Everyone has blind spots. It is philosophy’s ambition to cure these by canceling them out, through dialogue and scrutiny of assumptions. But even academic philosophy has its dogmas. One current example is support for same-sex marriage: To question it is to be anathematized by those occupationally averse to anathemas.

So I was both pleased that Alex Worsnip reviewed my co-authored book What Is Marriage? Man and Woman: A Defense, and unsurprised that he misunderstood it. My former classmate in Oxford’s philosophy master’s (B.Phil.) program, Worsnip is sharp and serious about arguments, and consistently blind to arguments of certain sorts.

Though talented and well-trained, he makes claims clearly contradicted by a quick read of our short book. His objections are almost offhand. And he is mute on the most serious—we think decisive—challenges we pose to those who favor redefining marriage. From another pen on another topic, I doubt such a treatment would survive scrutiny by other philosophers—or even, perhaps, by Worsnip.

Our Argument in Brief

To orient readers, let me summarize the claims we defend in our book.

Marriage is a human good with its own structure, like knowledge or friendship. The present debate is not a debate about whom to let marry, but about what marriage (the human good that the law has reasons to track) really is. Two answers compete for legal enshrinement.

The first, driving the push for same-sex marriage, is that a certain emotional intimacy makes a marriage. But as our book shows, this answer can’t coherently distinguish marriage from companionship, an obviously broader category. So it gets marriage (the human good) wrong.

The second view of marriage begins from basics. Any voluntary form of community involves common action; it unites people toward common ends in the context of commitment. And in these respects, what sets marital community apart is its comprehensiveness: in (1) how it unites people, (2) what it unites them with respect to, and (3) how extensive a commitment it demands.

First, marriage unites people in their bodies as well as their minds. Just as your organs are one body by coordinating for the biological good of the whole (your survival), so a man and woman’s bodies unite by coordination (in sexual intercourse) for a biological good (reproduction) of the couple as a whole. No other activity makes of two people “one flesh.”

Second, as the act that makes marital love also makes new life, so marriage itself is uniquely enriched and extended by the bearing and rearing of children, and the wide sharing of family life.

Third, because of its comprehensiveness in both these senses, marriage alone requires comprehensive (permanent and exclusive) commitment, whatever the partners’ tastes.

The stability of marriage, so understood, best ensures that children will know the committed love of those whose union brought them forth. This gives them the best shot at becoming healthy and happy people, which affects every other social good. That is why every society with the merest ambition to thrive has socially regulated male-female sexual bonds: to shore up the stabilizing norms of marriage, on which social order rests.

If marriage is redefined (in law, and hence in public opinion and practice) as simple companionship for adult fulfillment, then, for reasons to be explained, it will be harder to live by its norms and urge them on others. And this will harm the social goods that hook society into regulating marriage in the first place.

Besides defending these claims, my coauthors and I answer the most common objections to the historic view of marriage. And we show how society can uphold that view without ignoring the needs, undermining the social dignity, or curbing the fulfillment of same-sex attracted people.


You can dispute our vision of marriage. What are indisputable—by any reader with ten minutes to spare—are the gross mistakes that Worsnip makes about what we argue.

First, he says that for us, unions can’t be marriages if they “do not involve procreation.” But we expressly deny this. Comprehensive union makes a marriage, we say, “with or without children.” The bodily union that completes marriage, we argue, involves “coordination toward” a bodily end, whose realization would “deepen the union but is not necessary for it.” These points, though hardly peripheral to our argument, fell in Worsnip’s blind spot.

Second, he reads us to argue that “homosexual relationships cannot be permanent.” Again, we expressly reject this, and even flag the point for readers: “Our point here is subtle and sometimes misunderstood,” we warn, before affirming in plain English that same-sex partners can “wish for, promise, and live out permanent sexual exclusivity.”

Third, Worsnip calls it a “central failing” that we “never justify” the idea that marriage has objective requirements. In fact we label the opposite view (“constructivism”), sketch it sympathetically, quote from one of its defenders, and offer four arguments against it.

Fourth, Worsnip thinks we conclude that the same-sex marriage “rights-demand doesn’t even have to be engaged.” In fact—as you will by now be unsurprised to learn—we spend nearly half the book engaging it, as expressed in six arguments (based on equality; interracial marriage; infertility; and same-sex partners’ practical needs, dignitary interests, and personal fulfillment).

It’s as if Worsnip had lost sections and whole chapters of the book, but pressed on to criticize what he had expected us to say. In philosophy as in driving, negligence makes us think we already know what’s in our blind spots. I know Worsnip to be a careful thinker, but this is hardly a careful review.

Main Objections

These errors aside, let me address, in turn, Worsnip’s three worries about our argument: (1) that it benightedly assumes that marriage has an essence; (2) that it doesn’t explain why any such essence would matter for law; and (3) that it implies that gay sex should be banned.

Does Marriage Have Essential Features?

Worsnip seems baffled by the idea that marriage, the human good, might have essential features.

So let’s back up. Many ethical systems (what Derek Parfit calls “objective-list” theories) begin from several bedrock values or human goods: conditions or activities that in themselves make us better off. Such goods—like health, knowledge, or friendship—can be realized in many ways but have some core requirements.

In the book we cite friendship. For all its cultural variety, it has an objective core, fixed by our social nature: mutually acknowledged good will and cooperation. Without that, two people’s connection simply lacks the value (and special duties) of friendship. To overlook this is to err about a human good, not just the label for a construct.

So too, we argue, for marriage. For all the variety of its cultural supports, shaped by shifting historical demands, marriage as a human good has an objective core, fixed by the demands of our nature as sexual-reproductive beings; to deviate from it is to miss a crucial part of a basic human good. It’s this idea that Worsnip rejects.

First, he says, essences make no sense on a “disenchanted, scientific” or “enlightened” worldview. Would Worsnip then exclude anything inaccessible to science? (In that case, sure, essences would have dim prospects, but so would logical and moral truths, and Worsnip’s and my careers.) No, Worsnip is here just making a barely veiled appeal to authority.

So let me respond in kind. Most of Worsnip’s and my colleagues—contemporary analytic philosophers—agree that things can have essential properties. Theories of essence have been developed by thinkers like Kit Fine of NYU, Saul Kripke of Princeton and CUNY, Hilary Putnam of Harvard, David Wiggins of Oxford, Stephen Yablo of MIT, and Alvin Plantinga of Notre Dame. This doesn’t prove that there are essences (discussed here, more in the book). But it does defeat Worsnip’s bald appeal to “scientific,” “enlightened” belief.

Nor does accepting essences require theology, as Worsnip suggests. It’s true that marriage law is whatever human beings make it; the human good of marriage is what has objective features. But accepting the latter only requires believing in a human nature that isn’t entirely constructed.

Moreover, many same-sex marriage proponents implicitly deny that marriage is just a construct—when they assume a natural right to marriage that unduly narrow laws might violate.

Finally, history reveals a remarkable confluence of different streams of thought on marriage, which can’t be traced just to religion, bigotry, or policy goals. Absent other explanations, doesn’t this support seeing marriage as rather like friendship, physical vitality, or artistic appreciation—a general and inherent good, whose contours can shape life and thought across continents and millennia because they’re knowable by reasoned reflection? Wouldn’t something similar be the natural explanation for such robust parallels of judgments on, say, the special value and criteria of theoretical knowledge?

Our book develops this at length. But consider that several ancient Greek and Roman thinkers variously implied, and some explicitly defended, a morally significant difference between (a) the mind-body union embodied in coitus and fit for family life, and (b) all other close bonds; and the idea that the first category could include infertile but not same-sex or group bonds.

These thinkers wrote without a Hebrew prophet or parchment in sight, and centuries before St. Paul struck up a correspondence with the Christians of Rome. They harbored no group animus, having lived in cultures that celebrated homoerotic activity, and long before the modern idea that gay identity defined a class of people. Finally, they were proposing a philosophy, not a policy to control property. Given their parallels with the view of marriage in Genesis and other (unconnected) cultural sources, can we really see their work as mere chronicles of a cultural construct? Or might they have latched on to something deeper—the structure of a bedrock human good?

But all this is lost in Worsnip’s blind spot. All he writes in “reply” is that if there are social constructs, “surely an institution like marriage is a prime candidate.” If he isn’t just mistaking our claim (for one about an “essence” of marriage law), this just amounts to an argument from “surely!” In other words, against our several points on this key question, Worsnip offers not a strong counterargument, nor a weak one, but none at all.

Does the Essence of Marriage Matter for Law?

Worsnip next objects that even if marriage had essential features, nothing would stop us from enacting a policy recognizing some other set of bonds, for other purposes.

We agree; not every legal accommodation would obscure the features of marriage. Again, our book discusses the practical needs of those sharing a home, whatever their relationship status. Though private arrangements often suffice, we consider when a public package might be needed. What we deny—and what Worsnip says nothing to defend—is that these should be limited to sexual partners. Why give them to two men in love but not, say, adult brothers who are (though not sexually involved) equally committed to sharing a home, equally bonded by the memories and sympathies that justify hospital visitation rights?

We can meet cohabitants’ practical needs without redefining civil marriage. Indeed, it would be unjust to limit such practical accommodations to pairs, in a sexual relationship, who aren’t closely related, and so on. But when pressed, that is what proponents of redefinition insist on.

As we argue in our book, social harms would result from enshrining in law and social practice the idea that emotional union is what sets marriage apart, and that mothers and fathers are interchangeable. The more these ideas took root, the weaker the internal motivations and social pressures would be for people to stick with a marriage once emotion fades, or to maintain exclusivity at the cost of emotional satisfaction. In the book, we cite scores of mainstream same-sex marriage leaders who agree, but embrace these results as liberating.

But think of the effects of no-fault divorce, which didn’t just ease the exit from high-conflict marriages but eroded permanence more broadly. A generation later, the results are in: It harmed children’s development, and thus (to some extent) every aspect of the common good. A more formal and final orientation of marriage toward adult emotional satisfaction would deepen or entrench these harms.

Worsnip announces that he disagrees, but without explanation. Instead he abruptly (perhaps unintentionally) changes the subject, to a logical point: One can, he says, support same-sex marriage without supporting polygamy (one man, several legal wives).

We agree, and (again!) say so in the book: Polygamy, unlike other policies, “would undermine women’s social and political equality.” What we challenged critics to produce is a single (non-circular) argument for recognizing same-sex bonds but not polyamory (group sexual units of whatever gender distribution).

Worsnip would answer using only cost-benefit analysis. But as our book also asks, what social cost is specific to recognizing polyamory? Does it justify violating, for an estimated 500,000 American polyamorist households, what for others is cited as a basic human right? Recognition might de-stigmatize polyamorists and their children, and give them the economic advantages enjoyed by monogamists. What is Worsnip’s utilitarian basis for denying these benefits unless people forfeit the partner-variety that some cite as crucial to their emotional fulfillment?

To judge by Worsnip’s silence on these many points (despite his confident overall dismissals of our book), his blind spot must have been wide indeed.

What About Freedom?

Finally, Worsnip thinks that consistency requires us to embrace criminal bans on gay sex.

But to get from “this sex act can’t consummate a marriage” to “ban it!”, you have to leap across a logical ocean (but one small enough, apparently, to fit in Worsnip’s blind spot):

1.  Assuming the act isn’t marital, you have to show its immorality (which depends on how your moral theory deals with tradeoffs).
2.  Assuming it’s immoral, you have to show its social harms (a political-theory question).
3.  Assuming it’s socially harmful, you have to argue that banning it would work (an empirical question).
4.  Assuming it’s non-marital, immoral, socially harmful, and effectively banned, you have to show that banning it is worth the costs (a moral, political, and empirical question).

So no, I don’t think we should ban consensual gay sex, polyamory, or other victimless sex acts. Worsnip’s argument that this makes me inconsistent is (to borrow a term from his review) nonsense. Refusing to redefine civil marriage bans nothing. Bigamy is banned; jail awaits its practitioners. But in all fifty states, two men or two women are and will remain free to share a bed and a life. This debate isn’t about curbing their freedom. It’s about which of the endless variety of deep bonds the state should regulate as marriages, and why.

At Oxford, I learned much from my classmates, including Worsnip, in Merton Street seminar rooms and pubs just off the High. And I expect to learn more from him in the future. But all I gained from his response to my book was an unfortunate new sense to the term “blind review.”