Is God an accident?

Is God an accident? That’s the title of an article by Paul Bloom, a Yale psychology professor, in the December issue of Atlantic Monthly.

Coming from cognitive neuroscience and evolutionary biology, Bloom points out the “built-in” systems often called “folk physics” and “folk sociology”, which he describes as “two different computers in the brain”. The sociology half of this hard-wired dualism, if you will, permits us to conceive of abstract human-like entities independent of a specific incarnation. The same social engine, one of whose jobs is to infer agency and intent, indiscriminately finds such agency and intent even where it does not exist in reality. Voilà: God and disembodied souls.

I haven’t read anything else by Bloom (he is the author of Descartes’ Baby: How the Science of Child Development Explains What Makes Us Human). His theory seems compelling as far as it goes, but…

Calling religion or God an “accident” certainly is catchy and looks good on the magazine’s cover. But is it really accurate? In fact, Bloom never uses the term accident again anywhere in the article. A feathery appendage a bird develops to protect itself from the cold turning out to be useful in flying—that’s an accident. That’s different from what’s happening here. The systems are functioning exactly as designed. The anthropomorphization module is doing exactly what it’s supposed to. All detectors yield false positives by their very nature. But they have mechanisms are in place to recover from those false positives. If we see a leaf moving, infer the presence of a predator, we freeze, then look again and finally realize it’s just the wind, and return to normal. The real question is: why does religious anthropomorphization bypass any similar corrective mechanism?

Second, from an evolutionary standpoint, we know there are spandrels, but one which decreases fitness will be selected away. If belief in God is such a spandrel (accident) and is a negative or waste of energy for the species, it should have been deselected, but it wasn’t. Why? Or are there evolutionary advantages to supernatural belief? If so, the story isn’t complete without identifying them.

A subset of this problem—probably impossible to ever answer—is why the dualistic body/mind model was selected for humans. Could there be fitness-decreasing aspects to an integrated body/mind model? If not, why wasn’t it selected? Or is evolution simply taking its time getting us there?

Finally, where do meditation or other spiritual practices—which, like Boyer (previous post) and Atran (previous post ), Bloom completely ignores—fit into this picture? Presumably, in terms of his folk physics vs. folk sociology dichotomy, practice has the effect of unifying the two engines. To borrow Dogen’s words: “cast off the gap between body and mind.” This raises the intriguing question of what kind of neurological implementation of the dual model might be amenable to such unification.

One Response to “Is God an accident?”

  1. Gruntled Says:

    I had a dialogue with Paul Bloom on why he only considered secular theories of belief in God. He expressed surprise that there were any other, religious theories of why we believe in God. I really don’t know what he thinks religion is, though I attempted to answer—see

Leave a Reply