One of the most important questions in the cognitive science of religion is, “Why do people believe in God or gods?” It seems to boggle the mind: how on earth can people seriously believe propositions that lack any concrete evidence? After all, we believe in chairs and dachshunds because those things obviously exist. We can see them, touch them, hear them. There’s no equivalent evidence for the resurrected Christ or an all-powerful God. But one philosopher of cognition, Neil Van Leeuwen, argues that this difference actually means that religious beliefs are different from normal beliefs. In fact, they’re a lot more like play.
In a 2014 paper in the journal Cognition, Van Leeuwen laid out a systematic argument that “religious credence is not factual belief.” Everyday factual beliefs, Van Leeuwen posited, are fairly straightforward. They rely on the evidence of our senses, apply across contexts, and invariably inform our actual decisions. For example, if you live on 7th Street in a two-story house that was built in 1899, then you probably believe that you live in a 118-year-old two-story house on 7th Street. When you have to write your address on a government form, you write down 7th Street. When your mother-in-law visits from out of town, you put her in the first-floor guest room, because you know she has trouble walking down stairs.
In other words, normal factual beliefs influence our everyday, practical decisions. What’s more, when we encounter new information, we change our factual beliefs. If you happen to go through your town’s tax records and discover that your house was actually built in 1901, not 1899, then you won’t keep telling people your house is 118 years old. You’ll adjust your belief according to the evidence, and tell people that your house is 116 years old.
Religious beliefs, Van Leeuwen argues, are fundamentally different. Instead of being vulnerable to new information, religious beliefs are likely to persist even in the fact of contrary evidence. Early Christians, like the Apostle Paul, really seemed to believe that Jesus would be coming back any day now. But as the years – then centuries! – went on and the Second Coming conspicuously failed to make its scheduled appearance, Christians didn’t give up or update their beliefs about the End Times. They kept reading the letters of Paul. They kept talking about the coming of the Kingdom.
This seems very different from normal beliefs about, say, the price of goods. Let’s say that a second-century Christian believed that the price of a bushel of wheat was a half-day’s wage. Depending on what specific year she was living in, that might have been a reasonable belief. But let’s say that she then went to the market and discovered that a bushel actually cost only third that amount. She would probably quickly update her belief about the price of wheat. No problem. But as she walked home with that bushel of wheat, her beliefs about the Second Coming wouldn’t change, despite the fact that by now several full generations had passed since Paul’s time.
Not only are religious beliefs relatively invulnerable to new evidence, but they’re also subject to what Van Leeuwen calls “free elaboration.” This means that religious believers often derive new propositions from prior religious beliefs in ways that don’t depend on logical induction or deduction. To support his case, Van Leeuwen cites the colonial-era preacher Jonathan Edwards’s infamous sermon “Sinners in the Hands of an Angry God.” In that sermon, Edwards claimed that God was more angry with some people in the very congregation he was preaching to than with many of the damned who were already in hell. This shocking claim was compatible with established Christian doctrine, but it wasn’t entailed by it. It was as if you knew that you owned a Volkswagen, but then extrapolated from that knowledge to claim that there was also a Land Rover in your garage. Sure, having a Land Rover is compatible with having a Volkswagen, but you can’t logically use the mere fact of having one as an convincing argument that you own the other. In religious beliefs, though, you can.
Finally, religious beliefs are subject to “special authority.” Van Leeuwen argues that the authorities we trust for our information on religious matters have a special moral status, conferred by the communities in which those authorities operate. The moral nature of religious authority is what sets it apart from other kinds of authority. We trust an electrician to be knowledgeable about the wiring in our house. If we find out that he cheated on his wife, we might dislike him, but this won’t make us doubt his expertise in electrical wiring. By contrast, a serious moral failing in a religious leader will usually undermine that leader’s credibility regarding religious issues.
We Moralize Religious Beliefs. We Don’t Moralize Factual Ones.
This brings us to a pivotal point in Van Leeuwen’s argument: religious beliefs have a moral force. That doesn’t mean that religious doctrines are objectively moral, of course. But it does mean that believers perceive them as having moral weight. Disbelieving them or rejecting them seems to be an immoral thing to do. This really is different from factual beliefs – you may believe there’s a toaster in your kitchen, but there’s no particular moral weight to that belief. If one of your roommates says, “No, actually, there’s no toaster in the kitchen anymore. Rob took it with him when he moved out,” then you won’t look at that roommate as if she were an evil person. You’ll just shrug and say, “Ah. I guess we need to get a new toaster.”
Now, let’s consider what morality actually does. We place moral prescriptions and proscriptions on actions or beliefs over which we have some level of control. In a democracy, it’s a moral duty to vote. Why? Because it’s perfectly possible not to vote. You could just stay home and play Xbox. It’s also possible to cheat on your taxes, litter, or collude with a foreign government to undermine elections. We have moral proscriptions against those things because it’s possible to do them. Cognitively and sociologically, moral overtones come into play when there’s a possibility of violating a norm or rule. So when religious believers feel moral pressure to affirm particular doctrines, this tells us that it’s possible not to affirm those doctrines. Already, then, religious beliefs seem to be different than regular beliefs. It’s really not possible to believe that objects fall up when dropped, or that snow is hot. Ergo, society doesn’t place moral pressure on us to believe that things fall down or that snow is cold. If religious beliefs really were the same as other kinds of beliefs, why would we need moral pressure to enforce them?So: religious beliefs aren’t vulnerable to contrary evidence, are subject to free elaboration, and depend on special authorities whose credibility is fundamentally rooted in their moral standing. To Van Leeuwen, these observations indicate that religious credences are something closer to play than to serious everyday facts. We can claim that Paris is the capital of France, and that Krishna is an avatar of the god Vishnu – but we don’t mean these claims in the same way. If a devout Hindu informs you that Krishna is really Vishnu, the creator and destroyer of all, she’s not expressing the same kind of proposition as when she tells you that the atomic number of beryllium is four. As a Hindu, she’s speaking in something closer to the mode of play, or of poetry. It’s almost as if there are impish scare quotes around the religious claim.
The Skeptics Strike Back
Of course, not everyone agrees with Van Leeuwen’s argument. The philosopher and skeptic Maarten Boudry, together with the biologist and atheist Jerry Coyne, published a critical response to Van Leeuwen in the journal Philosophical Psychology, in which they argued that religious believers really do believe the things they say they do. Otherwise, why would events like the September 11, 2001 terrorist attacks happen? If the terrorists on those airplanes had only “believed” that God wanted them to kill innocent civilians (instead of believing without scare quotes), wouldn’t they have pulled up at the last minute, saying “Just kidding?” Clearly, many religious people do things that would be difficult to justify if they didn’t actually accept the beliefs.
In fact, as I’ve recently written here, credibility enhancing displays, or CREDs, are ritual behaviors that only make sense if the performer actually believes the underlying tenets. If you’re a Jew who doesn’t really believe that God exists, it’s tough to keep motivating yourself to obey all the restrictive Shabbat regulations. The fact that people are willing to refrain from shopping, turning on lights, or carrying anything from one house to another seems to suggest that they really believe God cares about those things. Otherwise, why on earth would they go to all the trouble? Similarly, costly signaling theory predicts that, all things being equal, religious groups with very costly rituals will survive and thrive better than those with weak rituals, because costly requirements will weed out members who don’t really believe.
Van Leeuwen responds to such criticisms by arguing that religious beliefs are a kind of “secondary cognitive attitude,” more similar to patriotism or other “identity-constituting” credences. As such, religious beliefs really are more like make-believe – but we all know that make-believe games can get out of hand. So can patriotism. Extrapolating from Van Leeuwen’s point, when people do extraordinary or destructive things based on their religious beliefs, it may not be too different from a basketball player pushing his body to the very edges of endurance during a game. Does the player really believe that the court boundaries are “out-of-bounds” in some metaphysical sense, or that “points” (scored by the made-up act of putting a ball through a hoop) are the same kind of ontological category as “maple wood” (which really grows in forests)? No. As soon as the game is over, the player will walk off the court, ignoring the out-of-bounds barrier completely. But while the game is on, he’ll push his body as hard as a human body can physically be pushed – in service of a bunch of made-up rules.
Offering empirical support for his claim, Van Leeuwen cites a recent neuroimaging study. The authors of this study found that the neural processing of supernatural beliefs looks more like the neural processing of metaphorical concepts than like literal ones. In other words, brains seem to treat religious beliefs as metaphors, not literal facts.
Ritual is Evidence that Religious Beliefs Are Play-Like
I think Van Leeuwen is mostly right here, despite Boundry’s and Coyne’s useful criticisms. If religious beliefs really were the same as everyday beliefs, why would there be so many taboos and so much moral pressure surrounding them? And why would believers spend so much time conducting rituals and reciting creeds? You don’t need to recite the law of gravity to yourself daily to believe it. But many Christian churches recite the Apostle’s Creed every week, in unison. This ritual serves to make the tenets of Christianity more salient, reinforcing them in the minds of attendees. As the historian of religion Jonathan Z. Smith writes,
Ritual is a means of performing the way things ought to be in conscious tension to the way things are.
The fact that religious beliefs are something closer to play than to factual assertions doesn’t mean that people don’t take them seriously. It also doesn’t mean that religious beliefs can’t start wars – or inspire people to devote their lives to the poor. After all, patriotic beliefs are just as metaphorical and just as non-literal as religious ones (E Pluribus Unum!), and patriotism inspires people to commit all kinds of sacrifices, violence, and extreme acts. It just means that religious beliefs aren’t quite straightforward. They require more scaffolding than normal beliefs, and that scaffolding comes in the form of ritual, serious play, and social pressure. As usual, the story about religion is a lot more complicated, and a lot more interesting, than it looks to the casual outsider. And I really believe that.