Paul Seabright Responds!
Delving Deeper into the Divine Economy
My previous post review Paul Seabright’s new book The Divine Economy. Do go ahead and read that post first. Here I ask Paul some questions that came up doing my review and Paul has kindly agreed to respond. I hope you enjoy his detailed and insightful answers!
Mark: How did you land on the "platform" model of religion? Was it a well formed idea before you set out to write the book? Or did it emerge out of the process of researching and writing?
Paul: I work at the Toulouse School of Economics, where a lot of the theoretical research on platforms has been conducted by Jean Tirole, Jean-Charles Rochet, Bruno Jullien, Patrick Rey, Jacques Crémer and also by younger scholars such as Andrew Rhodes, Alex de Cornière and Yassine Lefouili. This is a fantastic environment in which to think about how platform relationships pervade our economic institutions. But the book has been a long time in the making, so the centrality of the platform model to my thinking has come about gradually. I was influenced by my empirical observations that the members of religious movements are simultaneously its greatest assets, an insight that is central to the notion of platforms. Seeing how often people would tend to find spouses among fellow members, for instance, made me realize that the congregations were also functioning as dating platforms. Even if it’s a key to the way the churches do this that you wouldn’t go to church to find a spouse whose only reason for going to church was to find a spouse.
Mark: A fascinating part of the book that I didn't explore in my review was your use of Robert Aumann's paper on agreeing to disagree to explore how much differences in beliefs matter for religious choice. Can you expand upon that argument?
Paul: I came across Aumann’s extraordinary theorem while working on my PhD many years ago. It takes two Bayesian decision makers who share a common prior about some event and who then observe private information, before sharing their posterior probabilities of that event. He shows that if their posterior probabilities are common knowledge they cannot diverge! In other words, two rational decision makers cannot “agree to disagree” about the truth of some proposition unless they have divergent priors (which raises the question where those divergent priors come from, since they cannot be the product of divergent previous information). Each would give weight to the fact that the other had observed information leading to a different conclusion, and the only equilibrium of this iterative process would cause them to converge. I use the theorem in the book to explore possible explanations of how religious people can disagree about so many questions they believe to be important (though obviously similar questions can arise about disagreement over secular matters).
Divergent priors might be one explanation (to the extent that people select into religious movements according to personality traits such as optimism, pessimism, or those traits described in Jonathan Haidt’s book The Righteous Mind: Why Good People Are Divided By Politics and Religion). But I don’t think it’s the most interesting. A second explanation is that it’s often hard to know what religious people believe as opposed to what they find it appropriate to say. There may be common knowledge about the latter without their being common knowledge about the former. For example, one third of American Catholics say they believe the communion bread and wine become literally the body and blood of Jesus. But while we can all see what they say (in response to a survey), it’s not clear how many of those would accept the corollary that if we took the communion wine to a medical laboratory it would test positive for the presence of haemoglobin. So it would be very hard to know to what precise beliefs people should converge.
A third possibility is that human beings are just not very Bayesian, in two ways I discuss at length in the book (noting that these non-Bayesian features of our decision-making would not have been particularly maladaptive for us in prehistory). One is that they don’t update their beliefs efficiently when they receive new information. The second is that they don’t condition their beliefs adequately on the circumstances through which they came to acquire them. For example, if they developed their beliefs by copying other people, they don’t condition on the circumstances that gave them the particular models they copied. Thus Catholics who believe in transubstantiation, for example, rarely take into account that if they had happened to grow up in non-Catholic households they would not even have begun to consider transubstantiation plausible. And most doctrinal beliefs are simply not very important in the process of recruitment to religious movements. As I put it in the book: “accepting doesn’t require believing, and believing is optional in practice for most members, most of the time (even while it passionately preoccupies some other members). It’s only after joining that most members start to shift their beliefs in the direction of the religion’s doctrines—and they do it because it comes naturally to them, not because their membership requires it.”
Mark: Were you ever frustrated that you couldn't find compelling evidence to answer the questions you were dealing with in the course of writing the book? And what areas would you like to see more research on?
Paul: Yes, often! For example, I wish it were easier to test in a rigorous comparative way the various hypotheses I discuss about how the political instrumentalization of religion can lead to a loss of legitimacy of religious movements. I use various case studies in the book, and I’m currently working on a comparative econometric paper with my former PhD student Julia Hoefer, but the data are frustratingly limited. I also wish we could find better data about the revenues of religious movements, so we could do econometric work on the determinants of demand for religious services. I’m convinced that there’s really no such thing as religiosity, in the sense of a single psychological trait that makes people more likely to belong to religious movements. Instead there are many traits to which different religious platforms respond in a variety of ways. But it would be easier to show this if there were better field data involving detailed measurements of how much people actually contribute to their religious communities.
Still, there are some great topics wide open for smart Ph.D. students to work on. One is the determinants of religious violence. There’s pioneering work by Sascha Becker and Luigi Pascali on anti-Jewish pogroms in medieval Europe, and by Saumitra Jha on Hindu-Muslim violence in India, but there’s still so much more to do. Another is whether the doctrine of religious movements has first-order economic and social consequences. My book tends to be skeptical, suggesting that religious leaders mostly develop doctrine to suit their own purposes, and to apply doctrine flexibly when that suits their purposes better. But scholars I respect have recently made strong cases that doctrine, once adopted, can constrain future development in good and bad ways. Timur Kuran’s excellent book Freedoms Delayed: Political Legacies of Islamic Law in the Middle East argues this for the Islamic waqfs, which he claims have held back both economic and political development in the Middle East (Mark: see my review here). I’d have been curious to see a comparison with Islamic countries outside the Middle East. Robert Eisen’s fascinating Jews, Judaism, and Success: How Religion Paved the Way to Modern Jewish Achievement makes a strong positive case for rabbinical Judaism as the foundation for Jewish economic, scientific and artistic achievements in the modern world (I have a more detailed review of this to appear shortly in Contemporary Jewry). More work sorting out the causality between ideas and economic or social conditions will be a wonderful area for future researchers to work on, particularly now we have sophisticated tools available to undertake textual analysis on a large scale.
4. From the perspective of the field, what is more important: more field work (say in developing countries or for religions that are much less well studied than Christianity/Islam i.e. indigenous folk religions), more historical studies, or more detailed investigation of how large-scale religions (like the Catholic Church) work in say the US or Europe?
Paul: All of these are important, but whichever type of study it is, it must be comparative. Even a detailed participant-observed study of an indigenous folk religion in one community has to ask questions about how this compares with religious movements elsewhere, or it’s a waste of everybody’s time and energy. Anthropologists like Eleanor Power (who works on ritual in Tamil Nadu in southern India, a region I know well) can provide all the detail an old-fashioned ethnography might demand, while still asking great questions about motivations and mechanisms that are of global relevance.
Mark: How were political leaders "weaned" from their reliance on religious legitimacy? How are religious leaders "weaned" from falling into an often fatal embrace of the political authorities?
Paul: We can ask this question at so many levels. Your great book Persecution and Toleration with Noel Johnson, about the evolution of political power from identity-based to a universal rule of law, asks the question at the level of European society in the early modern period. I discuss in The Divine Economy a number of national case studies in the last few decades, comparing the Catholic Church in Spain, Ireland and Poland for example as well as making observations about Iran and the US. There seems to be a difference between situations where the alliance between religious leaders and political leaders is undermined by the loss of political authority directly (as in Spain after Franco, and as may be happening in Poland today), and those where the religious leaders lose their authority because of scandals within their movements, as in Ireland. The US is a more complex story, about which I say quite a lot in the book, though I wish I had known about David Hollinger’s 2022 book Christianity’s American Fate: How Religion Became More Conservative and Society More Secular when writing that part.
At the level of individual leaders I suspect the weaning rarely if ever occurs. Religious leaders who have sold out to political leaders don’t usually repent, they just get sidelined (or die). Political leaders who have instrumentalized religion don’t see the error of their ways, they just leave office (or die). I don’t foresee either Vladimir Putin or Patriarch Kirill ever back-pedalling on their savage partnership of bloodshed in Ukraine. But pointing out the long-term consequences of such misalliances can perhaps serve a purpose in warning other leaders who might be tempted to imitate them.