Brian Leiter’s book Why Tolerate Religion? has received a fair amount of attention in recent months. Much of that is positive: The book has received enthusiastic reviews in a number of venues, though there are some, typically religious, dissenting voices. My interest here is neither to articulate the book’s overall argument, nor to evaluate it as a whole. Rather, I am interested in Leiter’s judgment that religious believers are guilty of a “culpable failure of epistemic warrant.”

That judgment is important to the question that serves as Professor Leiter’s title: Religion almost certainly can deserve no positive respect if it is based on wrongful belief for which believers are responsible, and so toleration is the best that is on offer.

The judgment that Leiter must pass to reach his description is in fact twofold, for a judgment that someone has culpably acted wrongly requires both the judgment that the action in question is wrong: unreasonable, illicit by some objective standard; and the judgment that the agent is responsible for having wrongly done that action: He either knew that it was wrong and willingly acted anyway, or was culpable for his belief that it was not wrong.

Thus, the judgment that a certain act is unreasonable in itself, and the judgment that an agent is culpable for having done that act, are themselves importantly different. The late philosopher Alan Donagan has characterized them as first-order and second-order moral judgments respectively.

Start your day with Public Discourse

Sign up and get our daily essays sent straight to your inbox.

Judgments of culpability, here and now, of our peers, fellow citizens, and friends are themselves important moral acts. Even when we think another agent has acted wrongly, in a first-order sense, we think it important—a matter of justice—to take care in passing judgment on that agent. And when we do, that act of judgment can itself be morally assessed.

Suppose, then, that Smith has judged his friend to be guilty of a betrayal—he has judged him both wrong in his actions and culpably so. We can then ask of Smith whether his act of judging was fully reasonable. Perhaps he made his assessment of Jones too quickly; perhaps he did not take into account mitigating factors in Jones’s action. Perhaps he was angry at Jones, and this led him to color Jones’s actions in an unfair negative light. And perhaps Smith was wrong about whether there was a betrayal at all: Error about the first-order question will surely lead to error in Smith’s second-order judgment of Jones’s culpability.

It is clear that failing to be responsible in one’s judgments of responsibility can itself be a very serious moral lapse: To judge another guilty of wrongdoing on the basis of partial evidence, for example, or because I dislike that person for independent reasons, is unreasonable, and involves an injustice done to the one unreasonably accused. The injustice is likely to be exacerbated by the actions we typically take in consequence of a judgment of culpability: At worst, those judged culpable of wrongdoing are punished; at best, they are merely tolerated when in fact they deserve better.

Now Leiter’s judgment regards an act of belief, and must accordingly involve two moments as articulated above. The first judgment concerns the reasonableness of belief, in and of itself. Objectively, and as a first order-matter, Leiter believes, religious belief is uniformly false and unreasonable to hold.

This is a strong judgment, for the belief that a creator deity is responsible for bringing into existence the world of contingent things is unlike certain perceptual or logical judgments. The evidence for and against that judgment is not so overwhelmingly on one side that belief of one sort is rational while belief of the other is simply irrational or unintelligible. That this is so is evidenced by Leiter’s willingness to make judgments of culpability at all, for we can only be culpable for believing what it was available for us to believe: One who believes it to be raining in the presence of a sunny sky is mad, and not culpable at all.

Is Leiter’s confidence in his judgment of the unreasonableness of faith justified? That question is beyond the scope of this essay, though I shall return to it below; but as others have noted, Leiter gives little in the way of reason to accept his judgment beyond a nod at the Scientific Revolution and the Enlightenment and an accusation of “sometimes desperately post-hoc” rationalization by philosophical religious believers. Leiter appeals as well to a philosophic consensus (of, obviously, non-believing philosophers); yet his unwillingness to engage the most rigorous and extended arguments of believers casts doubt on the seriousness with which his judgment should be taken.

What is of more immediate interest to me is Leiter’s second-order judgment that religious believers are culpable for their wrongful belief. For this act of judgment on his part is, as I have shown above, a moral act, one in which there is the possibility of injustice. That possibility would be realized were Leiter to have been unreasonable in his assessment of the responsibility of moral believers for their false beliefs.

The judgment of culpability in wrongful religious belief is one that has been made of religious believers in the past, often by other religious believers. Judgments of heresy, for example, are often judgments that another’s religious beliefs are not merely wrong but culpably wrong. On the basis of such judgments, punishment was frequently thought warranted, and almost as often meted out.

Yet even among these same believers, we find the distinction between material heresy (wrongful religious belief) and formal heresy (being guilty for that wrongful belief). And increasingly churches and religious believers have been unwilling to declare a judgment of formal heresy on others. It has come to be recognized that the norms governing correct belief in this area are difficult to apprehend, that the evidence is difficult to assess, that longstanding and inherited patterns of belief and worship create strong dispositions to continue worship and belief in that manner, and the like, obscuring the perhaps greater evidence for a different view.

The prospect of judging religious believers of this or that stripe to be guilty for their wrongful belief has, except in some fairly extreme cases, been found to be an act considerably lacking in charity and justice, and one that has tended to precipitate further acts of uncharity and injustice.

So it is puzzling that Leiter would be willing, and indeed apparently enthusiastic, to return to a time in which judgments of culpability in religious belief could be made, and made not just of this or that person, or even of this or that sect, but of the entire class of religious believers. And it is equally puzzling why Leiter should think that a practice of toleration could be sustained in the long run when such a judgment of culpable failure was apparently so easy to make. That is not the lesson to be drawn from history, nor is it, apparently, the direction currently taken by those who argue that religious conscience needs no respect when it comes up against more obvious secular rights: to contraceptive services for employees of religious institutions, to give but one example.

There is a further mystery, which brings us back to the first-order question concerning the reasonableness, in and of itself, of religious belief. This is the mystery of the normativity to which Leiter appeals in making his judgment of culpable failure of epistemic warrant. Leiter’s judgment of culpability requires both that there is a categorical norm governing belief, and that believing agents can follow or culpably fail to follow that norm.

This is an unusual normativity: It is not the normativity of the natural, for natural objects are not held culpable for failing to live up to species norms. It is not the normativity of logic, for, while categorical, it is impossible culpably—i.e., knowingly and willingly—to violate the norms of logic. And it is not the normativity of a means to an end, for that normativity is not unconditional: Perhaps religious believers have different ends from secular rationalists, in light of which their belief makes good sense.

Rather, the norm to which Leiter appeals must be one that governs belief categorically: Believers must, or ought, to believe in accordance with the evidence; yet this must be a norm agents can choose to abide by or not, for in the absence of such a capacity, there can be no judgment of culpability. So Leiter’s judgment of culpability is either a mere rhetorical flourish, or it points to both a categorical normativity, and that normativity’s correlative, a capacity for freely choosing in accordance with that normativity.

Yet neither reality—the reality of categorical normativity, nor the reality of free choice in accordance with categorical norms—is easily squared with an entirely naturalistic and materialistic universe. On the other hand, it is only within the framework of a belief in such a universe that one can be sure to the point of Leiter’s confidence that there is no case to be made for a nature-transcendent being of the sort that God is. Normativity and free choice thus point beyond the material world in ways that many believers have recognized as an important source of evidence for their reasonable belief.

So what appeared to be a reasonable worry about the injustice of Leiter’s judgment of culpability has become something more. It appears as well that his willingness to make such judgments undercuts the confidence with which he should judge that the case against religious belief is, in and of itself, without merit. Both his first-order judgment concerning religious belief, and his second-order judgment concerning religious believers, are highly suspect.