Is religious conscience special? And what kinds of claims (if any) does conscience warrant? These are two of the many questions Brian Leiter raises in his provocative book Why Tolerate Religion? (Princeton University Press, $24.95, 192 pp.).

Note that in principle one could answer the first question in the negative—by denying the distinctiveness of religion—while endorsing broad claims for conscience as such. Imagine a two-by-two table: In the upper left quadrant is an expansive notion of conscience coupled with a broad conception of conscientious claims; in the bottom right is conscience restricted to religion with few or no claims to which the law must yield. The two remaining quadrants are broad/narrow and narrow/broad, respectively.

In the middle decades of the twentieth century, the prevailing view combined a narrow conception of conscience (restricted to religion) with a capacious understanding of conscientious claims as warranting, in suitable circumstances, exemption from generally valid public laws. This view then came under pressure, from two directions. During the Vietnam era, in response to claims for exemption from the draft, the Supreme Court expanded the perimeter of conscience to include explicitly secular beliefs that "play the role of a religion and function as a religion in life" (Welsh v. United States, 1970).

In 1990, the other shoe dropped. In Employment Division v. Smith, a decision that has remained controversial ever since, Justice Antonin Scalia rejected a claimed exemption from drug laws for peyote used in Native American religious rituals. Granting this claim, he argued, would create a system "in which each conscience is a law unto itself." A society that did this would be "courting anarchy."

It was not the expansive concept of conscience that worried Scalia; it was the core meaning: "actions thought to be religiously commanded." The more religiously diverse a society, the more such actions there will be, covering an ever greater sphere of social life and public law. Acting through their elected representatives, the people may carve out exceptions for religious individuals and institutions. But religion does not enjoy exemption from law as a matter of constitutional or moral right.

I have summarized the Smith decision in some detail because Leiter embraces a nuanced version of its central holding. Does it make sense? I propose to test it in two ways—constitutionally and philosophically. Let's begin with perhaps the least successful constitutional experiment in American history.


The Eighteenth Amendment was ratified on January 16, 1920. It was widely understood that without the concurrent legislation authorized in Section 2, the general prohibition on the manufacture, sale, and transportation of alcoholic beverages would be too vague to enforce. On October 29, 1919, the National Prohibition Act (popularly known as the Volstead Act), which created the legal definition of "intoxicating liquor" and specified penalties for producing it, passed over President Woodrow Wilson's veto and stood as the law of the land until 1933.

The Volstead Act created a number of exemptions to the prohibition regime, of which two are especially noteworthy. First, the act allowed physicians to prescribe liquor to individuals for medicinal purposes and to employ it pursuant to treatment for alcoholism in certified treatment programs. Second, the act stated that nothing it contained should be construed as applying to "wine for sacramental purposes, or like religious rites," and it permitted the sale or transfer of wine to rabbis, ministers, priests, or an officer duly authorized by any church or congregation.

Suppose the act had not exempted physicians. The omission would have been subject to criticism on policy grounds, but no one would have suggested that it ran afoul of constitutional norms. If the act had failed to exempt wine for sacramental purposes, however, there would have been both a political firestorm and a First Amendment challenge that almost certainly would have succeeded.

The use of sacramental wine lies at the heart of more than one religion. The Code of Canon Law of the Catholic Church prescribes that "the most holy Sacrifice of the Eucharist must be celebrated in bread, and in wine to which a small quantity of water is to be added." For its part, Jewish law commands the drinking of wine during the Passover Seder, specifying not only the famous four cups but also a minimum quantity to be consumed. (There is no maximum.) Comprehensive prohibition without exemptions would have prevented faithful Jews and Catholics from behaving as their religion requires. The Constitution's presumption in favor of free exercise is designed to reduce to an avoidable minimum the circumstances in which such clashes are resolved in favor of the state. If free exercise means anything, it means the liberty to conduct the mandatory rites of one's faith.

As Leiter rightly points out, this liberty is not absolute; there are what he calls "side-constraints" on its scope and exercise. A neo-Aztec religion could not claim moral or constitutional protection for human sacrifice, however central to its beliefs that ritual might be. While adult Christian Scientists may spurn standard medical practices, parents may not withhold treatment when the life of their child is at stake. A denomination might claim that God commands them to evangelize, but free exercise doesn't give it the right to conduct a revival meeting at 2 a.m. in a residential neighborhood. In such circumstances, religious noise is to be treated the same way as its secular counterpart.

There are, in short, some bedrock civil concerns that the law may enforce, regardless of their effects on particular religions. But for most of our national history, legislators and jurists distinguished between such concerns and the more typical objects of legislation, which were thought not to be so fundamental as to outweigh religious free exercise. Despite the obvious importance of communal self-defense, many colonies exempted Quakers (.pdf) from serving in battles against the French and Native Americans, an exemption that some colonies continued during the Revolutionary War. Madison and the members of the First Congress, who drafted and endorsed the First Amendment, were well aware of this history.

Indeed, it was Madison who formulated the basis for religious accommodation in his famous Memorial and Remonstrance: "It is the duty of every man to render to the Creator such homage and such only as he believes to be acceptable to Him. This duty is precedent, both in order of time and degree of obligation, to the claims of Civil Society." Many believers consider themselves to be subject to two authorities, one human, the other divine. In cases of conflict between them, the faithful believe that God's authority is paramount. As we have seen, civil authority cannot accept every manifestation of this belief: political communities have minimum conditions that they must enforce, come what may. But when the claims of faith pull against civil law, there is a rebuttable presumption that the latter should give way.

It is hard to square the Smith decision with the most plausible understanding of our constitutional traditions. I can see no principled distinction between laws that have the effect of banning the use of peyote in Native American rites and versions of Prohibition that would have outlawed the use of wine in the Mass and the Seder. If the former is consistent with conscience, then so is the latter.


The First Amendment unmistakably singles out religion. Congress may not establish a religion, either by giving it a preferred institutional position or by using its distinctive doctrines as the basis for legislation. But as Leiter points out, there is nothing in the Constitution to stop Congress from establishing a secular doctrine. For example, it can create and fund an economic board whose membership is restricted to Keynesians (or supply-siders), and it can base legislation on its preferred economic theory, even though many experts and ordinary citizens reject it.

In the broadest sense, however, we must understand the U.S. Constitution as positive law. Rational analysis might lead us to conclude that there is nothing special about religion—that religion is a specific instance within a more general category of belief or commitment. But a philosophical question is not just the same as a constitutional question. The Constitution might explicitly affirm, or implicitly reflect, propositions that philosophical reflection would reject.

Leiter's main concerns are philosophical, not jurisprudential. He begins with what he calls the "central puzzle in this book"—why the state "should have to tolerate exemptions from generally valid laws when they conflict with religious obligations but not with any other equally serious obligations of conscience." A satisfactory answer would have to show, first, that there is a distinction between religious and nonreligious conscience, and, second, that this difference is such as to warrant disparate state treatment.

Leiter's point of departure is the proposition that "if there is something morally important about religious belief and practice that demands legal solicitude, it is connected to the demands of conscience that religion imposes upon believers." Other scholars are not so sure that this is the only such feature of religion, and neither am I. But that it is at least one such feature seems clear.

Building on the work of John Witte Jr. and Timothy Macklem, Leiter suggests that two things distinguish religion from other modes of belief. First, it issues in "categorical" demands that must be satisfied "no matter what an individual's antecedent desires and no matter what incentives or disincentives the world offers up." Second, it does not ultimately answer to evidence and reasons as ordinarily understood: "Religious beliefs, by virtue of being based on 'faith,' are insulated from [the] standards of evidence and rational justification…we employ in both common sense and in science." Religion is distinctive in conjoining these two features of belief.

The first feature of religion—categorical demands that contradict public law—is, I believe, the heart of the matter. In agreement with both the Supreme Court and the facts of human life, Leiter argues that the experience of being categorically commanded "does not track religious belief." "Here I stand, I can do no other" can be a sincere secular claim. The Supreme Court was not wrong to recognize the claims of secular creeds that "play the role of a religion and function as a religion in life." (Nor was the Court wrong to see religion as the paradigm for such claims.)

But at least religion identifies the source of the command and specifies the content of the command in ways that can be verified. When Quakers say that they cannot engage in armed conflict, or Jews that they cannot worship idols, they can point to the core texts and settled practices of their faith as proof. Religion offers conscience a measure of public objectification. Individualized claims of conscience detached from religion are harder to assess. That does not mean that they should be dismissed outright.

Still, inquiries into such claims are bound to be risky and intrusive. The external indicia of sincerity are less than reliable. And if courts try to reason from the credibility of belief to the sincerity of the believer, many religions would fail the test. By definition, all miracles defy the laws of nature, and it is hard to see what makes one purported miracle more or less credible than the next. Surely courts cannot "grandfather" religions whose miracles have been long and widely accepted while subjecting newer faiths to stricter scrutiny.

But let's set aside questions of proof and return to the main thread. Leiter argues, and I agree, that conscientious claims include, but extend beyond, religion, and that honoring only religious claims is indefensible, at least on the plane of principle. We agree that all conscientious claims should be treated equally. But we disagree about what that uniform treatment should be.

If I understand Leiter correctly, he endorses a generalized version of the position Scalia espoused in Smith: As long as the state is pursuing generally valid public purposes and is not directly targeting or burdening claims of conscience, it need not accommodate conscientious claims for exemptions from the law, and (Leiter adds) it must not do so if accommodation would have the effect of transferring burdens to others or of undermining law's capacity to promote the common good.

The nub of the matter is this: Leiter believes that even when his two conditions—no transferring of burdens and no impeding the common good—are satisfied, the state has no obligation to accommodate conscientious claims. I disagree. Unless the state can credibly argue that making an exception for sacramental wine or sacramental peyote violates one of Leiter's conditions (and I don't think it can), the inherent moral weight of allowing individuals to act in accordance with their deepest convictions should trump the application of the law to those with conscientious objections against it—especially when the law prevents believers from practicing core rituals of their faith. If free exercise means anything, surely it means that.

But what of the fear that recognizing claims of conscience courts anarchy? My response is simple: In the real world, claims of conscience have not had, and will not have, the consequences the objectors fear. The law is capable of establishing templates to distinguish between real and spurious claims, and courts and agencies are capable of applying them. Even when the stakes are very high, as they are in wars of total mobilization, authorities are able to accommodate conscientious claims without undermining military effectiveness. And consistent with specific accommodations, states may legitimately require those receiving accommodations to perform alternative services that compensate for whatever burden may have been shifted. Given the risks and costs of seeking accommodations—the time and money needed to meet strict tests, plus the likelihood of social disapproval—it is no wonder that relatively few people choose to run the gauntlet. This is not anarchy, unless every limit to state authority implies anarchy, in which case liberal democracy is by definition anarchic.


The second distinctive feature of religion—its supposed imperviousness to ordinary evidence and argument—is more controversial than Leiter thinks. Theologians in more than one religion reject its application to their faith. Every religion that rests on a historical narrative, as many do, is in principle exposed to the possibility of discoveries that may challenge its core beliefs.

As the book proceeds, Leiter's misgivings about the epistemic dimension of religion gradually come into focus. At one point he says that in the Middle Ages religious belief was "neither irrational nor unwarranted—and thus not culpably false belief—but after the Scientific Revolution and the Enlightenment, it is less clear." A few pages later, after animadversions against contemporary philosophers who dare to defend the rationality of religious belief, he is prepared to go further: "Let us suppose, as seems most plausible, that religious belief in the post-Enlightenment era involves culpable failures of epistemic warrant." Unlike their medieval counterparts, in short, modern believers are being unreasonable, and they ought to know better. Faith today is morally as well as epistemically deficient.

I am hardly the first to discern something approaching tautology in claims of this sort. Yes, if science and logic are the only modes of apprehending what there is, then faith is a species of illusion that claims to discern what in fact does not exist. But doesn't faith challenge the premise that science and logic are epistemologically exhaustive? Is it Leiter's position that science and logic suffice to disprove the existence of God and the possibility of miraculous divine eruptions into the natural world? Are the claims of atheism epistemologically stronger than the claims of theism, or are they on a par? There may be convincing answers to these questions, but Leiter hasn't offered them.

So what, then, is "culpable" about religious belief? Early on, Leiter observes that religious beliefs "render intelligible and tolerable the basic existential facts about human life, such as suffering and death." Religion offers "existential consolation," a feature that "explains its central importance in so many human lives." One might argue (many have) that faith is culpable because the consolation it offers is false and thwarts the development of the courage that can result only from seeing the human situation as it really is. But this is not Leiter's view. On the contrary, he says:

I have adopted throughout [this book] what seems to me the clearly correct Nietzschean posture—namely, that the falsity of beliefs and/or their lack of epistemic warrant are not necessarily objections to those beliefs; indeed, false or unwarranted beliefs are almost certainly, as Nietzsche so often says, necessary conditions of life itself, and so of considerable value, and certainly enough value to warrant toleration.

It is hard to see how these beliefs could be humanly valuable in Nietzsche's sense if those who hold them know (or believe) them to be false. Indeed, one can hardly believe something while believing that it is false. So illusion is a precondition of life.

We are left with a puzzle. Leiter clearly believes in the validity of evidence and argument as employed in ordinary life and in the sciences, and uses them to criticize faith as culpably irrational. For his part, Nietzsche views the epistemic distinction between faith and reason as the latest iteration of metaphysical illusion—or so most interpreters of Nietzsche believe. The fundamental distinction for Nietzsche is not between truth and falsity but rather between beliefs that affirm life and those that harbor a life-denying ethic of ressentiment. Can one coherently adopt "the clearly correct Nietzschean posture," as Leiter does, while continuing to affirm the epistemological validity of conventional empiricism?

Funding for this essay has been provided by a grant from the Henry Luce Foundation.


William Galston is Ezra Zilkha Chair and Senior Fellow in Governance Studies at the Brookings Institution. He is the author of Liberal Purposes and Liberal Pluralism, both published by Cambridge University Press. Galston served as deputy assistant for domestic policy under President Bill Clinton, 1993–95.
Also by this author

Please email comments to [email protected] and join the conversation on our Facebook page.

© 2024 Commonweal Magazine. All rights reserved. Design by Point Five. Site by Deck Fifty.