Artikel als PDF herunterladen

Zeitschrift für philosophische Literatur 1. 1 (2013), 93–100

Leiter, Brian: Why Tolerate Religion? Princeton: Princeton University Press 2012. 192 + xv Seiten [978-0-69115361-2]

Reviewed by Martin Sticker, Universities of St Andrews and Stirling

Brian Leiter’s new book aims to be accessible to scholars outside of philosophy as well as to “educated laypeople” (xi). In my view, he succeeds in this endeavour. His book is very readable, and avoids unnecessary technicalities. The question Leiter addresses in his book, “whether there is a moral reason to single out matters of religious conscience for special legal consideration and solicitude” (ix), is of interest not only to academic philosophers, but to everyone who is curious about questions concerning the societal function and role of religion, toleration, minority rights, and conscience. Leiter sets the stage as follows: Sikhs are in many jurisdictions exempt from laws banning weapons in schools. They can bring their ritual dagger – a symbol of their religious devotion. What if a boy for non-religious reasons wants to bring a dagger? Even if he can

“appeal to a century-old tradition, central to his identity […] he is out of luck. The central puzzle in this book is why the state should have to tolerate exemptions from generally applicable laws when they conflict with religious obligations but not with any other equally serious obligations of conscience.” (3).

Leiter begins chapter 1 by noting that, surprisingly, no one has been able to articulate “an argument that would explain why, as a matter of moral principle, we ought to accord special legal and moral treatment to religious practices” (7).

Leiter is interested in what he terms “principled toleration” of conscience and actions based on conscience: A group disapproves of something, has the means to end it, but for moral or epistemic reasons refrains from doing so. There are two types of argument for principled toleration: moral (i) and epistemic (ii). Ad (i): Both deontologists and utilitarians can advance arguments for the right to deviant beliefs, decisions, and actions based on conscience. Deontologists can, for instance, ground this right in an original position in the Rawlsian sense, utilitarians, for instance, in the assumption that having a “private space” can improve our lives. Ad (ii): The epistemic argument is drawn from John Stuart Mill, who argues that we can only discover truth when different beliefs and practices are permitted to flourish, and discovering the truth contributes to utility. In addition, the experience that comes with being exposed to different ways of life helps us find out how best to live our own lives.

All of these positions recognize side-constraints on the scope of toleration. For Rawls, the limits of toleration are damages to liberty worse than non-toleration, and damage to public order; for Mill it is the Harm Principle: power may only be rightfully exercised against the will of an agent to prevent harm to others (see Mill 1978: 9).

In chapter 2, Leiter aims to clarify our notion of religion by finding “features that all and only religious beliefs have, either as a matter of (conceptual or other) necessity or as a contingent matter of fact” (27). Leiter initially suggests two features. Firstly, religious beliefs can demand something to be done no matter what. This categoricity of religious beliefs is “one of the most admirable and one of the most frightening aspects of religious commitment” (36). Secondly, religious beliefs “in virtue of being based on ‘faith’, are insulated from ordinary standards of evidence, and rational justification” (34). Leiter calls this the “insulation from evidence”.

Leiter then discusses whether his account of religion based on these two features is under- or over-inclusive. On the side of under-inclusiveness, it seems that Leiter’s conception omits rationalist conceptions of religion, and Buddhism as a religion. On the over-inclusive side, it appears that his conception encompasses personality cults, and morality as a religion.

Rationalist conceptions of religion (e.g., Paley’s Natural Theology, and Neo-Thomism) seem to be responsive to evidence. Leiter, however, argues that proponents of these conceptions never in fact revise beliefs in light of evidence. Rather, they rationalize ex-post to find evidence for something they hold independently of evidence. He makes essentially the same point against reformed epistemology in chapter 4. Leiter’s brief dismissal of reformed epistemology is polemical and seems unfair, given that this is one of the main sources of opposition to his claim that religion is insulated from evidence. He should have at least discussed reformed epistemology and rationalist conceptions of religion in greater detail.

Concerning over-inclusiveness, Leiter quickly dismisses the idea that Kantianism and Marxism count as a religion in his sense. In a (polemical) endnote (152) and on p. 53, Leiter, however, admits that he would be happy to count Kantianism as a religion. In order to deal with personality cults, Leiter introduces a third feature of religious beliefs: they involve, “explicitly or implicitly, a metaphysics of ultimate reality” (47). “Ultimate reality” refers to what is of ultimate value in life. Personality cults often have only indirect connection to ultimate values, namely, only insofar as the leader is interested in these matters.

The biggest challenge for Leiter’s account is morality. Leiter grants that moral commands, on a standard picture, are categorical. He distinguishes between cognitive realism and non-cognitivist anti-realism. The former is not a religion because it answers to evidence from science and reason, since its claims can be true or false. The latter holds that moral judgments do not express beliefs. Religious judgments, by contrast, do express beliefs and are hence in principle answerable to science and reasons, but they are insulated from it. Morality, in both its cognitivist, and non-cognitivist form, therefore is different from religion.

Leiter’s argument here is very brief, his distinction between cognitive realism and non-cognitivist anti-realism not exhaustive, his treatment of these positions schematic, and in an endnote (152–3) he somewhat swiftly dismisses Kantian constructivism as a viable account of morality. It would have been more promising, I think, to distinguish between pre-theoretical conceptions of morality, and moral philosophy. The former might be insulated from evidence, since they show relatively little reflection. It might count as a religion, but that should not be a problem for Leiter, since many pre-theoretical notions of morality are presumably interwoven with religious conceptions, or even based on them. A more reflected, philosophical conception of morality (cognitivist or not) is, unless exceedingly poor, responsive to evidence (natural facts, reasons, intuitions, peer disagreement, and peer objections, etc.). This form of morality is not a religion.

Leiter concludes the chapter by introducing a fourth condition for religious beliefs, supplementing his framework of categoricity, insulation, and ultimate reality: religion renders “intelligible and tolerable the basic existential facts about human life, such as suffering and death” (52). This existential consolation is supposed to reinforce that personality cults and morality are not religions.

In chapter 3, Leiter discusses whether the specific features of religious beliefs demonstrate that there is a principled reason to tolerate religion because it is religion. According to Leiter, the Rawlsian argument can be dispensed with, since Rawls does not single out religious beliefs for consideration behind the veil of ignorance. Rawls’ argument cannot, therefore, single out religion for special toleration. Leiter is a bit too dismissive here, I think. He offers no principled reason as to why Rawls’ theory cannot be amended to take religious beliefs, and religious liberties in particular, into account. It may simply be the case that Rawls did not have Leiter’s specific question in mind when he construed his general theory. Adams (2013), for instance, in his review of Leiter’s book, argues that a Rawlsian account can make significantly more space for religious conscience than Leiter assumes.

Leiter then turns to the Millian epistemic argument. Leiter argues that religion fares badly on the epistemic argument, due to religion’s insulation from our best conception of evidence and science. Hence, there is no special epistemic reason for religion to be tolerated. An alternative Millian argument for a specific toleration of religion is that religion helps us determine how to live. It is, however, a problem for every argument for tolerating religion qua religion on utilitarian grounds that one could develop in the Millian spirit that categorical commands in combination with insulation from evidence have the potential to infringe on others and to cause harm, i.e., to violate the Harm Principle.

Utilitarian arguments for toleration of religion can seek support in the fourth feature of religious belief: existential consolation. Existential consolation, no doubt, generates utility. There are, however, two questions: Does the utility of consolation outweigh the potentially harmful effects of religion? Is there another way to bring about consolation without a combination of insulation and categoricity? Leiter answers the second question in the affirmative: There is philosophy, meditation, therapy, etc. Thus, there can be a rationale for tolerating religion qua religion

“only if we are willing to speculate that the existential consolation functions of religion produce more utility than the harm produced by the conjunction of categoricity and insulation from evidence; and only if we are willing to speculate that the preceding net gain in utility would be greater than the alternative ways of producing existential consolation without the conjunction of categoricity and insulation from evidence” (63).

Why one would be willing to engage in this speculation “absent an antecedent bias in favour of religion” (63) is, according to Leiter, not clear at all. Leiter concludes: There is no principled reason to single out religion for legal or constitutional toleration. If matters of religion deserve toleration, they do so qua being matters of conscience not qua being matters of religion.

After arguing that there is no principled reason to tolerate religion qua religion, chapter 4 is dedicated to critical discussion of Martha Nussbaum’s claim that religion should not be merely tolerated but respected. The most important reason why religion, according to Leiter, does not merit a special kind of respect, called “appraisal respect”, is that religious beliefs are (epistemically) culpably false: They are unwarranted and the believers know this. According to Leiter,

“[o]nly if there were a positive correlation between beliefs that were culpably without epistemic warrant and valuable outcomes would it seem that we should think them proper objects of appraisal respect. But the evidence on this score is, as we have already had occasion to note, mixed” (84–5).

This holds even if religion can lead, and historically has led, to “moral gems” (86), such as resistance to racism and fascism. It is a contingent, empirical question whether or not religion often enough produces these gems to outweigh its negative effects.

In a last paragraph, Leiter allows Nietzsche to come to the aid of religion. Leiter endorses Nietzsche’s idea that falsity of belief or lack of warrant is not necessarily a reason to object to a belief. Some beliefs may be necessary conditions for life itself, regardless of their truth and warrant. In this spirit, Leiter does not think that we should withhold toleration from religion or disrespect it for its epistemic deficiencies.

In chapter 5, Leiter applies his conclusions to address the de facto legal status of religion. There are principled arguments for the toleration of conscience, but these arguments do not pick out religious conscience as distinct or special. Yet, in western societies religious conscience enjoys special exemptions from general applicable laws.

The first reaction to this disparity between what arguments can establish, and how legal systems in fact work, is to afford the same toleration to all claims of conscience. However, protecting all claims of conscience would be tantamount to constitutionalizing a right to civil disobedience in potentially all cases, or to “a legalization of anarchy” (94). An advantage of religious claims for toleration over claims of toleration based merely on conscience is that it is easier for courts to check whether these are genuine claims of conscience. Privileging liberty of religious conscience gives courts a “more robust evidential base for their determination” (95). Instead of looking into someone’s soul, courts can look into scripture, oral evidence, ask experts, etc. Extending toleration to all groups that are like religions in the sense that they are organized, have traditions, practices, etc., however, would not be sufficient. Some claims of conscience would still be discarded merely because the public evidence for their genuineness is too weak. This seems unfair.

There is another objection to general exemptions for conscience: Exemptions usually (though not always) shift the burden on to others who lack the exemption. Leiter calls these exemptions “burden-shifting” (100). Even if everyone who meets the criteria for an exemption is exempt, there is something unfair about these exemptions, since some people will not meet the criteria and have to bear a greater burden. An example for this is military service: The members of a religious group might be exempted from being drafted into the army. This presumably entails that more of the non-members will be drafted, whether they want it or not.

The second reaction to the problem of exemptions for religious conscience only, and to the inability to establish principled toleration for religion qua religion, is to never allow burden-shifting exemptions from general laws. This no-exemptions approach would eliminate the unfairness resulting from a situation in which non-religious matters of conscience are not exempt, whereas religious matters are (or are more likely to obtain an exemption). Leiter discusses the example of the French Burka ban in public places for the sake of laïcité as a seeming example for the no-exemptions approach (100–114). Leiter dismisses laïcité as an unnecessary strong candidate of the no-exemptions approach, because it wants to ban religion entirely from the public sphere. One can realize the no-exemptions approach without laïcité.

Leiter argues that it is legitimate for a liberal state to promote a notion of the good that violates the conscience of some of its members, if it is not the goal of this procedure to violate their conscience or to coerce consciences of minorities beyond what the Harm Principle would licence. Leiter concludes: The no-exemptions approach is “the one most consistent with fairness (given the practicalities of enforcement)” (130-131). Leiter acknowledges that there will always be some people (religious or not) who, based on their conscience, will refuse to comply with general laws. These people might be right or wrong, still that is no argument against No-Exemptions: “Toleration may be a virtue, both in individuals and in states, but its selective application to the conscience of only religious believers is not morally defensible” (133).

Besides the more specific critical points which I have already made, a more general critical remark is in order. According to Leiter, one of the crucial features of religious belief, which in combination with insulation from evidence fuels Leiter’s sceptical take on the role of religion in society, is categoricity. Categoricity is usually discussed as a property of moral commands (hence Leiter’s discussion whether morality counts as religion). First of all, one can have doubts that moral claims or beliefs are always, or even usually, accepted as categorical even by reasonably moral persons. Leiter seems to acknowledge this when he objects to Kant that there are very few people who really orient their life on a categorical conception of morality (for some classical challenges to the notion that morality is categorical, overriding, or normatively superior see Foot 1978, Williams 1981, Wolf 1982, Slote 1983).

Categoricity is, I think, even more debatable in the case of the phenomenon of religion. A central feature of religion is that it offers a way to come to terms with evil, insecurity about the future, finitude, fallibility, questions of meaning, and morality. Leiter captures this partly when he acknowledges the existential consolation function of religion, but he does not see that one of the consoling aspects of religion for many believers is that they will not have to sacrifice too much to obtain the benefits of religion. After all, Gods are, in most religions, generally quite forgiving, and aware of human imperfection. Though some people are willing to become martyrs, I doubt that the ordinary believer will usually heed religious commands when their life or livelihood is at stake. This is partly because institutionalized forms of religion tend to have traditions of rationalization: more or less sophisticated means to allow oneself to not follow onerous commands, whilst maintaining a self-conception as a reasonably pious person. Human imperfection is already built into the framework of many religions. Whilst this, strictly speaking, does not show that religious commands are not categorical, it casts doubt on whether religious commands really figure in the mental life of many ordinary believers in a categorical way in any meaningful sense and as envisaged by Leiter.

It seems to me that Leiter’s paradigm of a believer is a zealot who is willing to follow (supposed) divine commands regardless of what this would mean for him or herself, and others, and no matter what the evidence suggests. Most believers, I take it, have a more distanced relation to their deities and their religious establishment, and they are able to keep their own interests well in mind. Categoricity seems too strong and crude to describe such a susceptibility to religious commands. Closer to the truth is something like the following: religious beliefs or commands are important for religious people, they provide them with reasons for actions, and believers are usually unwilling to openly violate their religious commands and beliefs without something they take to be a sufficient justification or excuse, preferably drawn from resources of their particular religion (papal dispenses, lenient passages in holy scripture, reinterpretation of passages, figures of religious authority, etc.). In addition, believers of pretty much any religion are aware that if they are to violate religious commands, there are many roads to redemption, forgiveness, atonement, etc. afforded by their religion (hence going against the will of God(s) is hardly the end of the world).


Adams, Robert. “Brian Leiter: Why Tolerate Religion?” In Notre Dame Philosophical Reviews (2013), ‹› (8 December 2013).

Foot, Philippa. “Are Moral Considerations Overriding?” In Virtues and Vices. Berkley: University of California Press, 1978.

Mill, John Stuart. On Liberty. Indianapolis: Hackett, 1978.

Slote, Michael. “Admirable Immorality.” In Goods and Virtues. New York: Oxford University Press, 1983.

Williams, Bernard. “Moral Luck.” In Moral Luck. New York: Cambridge University Press, 1981.

Wolf, Susanne. “Moral Saints.” The Journal of Philosophy 79.8 (1982), 41939.

© 2013 Zeitschrift für philosophische Literatur, ISSN 2198-0209, lizenziert unter CC-BY-ND-3.0-DE


  • Im Moment gibt es keine Refbacks



ISSN 2198-0209  -  Twitter - Facebook - Datenschutzerklärung