Support First Things by turning your adblocker off or by making a  donation. Thanks!

There is a great deal to be said for grand mistakes, clearly and prolifically expounded; they force us to think. Rodney Stark’s rational choice theory of religion has energized the social scientific study of religion by being big and wrong. Where others have set themselves the humbler task of trying to explain this or that aspect of religious belief or behavior, Stark stands out by emulating Marx, Durkheim, and Freud in proposing a grand theory of religion, which he and his colleague Roger Finke set forth clearly and succinctly in their summa, Acts of Faith .

In the 1970s Stark argued for a model of religious behavior based on rewards and compensators. People seek rewards but usually cannot get them and so are in the market for compensators. Religious compensators triumph over secular ones because they can invoke God and make promises about the next world. Since the desire for rewards always outstrips their availability, the demand for religion will be a constant. Hence long“run secularization is impossible. In the 1980s Stark became attracted by the economics of the Chicago school, which argued that models of human behavior that assumed a desire to “maximize” (if you can buy the same box of cornflakes in two outlets at a different price, you buy the cheapest) worked not just for the traditional concerns of economists but for all types of human action. Including religion.

One application of rational choice theory to religion concentrated on the microeconomics of human action, producing such gems as the claim that couples who shared a common religion went to church more often than couples who differed in religion because such church involvement is cheaper (you only need one car). There are, of course, more obvious explanations. The causal connection could be the other way round: committed believers are more likely to want to marry within their faith than waverers. Or the link could be strength of belief: a same“religion couple will reinforce each other’s beliefs.

More plausibly, rational choice was applied to the religious complexion of societies. Starting from Tocque­ ville’s observation that France had a single dominant religion that was only weakly supported while America had a plurality of competing sects and a vibrant religious culture, “supply“side” economists argued that, just as competition is virtuous in driving down prices, improving products, and increasing consumption in such material matters as the market for cars, so too is it virtuous in religion. Apparent differences in the religiosity of various societies were not to be explained (as sociologists had traditionally assumed) by differences in the demand for religion or in social circumstances that made religious beliefs more or less plausible. As Stark had argued, the demand for religion was a constant. Differences in religious cultures therefore were to be explained by the structure of the market and the nature of the supply of religious products. Diversity and competition were good; religious monopoly (especially when supported by the state) was bad.

One problem in assessing this radical and iconoclastic treatment of religion is that, beyond general impressions, the evidence is difficult to judge, and of necessity much of the critical response to Stark and Finke’s work has been so technical as to exclude anyone who does not comprehend multivariate statistical analysis. Suffice it to say that, although Stark and a small group of acolytes have been highly productive, the general weight of research publication has been against them. Even for the United States, which has provided the evidence closest to Stark’s claims, church involvement has generally been greatest in times and places where there has been the least diversity, not the most. And an acquaintance with the world beyond the United States quickly undermines the plausibility of a universal link between competitive markets in religious products and the popularity of religion. One only has to look at the history of resistance to communism in Eastern Europe to appreciate that it was the overwhelmingly Catholic countries (Poland and Lithuania, for example) that remained most religious.

Stark and his colleagues are fond of claiming that there is little published refutation of their supply“side theories. Not only is this self“aggrandizing claim factually incorrect, it misses the larger point that the vast majority of historians, sociologists, and political scientists ignore their work because they know that it cannot encompass the complexity of the real life of Bulgaria or Ireland or Australia.

Even if we somehow persuade ourselves that we can leave aside history and politics and explain human action by the autonomous choices of maximizing individuals, there is still the problem that opportunities for “choice” are not easy to identify or measure, and hence compare. It may appear that the United States has a greater variety of religious products than, say, England, but much of that apparent difference is a result of the U.S. having duplicate ethnic and linguistic variants of the same religious tradition. If an American town has a German and a Swedish Lutheran church and a black and a white Baptist church, it may well have greater diversity than an English town with only an Anglican and a Methodist church, but it is hardly sensible to see the U.S. town as offering more choice to most of its citizens.

Even those of us who are skeptical of Stark’s thesis will nonetheless find much to agree with in the enormous array of propositions that Stark and Finke develop, and this collection is worth reading for those nuggets of good sense. The disagreement comes when the reasonable points are claimed as support for an implausible general theory. For example, in this volume Stark again repeats the various deficiencies of a church supported out of public taxation. If the clergy do not have to win and keep a congregation in order to live, they will not bother. Familiarity with Anthony Trollope’s Barchester Chronicles will provide many supporting examples of clerics who took the income from their many benefices, paid a fraction of it to miserable curates to do their work for them, and headed off to the warmer climes of Italy.

But do such examples confirm a general theory? Does Rodney Stark publish only for the money? Would he be happy to be paid well but have no students enroll for his classes and have no audience for his guest seminars? Such narrowness of motivation is precisely what he imputes to Scandinavian Lutheran clerics. There seems no space here for any notion of being called, or driven by ideology, or concerned to save souls.

While treating religion as if it were a product like any other can generate fresh insights, such reductionism in the end destroys the theory because it ignores too much that we know of religion. For most people religion is not a matter of choice in which opportunities to maximize or economize can be sought. Most people are born into a particular religion and so thoroughly socialized in it that alternatives are not economizing opportunities, but implausible heresies. A Ford tractor dealer whose sales decline can switch to a John Deere dealership. If agribusiness goes down the tubes, he can get into cars or even sell his lot for a housing development. The Calvinist Free Presbyterians of Scotland’s Hebrides know that their faith is declining in popularity and they can argue amongst themselves about what small changes might make it more attractive, but they cannot announce a return to Rome.

The point can be seen very clearly in how believers deal with failure. They persuade themselves that everyone else is too mired in sin to hear God’s truth. When the Ford motor company realized that the Edsel was a dog, it did not persist in making it and console itself with the view that the ordinary Joe was too dumb to know a good car when he saw one. It dropped the Edsel and started again.

Rational choice works well for cars because the car market has features that are mostly absent from the world of religion: there is widespread demand for the general product; brand loyalty is relatively weak; car choices are largely free from ethnic, racial, or national constraints; few people get ostracized, or even shot or raped, because they switch models. In addition, we can make rational choices between cars because the rewards (seat comfort, speed, fuel economy, and the like) are fairly readily identifiable and comparable, as is the cost. But in this life I cannot know whether it is Islam, Catholicism, or snake“handling fundamentalism that will guarantee eternal salvation. In a nutshell, rational choice does not work for religion because there are enormous constraints on choice and because the information required to make the choice rational is not available.

This is not to say that religious choices are irrational, as Stark alleges the other approaches must assume. Here, as in so many other elements of his approach, Stark enhances the plausibility of his approach by falsifying the alternatives. In fact, the other explanations for religious behavior simply have a broader notion of what counts as sensible grounds for belief (such as having been taught certain things by parents and others one trusts and having found them a sufficient guide to a pleasant life). Indeed, a very general criticism of Stark’s work is that he does not deal fairly with scholars who disagree with him.

For example, he repeatedly summarizes the secularization thesis (which supposes that various features of modernization in certain settings combine to reduce the plausibility of religious ideas and practices) with the same quotation from an undergraduate textbook written in 1966 by an anthropologist. Marx is cited to stand for the work of hundreds of us who have never been influenced in the least by the man. Scholars are routinely parodied or assigned views they have clearly disavowed. I am cited, for example, as believing that religion in Northern and Western Europe is “soon to disappear.” I certainly believe that mainstream and liberal denominations are in severe trouble, but I have also written extensively on the survival of sects, the appeal of the expressive spirituality of the New Age, and the continued vitality of religion where it is part of contested religio“ethnic identities. Bryan Wilson, like David Martin, if forced to choose one key cause of secularization, would probably cite our modern concern with engineered change and our purported mastery over our fate. He would not, as Stark claims, suggest that “it is science that has the most deadly implications for religion.” Science is not even mentioned in Wilson’s two classics, Contemporary Transformations of Religion and Religion in Secular Society .

What is more, Stark is unable to admit he has learned from his critics. For example, his early versions of the supply“side propositions permitted no serious qualifications to the centrality of individual choice. After enough of us had said it often enough, he came to accept that many people’s ability to economize in religious behavior is indeed heavily constrained. But this concession was slipped into yet another list of propositions with no recognition that its initial absence had any implications for the virtue of the general theory.

These ad hominem criticisms of Stark would be out of place were it not for that fact that his sectarian insistence on dividing the world of serious students of religion into his people and the rest severely limits the value of his scholarship. These essays give us a very clear notion of what Stark thinks, some of which is useful, but we can learn very little from the vast range of sources he cites because we can have little faith in his accounts of the work of others.

A final thought. Stark sees himself as disproving the claim that most modern societies are patently less religious than they once were. Yet if we ask ourselves what sort of society it would be in which people switched readily between religions as they sought to maximize their utility, the answer would have to be a society in which religion did not much matter.

Steve Bruce is Professor of Sociology at the University of Aberdeen, Scotland, and author of Choice and Religion: A Critique of Rational Choice Theory (Oxford University Press).