If religion is a virus, then perhaps the spread of religion can be understood through the lens of evolutionary theory. Perhaps cultural evolution can be modelled using the same mathematical tools applied to genetic evolution.
Well, that's overly simplistic, of course - as anyone who's followed the 'meme' controversy over the years will know. In fact, the authors of the paper I'm writing up today - Michael Doebli and Iaroslav Ispolatov at the University of British Columbia - studiously avoid using the term.
Want they set out to model was the development of religious schisms. Such schisms are a recurrent feature of religion, especially in the West. The classic example is the fracturing of Christianity that occured after the reformation.
Their model made two simple assumptions. Firstly, that religions that are highly dominant actually induce some people to want to break away from them. When a religion becomes overcrowded, then some individuals will lose their religion and take up another.
That could happen, they say, when religions are hierarchical, providing greatest benefit to a few at the top. Eventually, for those cut out from the power structure, the benefit of striking out with their own, new religion might outweigh the costs.
Second, they assume that every religion has a value to the individual that is composed of it's costs and benefits. That value varies between religion, but is the same for all individuals. It's a pretty simplistic assumption, but even so they get some interesting results.
At each generation, the religion can mutate a little, just like a virus.
With these simple assumptions, they get a range of results depending on how they tweak the parameters. In one, they end up with a wide, even spread of religions - and infinite range of religions, clustered around a central, average, archetype.
But tweak the parameters a bit more,and you get a discrete number of stable schisms - just like the religious sects seen in real life.
Now, this is a very simple model, and so the results shouldn't be over-interpreted. But it's a fascinating result for a couple of reasons.
It shows how new religious 'species' can come into being in a mixed population - no need for geographical separation. That's such a common feature of religion - from the Judaeo-Christian religions to examples from Papua New Guinea - that it's worth trying to understand what drives it.
What's more, this is the first time that anyone has attempted to model the transmission of religious ideas in evolutionary terms. It's a first step, to be sure, but just showing that it can be done is a significant achievement.
The value comes because it shifts the focus from thinking about how culture benefits the host, and instead asks how the cultural trait is adaptive in it's own right. What is important is not whether or not the human host benefits from the trait, but rather whether the trait can successfully transmit and reproducing itself (see Bible Belter for an example of how this could work).
Even more intriguing is the implications for understanding cultural-genetic co-evolution. After all, we know that viruses and their hosts co-evolve in a kind of arms race - sometimes ending up in a relationship that benefits both.
Genetic evolution in humans occurs in an environment shaped by culture - and culture, in turn is shaped by genetics. There are clearly some very deep relationships here, and the kinds of models introduced by Doebeli offer a way to untangle them.
Doebeli M, & Ispolatov I (2010). A model for the evolutionary diversification of religions. Journal of theoretical biology, 267 (4), 676-84 PMID: 20854828
This article by Tom Rees was first published on Epiphenom. It is licensed under Creative Commons.
Information and Structure in Complex Systems
5 hours ago in PLEKTIX