1 The first draft of this paper was written at the University of Montreal's Centre de recherche en ethique (CREUM), courtesy of the financial support of the Fonds quebecois de la recherche sur la société et la culture (FQRSC). I would like to thank Charles Blattberg for asking the question that prompted me to write this paper. I would also like to thank the members of the Queen's Political Philosophy Reading Group for comments and criticisms that led me to revise the paper substantially. Thanks to Rahul Kumar in particular for helpful discussions about the idea of reconciliation. Finally, I received detailed, constructive criticism from the paper's two anonymous referees. Responsibility for the remaining errors is mine.

2 For criticism based on this kind of interpretation, see, for example, Hampton, Jean 'Should Political Philosophy Be Done Without Metaphysics?’ Ethics 99 (1989) 791–814,CrossRefGoogle Scholar and Raz, Joseph ‘Facing Diversity: The Case of Epistemic Abstinence’ Philosophy & Public Affairs 19 (1990) 3–46.Google Scholar For praise based on a similar interpretation, see Rorty, Richard ‘The Priority of Democracy to Philosophy,’ reprinted in Objectivity, Relativism and Truth: Philosophical Papers I (Cambridge: Cambridge University Press 1990), 175–96.CrossRefGoogle Scholar

3 Rawls, J. Political Liberalism (New York: Columbia University Press 1996), 225,Google Scholar cited hereafter also as PL.

4 PL, 11, 147. I follow Charles Larmore in thinking that public reason is the fundamental idea in Rawls's thought, one that underlies the ideas of a political conception of justice and an overlapping consensus. Public reason governs the relation in which we stand to one another as Citizens, and so encompasses all the different elements that make up the ideal of a constitutional demoeraey — a paraphrase of Larmore, C. 'Public Reason’, The Cambridge Companion to Rawls (Cambridge: Cambridge University Press 2003), 368.Google Scholar

5 Benjamin, M. Splitting the Difference: Compromise and Integrity in Ethics and Politics (Lawrence: Kansas University Press 1990).Google Scholar

6 Public reason has been described as a principle of restraint (Raz, J. The Morality of Freedom [Oxford: Oxford University Press 1986], 136Google Scholar), epistemic abstinence (Raz, J. ‘Facing Diversity: The Case of Epistemic Abstinence’, Philosophy and Public Affairs 19 [1990], 4Google Scholar), preclusion (Gutmann, A. and Thompson, D. ‘Moral Conflict and Political Consensus’, Ethics 101 [1995], 64CrossRefGoogle Scholar), or disagreement-avoidance (Raz, J. ‘Disagreement in Politics’ American Journal of Jurisprudence 43 [1998] 30CrossRefGoogle Scholar).

7 These alternatives are not necessarily nested, since it is conceivable to maintain that the principle of public reason holds only for non-basic policies. As Steven Wall points out, one might think that it is never a legitimate reason for public policy that the policy in question promotes belief in or living in accordance with the correct ideal of human flourishing, but nonetheless claim that in determining the content of our basic rights and liberties, appeals to the full moral truth are appropriate: Wall, S. Liberalism, Perfectionism, and Restraint (Cambridge: Cambridge University Press 1998), 75.CrossRefGoogle Scholar We should never use the State to promote a particular way of life simply for its own sake, but when we define what is going to count as murder, one might argue, we ought to base our decision on the truth about whether fetuses are persons, even if this topic is the subject of inevitable reasonable disagreement.

8 The ideal of public reason holds ‘for Citizens when they engage in political advocacy in the public forum’ and ‘for how Citizens are to vote in elections when constitutional essentials and matters of basic justice are at stake’ (PL, 215).

9 As Stephen Macedo puts it, ‘no one is suggesting that the contours of liberal public reason should be used to define the limits of constitutional rights to free speech. Public reason helps define a moral ideal, not a legal requirement’ (Macedo, S. ‘In Defense of Liberal Public Reason: Are Slavery and Abortion Hard Cases?’ American Journal of Jurisprudence 42 [1997], 20–2CrossRefGoogle ScholarPubMed).

10 ‘Public reason is the reason of equal Citizens who, as a collective body, exercise final political and coercive power over one another in enacting laws and in amending their constitution’ (PL, 214). Much of the discussion of public reason in the context of debates over deliberative democracy wrongly assumes that the principle restricts what Citizens can say to each other ‘in public'. As Macedo and Larmore argue, if it is the coercive character of political authority that makes public reason essential to a constitutional democracy, then public reason ought to be understood as applying to the reasoning by which Citizens take part in political decisions (Larmore, ‘Public Reason’, 383; Macedo, ‘In Defense of Liberal Public Reason’, 21). Public reason does not apply to just any discussion that takes place in public; there is nothing wrong with debating the existence of god or the ethics of sexual orientation in the opinion pages of a newspaper. What is supposed to be wrong is to make a decision about public policy that can be justified only if one assumes that god does (or does not) exist, or only if one assumes that homosexual conduct is (or is not) sinf ul.

11 There is little doubt that Rawls adopted this view, but there has been much debate as to why and whether he was right to do so; see, for example, Quong, J. ‘The Scope of Public Reason’, Political Studies 52 (2004) 233–50CrossRefGoogle Scholar

12 PL, 135-6, 216-17; see also Nagel, Thomas ‘Moral Conflict and Political Legitimacy’, Philosophy and Public Affairs 16 (1987), 223,Google Scholar and Larmore, Charles ‘The Moral Basis of Political Liberalism’, The Journal of Philosophy 46 (1999), 606.Google Scholar

13 Larmore, ‘The Moral Basis of Political Liberalism’, 600-5

14 Ibid., 600

15 Ibid., 607-8

16 PL, 37, 78; Larmore, ‘Political Liberalism’, 343.Google Scholar Rawls's reasons for denying that public reason involves compromise can therefore not be the same as his critics' reasons for denying that public reason involves compromise.

17 See, for example, Sher, George Beyond Neutrality: Perfectionism and Politics (Cambridge: Cambridge University Press 1997), 47.CrossRefGoogle Scholar

18 Larmore, ‘Political Liberalism’, 349Google Scholar

19 Thus, for example, one might argue that the claim that homosexual conduct is sinful ought not to be counted in the factors influencing our decision about same-sex marriage because God approves of homosexual relationships.In the 2003 Parliamentary hearings on same-sex marriage, Choice Okoro testified to this effect on behalf of the United Church of Canada. ‘Human sexual orientations, whether heterosexual or homosexual, are a gift from God and part of the marvelous diversity of creation' (Standing Committee on Justice and Human Rights, 20 February 2003 hearings, available at <http://www.parl.gc.ca/committee/CommitteePublication.aspx?SourceId=24531)>.

20 See, for example, Ackerman, Bruce Social Justice in the Liberal State (New Haven: Yale University Press 1980), 9.Google Scholar In Ackerman's view, scepticism is just one of a number of possible justifications for neutrality. See also Dworkin, A Matter of Principle, 205,Google Scholar where Dworkin distinguishes liberalism based on neutrality, which finds its ‘most natural defense’ in moral scepticism, from liberalism based on equality.

21 For example, Hollis and Bellamy liken the self-imposed constraints of public reason to the policy of neighbours who keep the peace by not talking politics, so that 'baby-sitting gets done and cats are fed when their owners go on holiday’ (Bellamy, R. and Hollis, M. ‘Consensus, Neutrality and Compromise’, Pluralism and Liberal Neutrality [Illford: Frank Cass 1999], 61Google Scholar).

22 Proponents of liberal neutrality commonly characterize this neutrality as justificatory rather than consequential, in order to insist that neutrality involves only a presumption against differential treatment, not a guarantee of equal outcomes. See, for example, Kymlicka, Will ‘Liberal Individualism and Liberal Neutrality’ Ethics 99 (1989), 883–4.CrossRefGoogle Scholar Freedom of conscience and association will have nonneutral consequences, Kymlicka argues, since some groups will be more successful than others in attracting adherents, which shows that Rawls could not have endorsed 'consequential neutrality’ (883). This emphasis on reasons rather than outcomes gives liberal neutrality the form of a doctrine of public reason. The justification for justificatory neutrality may be consequentialist, however, since it is possible to advocate neutrality in the dimension of reasons for policy on the grounds that f ollowing such a rule has good consequences. My description of such a view f ollows Thomas Hurka's account of Kymlicka's position (‘Indirect Perfectionism: Kymlicka on Liberal Neutrality,’ The Journal of Political Philosophy 3 36-57).

23 Hurka, George Sher, and others have argued that such consequentialist considerations only justify a quite limited form of State neutrality — see, for example, Sher, Beyond Neutrality (1997), 45-72. The point here is simply to distinguish the idea of public reason from the quite different idea that the diversity of good lives and the need to live one's life ‘from the inside’ limits governmenf s ability to promote the true good directly.

24 To describe the reasonable pluralism that gives rise to the problems political liberalism addresses, Macedo talks of ‘ultimate religious and philosophical ideals’ and ‘religious beliefs and philosophical ideals of life’ (Macedo, ‘In Defense of Liberal Public Reason’, 3,10). Nagel talks about disagreements ‘so deep and so acute’ that no decision-procedure admitting the rival views in question as inputs could command the reasonable acceptance of the losers. Conspicuous among these disagreements are religious dif f erences, Nagel says, ‘but other convictions about the ultimate meaning of life or the sources of its value should be included as well’ (Nagel, T. Equality and Partiality [New York: Oxford University Press 1991], 154–5;Google Scholar and see also 167). Larmore talks about ‘deep questions concerning how we should live’, such as the question about the correct path to salvation, but also ‘the meaning of life and certain deep aspects of morality’ or just ‘ultimate questions of value’ (Larmore, C. The Marals of Modernity [Cambridge: Cambridge University Press 1996], 168–9CrossRefGoogle Scholar).

25 Rawls accepted that public funds for the arts and sciences could be provided through what he called ‘the exchange branch’, but only under very restrictive conditions (Rawls, J. A Theory of Justice [Cambridge, MA: Harvard University Press 1971], 282–3,Google Scholar 332; cited hereafter also as TJ). The exchange branch was a mechanism that would provide goods otherwise undersupplied by the market (goods whose benefits accrue to all while the costs of producing them accrues only to those who contribute). For any good that was truly public in this sense, there ought to some distribution of the extra taxes necessary to provide the good that could gain unanimous approval. Since no one would be taxed unless all consented, there would be no objection to defining the public good in perfectionist terms. However, the principles of justice did not permit subsidizing universities, opera or theatre, simply on the grounds that the majority believed these institutions to be intrinsically valuable.

26 TJ, 207

27 TJ, 327

28 See note 24 for an explanation of Rawls's account of how public funds might be devoted to arts and culture, subject to the requirement that some distribution of the necessary taxes receive unanimous consent.

29 PL, 224-5; emphasis added. Rawls's case against perfectionism was made easier by the f act that he def ined perfectionism as a teleological principle, meaning a principle holding that institutions should be designed in whatever manner is necessary to maximize some valued end — in this case, virtue or human excellence. Such a principle might require us to sacrifice the perfection of some (those not easily perfectible), for the sake of the greater perfection of others (those for whom investments in perfection are more cost effective). This objection targets the property of maximizing an aggregate good, however, not the property of defining the good in terms of human excellence or perfection. It is possible to be an egalitarian perfectionist — someone who believes that the good of each person merits equal concern, but that this good is to be defined in terms of virtue. A strictly aggregative notion of perfectionism is implausible because it seems aesthetic rather than moral. An egalitarian perfectionism, in contrast, would not permit that individuals be sacrificed for social or human beauty.

30 PL, 3-4

31 Joseph Chan points out that arguments for neutrality between conceptions of the good life gain in plausibility if such conceptions are treated on a par with religious doctrines (Chan, J. ‘Legitimacy, Unanimity, and Perfectionism’, Philosophy and Public Affairs, 10Google Scholar). But Chan argues that, unlike views about the existence of God and the nature of the soul, many perfectionist judgments about what makes a good life are not inevitably reasonably contestable.

32 Another way to make room for some perf ectionism within the context of a commitment to the principle of public reason would be to point out that since public reason applies only to fundamental political decisions, perfectionist ideals are legitimate in other areas of policy and public life. In Rawls's view, for example, ‘the limits imposed by public reason do not apply to all political questions but only to those involving what we may call ‘constitutional essentials’ and ‘questions of basic justice’ (PL, 214). Larmore suggests that in virtue of this restriction in scope, Quebec's present System of Special support and protection for the French language might be consistent with the principle of public reason (Larmore, ‘Public Reason,’ 381). As Larmore admits and others have pointed out, however, Rawls did not clearly explain the rationale for this restriction of scope (see, for example, Chan, ‘Legitimacy, Unanimity and Perfectionism,’ 7-8, note 7). One possibility is that we restrict the scope of public reason to fundamental political decisions as a compromise with perfectionists or as a concession to practicality, but that the principle would ideally, govern all political decision-making. Compromising public reason would not make public reason a compromise, however. Another possibility is that the fundamental value or values that underlie the principle of public reason themselves dictate the restriction in scope to fundamentals. The case for excluding perfectionist reasons from public decision-making certainly seems stronger when the decision involves use of criminal law to prohibit conduct, when it is entrenched by supermajoritarian amendment procedures, and when it has deep and pervasive effects on people's opportunities and aspirations, than it does when the policy in question is enacted by ordinary law and establishes some small local subsidy (though this set of considerations does not support a clear distinction between fundamental and non-fundamental questions). My argument that public reason should be understood as a form of moral compromise is intended to be independent of this issue of the scope of the principle's application.

33 PL, 141-2

34 PL, 39

35 PL, 39

36 PL, 40

37 PL, 142

38 PL, 171

39 TJ,6

40 Arnsperger and Picavet suggest that Rawls dismisses compromise ‘because he equates compromise with mere modus vivendi’ (Arnsperger, C. and Picavet, E.B. ‘More than Modus Vivendi, Less than Overlapping Consensus: Towards a Political Theory of Social Compromise’, Social Science Information 43 [2004], 177CrossRefGoogle Scholar). While it is true that Rawls's main concern is to avoid the misinterpretation of public reason as a mere modus vivendi, I do not think it true that Rawls simply disregarded the possibility of moral compromise. In the fourth section of the paper, I articulate (and criticize) what I take to be Rawls's deeper reasons for rejecting the characterization of public reason as a form of moral compromise.

41 Golding, M. ‘The Nature of Compromise: A Preliminary Inquiry,’ Compromise in Ethics, Law, and Politics, Pennock, J. and Chapman, J. eds. (New York: New York University Press 1979), 3–25;Google Scholar A. Kuflik, ‘Morality and Compromise,’ Compromise in Ethics, Law, and Politics, 38-65; T. Benditt, ‘Compromising Interests and Principles,’ Compromise in Ethics, Law, and Politics, 26-37; Benjamin, Splitting the difference; Leydet, D. ‘Pluralisme et compromis,’ Les raisons du compromis, 81–106 (Toulouse: Octares 2004).Google Scholar

42 If we take ‘policy’ in the broadest sense, the conflict could be over any public decision, including everything from concrete policies or public actions to questions of decision-making procedure to the question of what reasons should count as public. To illustrate the idea of moral compromise, I will focus on concrete policies, such as those surrounding marriage. The question will then be whether the idea of moral compromise so described also applies to the more abstract policy of restricting the set of legitimate reasons for public decisions.

43 Here, I follow Dominique Leydef s characterization of compromise (Leydet, ‘Pluralisme et compromis’, 84-8).

44 Something like traditionalist same-sex marriage has been advocated by leading American proponents of same-sex marriage (A. Sullivan, ‘Here Comes the Groom: A Conservative Case for Same-Sex Marriage’, The New Republic [Aug. 28, 1989]; Rauch, J. Gay Marriage: Why It Is Good for Gays, Good for Straights, and Good for America [New York: Henry Holt 2004];Google Scholar see also Macedo, S. ‘Sexuality and Liberty: Making Room for Nature and Tradition’, Sex, Preference, and Family, Estlund, D. and Nussbaum, M. eds. [Oxford: Oxford University Press 1997], 86–101).Google Scholar

45 Benjamin, Splitting the Difference, 34Google Scholar

46 Ibid., 34; see also Benjamin, M. ‘Compromise and Integrity in Ethics’, The Practice and Theory of Ethics, Kent, T. and Gentry, M. eds. (Indianapolis: University of Indianapolis Press 1995), 31.Google Scholar

47 This distinction between ‘the best policy’ and ‘the best policy in the circumstances’ may seem to introduce an arbitrary asymmetry. For what we call ‘the best policy’ is only best outside of the conditions in which the compromise policy is best. The reason we call one ‘best’ and the other only ‘best in the circumstances’ is that we prefer to be in the circumstances that allow for the ‘best’ policy, rather than in the circumstances that require the compromise policy, which is only best under these unfortunate conditions.

48 PL, 170-1

49 PL, 170-1

50 PL, 37,144

51 The fact of reasonable pluralism is not, to be clear, the hopeful view that only reasonable doctrines will persist in an approximately just society. It is, rather, the apparently more pessimistic view that even fully reaonable people will espouse a wide ränge of different and conflicting, though reasonable, doctrines, in such a society. Rawls speaks for example, of ‘the fact of profound and irreconcilable differences in Citizens’ reasonable comprehensive religious and philosophical conceptions of the world, and in their views of the moral and aesthetic values to be sought in human life’ (Rawls, J. Justice as Fairness: A Restatement [Cambridge, MA: Harvard University Press 2001], 3Google Scholar).

52 Buchanan, A. ‘Assessing the Communitarian Critique of Liberalism’, Ethics 99 (1989), 877CrossRefGoogle Scholar

53 Justice as Fairness, 3. The idea of reconciling us to our institutions sounds deeply conservative, but this impression is misleading. What we are supposed to be reconciled to is our ‘social world’, by which Rawls means the background conditions that set the limits of the possible, not the specific characteristics of the institutions that exist today (Rawls, J. The Law of Peoples, 124Google Scholar).

54 Rawls, J. Justice as Fairness, 3Google Scholar

55 Michael Hardimon makes this point by contrasting reconciliation with consolation. Both are responses to prior disappointment and both involve a form of acceptance, rather than just passivity in face of the impossibility of change, but consolation involves continuing disappointment. ‘Your spouse dies but you find consolation in the continuing existence of your children. Their existence does not negate the loss, but it does console you…. If, however, you are truly reconciled … you have no need for consolation, because, in being reconciled, you have fully accepted the Situation. There is no thought: ‘Things would be better were they otherwise’ (Hardimon, M. 'The Project of Reconciliation: Hegel's Social Philosophy’ Philosophy and Public Affairs 21 [1992], 174Google Scholar). Hardimon also insists, however (as explained below), that Hegelian reconciliation does not involve wishing away or ignoring conflict and other problematic features of our social world, such as divorce, poverty, and war (175). Rather, it involves recognizing how the possibility of certain kinds of imperfection is essential to the realization of specific goods.

56 Rawls, Justice as Fairness, 3Google Scholar

57 Ibid., 5

58 Ibid., 3

59 Ibid., 4

60 Ibid., 5

61 Rawls, J. The Law of Peoples (Cambridge, MA: Harvard University Press 1999), 12.Google Scholar The passage from which this quote is taken is not entirely clear. ‘Though we can imagine what we sometimes think would be a happier world — one in which everyone, or all peoples, have the same faith that we do — that is not the question, excluded as it is by the nature and culture of free institutions. To show that reasonable pluralism is not to be regretted, we must show that, given the socially f easible alternatives, the existence of reasonable pluralism allows a society of greater political justice and liberty. To argue this cogently would be to reconcile us to our contemporary political and social condition.’ In the first sentence, Rawls seems to say that questions about Community or other goods are not relevant to the issue of reconciliation. Yet to be reconciled to something means to recognize the good it involves, such that one would not change it if one could. By itself, recognizing the impossibility of changing an unfortunate condition (without oppressive use of State power, that is) leads only to resignation. To show that reasonable pluralism allows greater justice and liberty we have to compare what is possible given the fact of reasonable pluralism with what would be possible if major differences in comprehensive doctrines were not inevitable between reasonable people.

62 This path to reconciliation amounts to simply denying that the world contains defects and imperfections. Hegel's view, again according to Hardimon, was that true reconciliation could not be achieved in this manner, by putting on rose-coloured glasses (Hardimon, ‘The Project of Reconciliation’, 175Google Scholar), and believing the world to be in a State of perfect harmony (178). Reconciliation instead required an understanding of how the defects and imperfections that do exist are necessary for the realization of greater goods.

63 Ibid., 126

64 ‘Our hope for the future rests on the belief that the possibilities of our social world allow a reasonably just constitutional democratic Society living as a member of a reasonably just Society of Peoples. And essential step to being reconciled to our social world is to see that such a Society of Peoples is indeed possible’ (Rawls, The Law of Peoples, 124Google Scholar). The possibility of a just society is only a step towards reconciliation.

65 The following discussion builds on Hardimon's discussion of Hegel (Hardimon, 'The Project of Reconciliation’ 176-7).

66 Some of the reasons for thinking Catholic doctrine not unreasonable is that it condemns a whole ränge of sexual activity as sinful, not just homosexual conduct, and that it does so on the basis of a complex set of views about personal integrity and openness to the divine gift of the potential to bring a new life into the world. Catholic doctrine also Claims to be concerned for the true good of all, gays and lesbians included, and criticizes expressions of malice and hate directed towards gays and lesbians. ‘Homosexual persons, as human persons, have the same rights as all persons including the right of not being treated in a manner which offends their personal dignity’ (1986 ‘Letter to the Bishops of the Catholic Church on the Pastoral Care of Homosexual Persons’, available at <http://www.vatican.va’,roman_curia/congregations/cfaith/documents/rc_con_cfaith_doc_20040731_ collaboration_en.html>).

67 This is not to say that every official pronouncement of the Congregation for the Doctrine of the Faith is reasonable. The Vatican's ‘Considerations Regarding Proposals To Give Legal Recognition To Unions Between Homosexual Persons’ states that ‘[t]here are absolutely no grounds for considering homosexual unions to be in any way similar or even remotely analogous to God's plan for marriage and family' (emphasis added; available at <http://www.vatican.va’,roman_curia/congregations/cfaith/documents/rc_con_cfaith_doc_20030731_ homosexual-unions_en.html>). Even if one grants the truth of Catholic doctrine, it seems to me to display an almost wilful moral blindness to suggest that no homosexual relationship bears any similarity to a true marriage in any of the goods it realizes.

68 For example, Robert George suggests that public reason is a doctrine promoted by comprehensive liberals to advance their political ends without having to justify them philosophically. ‘If we observe that whenever the doctrine of public reason is deployed, the result is that ‘we’ (i.e. those of us who dissent from the prevailing liberal orthodoxy on questions such as abortion, euthanasia, homosexuality, and human cloning) are declared to have lost even bef ore the argument begins, perhaps our suspicion is not entirely unwarranted’ (George, R. and Wolfe, C. ‘Natural Law and Liberal Public Reason’, American Journal of Jurisprudence 42 [1997], 32CrossRefGoogle Scholar).

69 This is the Intuition behind one of the main constitutional arguments for same-sex marriage in the United States, the view that marriage is one of the fundamental rights not enumerated but still protected by the Constitution: Gerstman, E. Same-Sex Marriage and the Constitution (Cambridge: Cambridge University Press 2004).Google Scholar Of course, gays, and lesbians are permitted by law to marry — they simply cannot marry someone of the same sex, no more than can heterosexuals marry someone of the same sex. Clearly, however, sexual orientation is, for most people, unchosen and virtually impossible to change. Many gays and lesbians therefore cannot honestly marry someone of the opposite sex, which gives the comparison with the rights of felons its force. I do not claim that this argument is convincing as a constitutional argument, but rather that it is a powerful moral argument. If the right to marry is so important that one can't lose it, no matter what crimes one commits, then how can we define marriage so that for some Citizens the right to marry is virtually empty (since these Citizens are allowed to marry only people they could not possibly want to marry)?

70 Such reasoning could be used to support the Supreme Court of Canada's 2002 decision in the Surrey, BC, book-banning case, Chamberlain v. Surrey School District No. 36,4 S.C.R. 710, although the Court did not distinguish between (a) putting the goal of teaching non-discrimination and civic equality above parents’ ability to guide the moral education of their children, given the inevitable conflict between the two concerns at the level of primary school, and (b) teaching children (of all ages) that parents with conservative religious views are wrong about sin and sexual orientation.

71 Such a response seems appropriate, for example, with respect to the fanatical, hateful views of the Reverend Fred Phelps of the Westboro Baptist Church of Kansas. Phelps’ views are clearly unreasonable, both in the theoretical sense of not being a possible conclusion of undistorted human reason, and in the practical sense of not being a possible component of a fair and stable System of social cooperation.