The Evolution of God by Robert Wright


excerpt from

How Human Nature Gave Birth to Religion

When something appears in every known society, as religion does, the question of whether it is “in the genes” naturally arises. Did religion confer such benefits on our distant ancestors that genes favoring it spread by natural selection? There are scientists who believe the answer is yes—enough of them, in fact, to give rise to headlines like this one, in a Canadian newspaper: “Search continues for ‘God gene.’”

Expect to see that headline again, for the search is unlikely to reach a successful conclusion. And that isn’t just because, obviously, no single gene could undergird something as complex as religion. Things don’t look good even for the more nuanced version of the “God gene” idea—that a whole bunch of genes were preserved by natural selection because they inclined people toward religion.

Oddly, this verdict—that religion isn’t in any straightforward sense “in the genes”—emerges from evolutionary psychology, a field that has been known to emphasize genetic influences on thought and emotion. Though some evolutionary psychologists think religion is a direct product of natural selection, many—and probably most—don’t.

This doesn’t mean religion isn’t in any sense “natural,” and it doesn’t mean religion isn’t in some sense “in the genes.” Everything people do is in some sense in the genes. (Try doing something without using any genes.) What’s more, we can trace religion to specific parts of human nature that are emphatically in the genes. It’s just that those parts of human nature seem to have evolved for some reason other than to sustain religion.

The American psychologist William James, in his 1902 classic The Varieties of Religious Experience, captured the basic idea without referring to evolution: “There is religious fear, religious love, religious awe, religious joy, and so forth. But religious love is only man’s natural emotion of love directed to a religious object; religious fear is only the ordinary fear of commerce, so to speak, the common quaking of the human breast, in so far as the notion of divine retribution may arouse it; religious awe is the same organic thrill which we feel in a forest at twilight, or in a mountain gorge; only this time it comes over us at the thought of our supernatural relations.”

If you want to put James’s basic point in the language of evolutionary biology, you have to drag in the concept of an “adaptation.” An adaptation is a trait whose underlying genes spread through the gene pool by virtue of their giving rise to that trait. Love, for example, seems to be an adaptation. Love of offspring, by inspiring nurturance of those offspring, can help genes get into future generations; as a result, genes underlying parental love seem to have spread by virtue of their conduciveness to love. You can similarly make arguments that awe and joy and fear—the other sentiments James cites—were, in themselves, adaptations. (Fearing a big aggressive animal, or a big aggressive human being, could save your skin and thus save the genes underlying the fear.) But that doesn’t mean religion is an adaptation, even though religion may involve love, awe, joy, and fear and thus involve the genes underlying these things.

To shift back into less technical terminology: you might say that we were “designed” by natural selection to feel love and awe and joy and fear. (So long as you understand that “designed” is a metaphor; natural selection isn’t like a human designer who consciously envisions the end product and then realizes it, but is rather a blind, dumb process of trial and error.) But to say that these emotions are a product of “design” isn’t to say that when they’re activated by religion they’re working as “designed.”

Similarly, humans were “designed” by natural selection to be able to run and were also “designed” to feel competitive spirit, but that doesn’t mean they were “designed” to participate in track meets. Religion, like track, doesn’t seem to be an “adaptation.” Both seem to be what the paleontologist Stephen Jay Gould called a “spandrel”—a phenomenon supported by genes that had become part of the species by doing something other than supporting that phenomenon. A spandrel is an incidental by-product of the organic “design” process, whereas an adaptation is a direct product. Religion seems to be a spandrel.

And yet, you might say, religion does have the hallmarks of design. It is a complex, integrated system that seems to serve specific functions. For example, religions almost always handle some key “rites of passage”—getting married, getting buried, and so on—whose ritualized handling is probably good for the society. How do you explain the coherence and functionality of religion without appealing to a designer—or, at least, a “designer”?

You don’t. But biological evolution isn’t the only great “designer” at work on this planet. There is also cultural evolution: the selective transmission of “memes”—beliefs, habits, rituals, songs, technologies, theories, and so forth—from person to person. And one criterion that shapes cultural evolution is social utility; memes that are conducive to smooth functioning at the group level often have an advantage over memes that aren’t. Cultural evolution is what gave us modern corporations, modern government, and modern religion.

For that matter, it gave us nonmodern religion. Whenever we look at a “primitive” religion, we are looking at a religion that has been evolving culturally for a long time. Though observed hunter-gatherer religions give clues about what the average religion was like 12,000 years ago, before the invention of agriculture, none of them much resembles religion in its literally primitive phase, the time (whenever that was) when religious beliefs and practices emerged. Rather, what are called “primitive” religions are bodies of belief and practice that have been evolving—culturally—over tens or even hundreds of millennia. Generation after generation, human minds have been accepting some beliefs, rejecting others, shaping and reshaping religion along the way.

So to explain the existence of “primitive” religion—or for that matter any other kind of religion—we have to first understand what kinds of beliefs and practices the human mind is amenable to. What kinds of information does the mind naturally filter out, and what kinds naturally penetrate it? Before religion appeared and started evolving by cultural evolution, how had genetic evolution shaped the environment in which it would evolve—that is, the human brain?

To put the question another way: What kinds of beliefs was the human mind “designed” by natural selection to harbor? For starters, not true ones.

At least, not true ones per se. To the extent that accurate perception and comprehension of the world helped humanity’s ancestors get genes into the next generation, then of course mental accuracy would be favored by natural selection. And usually mental accuracy is good for the survival and transmission of the genes. That’s why we have excellent equipment for depth perception, for picking up human voices against background noise, and so on. Still, in situations where accurate perception and judgment impede survival and reproduction, you would expect natural selection to militate against accuracy.

Truth and Consequences

In 1974, San Francisco newspaper heiress Patty Hearst was kidnapped by a radical group called the Symbionese Liberation Army, whose goals included “death to the fascist insect that preys upon the life of the people.” After being kept in a closet for a while, she came to identify with her new peer group. Before long, she was enthusiastically helping them generate income, at one point brandishing a machine gun during a bank robbery. When left alone, with an opportunity to escape, she didn’t take it.

She later described the experience: “I had virtually no free will until I was separated from them for about two weeks. And then it suddenly, you know, slowly began to dawn that they just weren’t there anymore. I could actually think my own thoughts.” Hearst didn’t just accept her captors’ “subjective” beliefs, such as ideology; she bought into their views about how the physical world works. One of her captors “didn’t want me thinking about rescue because he thought that brain waves could be read or that, you know, they’d get a psychic in to find me. And I was even afraid of that.”

Hearst’s condition of coerced credulity is called the Stockholm syndrome, after a kidnapping in Sweden. But the term “syndrome” may be misleading in its suggestion of abnormality. Hearst’s response to her circumstances was probably an example of human nature functioning properly; we seem to be “designed” by natural selection to be brainwashed.

Some people find this prospect a shocking affront to human autonomy, but they tend not to be evolutionary psychologists. In Darwinian terms, it makes sense that our species could contain genes encouraging blind credulity in at least some situations. If you are surrounded by a small group of people on whom your survival depends, rejecting the beliefs that are most important to them will not help you live long enough to get your genes into the next generation.

Confinement with a small group of people may sound so rare that natural selection would have little chance to take account of it, but it is in a sense the natural human condition. Humans evolved in small groups—twenty, forty, sixty people—from which emigration was often not a viable option. Survival depended on social support: sharing food, sticking together during fights, and so on. To alienate your peers by stubbornly contesting their heartfelt beliefs would have lowered your chances of genetic proliferation.

Maybe that explains why you don’t have to lock somebody in a closet to get a bit of the Stockholm syndrome. Religious cults just offer aimless teenagers a free bus ride to a free meal, and after the recruits have been surrounded by believers for a few days, they tend to warm up to the beliefs. And there doesn’t have to be some powerful authority figure pushing the beliefs. In one famous social psychology experiment, subjects opined that two lines of manifestly different lengths were the same length, once a few of their “peers” (who were in fact confederates) voiced that opinion.

Given this conformist bias in human nature, it’s not surprising that people born into “primitive” religions—or any other religions—accept an elaborate belief system that outside observers find highly dubious. But the question remains: How did the elaborate belief system ever come to exist? Granted that people are inclined to accept their community’s official edifice of belief and ritual (especially if no alternatives are on offer). But how did the edifice come to exist in the first place? How did religion get built from the ground up?

God Bites Man

To answer this question we have to view cultural evolution at a fine-grained level. We have to think about individual units of culture—beliefs and practices, in this case—and how they spread. The biologist Richard Dawkins coined the term “meme” for units of culture, in part because it sounds a bit like “gene,” and he wanted to stress some parallels between cultural and biological evolution. For example: just as genes are transmitted from body to body, down the generations, memes are transmitted from mind to mind. And just as newly minted genes “compete” for a place in the gene pool, newly minted memes “compete” for the finite space in the world’s supply of human brains. In this constant struggle of meme against meme, what kinds of memes will have a “selective advantage”?

Newspapers are a good place to look for clues. Newspaper editors work hard to figure out what kinds of information people want, and to fill that demand. They are accomplished meme engineers, and thus students of human nature. One thing you’ll notice about newspapers is that they have a bias toward good things and bad things. The headlines “Stock market rises by 5 percent” and “Stock market drops by 5 percent” will get better play than the headline “Stock market does nothing in particular.” Here religions, and certainly “primitive” religions, are like newspapers. In every hunter-gatherer society, religion is devoted largely to explaining why bad things happen and why good things happen—illness, recovery; famine, abundance; and so on.

There is also devotion to raising the ratio of good to bad. The Andaman Islanders, convinced that whistling at night attracts bad spirits whereas singing repels them, do more singing in the dark than whistling in the dark. People naturally try to exert control over their environment, and believing that they have such control naturally makes them feel good. So people’s minds are open to ideas that promise to give them such control. This doesn’t mean people uncritically embrace every such idea that comes their way. But it does mean that these ideas get their attention—and for a meme, that’s the first step toward acceptance. While the Andaman Island meme asserting that thunderstorms are divine punishment for melting beeswax was hardly guaranteed a place in the society’s religion, it had a big head start over memes saying, “Thunderstorms just happen—there’s nothing you can do about it.”

Another thing you’ll notice in newspapers is that the strange and novel wins out over the ordinary and expected. Tuberculosis and the West Nile virus are both bad news, and in terms of the number of people killed, tuberculosis is the worse of the two. Yet the headline “Outbreak of deadly new virus puzzles experts” easily crowds out “Usual number of people expected to die of tuberculosis this year” (except, perhaps, in the humor magazine The Onion, which earns its laughs by violating this pattern). As journalism sages famously put it: “Dog bites man” is not a story; “Man bites dog” is a story.

It makes sense that human brains would naturally seize on strange, surprising things, since the predictable things have already been absorbed into the expectations that guide them through the world; news of the strange and surprising may signal that some amendment of our expectations is warranted. But one property of strange, surprising claims is that they’re often untrue. So if they get preferred access to our brains, that gives falsehood a kind of advantage—if a fleeting advantage—over truth. In the days after the terrorist attacks of September 11, 2001, one widely circulated story was that a man at the top of one of the twin towers had survived by sliding down the rubble as it formed. It was a story so incredible that it virtually compelled you to click the “forward” icon on your e-mail—and a story so incredible that it wasn’t true. It was an example of the famous dictum that a lie can get halfway around the world before the truth has a chance to get its boots on.

Of course, in the long run, the truth often does get its boots on, and people often welcome it upon its arrival. Indeed, if the attraction to surprising news weren’t balanced by an attraction to claims that survive subsequent scrutiny, the average human ancestor wouldn’t have lived long enough to become a human ancestor. Imagine a local sage, 200,000 years ago, saying that eating a certain berry will let you live forever. Now imagine that the first two people who follow his advice drop dead. Genes that counseled continued faith in advice thus besieged by countervailing evidence would not long remain a part of the species, whereas genes that inclined the brain to take account of such evidence might. This natural human respect for evidence is the reason convincing someone that one plus one equals three, or that water flows uphill, takes real work.

But some kinds of beliefs are harder to test than those two. And hard-to-test beliefs could do well in the process of cultural evolution that gave birth to religion. Indeed, hunter-gatherer religious belief—like religious belief generally—consists largely of claims that resist falsification. The Haida, a people indigenous to the northwest coast of North America, when caught in a storm while out at sea, would try to appease the relevant authorities (killer whale deities) by pouring a cup of fresh water into the sea or putting some tobacco or deer tallow on the end of a paddle. Many people no doubt returned from sea to report that these measures had kept them from drowning. No one, presumably, ever reported that they had taken these measures but drowned anyway.

To be sure, some religious beliefs can be put to a clearer test. If the Andaman Islanders were right, and melting beeswax was a leading cause of thunderstorms, then a melting moratorium should cut down on thunderstorms. But how can you be sure that, in the days preceding a thunderstorm, no one in your village melted a smidgen of wax—or engaged in some other thunder-inducing activity, such as making a loud noise while the cicadas were singing?

Such loopholes are found in modern religions, too. If you pray for someone to recover from illness, and they don’t, then prayer would seem to have lost credibility. But religions usually have ways of explaining such failure. Maybe you or the sick person had done something horribly wrong, and this is God’s punishment. Or maybe God just works in mysterious ways.

So far, then, we would expect the following kinds of memes to be survivors in the dog-eat-dog world of cultural evolution: claims that (a) are somewhat strange, surprising, counterintuitive; (b) illuminate sources of fortune and misfortune; (c) give people a sense that they can influence these sources; (d) are by their nature hard to test decisively. In this light, the birth of religion doesn’t seem so mysterious.

But doesn’t our attraction to strangeness have its limits? It’s one thing to believe that a man could survive a slide down a crumpling skyscraper through a series of lucky breaks. It’s quite another to believe, with the Inuit (in chapter 1), that a sudden shortage of game is the work of a pouty female deity who lives at the bottom of the sea. In other words, “Man bites dog,” however unlikely, seems more plausible than “God bites man.”

Lord of the Chimps

But, actually, the idea of a personal god or spirit who peevishly withholds food, or maliciously hurls lightning, gets a boost from the evolved human brain. People reared in modern scientific societies may consider it only natural to ponder some feature of the world—the weather, say—and try to come up with a mechanistic explanation couched in the abstract language of natural law. But evolutionary psychology suggests that a much more natural way to explain anything is to attribute it to a humanlike agent. This is the way we’re “designed” by natural selection to explain things. Our brain’s capacity to think about causality—to ask why something happened and come up with theories that help us predict what will happen in the future—evolved in a specific context: other brains. When our distant ancestors first asked “Why,” they weren’t asking about the behavior of water or weather or illness; they were asking about the behavior of their peers.

That’s a somewhat speculative (and, yes, hard-to-test!) claim. We have no way of observing our prehuman ancestors one or two or three million years ago, when the capacity to think explicitly about causality was evolving by natural selection. But there are ways to shed light on the process.

For starters, we can observe our nearest nonhuman relatives, chimpanzees. We didn’t evolve from chimps, but chimps and humans do share a common ancestor in the not-too-distant past (4 to 7 million years ago). And chimps are probably a lot more like that common ancestor than humans are. Chimps aren’t examples of our ancestors circa 5 million BCE but they’re close enough to be illuminating.

As the primatologist Frans de Waal has shown, chimpanzee society shows some clear parallels with human society. One of them is in the title of his book Chimpanzee Politics. Groups of chimps form coalitions—alliances—and the most powerful alliance gets preferred access to resources (notably a resource that in Darwinian terms is important: sex partners). Natural selection has equipped chimps with emotional and cognitive tools for playing this political game. One such tool is anticipation of a given chimp’s future behavior based on past behavior. De Waal writes of a reigning alpha male, Yeroen, who faced growing hostility from a former ally named Luit: “He already sensed that Luit’s attitude was changing and he knew that his position was threatened.”

One could argue about whether Yeroen was actually pondering the situation in as clear and conscious a way as de Waal suggests. But even if chimps aren’t quite up to explicit inference, they do seem close. If you imagine their politics getting more complex (more like, say, human politics), and them getting smarter (more like humans), you’re imagining an organism evolving toward conscious thought about causality. And the causal agents about which these organisms will think are other such organisms, because the arena of causality is the social arena. In this realm, when a bad thing happens (like a challenge for Yeroen’s alpha spot) or a good thing happens (like an ally coming to Yeroen’s aid), it is another organism that is making the bad or good thing happen.

To be sure, other kinds of bad and good things happen to chimps: droughts, banana bonanzas, and so on. But there’s no reason to think chimps are anywhere near consciously puzzling over those things—trying to anticipate droughts the way they try to anticipate the behavior of their neighbors. And there’s no reason to think that our prehuman ancestors were, either. The best guess is that when natural selection built the mental machinery for predictively pondering causality, the causal agents in question were peers—fellow prehumans. (Is he going to punch me? Is she going to betray me?) Moreover, when our ancestors first started talking about causality, they were probably talking about peers. (Why did you punch me? Do you know why she betrayed me?)

I’m not just talking about a habit. I’m not saying our ancestors were used to pondering questions of “Why?” by thinking about human beings. I’m suggesting that the human mind is built to do that—was “designed” by natural selection to do it.

So it’s no surprise that when people first started expanding their curiosity, started talking about why bad and good things emanate from beyond the social universe, they came up with the kinds of answers that had made sense within their social universe. To answer a “why” question—such as “Why did the thunderstorm come just as that baby was being born?”—with anything other than a humanlike creature would have been kind of strange.

More than one hundred years ago Edward Tylor wrote that “spirits are simply personified causes,” but he probably didn’t appreciate, back then, how deeply natural personification is. Indeed, to talk about “personifying” causes is in a sense to get the story backward. Better to say that the modern scientific notion of a “cause” is a depersonified human being—or a depersonified god.

Even in modern science, the depersonification process may not be complete. Some philosophers believe that to chop the world up into “causes” and “effects” is to impose a falsely binary scheme on what is in fact a seamless reality. It may be that our “modern” way of thinking about causality still carries the vestiges of our primitive brains, still falsely reflects a social arena of causality, in which “causes” are distinct agents.

Spirits with Legs

The idea that gods and spirits got their start as a supernatural version of people encounters one obvious objection: In hunter-gatherer societies, aren’t some supernatural beings thought of as animals, not humans? And aren’t some supernatural beings—especially the ones anthropologists call “spirits”—too vague a life form to qualify as either human or animal? Why should we, along with Tylor, talk of personified causes when Tylor’s own terminology—animated causes—would fit better?

For starters, however unlike a human a “spirit” may sound, when anthropologists ask people to draw pictures of spirits, the pictures usually look more or less like humans: two arms, two legs, a head. Similarly, a great god of the ancient Chinese was called Tian, or “heaven,” which sounds quite unlike a person—yet the earliest written symbol for that god is a stick figure: two arms, two legs, a head.

Even when a supernatural being looks like an animal—as those snowmaking birds of the Klamath (chapter 1) presumably did—it doesn’t act like an animal. It may have wings or fur or scales, and it may lack various parts of a normal human being, but it won’t lack the part that explains why human beings do the things they do. As the anthropologist Pascal Boyer has observed, “the only feature of humans that is always projected onto supernatural beings is the mind.”

Boyer believes that the genetic architecture of human cognition helps explain why people conceive of gods as they do. The mind, he says, comes with built-in assumptions about reality. People naturally divide the world into a few basic “ontological categories”—such as plants, animals, human beings—and attribute certain properties to beings that fall into a certain category. In other words, we have a mental “template” that helps us think about plants, and a different template for people, and so on. We assume that if we walk up and whack a person, the person won’t like it and may retaliate, whereas whacking a plant is less hazardous. In Boyer’s view, when people think about a god or spirit, their brains are invoking the template for human beings, but in amended form, with a few of the template’s normal properties changed. Thus, the Abrahamic God is a lot like a person—capable of love, anger, disappointment, jealousy—except that he knows everything and can do anything.

To some people, this last part—omniscience and omnipotence—strains credulity. In a modern scientific culture, that’s no surprise. But Boyer’s work suggests that such scarcely credible features would have been an asset to a god meme that was just getting off the ground tens of thousands of years ago. His experiments show that things with starkly counterintuitive features—things with properties that aren’t part of their template—are especially memorable. If you tell someone about a table “that felt sad when people left the room,” they are more likely to remember it months later than if you tell them about a normal table, one with the unshakable stoicism generically associated with furniture. Presumably they are more likely to tell people about it as well. So memes that depict gods as unlike anything you’ve ever seen would have a kind of advantage over more “plausible” memes.

So long as the strangeness of these gods didn’t get excessive, that is. Boyer says that the meme most conducive to spreading would be one that is strange but easy to think about: it might have one or two basic “ontological violations,” such as omniscience and omnipotence, but these violations wouldn’t be so numerous, or so quirky, that imagining the behavior of such a deity was unwieldy.

In fact, even traits like omniscience and omnipotence seem to press against the limits of imagination. When two psychologists quizzed people about the properties of a supreme being, the answers were overwhelmingly “theologically correct”—omniscience, ubiquity, and so on. But then these same people were led to think more concretely, to imagine God actually exerting influence in specific situations. Suddenly they conjured up a more human deity. They thought of God as occupying a single point in space and being unable to do two things at once, and, in the words of one of the psychologists, “needing to see and hear in order to complete otherwise fallible knowledge.”

This points to a problem for modern theology: as divinity is defined more abstractly to fit more comfortably into a scientific worldview, God becomes harder for people to relate to. In the mid-twentieth century, when Paul Tillich defined God as “the ground of being,” some fellow theologians approved, but he was also met with dismay, incomprehension, and the occasional charge of atheism. Still, he could rightly have replied that his critics suffered from innately narrow vision; being humans, they labored under the handicap of minds designed to fathom the social universe, not the universe writ large.

Dealing with the Supernatural

According to the book of Genesis, “God created man in his own image.” According to Aristotle, “men create the gods after their own image.” As should be clear by now, Aristotle seems to have been onto something, especially when it comes to the minds of gods. So, in theory, some of the more basic features of the human mind should be fairly standard equipment in gods, especially the gods of “primitive” religions.

That seems to be the case, and one of these features deserves special consideration: the part of the human mind shaped by the evolutionary dynamic known as “reciprocal altruism.” In light of this dynamic, much about the origin of religion, and for that matter much about contemporary religion, makes a new kind of sense.

Thanks to reciprocal altruism, people are “designed” to settle into mutually beneficial relationships with other people, people whom they can count on for things ranging from food to valuable gossip to social support, and who in turn can count on them. We enter these alliances almost without thinking about it, because our genetically based emotions draw us in. We feel gratitude for a favor received, along with a sense of obligation, which may lead us to return the favor. We feel growing trust of and affection for people who prove reliable reciprocators (aka “friends”), which keeps us entwined in beneficial relationships. This is what feelings like gratitude and trust are for—the reason they’re part of human nature.

But of course, not everyone merits our trust. Some people accept our gifts of food and never reciprocate, or try to steal our mates, or exhibit disrespect in some other fashion. And if we let people thus take advantage of us day after day, the losses add up. In the environment of our evolution, these losses could have made the difference between surviving and not surviving, between prolifically procreating and barely procreating. So natural selection gave us emotions that lead us to punish the untrustworthy—people who violate our expectations of exchange, people who seem to lack the respect that a mutually beneficial relationship demands. They fill us with outrage, with moral indignation, and that outrage—working as “designed”—impels us to punish them in one way or another, whether by actually harming them or just by withholding future altruism. That will teach them! (Perhaps more important, it will also teach anyone else who is watching, and in the ancestral hunter-gatherer environment, pretty much everyone in your social universe was watching.)

This is the social context in which the human mind evolved: a world full of neighbors who, to varying degrees, are watching you for signs of betrayal or disrespect or dishonesty—and who, should they see strong evidence of such things, will punish you. In such a social universe, when misfortune comes your way, when someone hits you or ridicules you or suddenly gives you the cold shoulder, there’s a good chance it’s because they feel you’ve violated the rules of exchange. Maybe you’ve failed to do them some favor they think they were due, or maybe you’ve shown them disrespect by doing something that annoys them.

Surely it is no coincidence that this generic explanation of why misfortune might emanate from a human being is also the generic explanation of why misfortune emanates from gods. In hunter-gatherer religions—and lots of other religions—when bad things happen, the root cause is almost always that people in one sense or another fail to respect the gods. They either fail to give gods their due (fail, say, to make adequate sacrifices to ancestral spirits), or they do things that annoy gods (like, say, making a noise while cicadas are singing). And the way to make amends to the aggrieved gods is exactly the way you’d make amends to aggrieved people: either give them something (hence ritual sacrifice), or correct future behavior so that it doesn’t annoy them (quit making noises while cicadas are singing).

In this light, bizarre superstitions seem less bizarre. The Ainu, the indigenous hunter-gatherers of Japan, scrupulously refrained from spitting into fires. Strange! But if you accept their premise that hearth fires come courtesy of the fire goddess, the rest follows. You don’t do things that insult people who give you gifts, because if you do they’ll get in a snit and stop giving. And one thing that might well be taken as an insult is to spit on the gift.

Boyer believes that much of religion can be explained this way—a result of our attributing to supernatural causal agents the very human emotions that evolved to regulate reciprocal altruism; like our fellow human beings, gods are bent on enforcing their deals with us. This doesn’t mean that the grievances of gods are always just. Evil deities, Boyer says, are “enforcers of unfair deals.” But it’s only natural that there should be such unfair gods; there are, after all, unfair people. (And people who can get away with being unfair—that is, can get more than they give—tend to be powerful, like gods.)

Two and a half millennia ago the Greek poet Xenophanes speculated that if horses had gods, these gods would be horses. Could be, but we’ll never know, and in any event that’s not quite the point being made here. It isn’t that any imaginable intelligent species, in trying to explain mysterious things, would attribute them to beings like itself. It’s that the history of the human species—notably including the evolution of the human brain in a context of reciprocal altruism, of social exchange—pointed it in that direction. A law of the social jungle in which the human brain evolved is this: when bad things happen to you, it often means someone is mad at you, maybe because you’ve done something to offend them; making amends is often a good way to make the bad things stop happening. If you substitute “some god or spirit” for “someone,” you have a law that is found in every known hunter-gatherer religion.

Back into Time

That religious ideas naturally appeal to the human mind doesn’t, by itself, explain how religion got off the ground. Granted that religious “memes” have a “selective advantage” in cultural evolution, how exactly would a given meme—a particular religious belief—first take shape and gain momentum? We’ll never know for sure, but human nature makes it easy to sketch a plausible scenario.

First, people like to command attention, and one way to do that is to place yourself at the center of dramatic events. In Mark Twain’s The Adventures of Tom Sawyer, Tom Sawyer runs away with his friends Huckleberry Finn and Joe to play pirates on the Mississippi River, and the townspeople conclude that the boys have drowned. Twain describes their friends gathering and

talking in reverent tones of how Tom did so-and-so, the last time they saw him, and how Joe said this and that small trifle (pregnant with awful prophecy, as they could easily see now!)—and each speaker pointed out the exact spot where the lost lads stood at the time, and then added something like “and I was a-standing just so—just as I am now, and as if you was him—I was as close as that—and he smiled, just this way—and then something seemed to go all over me, like—awful, you know—and I never thought what it meant, of course, but I can see now!”

     Then there was a dispute about who saw the dead boys last in life, and many claimed that dismal distinction, and offered evidences, more or less tampered with by the witness; and when it was ultimately decided who did see the departed last, and exchanged the last words with them, the lucky parties took upon themselves a sort of sacred importance, and were gaped at and envied by all the rest.

There is no reason to think that this incentive to claim special witness to high drama was any less powerful among hunter-gatherers circa 30,000 BCE than among midwestern Americans circa 1900 CE. Imagine that you are one of those hunter-gatherers and you walk past a place where someone died mysteriously, and you hear leaves rustle eerily. That’s a story that will get people’s attention, and you can heighten the attention by stressing how exquisitely timed the rustling was. And, by the way, didn’t you catch sight of a shadowy—almost ethereal—creature out of the corner of your eye?

The anthropologist Stewart Guthrie has suggested that hunter-gatherers would be encouraged to make just such false sightings by standard human mental equipment—something called a “hyperactive agent-detection device.” Because the costs of failing to detect a predator lurking in the woods are much higher than the costs of detecting one that isn’t there, natural selection, he plausibly argues, may have biased our brains toward “false positives”: you hear a rustling, your mind flashes the vivid hypothesis of some generic animal that’s doing the rustling, and you turn toward it expectantly. Did you actually see something? Kind of.

In any event, if upon recounting your eerie encounter you get caught up in the spirit of the story and say you saw an ethereal being, then you may convince not just your audience, but yourself. One notable finding of modern psychology is how systematically misleading memory is. People often remember events wrongly from the get-go, and even when they don’t, their memory can later be steered toward falsehood. In particular, the act of reporting false details can cement them firmly in mind. You don’t just recount what you remember; you remember what you recount. (Football star O. J. Simpson’s former agent was sure Simpson had killed his ex-wife and also sure that Simpson believed he didn’t.) This built-in fallibility makes sense from a Darwinian standpoint, allowing people to bend the truth self-servingly with an air of great and growing conviction. And, clearly, bent truths of a religious sort could be self-serving. If you were a close friend or relative of the deceased, then the idea that his powerful spirit is afoot may incline people to treat you nicely, lest they invite his wrath.

Another gem from social psychology: publicly espousing something not only helps convince you of its truth; it shapes your future perception, inclining you to see evidence supporting it but not evidence against it. So if you speculate that the strange, shadowy creature was the disgruntled spirit of the deceased, you’ll likely find corroboration. You may notice that one of his enemies fell ill only a week after your sighting, while forgetting that one of his friends fell ill a few days earlier.

If you’re a person of high status, all of this will carry particular weight, as such people are accorded unusual (and often undue) credibility. If, in a hunter-gatherer band of thirty people, someone widely esteemed claims to have seen something strange—and has a theory about what it was—twenty people may be convinced right off the bat. Then the aforementioned tendency of people to conform to peer opinion could quickly yield unanimity.

The number of mental tendencies involved in the creation and nourishment of religious falsehoods shouldn’t surprise us. After all, the mind was built by a process that is, strictly speaking, indifferent to truth. Natural selection favors traits that are good at getting their bearer’s genes into the next generation, period. If saying something false, or believing something false, often furthered that goal during human evolution, then the human mind will naturally encourage some kinds of falsity. This systematic muddle isn’t an exclusive property of the “primitive” mind, as John Lubbock (chapter 1) suggested; all of the above delusory tendencies have been documented in people living in modern societies—many of them students at fine universities!

So why are people in modern societies so often aghast at “primitive” religion, so unable to comprehend how “primitive” belief got started? In part, it is the classic human failure of objectivity—an inability to see that your own beliefs may seem as strange to others as theirs seem to you. (An African Pygmy once responded to a missionary’s description of heaven by asking, “How do you know? Have you died and been there?”) And in part it is a failure of imagination. Imagine that you are living in a small encampment surrounded by jungle or woodland or desert, entirely untouched by science and modern technology. Within the encampment, the social universe operates by largely intelligible laws; people don’t generally, say, fly into a rage and assault their neighbors without a cause of some sort or another. But from outside this universe come mighty and momentous forces—storms, droughts, deadly animals, fatal illness. You are viscerally interested in explaining and controlling these things; you readily absorb and repeat any news or conjecture bearing on this goal. And, above all, you are only human. The rest is history.

Thinking and Feeling

This view of religion’s origins—the view from modern psychology—is in some ways just an updated version of Edward Tylor’s view: people first conceived of gods and spirits to explain the unexplained. Indeed, Tylor even seemed to vaguely anticipate the modern focus on reciprocal altruism: “Spiritual beings are held to affect or control the events of the material world, and man’s life here and hereafter; and it being considered that they hold intercourse with men, and receive pleasure or displeasure from human actions, the belief in their existence leads naturally, and it might almost be said inevitably, sooner or later to active reverence and propitiation.”

Still, there is a difference of emphasis. When Tylor says belief in gods “leads naturally” to their propitiation, he seems to mean that this progression was logically natural—that extended reflection led eventually to the conclusion that giving gods respect and foodstuffs would satisfy them. An evolutionary psychologist, in contrast, might stress how viscerally natural this propitiation is; it feels like the right thing to do. Tylor’s oft-ridiculed reference to “ancient savage philosophers” (see chapter 1) does indeed connote more in the way of cool, detached reflection than was probably operative—and than is operative generally in human beings. Some features of the mind that undergird religious belief are “cognitive” traits that guide our “intellectual” lives but are also shot through with feeling.

Varieties of Religious Experience

In addition to our mental machinery for thinking consciously about causality—the machinery shaped by the evolution of reciprocal altruism—there are other innate tools for taking causality into account, and some of them operate almost entirely at the level of feeling.

For example, back when our ancestors didn’t know that disease travels by microscopic organism, natural selection seems to have filled this knowledge gap, installing in our lineage an aversion to disease-carrying things. That is the conclusion the psychologist Paul Rozin reached by studying disgust. It’s no coincidence, he believes, that things which fill people everywhere with disgust—rotting corpses, excrement, putrid meat—are hazardous to our health.

However unsophisticated a feeling disgust may seem like, it actually entails a kind of metaphysics: a sense that some things are deeply impure and emit an invisible aura of badness, creating a dread zone. Pascal Boyer has suggested that disgust—our “contagion inference system”—may thus energize notions of ritual pollution that figure in many religions. (Recall the sin that so peeved the sea goddess in chapter 1: failing to throw out items contaminated by proximity to a miscarriage.)

There is another feature of the human mind that may be involved in religious experience and that, like the “contagion inference system,” is a way of taking account of causality without thinking consciously about it. In fact, it entered our lineage so long before consciously rational thought that it exists in all mammals. It is called “associative learning.”

If a dog burns itself on rocks that surround a dying campfire, it will thereafter avoid such rocks. What is going on in the dog’s mind is hard to say, but it probably isn’t extended reflection on the causal link between fires and hot rocks, or between hot rocks and singed fur. Presumably the dog has just acquired something like a fear of those rocks, a fear that leads it to behave as if it understood the connection between rocks around dying campfires and singed fur. I once tried to walk a golden retriever past an intersection where, weeks earlier, she had been hit by a car. As we approached the intersection, she walked more and more slowly and warily until finally she came to a halt and started desperately resisting attempts to move her farther. It was as if, in her mind, the intersection was giving off a kind of spooky aura, and the closer she got to it, the stronger the aura felt.

Vestiges of this kind of crude learning mechanism in the human brain may incline people to see objects or places as inhabited by evil, a perception that figures in various religions. Hence, perhaps, the sense of dread that has been associated by some anthropologists with primitive religious experience.

And what of the sense of awe that has also been identified with religious experience—most famously by the German theologian Rudolf Otto (who saw primordial religious awe as often intermingled with dread)? Was awe originally “designed” by natural selection for some nonreligious purpose? Certainly feelings of that general type sometimes overtake people confronted by other people who are overwhelmingly powerful. They crouch abjectly, beg desperately for mercy. (In the Persian Gulf War of 1991, after weeks of American bombing, Iraqi soldiers were so shaken that they knelt and kissed the hands of the first Americans they saw even when those Americans were journalists.) On the one hand, this is a pragmatic move—the smartest thing to do under the circumstances. But it seems fueled at least as much by instinctive emotion as by conscious strategy. Indeed, chimpanzees do roughly the same thing. Faced with a formidable foe, they either confront it with a “threat display” or, if it’s too formidable, crouch in submission.

There’s no telling what chimps feel in these instances, but in the case of humans there have been reports of something like awe. That this feeling is naturally directed toward other living beings would seem to lubricate theological interpretations of nature; if a severe thunderstorm summons the same emotion as an ill-tempered and potent foe, it’s not much of a stretch to imagine an ill-tempered foe behind the thunderstorm.

Even chimpanzees may at times make a dim version of this conceptual leap. The primatologist Jane Goodall has observed chimps reacting to a rainstorm or a waterfall by making a threat display. She speculates that the “awe and wonder” that “underlie most religions” may be grounded in “such primeval, uncomprehending surges of emotion.”

None of this is meant to deny the possibility of valid religious experience. The prospect that some states of consciousness move us closer to what mystics call “ultimate reality”—or even toward something worthy of the name “divine”—is hardly excluded by a scientific worldview. But defenders of religion would be ill advised to stake its validity on the claim, as Otto suggested in The Idea of the Holy, that at the dawn of religious history lies some mystical or revelatory experience that defies naturalistic explanation. Because the more we learn about the labyrinthine and sometimes irrational character of human nature, the easier it is to explain the origin of religion without invoking such a thing. Religion arose out of a hodgepodge of genetically based mental mechanisms designed by natural selection for thoroughly mundane purposes.

At times Otto himself seemed to doubt that religious experience could defy scientific explanation. In The Idea of the Holy, after discussing such things as spirit worship, ancestor worship, and primitive magic, he wrote,

Different as these things are, they are all haunted by a common—and that a numinous—element, which is easily identifiable. They did not, perhaps, take their origin out of this common numinous element directly; they may have all exhibited a preliminary stage at which they were merely “natural” products of the naive, rudimentary fancies of primitive times. But these things acquire a strand of a quite special kind, which alone gives them their character as forming the vestibule of religion, brings them first to clear and explicit form, and furnishes them with the prodigious power over the minds of men which history universally proves them to possess.

Otto’s exact meaning is debatable, but the general drift is intriguing: that elements of early religion, though themselves of mundane origin, could through subsequent cultural evolution come to acquire a deeply, validly spiritual character. This idea isn’t implausible. But how far humanity has traveled along the path of spiritual evolution is another question altogether.







“One World, Under God”
(The Atlantic article)