What makes people believe in God?

The relatively new research field cognitive science of religion is busy trying to answer this question. And it’s come up with some powerful answers so far. Importantly, its answers are psychological. They focus on the mental processes that cause religious belief—or religious credence, as I call it.

But the existence of this research program raises an important philosophical question. What should understanding the psychology of belief in God do to that very belief? In other words, once we know where religious credence comes from, should we be inclined to reject that credence, or not?

Now, one partial answer to the question of what makes people believe in God is the Minimally Counterintuitive Hypothesis (MCH), which has been advocated by Pascal Boyer in his book Religion Explained and by Scott Atran and Ara Norenzayan in this seminal paper. The idea is that what gets represented as gods in people’s minds across cultures are beings that are minimally counterintuitive. That is, they involve just one violation (or only a few violations) of intuitive psychology, intuitive biology, or intuitive physics.

A boy born of a virgin, for example, is minimally counterintuitive in this sense: it involves one violation of intuitive biology (virgins can’t conceive). A snake with feathers also involves one violation of intuitive biology, as do human figures that never die. These minimally counterintuitive ideas are the bases for Jesus, Queztacóatl, and the Greek gods, respectively.

The empirically supported explanation for why minimally counterintuitive representations become culturally widespread is that they are more memorable and hence likely to be transmitted. A being that has no intuitive violations—say, a plain old cat—is not that memorable. A being that has too many—a talking mountain that is also an invisible horse—is just too confusing to understand. The minimally counterintuitive representations hit that psychological sweet spot that allows them to be remembered and function as divinities. (For aficionados: I am setting aside the famous Mickey Mouse problem. See Justin Barrett’s paper.)

So MCH identifies a psychological tendency to believe. But now to our main question. What if you currently have a belief/credence that, say, a boy born of a virgin is God? What should learning about the MCH do to that credence? Should you respond by rejecting the credence?

Here are two schools of thought.

“Give it up!” says one school, including the so-called New Atheists. The idea is that if you know your belief just came from a psychological quirk, then you’re not left with any good reason to hold the belief. It’s like a visual illusion: once you know where it comes from, you have no reason to believe what you “see.”

“Not so fast!” says the other school, associated with apologists for religion such as William Lane Craig. This school would say that the argument of the last paragraph commits the Genetic Fallacy. The fact that you can explain how (psychologically or otherwise) someone’s belief came about does not entail that the belief is false. The fact that there are psychological mechanisms behind belief in, say, Vishnu—as MCH suggests—does not entail Vishnu does not exist. After all, my belief there is a chair in front of me is also the result of psychological processes, but that doesn’t show there is no chair! New Atheists, claims this school, are committing a fallacy.

What school should we go with?

I don’t propose to resolve the matter. But I wish to make two points.

First, the details of the psychological processes matter. In particular, it matters what triggers the psychological processes that cause a belief. Part of the psychological explanation for why I believe there is a chair is that there is an object before me with photons bouncing off it that hit my retina, which causes signals to travel along my optic nerve, etc. A good psychological theory of normal vision includes positing the object (the chair) in whose existence I come to believe. Notably, the MCH explanation (granting it is only a partial explanation) of belief in any given god does not involve positing the existence of the god. So we should look at the details of the psychological processes that have been empirically demonstrated to see if they include contact with the believed-in entity.

Second, if none of the psychological processes—as far as we can tell empirically—involve actual cognitive contact of some sort with a divinity, then we may fairly consider invoking Occam’s Razor. This is the idea, useful in both science and philosophy, that we shouldn’t posit more entities than are required to explain a phenomenon. If we can explain an event with a simple ontology, why prefer a complex one? If we can give an end-to-end explanation of belief in God without positing God, then that’s one less thing we need to posit in order to explain the religious belief.

So it seems to me that the dialectical situation is this. Craig is right in one sense: showing there are psychological causes of belief in God doesn’t by itself show that God (or a god of any sort) does not exist. But that does not make findings in cognitive science of religion irrelevant to the ontological question of whether God exists. Rather, we’re in the following scenario. There are many things that, prior to a scientific view of things, we might have thought were brought about by a real, existing divinity. One of those many things is credence itself in the divinity. But as science shows, one by one, that each such phenomenon can be explained without positing a divinity, we get closer and closer to the point where it simply makes sense to use Occam’s Razor to remove the divinity from our ontology—from the set of things we think exist.

Religious credence and experience are, perhaps, some of the last holdouts in terms of what actually has been explained without positing a divinity. But cognitive science of religion is making rapid strides. And that’s why it makes theists nervous.