Same-Sex Marriage in a Liberal State: Why there is Nowhere Between Rejection and Recognition Peter Brian Barry Associate Professor of Philosophy Saginaw Valley State University pbbarry@svsu.edu draft: 4/27/2013 No small number of gay scholars, queer theorists, and advocates of gay and lesbian persons have denied that there is much comfort or hope to be found in the liberal tradition. No comfort, because no small number of gay and lesbian persons and their allies insist that liberal states have a long, sordid, and unfriendly history to large groups of people who have been disenfranchised, often by the liberal state that was supposed to protect them, based on their race, ethnicity, age, class, body, or sexual orientation. No hope, because at least some proponents of liberalism insist that, in a liberal state, judicial conservatism must reign such that courts and judges should, other things being equal, refrain from expanding current rights and liberties. This sort of commitment is especially problematic partly because the current distribution of rights and liberties in liberal states that do not recognize same-sex marriage works dramatically to the disfavor of homosexual persons (along with bisexual and transgendered persons, among others). Further, at least some citizens of some liberal states cannot realistically hope that things will change without intervention by the courts. While same-sex marriage was effectively recognized in Canada when Canadian parliament enacted the Civil Marriage Act in 2005, things are rather different in the United States. As I write, only a handful of United States legally recognize same-sex marriage as a result of a public referendum or legislative action and there is simply no realistic hope for relief in the overwhelming majority of states that do legally prohibit same-sex marriage absent judicial intervention. Similarly, as I write, the Defense of Marriage Act which defines marriage as the union of one man and one woman for purposes of federal law remains law and it remains unlikely that there is sufficient support in the United States Congress to overturn it. For all that, I do not think that advocates of SSM should find contemporary liberalism to be a philosophically inhospitable place. However slow liberal states have been to enfranchise homosexual persons, things can change. And since I doubt that liberal states really are committed to endorsing judicial conservatism, I doubt that liberal courts and judges must refrain from expanding rights and benefits that would protect and enable homosexual persons in liberal states. More problematic, however is the claim advanced by some philosophers that recognition of same-sex marriage (henceforth: SSM) conflicts with fundamental commitments of liberalism—that civil recognition, and not the prohibition, of SSM is illiberal. For example, Jeff Jordan declares that: Liberalism does not require legal recognition of same-sex marriage. In fact, it cannot, since legal recognition is incompatible with… principles of liberalism… The push for legal recognition of same-sex marriage, though often packaged as being motivated or required by liberal reasons, is in fact illiberal. Jeff Jordan, “Contra Same-Sex Marriage,” in Taking Sides: Clashing Views on Moral Issues, 11th edition, Stephen Satris, ed. (Dubuque, IA: McGraw Hill, 2008) p. 123. See also Francis Beckwith, “Legal Neutrality and Same-Sex Marriage,” Philosophia Christi, vol. 7, no. 1 (2005), pp. 19-25; Robert George, “’Same-Sex Marriage’ and ‘Moral Neutrality’,” in his The Clash of Orthodoxies: Law, Religion, and Morality in Crisis (ISI Books, 2001); Jordan, “Is it Wrong to Discriminate on the Basis of Homosexuality?”, “Why Homosexuality is Abnormal,” Monist, vol. 67 (1984), pp. 251-83, reprinted in Hugh LaFollette, Ethics in Practice, 3rd ed, pp.290-97 and “Homosexuality, Abnormality, and Civil Rights,” Public Affairs Quarterly, vol. 10, no. 1 (Jan., 1996), pp. 31-48; Matthew O’Brien, “Why Liberal Neutrality Prohibits Same-Sex Marriage: Rawls, Political Liberalism, and the Family,” British Journal of American Legal Studies 1 (2012), pp. 411-66. Susan Shell argues that the “liberal sword cuts both ways,” insisting that: American citizens should not have the sectarian beliefs of gay-marriage advocates imposed on them unwillingly… The requirement that homosexual attachments be publicly recognized as no different from, and equally necessary to society as, heterosexual attachments is a fundamentally illiberal demand. Susan M. Shell "The Liberal Case Against Gay Marriage,” Public Interest (Summer 2004). If this is right then committed liberals should not just be disinterested in advocacy of SSM; they should be adamantly opposed to it. I contend that this liberal opposition to recognizing SSM is seriously misguided and that the arguments offered to demonstrate that the legal recognition of SSM is illiberal are seriously flawed, or so I have argued. Omitted for consideration. Further, I contend, faithful liberals should not rest content with compromises that purport to provide same-sex couples with some pale imitation of civil marriage in lieu of granting them the right to civil marriage simpliciter. This contention matters because David Gilboa has recently offered a thoughtful discussion of SSM and attempts to stake out a compromise position between those who are “completely in favor of SSM or completely against it.” David Gilboa, “Same-Sex Marriage in a Liberal Democracy: Between Rejection and Recognition,” Public Affairs Quarterly 23:3 (July 2009), pp. 245-60. Gilboa argues that “the best possible solution to the problem of same-sex marriage” falls somewhere between unqualified rejection and unqualified recognition of SSM. On his view: Same-sex marriage ceremonies ought to be lawful, hence tolerated, but same-sex marriage need not be recognized; the benefits and protections of the marriage license may be denied to same-sex couples without discriminating against them. Ibid., p. 245. But Gilboa’s solution is no solution at all and he seeks middle ground where there is none. A consistent liberal theorist will reject Gilboa’s arguments for his compromise position and, for related reasons, will demand full legal recognition of SSM if marriage is recognized at all. My hope is that a successful critique of Gilboa’s argument will also serve as a plausible defense of the thesis that liberal states must legally recognize SSM if they recognize marriage at all. In section I, I critique Gilboa’s purported compromise. In section II, I offer what I take to be the strongest liberal argument in favor of recognition of SSM while in section III I defend a crucial premise of that argument—that there is nothing that a liberal would regard as a rational basis for SSM prohibitions. In section IV, I reject Gilboa’s claim that prohibiting SSM need not deny any equal access to public services and resources and thus not amount to unjust discrimination. Section I: The Purported Compromise Clearly, Gilboa sees some conceptual cleavage between “the right to marry someone of the same sex on one hand, and the right to obtain public recognition of such marriage, on the other hand” since he distinguishes the two. The difference seems to involve the familiar legal benefits and protections associated with civil marriage insofar as Gilboa thinks that a liberal state may deny same-sex couples those familiar benefits and protections, although he insists that “SSM ceremonies ought to be lawful” and “SSM ought to be legally protected,” and so forth. Ibid. It is worth wondering, then, just what the right to marry really amounts to given Gilboa’s compromise position if it does not require conferring the familiar benefits and legal protections of civil marriage. Whatever else the right to marry secures, Gilboa thinks it secures the right to perform and participate in “the marriage ceremony.” In the following passage, he illuminates just what the marriage ceremony is supposed to consist in: In this paper, the debate about the essence of marriage is therefore circumvented in favor of focusing on the marriage ceremony as a service that is provided in either a religious or a civil (nonreligious) setting… Viewing the marriage ceremony as an occasion for delivering an essentially private service, we may focus on the actions performed at the ceremony by the participants. To make a long story short, after the people who are invited to the ceremony assemble, an official declares that two persons among the assembled are married. Ibid., p. 246. Making it illegal to perform or participate in SSM ceremonies, so described, would require violating some right of the participants involved that liberals regard as fundamental—say, the right of peaceful assembly or to freedom of speech in general. So understood, the right to marry is a derivative right and ultimately reducible to other rights cherished and protected by liberal thinkers. But that is all that the right to marry a person of the same-sex amounts to: the right to perform and participate in SSM ceremonies. A liberal state need not, on Gilboa’s position, legally recognize SSM and thereby confer the familiar legal benefits and protections associated with civil marriage on same-sex couples that participate in such ceremonies. Gilboa’s position is arguably a compromise position insofar as he does not advocate that liberal states altogether reject SSM but he does not commend fully recognizing them either. Gilboa is not the only philosopher inclined to suppose that a genuine commitment to liberalism requires taking up some sort of compromise position with respect to SSM. Jeff Jordan supposes that when faced with a “public dilemma”—that is, a matter of public policy in which, at most, one conflicting party can be correct—liberal states must stake out some “middle ground” that “entails that there are no absolute victors and no absolute losers,” a perfectly ordinary sense of the term ‘compromise.’ Jeff Jordan, “Is it Wrong to Discriminate on the Basis of Homosexuality?”, Journal of Social Philosophy 26:2 (1995), p. 43. More precisely, Jordan thinks that liberal states must seek such resolution so long as there is no overriding reason to prefer a resolution that makes one party in the dispute an absolute winner, a reason he thinks is not in play in debates about SSM. To resolve the public dilemma concerning SSM, Jordan proposes that a liberal state should refuse to sanction SSM but tolerate private consensual sexual acts performed by persons of the same sex, presumably by declining to criminalize them. Ibid., p. 46. But if the relevant public dilemma concerns SSM, then Jordan’s resolution is no compromise at all since it does result in absolute winners and losers: proponents of SSM would not get what they want with respect to SSM—that is, its legal recognition—while opponents of SSM would get everything they want with respect to SSM. David Boonin, “Same-Sex Marriage and the Argument from Public Disagreement,” Journal of Social Philosophy 30:2 (1999), pp. 251-59. But Jordan’s proposal is no compromise for another reason: it only affords same-sex couples some right that they already have and only affords them legal protection that they are already entitled to. A long line of liberal thinkers from Mill to Feinberg have supposed that private sexual behavior between consenting adults is beyond legal prohibition and that consenting adults can reasonably expect a liberal state to protect their rights to engage in said conduct. Thus, Jordan’s “compromise” is no compromise in the ordinary sense of the term, insofar as opponents of SSM have to give up nothing that they cherish while proponents of SSM gain nothing they did not already possess. In fairness to Jordan, at the time he wrote his paper, the Supreme Court of the United States had declared that there is no constitutionally protected right to engage in sodomy and that states could criminalize even consensual acts of sodomy performed in private: Bowers v. Hardwick, 478 U.S. 186 (1986). This decision was not overturned until 2003 in Lawrence v. Texas, 539 U.S. 558 (2003). That said Bowers is widely regarded as a mistake—as evidenced by the fact it was overturned less than 20 years later—and could only be embraced by a proponent of a substantially illiberal position that I discuss below. Gilboa’s position fails as a compromise for the same reasons that Jordan’s position fails. First, Gilboa’s proposal to permit SSM ceremonies but refuse to recognize SSM really does make one party in the debate an absolute loser, insofar as the relevant debate is a debate about marriage. Note that Gilboa’s marriage ceremony differs from the familiar sort of marriage ceremony in which some legal official performs the illocutionary act of, well, marrying. On the familiar view of things, in uttering certain words in certain circumstances, a particular legal official is making it the case that some couple that was not previously to be treated as being legally wed is now to be treated as being legally wed. On this view, in uttering certain words in certain circumstances, a particular legal official is making it the case that this couple is now eligible to claim certain legal benefits, guaranteed certain legal protections, and subject to certain legal requirements. Of course, the felicity conditions necessary for performing the relevant illocution—that is, marrying—might fail to obtain; a legal official who declares “I now pronounce you man and wife” in the familiar sort of circumstances but who had his legal powers suspended by the state that morning is not marrying, whatever else he is doing. The point is simply this: either the legal official who “declares that two persons among the assembled are married” performs the familiar illocutionary act or not. Insofar as Gilboa denies that liberal states must recognize SSM, he must deny that the relevant legal official really is performing the familiar illocutionary act in “the marriage ceremony.” But that is surely what at least very many proponents of SSM want. As one proponent of SSM declares: “we don’t want ‘gay marriage,’ we want marriage—the same freedom to marry, with the same duties, dignity, security, and expression of love and equality as our non-gay brothers and sisters have. Evan Wolfson, Why Marriage Matters: American, Equality, and Gay People’s Right to Mary (New York: Simon and Schuster, 2004), p. 17. If proponents of SSM really do want, not just SSM ceremonies, but SSM ceremonies constituted by this familiar illocutionary act that result in marriage with all of its consequent and constitutive legal trimmings, then Gilboa’s position really does deny proponents of SSM everything that they want: it makes them absolute losers. Now, if Gilboa’s legal officials are not performing the familiar illocutionary act of marrying during “the marriage ceremony” they might still successfully perform some other sort of speech act—say, some perlocutionary act. Perhaps by announcing that a couple is wed, the assembled audience would be caused to have the belief (albeit a false belief) that the couple really is married. Even if Gilboa’s compromise position does not enable and protect the right to perform illocutionary acts like the kind discussed in the previous paragraph, it would enable and protect the right to engage in this sort of perlocutionary act. And surely the liberal tradition does require protecting the right to engage in this sort of perlocution; at least, making it illegal to assemble and participate in marriage ceremonies would clearly violate rights of free speech and free association that are rightly well established in liberal states. But then Gilboa’s position only affords same-sex couples with something that they already could reasonably expect and expect to have protected: the right to assemble and publicly declare that two people among the assembled are married. Again, this is hardly a compromise in the familiar sense of the term: here too, opponents of SSM have to give up nothing that they cherish while proponents of SSM gain nothing they did not already possess. If all this is correct, then Gilboa is not really staking out some compromise position, some middle ground between rejection and recognition of SSM. Rather, he is better understood as arguing that liberal states are not obligated to recognize SSM—that they can legitimately refuse to extend the legal benefits and rights and protections associated with civil marriage to same-sex couples. This is especially evident given his concern with demonstrating that refusing to recognize SSM need not amount to unjust discrimination in violation of a liberal’s commitment to equality. I shall have more to say about Gilboa’s discussion of SSM and equality below, but initially I want to suggest that Gilboa misunderstands the equal protection claim that (at least some) liberal proponents of SSM mean to advance. Section II: Equal Protection and Rational Basis Review Liberalism’s commitment to equality does not require that everyone is treated exactly the same; if it did, then our seemingly innocuous treatment of felons and children would be unacceptable, given we do treat them differently than non-felons and adults, respectively. This observation is necessary given the apparent pervasiveness of the objection that SSM prohibitions are inconsistent with equality before the law because only homosexuals are legally prohibited from marrying someone of the sex that they find attractive as a result of SSM prohibitions. Gilboa ascribes this view to Jason A. Beyer, “Public Dilemmas and Gay Marriage: Contra Jordan,” Journal of Social Philosophy 22: 1 (2002), pp. 12-13 and Ralph Wedgwood, “The Fundamental Argument for Same-Sex Marriage,” Journal of Political Philosophy 7:3 (1999), p. 229. Gilboa responds by making the following general point: One problem with this argument is that the inequality to which the argument alludes is a consequence of the uniqueness of the gay sexual preference (in comparison with the heterosexual preference). When gays are excluded from marrying a member of the sex they prefer, this is because gays prefer to marry a member of the same sex, while heterosexuals prefer to marry a member of the opposite sex. However, this sort of inequality does not seem to be inherently wrong, for it is not always wrong for the law to exclude a group of people as long as the exclusion is a consequence of the uniqueness of the preferences of the group. Gilboa, “Same-Sex Marriage in a Liberal Democracy,” pp. 246-7. Later, Gilboa argues that the typical appeal to equal protection fails to establish the wrongness of SSM prohibitions because SSM prohibitions do not fail to provide access to public resources, an argument I consider in Section IV. However, liberal opponents of SSM who object that SSM prohibitions are inconsistent with a liberal commitment to equality may a very different sort of argument in mind. In American constitutional law, one way for legislation to violate equal protection is for it to illicitly utilize a suspect classification—say, race or religion or national origin. Legislation that utilizes a suspect classification must pass strict scrutiny: the state must demonstrate that use of the suspect classification is part of a necessary and narrowly tailored means for promoting some compelling state interest. If the state fails to make its case, that legislation is struck down as violating constitutional guarantees of equal protection. But there is another way that legislation can run afoul of equal protection: it can fail to survive rational basis review. For legislation to survive rational basis review, the state must merely show that the legislation bears a rational relationship to some legitimate state interest. Rational basis review is a much weaker test of constitutional legitimacy and while not much legislation survives strict scrutiny, most survives rational basis review. But not always. For example, the Supreme Court of the United States struck down as unconstitutional a Colorado law that would have prohibited all future legislation protecting homosexuals from discrimination on the grounds that the Colorado law “lacks a rational relationship to legitimate state interests.” Romer v. Evans 517 U.S. 620 (1996). Thus, legislation can violate Constitutional guarantees of equal protection either by illicitly utilizing some suspect classification or by lacking a rational basis. Similarly, one way that legislation can run afoul of liberalism’s commitment to equality by denying a legitimate claim to equal access to public resources because of an individual’s membership in a protected group, as Gilboa suggests. Gilboa, “Same-Sex Marriage in a Liberal Democracy,” p. 247. But there is another way that legislation can run afoul of liberalism’s commitment to equality: it can fail to bear a rational relationship to something that a liberal would regard as a legitimate state interest. It is beyond the scope of this present paper to make the case that SSM prohibitions must fail a liberal version of rational basis review. Footnote omitted for consideration. It will do for present purposes to note that Gilboa himself seems committed to supposing that SSM prohibitions will fail a liberal version of rational basis review. To demonstrate that this is so, consider the plausible and influential characterization of liberalism defended by Joel Feinberg. Feinberg follows a long line of liberal thinkers in supposing that only certain purported justifications of legislation that limit a person’s liberty are at all plausible. Liberals will recognize something like Mill’s “harm principle” as a plausible liberty-limiting principle, such that liberals will allow that if limiting A’s liberty will prevent (or sufficiently reduce the likelihood of) B’s being harmed—that is, having her interests wrongly set back—then there is a good reason in favor of legislation that will limit A’s liberty. Joel Feinberg, Harm to Others (Oxford: Oxford University Press, 1984). Similarly, liberals will (perhaps tentatively) allow that if limiting A’s liberty will prevent (or sufficiently reduce the likelihood) of B’s being seriously offended then there is a good reason in favor of legislation that will limit A’s liberty. Joel Feinberg, Offense to Others (Oxford: Oxford University Press, 1985). But liberals will recognize no other liberty-limiting principles as legitimate. Thus, “the liberal position” is neatly captured as follows: limiting A’s liberty is permissible only if doing so would either prevent B from being harmed or offended, but otherwise not. As Feinberg puts things: “The harm and offense principles, duly clarified and qualified, between them exhaust the class of good reasons for criminal prohibitions.” Ibid., p. xiii. But what is true of criminal prohibitions should be true of liberty-limiting legislation quite generally. The liberal position puts substantive limits on what sort of interests a liberal will regard as legitimate ones: liberally legitimate state interests include the prevention of harm and offense and interests crucially tied up with the prevention of harm and offense. But that is about all. So, if it can be shown that SSM prohibitions are not rationally related to anything akin to the prevention of harm or offense, it follows that SSM prohibitions will not survive a liberal version of rational basis review. Interestingly, by Gilboa’s own lights SSM prohibitions are bound to lack anything that a liberal would regard as a rational basis. In a footnote Gilboa concedes that “There is insufficient evidence… to show that marrying someone of the same sex is, in principle, imminently and remarkably harmful to any person.” Gilboa, “Same-Sex Marriage in a Liberal Democracy,” p. 258. That concession effectively amounts to admitting that SSM prohibitions cannot be justified by appeal to the harm principle. And it is simply unclear how the legal recognition of SSM could offend anyone, properly understood. For to be offended, in the relevant sense, is not simply to be made to suffer disliked mental states but to be made to suffer disliked mental states wrongly—say, in violation of one’s privacy rights. Feinberg, ibid., pp. 22-4. It is at least difficult to explain how two consenting adults of the same gender could violate a third party’s privacy rights simply by marrying one another. But that means that SSM prohibitions cannot be justified by appeal to the offense principle either. So, the consistent liberal will insist that there is simply no legitimate state interest rationally related to SSM prohibitions. The liberal concerned with the offense principle might allow that there can be restrictions on the manner and mode in which SSM is administered and provided. Just as liberal states can take measures to ensure that my consumption of pornography does not violate the privacy rights of those who wish to avoid the stuff, a liberal state could make sure that, say, the performance of SSM ceremonies does not violate the rights of those who wish to be left alone. But these would surely be fairly modest restrictions that fall well short of outright prohibition. So, whether or not SSM prohibitions discriminate based on something like a suspect classification or membership in a protected group, the liberal has a powerful argument at hand that they are illegitimate. Gilboa’s discussion suggests that he would contend that there are legitimate state interests served by SSM prohibitions—at least, he argues that there are genuine nondiscriminatory reasons for refusing to recognize SSM. I consider this line of argument in the following section. Section III: A Rational Basis for SSM Prohibitions? Gilboa notes that even if some legislation has a “disparate impact” on some particular group—and surely SSM prohibitions do have a disparate impact homosexual persons and same-sex couples—that disparate impact might be justifiable so long as it results only from nondiscriminatory reasons. Gilboa, “Same-Sex Marriage in a Liberal Democracy,” pp. 250-1. And he suggests two ostensibly nondiscriminatory reasons justifying the disparate impact that emerges from prohibiting SSM: the belief that homosexual actions are morally wrong and a concern for the social consequences of publicly recognizing SSM. Ibid., p. 251. Given I have insisted that there is no liberally legitimate rational basis for SSM prohibitions, it is necessary to consider these proposed nondiscriminatory reasons. Consider first Gilboa’s suggestion that “moral disapprobation of homosexual sex is naturally a reason for people to oppose recognition of SSM.” Ibid. I confess that it is not entirely clear to me why moral disapprobation of private sexual behavior is a reason to oppose the public legal recognition of civil marriage. Admittedly, the thought that a marriage is legally voidable if it is not consummated is not entirely unfamiliar to liberal states, but this consummation requirement is hardly essential to marriage law and perhaps should be shed by a liberal state: serious enforcement of the consummation requirement would require some fairly egregious intrusion into the private sexual lives of married partners after all. Perhaps the thought is that moral disapproval of homosexuality, quite generally, would naturally lead someone to oppose the legal recognition of SSM since it is homosexual persons who will most likely, if not exclusively, choose to exercise the right to marry someone of the same sex. But refusing to recognize SSM out of genuine moral disapprobation of either homosexual sex or homosexuality quite generally is, apparently, not the same thing as targeting homosexuals for exclusion due to their membership in a protected group and only the latter depends on discriminatory reasons. Ibid. To be clear, Gilboa does not rest his argument on the supposition that homosexual actions are, in fact, morally wrong. Instead, his position is that the failure to recognize SSM given the (perhaps false) belief that homosexual acts are morally wrong need not be based on discriminatory reasons. Further, he contends that even if the moral belief that constitutes the basis for moral disapprobation of homosexuality is false, it would still not follow that the failure to recognize SSM is based on discriminatory reasons. After all, many jurisdictions prohibit the sale or possession of violent pornography given the belief that there is some causal connection between consumption of violent porn and violence towards women, but “one could not argue that these ordinances… discriminated against consumers of pornography” if that belief turns out to be false. Ibid. By analogy, SSM prohibitions need not discriminate even if they are based upon false beliefs. I suspect that Gilboa’s first defense of SSM prohibitions based on an appeal to moral disapprobation falls victim to a fatal analogy concerning race that he himself is aware of, but I return to that problem below. His defense of SSM prohibitions by appeal to moral disapprobation fails for at least two other reasons. First, the alleged wrongness of homosexuality or homosexual acts suffices to justify prohibiting SSM only if some version of legal moralism is true, according to which, broadly, it can be morally legitimate for the state to prohibit certain types of action that cause neither harm nor offense to anyone on the grounds that such actions constitute or cause evils of other kinds. See Joel Feinberg, Harmless Wrongdoing (New York: Oxford University Press, 1990), p. 3 for this way of putting things. Again, Gilboa’s argument does not turn on homosexuality or homosexual sex actually being wrong, but the moral disapprobation that he appeals to is a reason for prohibiting SSM apart from harm or offense. So, his argument does depend upon the veracity of some version of legal moralism. But any version of legal moralism will be rejected by the consistent liberal; again, according to the liberal position, neither the actual wrongness nor the believed wrongness of some action (apart from harm or offense to others) constitutes a rational basis for its prohibition in a liberal state. Gilboa’s appeal to moral disapprobation in defense of SSM prohibitions never gets started. Second, Gilboa’s appeal to allegedly non-discriminatory prohibitions of violent pornography similarly fails to justify SSM prohibitions. For prohibitions of violent pornography, whether or not they are justified by an appeal to false beliefs, surely invoke some version of the harm principle: at least, the typical justification for prohibiting violent porn is that restricting men from consuming violent pornography prevent them from harming women. Since Gilboa is already on record as doubting that SSM prohibitions can be justified by appeal to the harm principle, the liberal legitimacy of anti-porn ordinances implies nothing about the legitimacy of SSM prohibitions. It is also worth wondering, I think, whether or not ordinances prohibiting the sale or consumption of violent pornography would be justifiable on liberal grounds if we learned that there is no causal connection between consumption of the stuff and violence towards women. After all, the skepticism of at least some liberal theorists about the legitimacy of ordinances prohibiting violent pornography is grounded in skepticism about the causal connection between consumption of the stuff and violence towards women. For one representative example, see Feinberg, Offense to Others, pp. 147-57. Consider next Gilboa’s suggestion that the fact that “some people believe that public recognition of SSM would have disastrous consequences” amounts to a nondiscriminatory reason for prohibiting SSM. Gilboa, “Same-Sex Marriage in a Liberal Democracy,” p. 253. What sort of disastrous consequences? The worry, apparently, is that recognizing SSM “may, in the course of events, invalidate and dismantle the set of rules adopted in our society to govern the institution of marriage as heterosexual monogamy” Ibid., p. 254. such that any number of new lifestyles and living arrangements will be available and attractive to many persons, perhaps “to the detriment of the values that the institution has historically helped to preserve.” Ibid., p. 255. Surely this is an argument that rather many opponents of SSM have offered, but does it suggest a liberally legitimate rational basis for SSM prohibitions? Gilboa’s invocation of “historically approved values” is elusive. On the one hand, the faithful liberal is surely not concerned with the preservation of historically approved values per se; that too would call for some version of legal moralism that liberals eschew. Of course, liberals might be concerned with preserving certain values if doing so is necessary for or conducive to the prevention of harm to others. But, again, Gilboa is already on record as denying that we have reason to suppose that there is any connection between harm to others and recognition of SSM; indeed, he admits that there is a lack of evidence that heterosexual monogamy would be preserved by prohibiting SSM. Ibid., p. 256. Nonetheless, Gilboa proposes one more strategy for locating a liberal rationale for SSM prohibitions: ..an action may be rational when no definite estimate is available of the probability that the action would lead to a desired outcome, if the actor seeks to avoid the worst outcome by following the maximin rule in making a decision under uncertainty... It may then be rational to prevent same-sex couples from being recognized as a means of protecting the dominant status of heterosexual monogamy, without knowing the probability that the action would have the desired outcome, if the opposition to same-sex marriage followed the maximin rule in making a decision under uncertainty. Ibid. Concerns about rationality aside, this line of argument suggests that SSM prohibitions do have a liberally legitimate rational basis, supposing that the worst outcome has something to do with harm to others: if there is no definite estimate of the probability that recognizing SSM would result in the worst of worst-case scenarios, then rationality compels us to refrain from altering the status quo and recognizing SSM. But then SSM prohibitions are rationally related to some liberally legitimate rational basis such that they would survive a liberal version of rational basis review. Gilboa’s reasoning here echoes Rawls’ invocation of the maximin principle in his defense his two principles of justice in his seminal A Theory of Justice, a principle that familiarly calls for ranking alternatives by their worst outcomes and adopting the outcome that has a worst outcome that is superior to all others—the best worse-case scenario, as it were. John Rawls, A Theory of Justice, revised edition (Cambridge: Harvard University Press, 1999), pp. 132-5. But as Rawls took pains to clarify, the maximin principle is a rule for choice under uncertainty: the situation must be one in which a knowledge of the relevant likelihoods is impossible, or at best extremely insecure. Ibid., p. 134. And it is far from clear that decisions about the recognition of SSM are being made in the conditions in which knowledge of the relevant likelihoods is impossible or extremely insecure. After all, we do have fairly substantial evidence that the recognition of SSM does not result in the sort of calamitous results that Gilboa gestures at; the experience of the Scandinavian nations suggests that recognition of SSM will not lead us to invalidate and dismantle the set of rules adopted in our society to govern heterosexual monogamy. See William Eskridge and Darren Spedale, Gay Marriage: For Better or for Worse? What We’ve Learned from the Evidence (New York; Oxford University Press, 2006) for an exhaustive review of this data. When considering the implications of recognizing SSM, we simply do not lack entirely the sort of information that agents lack behind Rawls’ veil of ignorance. Further, Rawls notes that invocation of the maximin principle is only appropriate if “the person choosing has a conception of the good such that he cares very little, if anything, for what he might gain above the minimum stipend that he can, in fact, be sure of by following the maximin rule” such that “It is not worthwhile for him to take a chance for the sake of a further advantage, especially when it may turn out that he loses much that is important to him.” Rawls, A Theory of Justice, p. 134. From the point of view of a liberal who cares about maximizing liberty within constraints of justice and fairness, the possibility of securing liberties for persons who are presently denied them is of significant value—something to be cared about greatly and something important to be lost. There is good reason, I submit, to deny that Gilboa’s invocation of the maximin principle is appropriate in the current context. But then, as far as I can see, Gilboa has no defense against the most plausible liberal argument in favor of the recognition of SSM: he has failed to show that there is any liberally legitimate interest served by prohibiting SSM. But things are actually even worse. For his discussion of equal access suggests that Gilboa’s position is prone to a fatal analogical argument, an argument that suggests that liberals must not only oppose SSM prohibitions but positively recognize SSM. Section IV: Equal Access and Discrimination Gilboa suggests a rationale for supposing that a failure to recognize SSM need not amount to an unjust failure to provide equal access to public services: When a service is provided publicly, the public, though tax money, would subsidize the individual’s purchase of the service either completely or partly. In a liberal society, a person who is eligible to receive a public service may prefer to receive the service privately, in which case the person would typically have to forgo the public subsidy for the service. Gilboa, “Same-Sex Marriage in a Liberal Democracy,” p. 248. For example, a family might choose to send its children to a private school and forgo publicly subsidized education. But a state that fails to subsidize the education of children attending private school need not fail to provide equal access to education: the family simply declines to exercise a legitimate publicly subsidized right claim that they could have exercised. While he rightly allows that marriage is not quite a service in the way that education is, Gilboa nonetheless thinks that public education and civil marriage are sufficiently similar such that an instructive analogy is available. Ibid., p. 249. For just as a family that sends its children to private school chooses to forgo exercising a publicly subsidized right, homosexual persons who want to marry someone of the same gender similarly choose to forgo exercising a publicly subsidized right. Homosexual persons may, after all, exercise the same publicly subsidized right that the rest of us can exercise—that is, the right to marry an eligible opposite-sex partner. If homosexual persons, like parents who prefer alternatives to public education, simply decline to exercise a legitimate publicly subsidized right that they could exercise, the failure of a state to subsidize SSM is similarly no failure to provide equal access. Thus, failure to subsidize SSM is not unjust discrimination if failure to subsidize private education is not. There are reasons for doubting that the purported analogy between SSM and education is strong enough to support the conclusions that Gilboa draws from it. One surely relevant difference is that a liberal state that prohibits SSM does not merely decline to subsidize SSM; it fails to recognize them at all—more on that below. But bracket that concern and suppose that marriage and education are sufficiently analogous such that Gilboa’s analogy goes through. I maintain that the analogy actually works against Gilboa’s position and implies instead that liberal states must recognize SSM. First, note that even if a liberal state permits families to opt out of sending their children to public school, it nonetheless takes significant steps to ensure that the alternatives available to those families meet a certain standard of quality. Private schools and parents who home-school are not free to teach just anything they wish, nor or they free to have classes taught and administered by just anyone. They are quite generally subject to all varieties of state regulation. A liberal state may also require that private schooled and home-schooled children engage in certain sorts of extracurricular activities to promote socialization and those activities too need to be regulated and perhaps even publicly subsidized. All this means that states which permit families to opt out of sending their children to public school do nonetheless subsidize the education of those children to some degree; at least, some public monies are dedicated to ensure equal access with respect to education. So, if ensuring equal access with respect to education actually requires regulating and subsidizing some alternative to public education, parity of reasoning suggests that ensuring equal access with respect to marriage actually requires regulating and subsidizing some alternative to opposite-sex civil marriage. But liberal states cannot very well regulate and subsidize some legal relationship without legally recognizing that relationship. Thus, ensuring equal access in a liberal state really does call for legally recognizing SSM. Further, it is important to note that states treat children whose parents opt out of publicly subsidized schooling on a par with children who are the product of public schooling. For example, children who are the product of legitimate private schools may compete for public jobs that require a degree of schooling, they may apply for admission to state subsidized universities that require completion of high school, and so forth. In other words, the state treats those who benefit from alternatives to public schools as being educated. But at present, there is no analogy with respect to those who benefit from alternatives to civil marriage. Part of the problem is that even the most generous legal relationships that fall short of marriage do not supply all of the rights and privileges associated with marriage. In the United States, there are some 1,138 federal rights and responsibilities associated with civil marriage that are not typically available to same-sex couples, As reported by the Government Accountability Office (formerly the General Accounting Office) of the United States Congress. My source here is Eskridge and Spedale, Gay Marriage: For Better or for Worse?, p. 166. many of which are not included in the package of rights and responsibilities provided by civil unions and domestic partnership alternatives. In 1999, the Supreme Court of Canada struck down Section 29 of the Ontario Family Law Act and held that “The exclusion of same-sex partners from the benefits of s. 29 promotes the view that… individuals in same-sex relationships generally, are less worthy of recognition and protection.” M. v. H., 2 S.C.R. 3 (1999). But that result had a limited impact and, like later provincial court decisions, failed establish a federal right of same-sex couples to marry and secure all of the rights and responsibilities associated with marriage. Same-sex couples in Canada only secured an equitable distribution of the rights and privileges and responsibilities associated with marriage with the passage of Canada’s Civil Marriage Act in 2005. By contrast, the Defense of Marriage Act is still federal law in the United States and same-sex couples lack an equitable distribution of the rights and privileges and responsibilities associated with marriage whatever individual state courts and legislatures have said about the matter. But that is only part of the problem and not even the gravest part of the problem. Arguably, standing marriage law could be revised such that the relevant bundle of rights and privileges and responsibilities associated with marriage could be incorporated into alternative legal relationships. Such a revision would have the odd result in creating two functionally equivalent legal relationships distinguished only in name, but never mind that. The problem is that the state does not treat those who benefit from alternatives to civil marriage, even revised alternatives with expanded bundles of rights and such, as being married. Appealing to Gilboa’s example involving public education is instructive here. Ensuring equal access with respect to education requires not simply providing children who opt out of public education with sufficient resources and benefits; it requires treating those who benefit from alternatives to public education as being educated. Suppose that some state provides children who opt out of public education with all the resources and opportunities that children who opt into public education are provided with, but adopts the policy of only recognizing children who have graduated from public schools as being educated; for example, children who are not educated in public schools are not permitted to enroll in state universities or compete for state jobs just because the state does not recognize them as having completed requisite educational requirements. Those children would certainly not be treated as though they had the legal status of being educated even if they had been provided with the most generous supply of benefits and resources along the way. Parity of reasoning suggests that ensuring equal access with respect to marriage requires those who benefit from alternatives to civil marriage as being married, a result that is not secured even if alternatives to civil marriage come with the most generous bundle of rights and privileges and responsibilities. But here too, liberal states cannot very well treat same-sex couples as being married without legally recognizing their relationship. Again, ensuring equal access in a liberal state really does call for legally recognizing SSM. I think this is enough to show that Gilboa’s appeal to public education fails to support his claim that prohibiting SSM need not violate the liberal demand for equality, but also that it implies that liberal states must recognize SSM. But still another component of his discussion of equal access deserves some attention. Gilboa anticipates an objection that, on his view, a state could legitimately refuse to recognize, say, Jewish weddings even though it recognizes Christian marriages. Gilboa, “Same-Sex Marriage in a Liberal Democracy,” p. 249. After all, if refusing to recognize SSM out of genuine moral disapprobation for homosexual sex does not illicitly target some group for exclusion due to their membership in a protected group then neither should refusing to recognize Jewish marriages out of genuine moral disapprobation for Jewish sex, and if refusing to recognize SSM need not violate liberalism’s commitment to equality then neither must refusing to recognize Jewish marriages, and so forth. In response, Gilboa argues as follows: The historically familiar type of anti-Jewish discrimination has never focused on one particular issue, such as marriage, but was rather comprehensive, excluding Jews from equal access to public resources, including legal rights and protections. Therefore, if counterfactually, Jewish weddings were denied public recognition, we have to expect it to be part of a pattern of exclusion encompassing various types of public resources and opportunities, which would clearly indicate anti-Jewish discrimination. If, however, incredibly, Jewish people enjoyed equal access to all resources except for the public recognition of their weddings, then we would have to consider the possibility that the exclusion is not discriminatory. Ibid., p. 250. We are to imagine that Jewish citizens of our state are either consistently denied equal access to public resources or that they are only denied access to one particular resource—that is, the public recognition of their marriages. If the former, Gilboa allows, like most of us, that prohibitions of Jewish marriages are clearly unjust. But if the latter, Gilboa denies that prohibitions of Jewish weddings must be unjust, presumably because there may well be some adequate rational basis for doing so. But I doubt that liberals will be as willing as Gilboa to grasp the second horn of this dilemma, as reflection on a real-life example will illustrate. Richard Perry Loving and Mildred Delores Jeter were residents of the Commonwealth of Virginia who married in the District of Columbia. Upon their return to Virginia, they were charged with violation of the state’s “Racial Integrity Act” which prohibited the marriages of white and non-white persons, a felony punishable by a prison sentence of up to 25 years. All of this occurred in the late 1950’s in the American South, such that it is plausible to suppose that a woman of African descent—like Jeter—was consistently denied equal access to public resources. But it is less plausible to suppose that a white man—like Loving—was similarly consistently denied equal access. Indeed, as far as I know, Perry was only denied access to one particular resource: the Commonwealth of Virginia refused to recognize his marriage to Jeter. But surely the faithful liberal should conclude that Perry was treated unjustly, prior to evaluating any ostensible rational basis for such a refusal. All this suggests that Gilboa has been impaled on the horn of the dilemma he hoped to grasp; at least, he seems committed to supposing that there could be a rational basis for prohibitions of inter-racial marriage given his comments about the possible justification of prohibiting Jewish marriages. But surely Virginia’s prohibition of inter-racial marriages is bound to shock the conscience of the liberal in a rather more direct and profound way. So, even if everything that I have suggested about rational basis and equal access is mistaken, Gilboa’s position is still subject to a fatal reductio: he is committed to the tenability of miscegenation laws that liberals are bound to reject on their face. At least, if the purported liberal case for prohibiting SSM yields a similar case for prohibiting inter-racial marriage, then so much the worse for that purported liberal case for SSM. Conclusion Gilboa’s position with respect to SSM is motivated by a commitment to the very principles of justice that I find plausible: those exemplified by liberal states. However, I a genuine commitment to liberal principles of justice does not lead to the “compromise” that Gilboa proposes, but instead calls for the recognition of SSM if civil marriage is to be recognized at all. At one point, Gilboa approvingly cites philosopher-judge Richard Posner who wonders out loud: Do we want homosexual couples to have the same rights of adoption and custody as heterosexual couples? Should we worry that a homosexual might marry a succession of dying AIDS patients in order to entitle them to spouse’s medical benefits? These questions ought to be faced one by one rather than elided by conferring all the rights of marriage in a lump on homosexuals willing to undergo a wedding ceremony. Richard Posner, Sex and Reason (Cambridge: Harvard University Press, 1992), p. 313. Posner’s request for a line-by-line review prior to extending the rights and benefits and protections of civil marriage to same-sex couples is not clearly imprudent, and if there is some liberally legitimate reason for providing only some of those rights and benefits and protections to same-sex couples, then—but only then—we would have found some conceptual space between rejection and recognition of SSM in a liberal democracy. However, I have endeavored to argue that there is no liberally legitimate justification for withholding those rights and benefits. As such, there is simply nowhere between rejection and recognition: liberal states must recognize SSM if they recognize opposite-sex marriage. PAGE 25