Testing major evolutionary hypotheses about religion

Posted on July 23, 2007


A workshop on the major transitions within evolution just finished here at the KLI. One of the speakers was Eors Szathmary who co-wrote with John Maynard-Smith the now classic book with first put together the necessary theory to properly express the idea. While it was very interesting to listen to some of the top people in that area the workshop did stop me from getting more work done. On the other hand, it did help me to round out some of my ideas.

Some of these relate to the work of David Sloan Wilson who has worked on the evolutionary roots of religion. I am yet to read his book on the topic – Darwin’s Cathedral – but have finished an article he wrote in which he tried to test the various evolutionary hypotheses regarding religion using a random sample of religions drawn from a encyclopedia of religion. The article, informatively entitled “Testing major evolutionary hypotheses about religion with a random sample”, looks at most important evolutionary theories that have been put forward and tries to judge their ability to explain the religious movements in question. The conclusion he reaches is that the group level adaptation view he puts forward is supported by a study of the random sample. My worry there is to what degree he avoids the possibility that, even though he chose a random sample, his evalutation of it was then biased by his own views.

Wilson’s work is all the more interesting for me as he also uses Durkheim as part of his theoretical background but only appears to consider the religion side of Durkheim’s distinction, focussing upon what Durkheim called their secular utility. The question I asked myself was whether Wilson’s methodology could be adapted to carry out a similar test of hypotheses regarding superstition. I have, therefore, contacted him about getting more detailed information about his methodology. I am particularly interested in what specific features he looked for in terms of support for the various hypotheses.