David Heyd’s careful chapter makes the double claim that toleration, often considered the “fundamental,” even “constitutive” virtue of liberal politics, is in fact neither political nor a virtue. Moreover, he claims, toleration is not strictly a duty but supererogatory, “beyond the call of duty.” His argument is absolutely valid given how he defines “political,” “virtue,” and “toleration.” My comments will, however, question the utility of all three of Heyd’s definitions. This may seem impertinent: one should not simply quarrel over terms. But a chapter so well reasoned leaves no choice: finding no flaw in an argument, one can only contest premises. I shall argue that in this case certain terms are really more productive than others in the sense of being more true to the circumstances that liberal democracies actually face and that make toleration necessary in the first place.
Heyd rightly sees his three claims as interrelated. So are mine. While Heyd defines politics as what pertains to the neutral state, toleration is in fact a political virtue because politics in a pluralistic democracy starts from the premise that the neutral state may not exist and certainly cannot be counted on. Democratic politics needs toleration to guide public decisions both because laws universal in scope admit of discretion in application and because the prospects of sustaining neutrality and other liberal values depend, in a democracy, on the attitudes and virtues of the ordinary citizens who control politicians’ jobs. As for virtues, political theory has long (since Machiavelli, Mandeville, and Montesquieu) drawn a distinction between the perfectionist or Aristotelian virtues that help people be good human beings and the more instrumental virtues that helps them be good citizens of their polity.2 Heyd must be right that toleration is not a virtue if that means a state of character based on trained natural impulses and conducive to an agent’s own good. But, as he recognizes, it might be a virtue in a broader sense relating to what a heterogeneous society needs to work well. If this is true, toleration, like the other liberal-democratic virtues (though few others are uncontroversial), means any attitude, emotion, reason, or habit that helps us get along with beliefs, behaviors, or people that we find morally objectionable, stipulating the anti-utopian assumption that no one is likely to become perfectly good or wise. Finally, given a realistic view of how liberal democratic politics works and how citizens must act to keep it working, toleration does start to appear necessary, not supererogatory. Essentially, toleration is what liberal democratic citizens need to get along, not only with the openly intolerant, but also with people—in fact, this is most people—who believe in and live by ideologies of toleration that strike “us” as obviously wrong and possibly dangerous.
For Heyd, the triumph of the neutral state has made the virtue of toleration less important than it used to be, in at least three ways. First, a “state,” being an “impersonal institution,” is not the kind of entity that can display toleration or any other virtue: it lacks feelings, beliefs, and wishes.3 Second, the officials of states operate by impartial and universal rules that rule out the kind of discretionary decisions in which virtue can be (or “ought to be”?) displayed. The idea, if I understand it, is that state agents, the kind of entities that could display virtues, ought not to if they are acting as they should: “it is the duty of the judge to ignore her personal moral views rather than to manifest toleration of other, incompatible views.” Heyd would apparently say that, if state officials do act on their attitudes or are governed by their virtues, they are illicitly freelancing and no longer count as state officials. Finally, the triumph of states that enforce neutrality means that the rest of us have less scope for exercising toleration: we refrain from burning witches not because we consider and (virtuously) reject this as a possible course of action, but because the act violates rights that the law protects.4
A full response would involve a theory of the state. Standing on one foot: it may be doubted whether such an entity exists, much less whether “the establishment of liberal democracy” has established it and its neutrality for all time. We should note competing traditions here. One tradition of talking about “the state” (and about toleration), drawing inspiration in various materialist and idealist forms from Hobbes, Bodin, some philosophes, Kant, and Hegel, stresses the duties of enlightened magistrates, who are to establish and enforce toleration precisely because popular passions on their own will do the opposite. Over time, an increasingly educated public can be trusted with more and more responsibility and authority.5 Another tradition, more pluralistic, democratic, skeptical towards elites—and therefore quintessentially though not exclusively American—follows instead Montesquieu, Madison, and Tocqueville, and has an affinity with political theory rather than philosophy in focusing on the actual practices of democratic polities rather than the ethos that animates their theoretical defenders. This tradition points out that when government officials are selected by the people, toleration as a government policy can only survive if popular attitudes support it, either directly or through complex institutional or social mechanisms. This second tradition tends even to define toleration differently—opposing it not to state partiality but to majority tyranny. And it stresses the need to make toleration consistent with personal liberty and equality of political power rather than redefining (or ignoring) the latter to accord with ideals of neutrality.
Arguments below will draw, as will surprise no one, on the second tradition. But even the first cannot eliminate the need for virtue among governing officials. Actual government officials will always have discretion, and toleration is among the many virtues that will guide that discretion. This is true even of judges, whose decisions may be free of personal or partisan interest but are still far from automatic. But it is even more true, and legitimately so, of legislators—who necessarily represent partial and controversial points of view—and even of executive officials who may be implementing laws that apply universally but who have as politicians stated, or implied, specific positions on what they will do with their inevitable discretion—which evils will be most sought out and what actions will be taken in fighting them.
Moreover, even an ethos of neutrality (stipulating for argument that we are governed by one) cannot settle difficult cases. Cases involving religion notoriously show that neutrality as a policy, refusal to actively favor or persecute a moral or religious outlook, does not always mean neutrality in effect. The rule that men and women in the U.S. Air Force must remove their headgear indoors was meant to ensure military uniformity, not to harm Jews—but still meant that observant Jews could not serve.6 The law forbids Native Americans and whites alike from using peyote, but forces only the former to abandon their religion or face the public disadvantages accruing to felons.7
In such cases, the virtue of toleration prompts a decent government to respond to these asymmetric disadvantages by not doing some of the things that might be constitutionally allowable. Toleration is a virtue that guides the governing power in making such accommodations precisely when constitutional neutrality and universality do not require this. And, in a democracy, that governing power ultimately rests in all of us. When we vote, organize, or persuade others, we have the choice of punishing officials who accommodate unusual religious and moral practices; and we must hope for the internal and social resources that overcome whatever tendency we may have to do so excessively. Far from making toleration understood as “mercy” or sovereign discretion superfluous, liberal democracy makes it universally necessary.8 And toleration, like most virtues, involves a mean. Too little toleration and we harm people who are trying to live as best they can by their own lights, in order to vindicate our love of abstraction; too much and we have no laws.9
Since collective decisions on such matters are not always politically salient and most people pay politics only occasional attention, this aspect of toleration will be intermittently useful for most citizens, a job requirement only for a few public officers. But toleration on the private scale remains crucial more generally because states cannot do everything. Many discussions of toleration imply, unrealistically, that the government has so much power as to eliminate all opportunities for private intolerance of the coercive variety.10 On the contrary: no police force can prevent all hate crimes, vandalism, and other acts of overt intolerance, and these acts cause substantial harm and social fear even when punished after the fact. Religious, ethnic, and national prejudice historically and currently lead to strong temptations to violate even quite stringent laws and norms of political behavior, especially those enjoining procedural neutrality and equal treatment. To the extent that those who harbor prejudice restrain themselves from acting on these temptations, they are exercising a virtue whose importance is clear. It would of course be better if nobody had the prejudice to begin with.11 But this does not make toleration a marginal subject for theoretical reflection. On the contrary, a theory of a prejudice-free society is a view towards nowhere.
Toleration is commonly defined as “the degree to which we accept things of which we disapprove.”12 This of course describes the practice of toleration, not the virtue. Even when a given act of toleration can be described without great controversy (which is not always), the kind of disapproval a given actor feels, and the motivations she has for not acting on it, can vary. Many combinations are logically possible.13 But the politically interesting question is which are likely to occur with some frequency and have a political manifestation.
Heyd defends in a highly articulated and subtle version a broad approach to toleration that he has helped make fairly standard among moral philosophers.14 This moral-philosophical view of toleration assumes, as most philosophers do, that “being judgmental with regard to beliefs and practices in the abstract is a desirable attitude.”15 It further holds that some sort of respect for the moral agents whose practices we condemn should prompt a switch in perspective that keeps us from putting this judgmentalism into practice (or at least changes the way we would think about doing so—the intended effects on action are sometimes ambiguous). On this account, toleration means refraining from doing what we would normally think praiseworthy, namely correcting a wrong. Hence the paradoxical tone, shared with many moral philosophers, of Heyd’s conclusion.
Both of the above assumptions can, of course, be challenged. Many people who are not moral philosophers regard judgmentalism about other people’s actions and practices as alien at best and impolite at worst; and few have a full-blown Kantian or similar theory of moral agency to ground whatever toleration they may practice. But more interesting than the theoretical possibility is the practical existence of a plurality of foundations for toleration.16 Besides moral-philosophical toleration, the common grounds for toleration currently in use include at least these:
(1) Libertarian/economistic: beyond a strictly limited list of actions that violate others’ concrete rights, most matters of private belief and action are a matter of individual choice, and it is wrong not only to interfere in others’ choices but to judge them morally.
(2) Nonviolent: given the horrors of violence and cruelty, almost nothing would justify coercive interference in others’ lives, regardless of the wrongness of what they do.
(3) Populist/solidaristic: democratic citizens ought to gain, through personal and egalitarian forms of interaction, an active and friendly appreciation for all the ways of life that a pluralistic society makes possible. One engages with others out of a camaraderie that renders cultural, religious, and ethnic differences irrelevant.17
(3a) Rationalist solidarism: we refrain from oppressing others out of “mutual respect” grounded in active engagement with, and reasoned debate about, one another’s moral beliefs and judgments.
(4) Freethinking-elitist: fanaticism or “enthusiasm” is simply silly or vulgar, below the dignity of an intelligent, enlightened person.18
(5) Religious free-conscience (sectarian Protestant, liberal Catholic, etc.): As no one can be saved against his or her will, religious coercion is useless and perhaps self-defeating. Moral agency is not therefore “autonomous”: we all retain a duty to pursue religious truth and obey God’s command.19
(5a) Secular-sectarian, e.g., some versions of feminism and environmental ethics that have nonviolence and/or noncoercion as part of their deeply held, activist, and non-skeptical creeds.
(6) Skeptical: one doubts that one can anyone can have firm reasons for their own convictions or for their disapproval of others.’
(7) Apathetic: one simply does not care much about the big questions that some people are willing to kill or die for.20
(8) Latitudinarian: life objectively presents us with big questions to which religions seek answers. Religiosity of an unspecified variety ought to be publicly encouraged to reflect our serious attitude towards them—but those questions are too mysterious for clear answers to be possible, let alone fit grounds for persecution.21
(9) Anti-clerical or militant secularist: strong religious beliefs are inherently dangerous and demonstrably false. Fortunately the forces of modernity and enlightenment—helped if necessary by state policies to undermine religion’s social power—make religion less common over time, resulting in toleration.22
(10) Principled individualist (Humboldt, one side of Mill, on some accounts Montaigne): Toleration, when politically viable, maximizes the human good because it is objectively and everywhere the case that individuals prosper most fully when they are allowed to make their own mistakes and develop in their own way.
(11) Utilitarian (another side of Mill): Error, in thought and action, should generally be tolerated because the social disputes it engenders help us to remember the reasons that ground currently known truths and to make progress towards learning new ones—but it is still error.
(12) Institutionalist: one may for whatever reason care deeply about moral questions and regard others’ choices as wrong, but one cares even more deeply about political institutions that establish free speech and dissent.
If this list has some surface plausibility both descriptively (those perspectives are out there) and normatively (they are prima facie tolerant rather than actively persecutory), it suggests two conclusions. First, the starting points other people have for thinking about toleration may be very different from what “we” first assume. Given that toleration involves a psychic fight within the agent, even a polity with common political values governing citizen behavior may embody many, very different, kinds of (internal) toleration among citizens. Each will consist in a combination of (1) a folk- or sophisticated belief or judgment that tempts a particular group of people to be intolerant, and (2) the countervailing considerations that stop members of that group from acting on their belief and bring them back to observing tolerant norms.
Toleration is indeed “an elusive virtue.”23 But this is not primarily because religion and other matters of intense moral commitment have become matters of public indifference (this is true in some tolerant polities but not others) nor because liberal states reliably enforce impartial laws that render toleration largely unnecessary (for even if people consistently wanted states to do this, it would not be possible). It is elusive, rather, because virtues—in either the pure or the politically instrumental sense—are always appropriate responses to particular temptations that human beings must habituate themselves out of.24 And while the virtues Aristotle treats respond to fairly universal human temptations (to eat too much, have sex when one should not, talk excessively, run from danger, and so on), the virtues of toleration respond to temptations that are not universal. The temptations involved are relative to a particular subgroup’s world-views and mores, and to how these views and mores characteristically clash (or don’t) with those of other groups in a given society. The countervailing attitudes, reasons, or moral demands that enable tolerant behavior are similarly multiform. There are many views about which moral violations should or should not be judged and attacked and why; and there are many dispositions of mind and character that restrain different people from doing so. Each of the latter can be called one among many virtues (or quasi-virtues) of toleration if it keeps a group of people, to some fairly large extent, from acting coercively against behaviors and beliefs that those people condemn but their practitioners value.
Crucially, some of these temptations, and even some of the considerations that lead to toleration, may strike “us,” or the practitioners of other toleration ideals in general, as unreasonable or even outrageous. Libertarian noninvolvement offends the matey (or deliberative) populist; elitist freethinking, militant secularism, principled individualism, and moral-philosophical toleration offend sectarians and latitudinarians alike; skeptical and utilitarian toleration offend (for different reasons) the institutionalist, the moral philosopher, the sectarian, and many others; apathy offends almost everyone who takes moral issues seriously. Other combinations also clash; and most of these forms of offense are mutual. Nor is the list of tolerations above close to comprehensive.
Moreover, “offense” here is a shorthand for something much more serious. People who hold each of these perspectives often think that only widespread, even near-universal acceptance of that perspective makes common life possible among people who disagree on other things. To reject their perspective is to proclaim oneself indifferent or hostile to the “clearly” necessary basis for social peace and cooperation. In other words, from each of these perspectives, some of the other perspectives—perspectives, again, that to those who hold them justify toleration—seem so anti-social and prone to cause harm (a useful but infinitely controversial concept)25 that it is hard to imagine letting them go unmolested. The question of tolerating the intolerant is often badly posed. Self-aware fascist or communist attacks on toleration as such are (now) rare. More often, people whom one group regards as intolerant appear very tolerant to themselves.
Toleration in a pluralistic society is therefore toleration of tolerations or meta-toleration. We put up not just with other people’s false and potentially dangerous world-views, but with other people’s false and potentially dangerous ways of thinking about why they should put up with ours.
On the account above, Bayle has never triumphed. As neutrality is not universally (or perhaps even widely) recognized as the governing principle of state action, it cannot reliably ground toleration. There is no unified “modern” or “liberal” ethos, described by Heyd as including “rights, respect for individuals, separation of state and church, state neutrality, value pluralism and skepticism,” that “the political culture of constitutional democracy” lives by and through (assuming that a “democratic” culture means one that most people actually recognize).26
In stable liberal democracies, religious, and even ethnic conflicts now rarely turn violent on a large scale. Most citizens lack even the impulse to settle their moral differences through direct physical violence, and the state has the legitimate power to suppress them if they do. This is a huge political gain that makes us immeasurably better off. But we should not therefore conclude that fundamental social conflicts no longer occur, nor that political order rests on a deep moral consensus. After repeated experience of sanguinary state-run or communal persecution of (and by) the intolerant, the reliable human sentiments of fear, conformity, and greed—all much more cunning than reason—have in many countries worn the harshest edges of previously popular worldviews.27 The motley remnants described above are as a result all tolerant in some sense. Inside our lucky countries, we can count on a minimal, “overlapping consensus” toleration under which only marginal individuals and groups want to settle conflicts of private belief with rifles and racks. But the pursuit of more consensus than this is quixotic.28 Only massive persecution of ideological diversity could achieve it; and all versions of protoleration world-views make this (fortunately) unattractive.
But there must be more to the story. Remember that certain of the above views appear to other perspectives not just strange but intolerable, because harmful. Therefore, any coalition of perspectives that commands a steady majority (or, in the less-than-democratic case, whatever other source of authority the state rests on) will be tempted to legally persecute—or at least undermine through de facto discrimination compulsory civic education, and so on—the perspectives it sees as dangerous. The temptation to achieve through official policy “the proper” conditions for toleration will often seem overwhelming, even to those who display the virtue of toleration in one of its many forms. For this course will seem to those tempted not partisan or narrow but on the contrary moral and justified and even necessary (and the other adjectives that a given perspective sees as reasons for state action: “liberal and enlightened” as opposed to superstitious and hypocritical; or “patriotic and public-spirited” as opposed to selfish and divisive; or “civilized and Western” as opposed to atheistic and materialist; or “democratic” as opposed to crypto-fascist; or “reasonable” as opposed to passionate, partial, or moblike).
These battles will shift as political and social power does. The postmodern attack on liberalism is wildly overblown, and the claim that some Nietzschean alternative would give everyone more freedom than liberalism does is absurd. But it remains true that the last two centuries provide myriad examples of what now look like official intolerance and unnecessary social strife occurring when one faction that saw itself as the engine of modern progress, freedom, civilization, reason, or enlightenment succeeded another such faction—and tried to establish a modern, just, and civilized society on a firm foundation by rendering the enemies of same permanently powerless.29 So one reason it is so attractive to posit that toleration relies on a single, stable, unified political culture is that it would seem logically necessary. For if there is more than one such culture, each will try to get rid of the others.
What in fact prevents this outcome is a social accident, though policy can help it along: partisans of various forms of toleration will refrain from persecuting one another if and only if none forms a stable majority and none knows whether its current opponents might be its future allies. This is a Madisonian solution: since people under conditions of free speech will always disagree about toleration, the various factions representing beliefs about toleration—like factions representing religious belief—must be allowed to multiply in the hope that none will become predominant.30 A virtuous circle is possible, though hardly guaranteed. As long as government intolerance does not lead to actual prohibition—or leads to inconsistent prohibitions against different groups at different times—we can expect world-views to multiply beyond measure, making durable persecutory majorities less and less imaginable over time.
Where this works, it has costs: we shall actually have more toleration in practice than any group would like. Many things will be tolerated that from the perspective of one group or another, possibly our own, look not like proper forms of toleration but like unwise and illogical allowances of wrongs or harms. Nor will those who believe in any particular version of toleration have any guarantee that they will gain over time. They will have no systematic reason to think that either the practices they think outside the legitimate bounds of toleration or the social groupings that lend support for such views will lose popularity over time. In fact, it is likely that most such views, and most such groups, will persist indefinitely.
This is, I think, what the United States has today. Hard-core pornography is tolerated, to the horror of religious conservatives (and some feminists) who in their eyes allow freedom of speech and sexual fantasy but draw the line at the “clear harms” that pornography causes to the family and our whole moral order. Gun ownership is tolerated, to the horror of the nonviolent, the utilitarian, and the moral-philosophical, who believe in personal liberty and are neutral towards the hobby of hunting but draw the line at allowing instruments that “must” be directly responsible for high murder rates. Nobody is satisfied; government policy is both incoherent and mocked; likely social harms tend to fester where other societies would probably attack them (because we refuse to act until every group with an ideology, a lawyer, and a lobbyist has had a chance to plead that it is not a harm but their reason for living). But no one can deny that persecution is minimal and liberty flourishes in the absence of clear agreement on what these mean.
This result is not the most attractive one imaginable. The above account entails the triumph of toleration, or meta-toleration, over another ideal, styled either moral progress or civilization. If toleration is an umbrella term for conflicting and partisan views, we can still support political reforms, and need not even renounce the belief that our favored reforms are morally justified. We must renounce, however, the hope that the reforms will ever be uncontroversial, that the justifications will ever be universally accepted. Everything that looks like reform to us will look like persecution to someone else.
But we could do worse. In practice, the attempt to settle publicly and authoritatively “the correct interpretation of constitutional principles and of the idea of the separation of state and religion” can unintentionally, as social forces mobilize and take sides, guarantee a bitterly contested rather than a neutral state. The headscarves issue in France (which Heyd describes as involving not toleration but the question of “the correct interpretation”) pits an American-style religious-pluralist conception of state neutrality directly against the conception embodied in republican laïcité. Because the state rather than society or the individual is regarded as authoritative in these matters, issues of religious, ethnic, and national identity become subjects of constant political mobilization. The quest for a universally valid principle of toleration is what rules out the policies that embody the most obvious and commonsensical forms of toleration: either individual choice, with no suggestion that the State either sanctions or opposes that choice,31 or else regulation by the option of local school authorities, which France practiced for many years (until recently) as politicians blessedly avoided “addressing the problem.”
This chapter has suggested alternatives to common philosophical assumptions about toleration: that the state through a policy of neutrality has largely solved the problem of toleration; and that the individual attitude of toleration only makes sense as a combination of moral judgmentalism towards beliefs and moral respect for human agents. Instead, the modern state (or rather, its agents) cannot help judging, and relies on official and citizen toleration to judge well. Many moral attitudes can be called toleration (reflecting a diversity of worldviews and worldly temptations). And the struggle among these several tolerations is just as potentially contentious as—though far less dangerous than—raw and self-conscious intolerance used to be. What contains this struggle is not moral consensus but the multiplying of toleration-based factions such that none has sufficient certainty of its own power to oppress the others.
We liberals are having to tolerate an awful lot, considering how much we know government policies and others’ social practices to fall short of moral requirements. When we confront those who, through their (politically influential) selfishness, sexism, racism, and propensity to environmental destruction, make the world morally worse than it should be, why should we have to keep in mind all these political and moral-sociological caveats, incapable—as Robert Frost said liberals were—of taking our own side in the argument?
One could respond with a piece of skepticism. Given that metaethics is not a deductively demonstrable science, how sure can we be of our moral judgments? But a political point is more likely to persuade.
Debates about toleration often start from the premise that people like us are the ones doing the tolerating. We should not assume this. True, reflective, open-minded, egalitarian, non-materialist people like us are being asked to tolerate as fellow citizens—immune to persecution, discrimination, and even compulsory reeducation—fundamentalist, fuel-guzzling, warmongering, anti-feminist, ethnocentric fanatics. But one must remember that hardworking, God-fearing, patriotic citizens with strong family values are also being asked to tolerate us: dangerously relativistic, offensively godless, snobby New Class intellectuals who endanger not just our immortal souls but others’ lives (since we clearly lack the moral fiber to fight wars in defense against our enemies). Given the numbers, “we” need toleration more than “they” do—especially in the United States, but to some extent in other countries too.32 There are also many places where they need toleration more than we do. The proper inference to draw is not cynicism but humility. While both we and they should feel free to take our own sides in argument, we should in doing so remember that others view us, not themselves, as the persecutors. And we should resist the temptation to engineer society so as to eliminate the other side to the argument. Toleration’s best friend is the fact of factionalism, not the ideal of neutral statism. And barbarism, in moderation, is to be preferred to persecution.
“Every one is Orthodox to himself”33 is not just a doctrine congenial to liberals but a warning that we liberals should heed ourselves. And the political virtue—or rather virtues—called toleration will seem superfluous or obsolete only if we forget that a democratic “state” can in the long run be tolerant, towards us as well as others, when its citizens are.
1. Dedicated to the memory of Judith N. Shklar. I would like to thank for their useful comments Ron Den Otter, William Galston, Chris Laursen, Melissa Williams, and the ASPLP conference participants generally. Very late in the drafting of these comments I came across what seem to be similar views, expressed very differently, in Jeremy Waldron, “Toleration and Reasonableness,” in Catriona McKinnon and Dario Castiglione, eds., The Culture of Toleration in Diverse Societies: Reasonable Tolerance (Manchester: Manchester Univ. Press, 2003). I have not had time to think through the relation between his argument and mine.
2. Political theorists who claim that toleration is a virtue do so, I think, because they take for granted that “political virtue” means not a full Aristotelian or Stoic virtue, directed at human happiness, but rather any disposition, attitude, or habit instrumentally productive of political goods. For a defense of the distinction see Andrew Sabl, Ruling Passions: Political Offices and Democratic Ethics (Princeton: Princeton Univ. Press, 2002), Chapter 2; idem, “Virtue for Pluralists,” Journal of Moral Philosophy, forthcoming. For claims that it characterizes liberal-democratic political theory both now and historically, see respectively Thomas A. Spragens, Jr., Civic Liberalism (Lanham, Md.: Rowman and Littlefield, 1999), 213 et passim, and John Parrish, “From Dirty Hands to the Invisible Hand: Paradoxes of Political Ethics,” Ph.D. dissertation, Harvard University, 2002. While it is possible to defend toleration for intrinsic rather than instrumental reasons, one does so at the cost of caring more about people’s internal dispositions than about their propensity to confer benefits or harms on others. I would also suggest that judgments about which qualities of mind and character are intrinsically attractive are even more controversial in a plural polity than judgments about which actions are justified. A separate question is which ends the virtue of toleration is supposed to be instrumental towards. This chapter takes for granted that toleration is meant to reduce conflict, coercion, or cruelty—and not, primarily, inequality, which toleration as such does not address one way or the other. I thank William Galston for comments on this last point.
3. David Heyd, “Is Toleration a Political Virtue?” 177.
4. This is a variation on Heyd’s theme: his brief example involves the right to decorate one’s own home in poor taste. But I assume his account of why we may not attack those we believe to be witches would be similar in form.
5. That Heyd follows this narrative broadly is suggested by the following passage: “An enlightened prince as well as a modern liberal state can and should promote the value of interpersonal toleration in society. The government has the power to inculcate standards of toleration by education, the support of institutions in which reason is freely exercised, and even to use its authority and capacity to enforce practices that advance communication and narrow the gap between the public and the private use of reason.”
6. Goldman v. Weinberger, 475 U.S. 503 (1986). While the Supreme Court upheld the military regulation described, Congress later reversed the decision through statute: 10 U.S.C. $774.
7. Employment Division, Department of Human Resources of Oregon, et al., v. Smith et al., 494 U.S. 872 (1990). In that case, contrary to common assumptions, the respondents faced no criminal penalties. Rather, the state had denied them unemployment compensation after they were fired by a private drug rehabilitation organization for using peyote. Thus the state could have accommodated their religious practices by giving them unemployment compensation without forcing their former employer to hire peyote users.
8. Toleration in the sense of accommodating those who violate neutral laws for conscientious and non-dangerous reasons seems (and has seemed to others whose contributions currently escape my notes) to call for similar intellectual and moral virtues to Aristotle’s equity or decency: the quintessential virtue of governing officials who must match laws—which necessarily lack full understanding of prospective cases—to particular instances. Aristotle describes equity as “a sort of justice” even though it goes beyond the decider’s technical entitlements: the person who practices it is “not an exact stickler for justice in the bad way, but tak[es] less than he might even though he has the law on his side.” Nicomachean Ethics, trans. Terence Irwin (Indianapolis: Hackett Publishing Company, 1985), 1138a; for the political official’s use see Politics 1308b27. Though the question of whether equity is supererogatory is anachronistic applied to Aristotle, he does say that it is part of justice. The granting of pardon when circumstances call for it is the just thing to do. John Tomasi, Liberalism Beyond Justice (Princeton: Princeton Univ. Press, 2001) makes a contemporary case for systematically accommodating non-neutral effects, though without using toleration language.
9. Heyd suggests something similar when he notes that toleration among social subgroups “ease[s] political tensions and decrease[s] the level of litigation in society.” But in slighting the dependence of official actions on public judgments, he downplays the extent to which the institutions and practices that he considers political—“constitutional arrangements, the rule of law, or the institutional relations of power and authority”—might depend on a democratized form of equity, perhaps something like the non-official equity that Aristotle calls “pardon” or “consideration” (Nicomachean Ethics 1143a20f.).
10. As noted by Randolph Head, “Religious Coexistence and Confessional Conflict in the Vier Dörfer: Practices of Toleration in Easter Switzerland, 1525–1615,” in John Christian Laursen and Cary Nederman, eds., Beyond the Persecuting Society: Religious Toleration Before the Enlightenment (Philadelphia: Univ. of Pennsylvania Press, 1998), 146.
11. As noted, inter alia, by Scanlon, “The Difficulty of Tolerance,” in David Heyd, ed., Toleration: An Elusive Virtue (Princeton: Princeton Univ. Press, 1996 [henceforth Elusive Virtue]), 226.
12. Bernard Crick, “Toleration and Tolerance in Theory and Practice,” Government and Opposition 6, No. 2 (1971): 144.
13. See the summary by Peter Laslett, “Political Theory and Political Scientific Research,” Government and Opposition 6, No. 2 (1971): 221–22.
14. See David Heyd, “Introduction” to Elusive Virtue, 3–17. Versions of this approach appear in Scanlon (op. cit., 226–39) and Albert Weale, “Toleration, Individual Differences, and Respect for Person,” in John Horton and Susan Mendus, eds., Aspects of Toleration: Philosophical Studies (London and New York: Methuen and Co., 1985), 16–35. Peter Nicholson describes it as “the most promising line of moral argument in favor of an extensive policy of toleration at both the personal and political levels” (“Philosophy and the Practice of Toleration,” in John Horton and Peter Nicholson, eds., Toleration: Philosophy and Practice [Aldershot, England and Brookfield Vermont: Ashgate, 1992], 5).
15. Heyd, “Introduction,” in Elusive Virtue, 15. As far as I can tell, Heyd’s chapter here is consistent with the same assumption: the state is not supposed to care about religious views, or moral ones so far as they are reasonable, but neither moral philosophers nor individuals generally are discouraged from doing so.
16. Note that not all of the below count as foundations for tolerance if one takes that to mean a very specific kind of internal moral disposition. But that begs the question, since I deny that it makes sense to talk about toleration so narrowly (and unpolitically). If a political virtue is a state that conduces to a (desirable) kind of political action, the same external action may be motivated by a great variety of internal states. The list below provides foundations for several virtues that may all be called toleration or tolerance because they lead people to accept things of which they disapprove. I thank Ron Den Otter for discussion on this.
17. Todd Gitlin describes the “Democratic Americanism” of the Popular Front and New Deal as involving “tolerance in the interest of composing a popular commons—a ‘people’s America’ against ‘the interests.’ … The war against fascism became a war of liberation in [sic] behalf of what was distinctly American: the diversity of the demos fused into a single, solid phalanx” (Twilight of Common Dreams [Henry Holt and Company, 1996]), 57, 60. Cf. Saul Alinsky, Reveille for Radicals (New York: Vintage, 1946), chapter 1. Australian political culture—with its ideal of “mateship” and the slogan “a Jack’s as good as his master”—expresses this ideal more durably and less breathlessly.
18. Voltaire’s Treatise on Toleration often expresses this argument, among others.
19. For arguments that the latter sentiment is key to sectarian Protestant arguments for toleration today, see Leif Wenar, “Political Liberalism: An Internal Critique,” Ethics 106, No. 1 (October 1995): 32–62, and for the historical case Joshua Mitchell, “Through a Glass Darkly: Luther, Calvin, and the Limits of Reason,” and Alan Houston, “Monopolizing Faith: The Levellers, Rights, and Religious Toleration,” both in Alan Levine, ed., Early Modern Skepticism and the Origins of Toleration (Lanham, Md.: Lexington Books, 1999), 21–50 and 147–64. On some accounts, Kantianism is a form of sectarianism, though Pelagian rather than Christian. But this seems too pat.
20. This is sometimes glossed, in a more positive vein, as “indifference” (e.g., by Bernard Williams, “Toleration: An Impossible Virtue?” in Heyd, Toleration, 20–21; and in other terms by Peter Laslett, “Political Theory and Political Scientific Research,” Government and Opposition 6, No. 2 [Spring 1971]: 219–20). The difference seems to be one of connotation: apathy means not caring about world-views or social practices that one should care about; indifferentism means not caring when caring would in fact be inappropriate. The distinction’s boundaries are naturally subject to debate.
21. Shirley Letwin, “Skepticism and Toleration in Hobbes’ Political Thought,” in Levine, 174–75, attributes this doctrine to Hobbes, and (somewhat more convincingly) to British Anglicanism up to this day. As she stresses, it cuts directly against many forms of philosophic rationalism and materialism (176).
22. This view is often attributed to Hobbes and Bayle and sometimes Spinoza. In various post-Spinozan and post-Marxist forms it characterizes much of the European Left as well as the “republican” or “radical” Center. For an account and endorsement, see Jonathan Israel, Radical Enlightenment: Philosophy and the Making of Modernity 1650–1750 (Oxford: Oxford Univ. Press, 2001); and more neutrally John Christian Laursen, “Baylean Liberalism: Tolerance Requires Nontolerance,” in Laursen and Nederman, 197–215. In the United States, Deweyan calls for a “common faith” grounded in democratic and scientific ideals rather than obsolete religious ones are the closest analogue, and again exist in both radical and centrist versions (both dedicated to undermining religious schools). Of course there are also moderate forms of anticlericalism which involve private disdain for religion and no particular political program. But these have few implications for toleration.
23. David Heyd, Elusive Virtue.
24. Aristotle, Ethics, 1109a.
25. Kirstie McClure, “Difference, Diversity, and the Limits of Toleration,” Political Theory 18, No. 3 (August 1990): 361–91; also John Horton, “Toleration, Morality, and Harm,” in Horton and Mendus, op. cit., 113–35 and esp. 132: “What liberalism represents as the neutral requirement of preventing harm to others will be perceived by those with different conceptions of what is harmful as the enforcement of a morality they do not share.”
26. One could, of course, define “the political culture of liberal democracy” as including only those who adopt a certain doctrine of toleration; others, whatever their numbers, will then not count as proper “liberal democrats.” Heyd does not make anything like this argument. When some other liberal theorists implicitly or explicitly do make it, they are justifying toleration through an argument that strains the label.
27. This whole argument assumes a skepticism towards rationalist (or crypto-rationalist) claims to the effect that only “valid” ideas, “capable of justification” in a non-relativistic sense, can persist in an open society. It does not require substantive relativism, skepticism towards the idea that some arguments are better than others—merely doubt that the metaphysical quality of an argument has much to do with its empirical popularity. I thank Melissa Williams for forcing me to clarify this.
28. “Certainly, it would be an untidy and unsatisfactory state of affairs if we had to construct a fresh line of argument for toleration to match each different orthodoxy that was under consideration.” Jeremy Waldron, “Locke: Toleration and the Rationality of Persecution,” in Susan Mendus, Justifying Toleration: Conceptual and Historical Perspectives (Cambridge: Cambridge Univ. Press, 1988), 63. With apologies, I claim that is the state of affairs that we face—except that the arguments for toleration need not be constructed “fresh” but exist in rough form embedded within most existing fairly tolerant orthodoxies.
29. See Hans-Ulrich Wehler’s account of how Bismarck’s politics relied on “negative integration,” defining the modern German Empire in terms of the successive groups that had to be persecuted to vindicate it. The Kulturkampf against Catholics, the campaign against “republican” or left-wing liberals, various official and unofficial policies against Jews, and the anti-socialist laws were pursued in the name of liberalism, modernity, and the state standing over partisan and ideological factions that endangered its unity. And all received the full support of the National Liberal party, the one group that was always part of the persecuting coalition. Hans-Ulrich Wehler, The German Empire 1871–1918, trans. Kim Traynor (Leamington Spa, England: Berg Publishers, 1985), 52–118. France’s Second Empire and Third Republic displayed some similar dynamics.
30. James Madison, The Federalist, No. 10.
31. As suggested by The Economist, December 13, 2003, 14—with characteristic contempt for both French statism and the United States’ alleged multiculturalism. To be sure, this kind of “live and let live” proposal slights the problem of private intolerance, of families or bands of radical Muslim youths forcing girls to wear headscarves on pain of ostracism or sexual violence. I am not qualified to assess accounts of how common such things are. But many defenses of French laicism do not stress such harms in any case.
32. As a rough measure: theism as commonly understood is no doubt an unusual attitude among moral, political, and legal philosophers. But fifty-eight percent of respondents in the United States—and about a third of Germans and Canadians—say that it is necessary to believe in God to be moral. (Pew Global Attitudes Project, “Views of a Changing World,” Pew Research Center for the People and the Press, June 2003, 115–16). Much of the toleration literature assumes a context in which “liberal,” “secular,” “neutralist,” “majority,” and “state” are synonyms or close cognates: the question is whether “we” will tolerate traditionalist cultural or religious “minorities” (see, e.g., Melissa Williams, “Tolerable Liberalism,” in Avigail Eisenberg and Jeff Spinner-Halev, Minorities within Minorities (Cambridge: Cambridge Univ. Press, 2005). This might be valid for Canada or Britain, or the United States in the era of Yoder—but seems a tricky assumption applied to Israel or the United States today. On these matters there is no ratchet but rather an unpredictable pendulum. Many no doubt thought that Italy had become “safely” secular until the Berlusconi regime’s policies on sex education and artificial reproduction in Italy—which orthodox Catholics of course defend not as persecutory but as humane and tolerant measures that safeguard the preconditions for respecting persons.
33. John Locke, Letter Concerning Toleration, first paragraph, trans. William Popple, ed. James H. Tully (Indianapolis: Hackett Publishing Company, 1983), 23.