In Darwin’s Cathedral: Evolution, Religion, and the Nature of Society (2002), David Sloan Wilson presents a case for the utility of religion. He claims that religious beliefs and practices arose and are maintained in human societies because such beliefs and practices are adaptive.

Something as elaborate—as time-, energy-, and thought-consuming—as religion would not exist if it didn’t have secular utility. Religions exist primarily for people to achieve together what they cannot achieve alone. The mechanisms that enable religious groups to function as adaptive units include the very beliefs and practices that make religion appear enigmatic to so many people who stand outside them. (p. 159–160)

Wilson’s argument depends on a controversial version of Darwinian natural selection, operating at the level of groups. He calls it multilevel selection, and quotes Darwin himself to vouch for its applicability to human cultural practices.

It must not be forgotten that although a high standard of morality gives but a slight or no advantage to each individual man and his children over the other men of the same tribe, yet that an advancement in the standard of morality and an increase in the number of well-endowed men will certainly give an immense advantage to one tribe over another. There can be no doubt that a tribe including many members who, from possessing in a high degree the spirit of patriotism, fidelity, obedience, courage, and sympathy, were always ready to give aid to each other and to sacrifice themselves for the common good, would be victorious over most other tribes; and this would be natural selection. At all times throughout the world tribes have supplanted other tribes; and as morality is one element in their success, the standard of morality and the number of well-endowed men will thus everywhere tend to rise and increase.

Thus did Darwin speculate, in The Descent of Man (1871), that natural selection, which generally operates through the differential survival of individuals, could operate on the level of groups, with fitter groups out-competing the others. And in the same, widely quoted paragraph, he reckoned that one way for a group to boost its fitness is for its members to be good. Tribes with a “high standard of morality” will be “victorious” over other tribes and therefore “supplant” them. This implies that the proportion of moral to immoral tribes in the world will increase over time, because moral tribes have superior evolutionary fitness.

It’s an appealing idea, but is this prediction true? Do the good guys tend to out-compete the bad guys? Has the average standard of morality in the world’s “tribes” risen over the long term? To test Darwin’s model we would have to understand, even quantify, what he intends by the word morality. On careful reading, the paragraph seems much too vague for this. It runs together altruism (giving aid to others) and self-sacrifice (“for the common good”), which are very different impulses. It seeks the source of these noble impulses in “patriotism, fidelity, obedience, courage, and sympathy”: a motley collection of other noble impulses. (Are they all good? Which of them leads to which? Do we have any idea?) It speculates that the combination of those two (or seven) traits will cause the tribe to be “victorious” over other tribes. Is it really true that more “moral” communities tend to wipe out the less “moral” ones? Why? What is the mechanism? And what’s so moral about slaughtering your neighbors?

This all seems especially weak in the light of the theoretical work done a hundred years later by George C. Williams and John Maynard Smith, who showed mathematically that group selection can only work under implausibly constrained conditions.

Also note that Darwin’s account does not mention religion. Darwin himself was an atheist, and certainly did not take for granted the idea that religion is, in general, beneficial. By quoting this passage in the context of an argument for the utility of religion, Wilson is implicitly claiming that if we replaced Darwin’s phrase a high standard of morality with the word religion, the passage would still have the meaning that Darwin intended. Darwin’s model—actually, it’s just a thought experiment, not a model—predicts a positive role for religion, only if religion is responsible for the winning groups’ high standard of morality—that is, only if religion makes people good. To assume that it does is to assume what Wilson wants to convince us of. Thus, his argument is circular.

This isn’t the only problem with Wilson’s argument. There are lots more.

First. That a process is natural does not make it a good idea. Maybe religion does help some tribes wipe out other tribes. Is that good? Is that something we want? Only if we have an independent reason for preferring the religious tribes over the secular-humanist tribes. (I wonder what David Sloan Wilson’s independent reason is.)

Second. Even if communities with lots of morality (or religion or parochial altruism or whatever) tend to defeat communities with less of those things, this does not tell us whether the people in the winning communities are happy. It could be that when it comes to warfare the most effective organization is a totalitarian misery-state where only the people who are not in the army are happy, and only twelve guys are not in the army. There is a difference between a moral society and a happy society, especially if morality is taken to mean following the rules, as it is in so many places—especially the religious ones.

Third. In the most-religious communities in this world, one’s participation is not voluntary, it is required on penalty of expulsion or death. Wilson assumes a quasi-economic model where people are free to choose their affiliations, but religion is in direct conflict with such freedom. The more powerful the religion, the less choice its “adherents” have. If such conditions do not violate Wilson’s assumptions, do we even care whether his model is predictive?

Finally, the model does not distinguish between “religion” and any other kind of strongly normative social structure based on persuasive falsehoods. How does religion come into it? Where does the religion part of his hypothesis come from? I will tell you. First, through the assumption mentioned a moment ago: that people are free to choose; therefore, they choose the religion that most benefits them. (But when religion is in the picture, they are not free to choose.) Second, through the assumption that “belief in God” is probably, in general, a good thing; the assumption that religion fosters both social cohesion and positive morality. This is circular logic, assuming what was supposed to be proved.

The brutal fact, D.S. Wilson’s own “problem of evil” if you will, is that what we see in vivo is nothing like this. We don’t see religion bringing out the best in everyone, or drawing people together in joyful brotherhood. What we see is amoral, charismatic leaders who leverage specialized memes and raw violence to control large populations for selfish reasons. Such leaders benefit from religion; no one else does.

Seemingly unaware of religion’s well-known (and not yet ended!) history of violence and injustice, within groups as well as between them, David Sloan Wilson has carefully built a case for the idea that religion is a Good Thing; that it is Good because it brings folks together; and that it brings them together because it is Good. I’m sure his funders at the Templeton Foundation are delighted.