Toleration
Religious toleration has long been the paradigm of the liberal ideal of toleration of group differences, as reflected in both the constitutions of the major Western democracies and in the theoretical literature explaining and justifying these practices. The American Constitution provides that “Congress shall make no law respecting an establishment of religion, or prohibiting the free exercise thereof.”1 As the German Constitution (or “Basic Law”) provides in Article 4, “Freedom of faith and of conscience, and freedom to profess a religious or philosophical creed, shall be inviolable,” adding, in a separate clause, “The undisturbed practice of religion shall be guaranteed.”2 The first of the four “Fundamental Freedoms” in the Canadian Charter of Rights and Freedoms is held to be “freedom of conscience and religion.”3 And Article 18 of the Universal Declaration of Human Rights declares,
Everyone has the right to freedom of thought, conscience and religion; this right includes freedom to change his religion or belief, and freedom, either alone or in community with others and in public or private, to manifest his religion or belief in teaching, practice, worship and observance.4
While the American Constitution only mentions religion,5 the recognition for claims of “conscience” in the other documents is more perfunctory than substantive: litigated cases overwhelmingly involve claims of religious conscience.6 Indeed, if claims of religious conscience were not really the primary concern in each case, then surely the explicit mention of religion (or the mention of special protections for religion) would appear redundant on the protection purportedly afforded claims of conscience more generally.
While the historical reasons for the special “pride of place” accorded religious toleration are familiar,7 what may be more surprising is that no one has been able to articulate a credible principled argument for tolerating religion qua religion—that is, an argument that would explain why, as a matter of moral principle, we ought to accord special legal and moral treatment to religious practices. There are, to be sure, principled arguments for why the state ought to tolerate a plethora of private choices and conscientious commitments, as well as related practices of its citizenry, but none of these single out religion for anything like the special treatment it is accorded in existing Western legal systems. So why tolerate religion? The answer in this book is: not because of anything that has to do with it being religion as such—or so I shall argue.
Principled Toleration
To see why this is so we will need to start, though, with some distinctions that make possible a more perspicuous formulation of the question. In particular, we need to state clearly what is at stake in something called a principle of toleration. I shall take as a point of departure a useful formulation of the issues by the late English philosopher Bernard Williams:
A practice of toleration means only that one group as a matter of fact puts up with the existence of the other, differing, group…. One possible basis of such an attitude … is a virtue of toleration, which emphasizes the moral good involved in putting up with beliefs one finds offensive…. If there is to be a question of toleration, it is necessary that there should be some belief or practice or way of life that one group thinks (however fanatically or unreasonably) wrong, mistaken, or undesirable.8
For there to be a practice of toleration, one group must deem another differing group’s beliefs or practices “wrong, mistaken, or undesirable” and yet “put up” with them nonetheless. That means that toleration is not at issue in cases where one group is simply indifferent to another. I do not “tolerate” my neighbors who are nonwhite or gay because I am indifferent as to the race or sexual orientation of those in my community. Toleration, as an ideal, can only matter when one group actively concerns itself with what the other is doing, believing, or “being.” Obviously, in many cases, the attitude of indifference is actually morally preferable to that of toleration: better that people should be indifferent as to their neighbors’ sexual orientation than that they should disapprove of it but tolerate it nonetheless.
But a practice of toleration is one thing, a principled reason for toleration another. Many practices of toleration are not grounded in the view that there are moral reasons to tolerate differing points of view and practices, that permitting such views and practices to flourish is itself a kind of good or moral right, notwithstanding our disapproval. Much that has the appearance of principled toleration is nothing more than pragmatic or, we might say, “Hobbesian” compromise: one group would gladly stamp out the others’ beliefs and practices, but has reconciled itself to the practical reality that it can’t get away with it—at least not without the intolerable cost of the proverbial “war of all against all.” To an outsider, this may look like toleration—one group seems to put up with the other—but it does not embody what Williams called a “virtue” of tolerance (or what I will call “principled tolerance”), since the reasons for putting up are purely instrumental and egoistic, according no weight to moral considerations. One group puts up with the other only because it would not be in that group’s interest to incur the costs required to eradicate the other group’s beliefs and practices.
Yet it is not only Hobbesians who mimic commitment to a principle of toleration. On one reading of John Locke,9 his central nonsectarian argument for religious toleration is that the coercive mechanisms of the state are ill-suited to effect a real change in belief about religious or other matters. Genuine beliefs, sincerely held, can’t be inculcated at gunpoint, as it were, since they respond to evidence and norms of rational justification, not threats.10 In consequence, says the Lockean, we had better get used to toleration in practice—not because there is some principled or moral reason to permit the heretics to flourish but because the state lacks the right tools to cure them of their heresy, to inculcate in them the so-called correct beliefs.
Locke, it is fair to say, did not fully appreciate the extent to which states and—in capitalist societies—private entities can employ sophisticated means to effectively coerce belief, means that are both more subtle and more effective than he imagined. That history offers up so many examples of societies in which the tyranny of the few over the many is accepted by the many as a quite desirable state of affairs is compelling evidence that states can successfully inculcate beliefs, even dangerously false beliefs. Locke’s “instrumental” argument for a practice of toleration should provide little comfort to the defender of toleration given his (understandable) failure to appreciate the full complexity of the psychology and sociology of belief inculcation.
Not only Hobbesians and Lockeans, however, mimic principled toleration. A variation on the Lockean instrumental argument for toleration is apparent in a popular theme in American political thinking—one that receives a well-known articulation in Frederick Schauer’s defense of free speech11—according to which government cannot be trusted to discharge the task of intolerance “correctly”—that is, in the right instances. Speech can harm, in all kinds of ways, notes Schauer, and the various rationales for putting up with these harms—from John Stuart Mill’s “marketplace of ideas” to Alexander Meiklejohn’s conception of free speech as essential to democratic self-government—almost all fall prey to objections of one kind or another. But, says Schauer, there is still a reason to demand that the state tolerate many different kinds of speech (even harmful speech), and that is because there is no reason to think the state will make the right choices about which speech ought to be regulated. Schauer calls this “the argument from governmental incompetence” and writes,
Freedom of speech is based in large part on a distrust of the ability of government to make the necessary distinctions, a distrust of governmental determinations of truth and falsity, an appreciation of the fallibility of political leaders, and a somewhat deeper distrust of governmental power in a more general sense.12
It is not, then, as in the Lockean argument, that government lacks the right means for bringing about intolerant ends; it is rather that government is not competent—that is, cannot be relied upon—to deploy its means in the right cases. Perhaps this kind of instrumental argument for state toleration is more plausible, but its justificatory structure makes it no different from that of the Lockean’s: it does not tell us why we, morally, ought not to eradicate differing beliefs or practices, it tells us only that we (through the instrumentality of the state) are unlikely to do it right.
Where a genuine “principle of toleration” gets its purchase is in the cases where one group (call it the “dominant” group) actively disapproves of what another group (call it the “disfavored” group) believes or does; where that dominant group has the means at its disposal to effectively and reliably change or end the disfavored group’s beliefs or practices; and yet still the dominant group acknowledges that there are moral or epistemic reasons (that is, reasons pertaining to knowledge or truth) to permit the disfavored group to keep on believing and doing what it does. That is “pure” or “principled” toleration,13 and the question, then, is whether there is such a reason to tolerate religion.
My concern here shall mainly be with the principled grounds of state toleration, as opposed to toleration in interpersonal relations, though the issues are often similar. Some contemporary “liberal” philosophers think the right posture for the modern state is one of neutrality, not toleration, with the disapproval the latter implies. But I reject the view that any state can really be neutral in this way; as I will argue in chapter 5, every state stands for and enacts what I call a “Vision of the Good”—even if most Western democratic states no longer endorse such a distinctively religious vision. The American state has decided not only that “liberty and equality for all” are fundamental values but that all children must learn Charles Darwin’s theory of evolution by natural selection, yet they do not need to “know” anything about the biblical view of creation. The American state is not at all neutral with respect to those sects that reject “liberty and equality for all,” let alone those that think creation myths are on a par with biology and deserve equal time in the public schools. Toleration thus remains a virtue for the liberal state, as it does for the individual. Even if one thinks states can aspire to more neutrality than I suppose, it is still the case that when particular minority claims of conscience, religious or otherwise, assert the need to be exempted from neutral laws of general applicability, what they are demanding is not neutrality but something like the virtue of toleration—that is, they are demanding that the state suspend its pursuit of the general welfare in order to tolerate (i.e., “put up with”) a conscientious practice of a minority of its citizens that is incompatible with it. That is why the central question in this book is: what are the principled reasons why the state should exempt religious claims of conscience from the burden of its laws? I frame the problem in these terms because, even though the historical problem about religious toleration was generated by conflict among religious groups, the contemporary problem, at least in the post-Enlightenment, secular nations (of which the United States may still be one) is different: it is why the state should tolerate religion as such at all.
Arguments for Principled Toleration
Before we consider religious tolerance in particular, it will be useful to consider the general structure of principled arguments for state toleration of group differences. The literature on the subject is voluminous, so necessarily I will be able to consider only a few themes here. Yet the themes I emphasize will, I believe, capture the main principled positions in the debates.
We can distinguish between two broad classes of principled arguments for toleration, which I will call moral and epistemic (though the latter ultimately rests on moral considerations as well). The strictly moral arguments for toleration claim either that there is a right to the liberty to hold the beliefs and engage in the practices of which toleration is required; or that toleration of those beliefs and practices is essential to the realization of morally important goods. The moral arguments divide, predictably enough, into Kantian and utilitarian forms.14
As paradigmatic of the broadly Kantian arguments, consider the Rawlsian theory of justice according to which “toleration … follows from the principle of equal liberty,”15 one of the two fundamental principles of justice that, Rawls argues, rational persons would choose in what he calls the “original position”—that is, a situation in which people choose the basic principles of justice to govern their societies, and in which they do so deprived of the kind of information about their place in society that would render their judgments partial and self-serving. As Rawls puts it,
[T]he parties must choose principles that secure the integrity of their religious and moral freedom. They do not know, of course, what their religious or moral convictions are, or what is the particular content of their moral or religious obligations as they interpret them…. Further, the parties do not know how their religious or moral view fares in their society, whether, for example, it is in the majority or the minority…. [E]qual liberty of conscience is the only principle that the persons in the original position can acknowledge. They cannot take chances with their liberty by permitting the dominant religious or moral doctrine to persecute or to suppress others if it wishes. Even granting … that it is more probable than not that one will turn out to belong to the majority (if a majority exists), to gamble in this way would show that one did not take one’s religious or moral convictions seriously, or highly value the liberty to examine one’s beliefs.16
Notice that nothing in this argument is specific to religion: the argument, as Rawls says quite clearly, is on behalf of rights securing “liberty of conscience,” which can include, of course, matters of conscience that are distinctively religious in character but are not limited to them.17 The argument depends only on the thought that persons in the “original position” know that they will have certain convictions about how they must act in certain circumstances—convictions rooted in reasons central to the integrity of their lives.
The utilitarian arguments have a similar feature—namely, that they do not obviously single out religion for special consideration as opposed to other important matters of conscience. These arguments come in many different varieties, but all share, in one form or the other, the core idea that it maximizes human well-being—however exactly that is to be understood—to protect liberty of conscience against infringement by the state.18
Why does it promote human well-being to protect liberty of conscience? Many arguments trade, at bottom, on a simple idea: namely, that being able to choose what to believe and how to live (within certain side-constraints, about which more shortly) makes for a better life. Being told what you must believe and how you must live, conversely, make lives worse. I shall gloss this simple thought as the “private space argument.” It maximizes human well-being, so the argument goes, if, within certain limits, individuals have a “private space” in which they can freely choose what to believe and how to live.
Is it true that granting individuals a private space maximizes human well-being? Could it be that many, perhaps even most, individuals make themselves miserable—that is, worse off—precisely because they make foolish choices about what to believe and how to live?—Or perhaps because they don’t make real choices at all, being hostage to social and economic milieux and enjoying only the illusion of choice? These illiberal thoughts—familiar to readers of Plato, Karl Marx, and Herbert Marcuse, among many others—have little purchase these days within the mainstream of English-speaking moral and political theory, though not, as far as I can tell, because they have been refuted systematically.19 For the sake of argument here, I shall put these doubts to one side and grant that the private space argument is plausible, and thus states a utilitarian ground for toleration. Notice, too, that it states a more capacious ground for liberty of choice than the Rawlsian argument, which seems delimited to matters of conscience (matters, for example, about which there are very weighty reasons, central to personal integrity and so on).
In contrast to the moral arguments for toleration that we have just considered, epistemic arguments for toleration emphasize the contribution that tolerance makes to knowledge. Such arguments find their most systematic articulation in the work of John Stuart Mill. According to Mill,20 toleration is necessary because (1) discovering the truth (or believing what is true in the right kind of way) contributes to overall utility; and (2) we can only discover the truth (or believe what is true in the right way) in circumstances in which different beliefs and practices are permitted to flourish. The first premise in the Millian argument for toleration is, quite obviously, a moral one: we should care about the truth (or believing the truth in the right kind of way) because of the contribution that makes to the morally valuable end of utility. Friedrich Nietzsche, among others, denies the moral premise: the “truth is terrible,” he says,21 by which he means precisely that sometimes knowing the truth is incompatible with life, a fortiori, with utility (though utility was not, of course, Nietzsche’s particular concern).
It is only the second premise of the Millian argument for toleration that is distinctively epistemic, for it is this premise that claims that toleration of divergent beliefs and practices contributes to knowledge of the truth. Note that the “truth” at issue for Mill concerns both truths about “facts” as well as truths about “value”—in particular, moral truths about the best kinds of lives available to creatures like us. From an epistemic point of view, both factual and moral truths have several features in common. First, in neither case are we justified in assuming that we are infallible: we may be wrong, and that is a reason to permit dissident opinions, which may well be true. Second, even to the extent our beliefs are partially true, we are more likely to appreciate the whole truth to the extent we are exposed to different beliefs that, themselves, may capture other parts of the truth. Third, and finally, even to the extent our present beliefs are wholly true, we are more likely to hold them for the right kinds of reasons, and thus more reliably, to the extent we must confront other opinions, even those that are false. For all these epistemic reasons, toleration of a wide array of expression of differing beliefs is warranted according to Mill.
Moral truths, however—that is, truths about how we ought to live—supply the ground for a wider scope of toleration, one that encompasses practices, not just beliefs. For the epistemic conditions for the discovery of moral truths require not only that we be exposed to differing beliefs, but that, as Mill puts it, “the worth of different modes of life should be proved practically” through what he calls “experiments in living.”22 In other words, to know how we really ought to live, it is not enough to hear differing opinions expressed on the subject; one must have the empirical evidence provided by lives actually lived in accordance with different guiding principles. It is only, for example, by seeing (or, better yet, experiencing) the lives of a pig satisfied and Socrates dissatisfied (in Mill’s famous example) that we can come to the knowledge that the latter life is better—that is, involves higher-quality pleasures—than the former.
Before we turn to the special case of religious toleration, we need to call attention to one more feature of principled arguments for toleration: namely, that they all recognize side-constraints on the scope of toleration.23 Even if there is a right to liberty of conscience that demands state tolerance of differing beliefs and practices, as Rawls holds; or even if toleration promotes overall utility or happiness—or facilitates a kind of knowledge that promotes overall utility—as the utilitarian arguments hold; it is still the case that there are limits on how much toleration the state must display toward acts of conscience.
For the Rawlsian, “The limitation of liberty is justified only when it is necessary for liberty itself, to prevent an invasion of freedom that would be still worse,” so “liberty of conscience is to be limited only when there is a reasonable expectation that not doing so will damage the public order which government should maintain.” “This expectation,” Rawls also says, “must be based on evidence and ways of reasoning acceptable to all….”24 For the utilitarian, by contrast, the side-constraints on toleration are typically set by some version of Mill’s famous Harm Principle, according to which “the only purpose for which power can be rightfully exercised over any member of a civilized community, against his will, is to prevent harm to others.”25
On either theoretical approach, the limits of tolerance are set by the liberty interests or well-being of others in the community, and these limits have their primary impact not on toleration of beliefs but on toleration of the practices or actions undertaken in accord with those beliefs. The state will still, on either the Rawlsian or Millian view, it seems, have to tolerate some religious group’s belief that adherents of all other religions are heretics, destined for damnation; but the state need not tolerate that same group’s desire to act on its beliefs by, for example, killing the infant children of the alleged heretics before their souls are corrupted, and thus eternally damned, by heresy.
Cases like these are, of course, the easy cases on any view of toleration and its limits. Much harder are two other kinds of cases: first, those involving the expression of beliefs that have as their probable (but not certain) consequence actions that infringe upon liberty or are otherwise likely to cause prohibited harms; and second, those involving practices or actions that have as their probable (but not certain) consequence the infringement of liberty or the causing of prohibited harms.
Mill is, of course, thinking of the first category of cases when he writes,
[E]ven opinions lose their immunity when the circumstances in which they are expressed are such as to constitute their expression a positive instigation to some mischievous [i.e., harmful] act. An opinion that corn dealers are starvers of the poor, or that private property is robbery, ought to be unmolested when simply circulated through the press, but may justly incur punishment when delivered orally to an excited mob assembled before the house of a corn dealer.26
That same thought is codified in American constitutional law by the doctrine that speech that poses what used to be called a “clear and present danger” can be suppressed by the state.27 This approach supposes that you can hold and express any belief unless there is a “tight” causal nexus between expression of the belief and forbidden acts. The Rawlsian view seems to come to the same thing, though the metaphors Rawls employs are different: the threat to liberty, he says, for example, “must be securely established by common experience.”28 The Rawlsian formulation does not as obviously incorporate a requirement that the resultant harm be as immediate or imminent as do the Millian examples or the American constitutional doctrine. It suffices on the Rawlsian view that the causal nexus between expression of belief and liberty-infringing act be “securely established.” To be sure, the criteria for securely establishing that nexus may only be satisfied in the same cases of immediate or imminent harmful conduct contemplated on the Millian view, which is reason to think they come to the same thing.29
The second category of cases presents the same issue: that is, there are practices based on beliefs that it seems ought to be tolerated (on either the Millian or Rawlsian view), about which we can ask whether those practices might stand in a causal nexus with harm that satisfies the applicable evidential standards. A 2006 Canadian case, Multani v. Commission scolaire Marguerite-Bourgeoys,30 involving the right of Sikhs to carry the kirpan (a ceremonial knife), as required by their religion—even in schools—illustrates this issue. Those opposing the practice argued, in part, that this religious practice poses too great a risk of harm, reflected on the general ban of weapons in school; the other side argued, by contrast, that the probability of harm was very slight, as evidenced, for example, by the fact that there was no known instance of a kirpan being used as a weapon. The Canadian Supreme Court, of course, opted for toleration of the practice of carrying the kirpan, given the importance of the practice to the believers, the putatively slight risk of harm, and the special value multiculturalism is assigned in the Canadian Charter. (We shall return to this case in chapter 3.)