In the 1950s and ’60s, social scientists predicted that organized religion would give way to secularism. Among the leading proponents of that theory was Peter Berger, well-known sociologist of religion, who in 1968 predicted that “people will become so bored with what religious groups have to offer that they will look elsewhere.” Simply put, the social scientists were all wrong. Religion did not give way to secularism, nor has it recessed to the hills, mountains, and the South. And, unlike the predictions of statistical mathematicians and physicists Daniel Abrams, Haley Yaple, and Richard Weiner, religion is not now on the verge of extinction either.
Abrams, et al., wrote a paper in January called “A mathematical model of social group competition with application to the growth of religious non-affiliation,” predicting the human tendency to migrate to larger and therefore more powerful social groups, in this case religiously unaffiliated. The paper is dense with applied mathematics and statistical equations, too thick for the mathematical layman, or anyone whose knowledge of partial derivatives is four years out of date (as mine now is). Since CNN posted the findings to their Belief Blog a little over a week ago, many people, from agreers to dissenters, the religious and nonaffiliated alike, have written their own opinions on the study. (This one, by a Unitarian Universalist minister, is my favorite.)
In simple terms, the authors took census data from 85 regions around the world and fit data models to them based on one premise: it’s better to be a part of a larger group than a smaller group. They write, “We assume the attractiveness of a group increases with the number of members, which is consistent with research on social conformity.” To a certain extent, I agree: one of the primary foundations of religion, after all, is that it is a social phenomenon, to channel Durkheim. A larger group can offer protection against outside forces, elevate political and social status, and give a sense of belonging. They assume “that society is highly interconnected in the sense that individual benefits stem from membership in the group that has an overall majority.” If this “perceived utility” of belonging to a larger group is the only requirement for religious identity, then their theory withstands mathematical rigor. They combine the concept that non-affiliation has a higher utility in Western countries with the older secularization theory to create a competition model with one group victorious over the other.
Mathematically, Abrams et al. have a strong correlation between their data and their model. Unfortunately, they’re off sociologically, anthropologically, religiously and theologically. I won’t criticize their use of a binary system (religious vs. nonreligious) because adding more variables to a statistical system can make it more complicated than useful, but it flaws the conclusion. Let’s start with their theological error, which they acknowledge as a mathematical error. Their system does not take into account abrupt changes. These changes can come from a new charismatic figure, a movement within a religion that sparks an Awakening, or any number of factors not imagined. Had Vatican II not happened, Catholicism may have continued to reject modernity driving itself towards extinction. Gay marriage may be one of the igniting issues now. Religious institutions that embrace the LGTBQ community are often flourishing, adapting to the needs of their communities and allowing the religion and faith evolve.
Religiously, Abrams, Yaple, and Wiener don’t seem to be considering non-Western religions, even in Western countries. They don’t talk about Christianity explicitly, but the census data they use—especially some of the older data—bases religious identity on whether or not the subjects belong to a congregation or tithe to a church. They explicitly do not address the migration of Islam into Europe, but according to Greene, Abrams “thinks the trend is robust enough to withstand some challenges.” While every religion requires some form of community, not every religion counts its members the same way a Christian Church might. From a traditional theological standpoint, everyone is Catholic (universal) and everyone is born Muslim (one who submits to Allah). I’ve been critical of “cafeteria religiosity” in the past, but many religions allow for a picking-and-choosing aspect of faith, belief, and practice; and those are the religions most likely to survive the future. There are millions of Hindus in the world, but most of those who live in South Asia wouldn’t consider themselves a practitioner of Hinduism because no one such religion exists.
This addresses the age-old academic question: “What is religion?” Is it a system of symbols, a la Clifford Geertz? Is it a model describing a set of practices, rituals, or beliefs? Is it a map—Jonathan Z. Smith‘s version of the “map is not territory” paradigm—of the physical universe or a map for understanding human social interaction? Is it Marx’s “opiate of the masses?” Or perhaps Peter Berger has the answer: “religion is the audacious attempt to conceive of the entire universe as being humanly significant.” Abrams, Yaple, and Wiener are talking about religious affiliation, and therefore organized religion, or the institutionalized version of all of the above definitions. But would Abrams et al. include Jonestown or Scientology as religions, or classify them as cults and ignore them? How we define religion is crucial because certain aspects of religion are vanishing, while others are becoming stronger.
Disregarded through equations is the increasing divide between the social and theological aspect of religion. A trend unobtainable through census data, many frequent churchgoers attend and participate for the social communitas of a religious experience rather than for express theological beliefs. The social/communitas churchgoers may not attend as frequently as those who have a greater theological investment. When attending church and receiving the Sacrament on a regular basis is no long requisite for salvation, even the devout are less likely to attend services habitually. And, let’s face it, there are more things to do on a Sunday now than there were 50-100 years ago. During the era of the Second Great Awakening in America, sermons were not only meant to convey religious morals and values, but were also entertainment. Today, if a preacher talks for more than twenty minutes, the last half may fall on deaf ears.
In the end, the conversation comes back to models. I referred earlier to J. Z. Smith’s model of/model for, or the “map is not territory” paradigm. A map can describe an area, but it is not the area itself; there are models that describe a phenomenon and models that predict a phenomenon. Abrams, Yaple, and Wiener have created a model of the religiously affiliated/unaffiliated population, but are using it as a way of predicting religious patterns we cannot yet imagine. In religious studies, we regard the multiple facets of the systems of religion, and understand that they do not all follow the same pattern as Protestant Christianity. I like to think the fields of religion and sociology have benefited and learned from the failure of secularization theory. Such grand statements as, “We will see the end of religion by the beginning of the 21st century,” are laughable in retrospect, looking at the American religious landscape today. Mathematically, the model is sound, but as UU Minister Daniel Harper remarks:
this is what happens when physicists try to do religious studies and sociology without learning the basics of the latter two fields: they mix good mathematical models with poor understanding of what it is they’re actually modeling. It’s as awkward as watching most theologians trying to talk about the Heisenberg Uncertainty Principle.
Needless to say, when it comes to modeling behavior, physicists should stick to particles.
Test 