John Wilkins, one of the most interesting philosophical bloggers around evolution, has finally noticed the most important thing about religion: that it doesn't exist. He's been attending a conference in Oxford with various scientific and philosophical luminaries, and writes:
I have decided that the term "religion" is polysemic. It refers to no single property or even cluster of properties. We "define" religion as those social and cognitive and psychological behaviours that happen to approximate our own exemplar of it, even if we are atheists: like the joke about the Irishman asking the atheist if he is a Catholic or Protestant atheist. It is simply not a natural phenomenon. There are a plurality of natural phenomena, such as ritual behaviours, certain cognitive dispositions such as the tendency to take an intentional stance to natural phenomena, and so on, that are natural, but the class, the category of religion, is a mishmash.
This has a lot of impact on both explanations given of religion, and also the moral and political implications. If we are dealing with, say, in-group and out-group behaviours, then religion is not privileged in that respect. If we are dealing with intentionality, neither is religion privileged there. It in the end evaporates in every respect save the social constructed.
It follows from this that the new atheism, with its constant use of "religion" as a term which means something (nasty) is an attempt at social construction. In particular it's an attempt to make fresh deep boundaries between ingroup and outgroup. Not that I expect Wilkins to agree with that.
(UPDATED) Wilkins turned up in the comments, protesting that he would not describe himself as an atheist. The standfirst has been amended accordingly. Welcome aboard; sorry for my error.
