or bodily union in coital acts. Currently, he seems to think that it is in principle limited to two persons. But he has given no coherent account of how to square these positions. Nor, we believe, can he.