I have an unusual job: I study human culture and religion through the lens of evolution.

Seen from this Darwinian perspective, humans are animals that happen to have language and culture. Fully embedded in the natural history of life on earth, our minds and behaviors have been shaped by millions of years of ruthlessly creative evolutionary selection. Our desires, our instincts, and our irrepressible drive to create symbolic cultures all stem from eons of striving for life on a planet where survival can never be taken for granted, and where fitness—passing down traits to the next generation—is the ultimate biological goal.

But I’m also a religious believer. I attend church with my wife and take part in the life of a local Benedictine monastery. It’s Lent as I write this, so I’m fasting. How does this work?

Well, it’s tricky.

There are plenty of religious scientists. World-renowned geneticist Francis Collins is a born-again Christian. He’s the former director of the Human Genome Project and current head of the National Institutes of Health. Astronomer Jennifer Wiseman is senior project scientist for the Hubble Space Telescope and a high-profile Christian believer who heads up the Dialogue on Science, Ethics, and Religion at the American Association for the Advancement of Science.

But in the evolutionary human sciences, where I work, believers seem particularly rare. Maybe that’s because these fields of research are more intimate than physics or biochemistry. They’re about how we humans came to be the way we are, how our minds carry the imprints of landscapes where our not-quite-human ancestors danced by firelight.

This is a beautiful story, but it’s a disconcerting and unsettling one as well. It disturbs us because, being about us, it stirs awake our vulnerable human sensibilities—then immediately slashes them down to explanation in terms of a physical and indifferent forces. What results is an uncanny mix of the personal and impersonal.

When we study the evolution of the human species, we dive headfirst into a flux that’s both intimately human and coldly inhuman, a story at once familiar and eerily alien and weird.

This disorientation haunts me, but I do my work anyway. I also keep going to church. But it takes serious work to reconcile the two. It doesn’t come automatically.

Religion as a product of evolution

If we’re truly a species shaped through and through by evolution, then evolution’s legacy touches everything about us. Startlingly, this includes religion itself: the universal human tendency to use stereotyped, culturally distinctive rituals in reference to empirically unfalsifiable persons and ideals.

From this perspective, “religion” is a catch-all term for the behaviors and beliefs by which we generate the self-regulating cultural niches that define particular tribes or societies. (In biological parlance, a “niche” is an environment to which an organism is well-adapted.) The ability to carve out unique cultural niches is what allows humans to adapt so remarkably well to a vast range of environments. But it’s also what divides in-groups from one another.

One mechanism undergirding these functions is ritual. By taking part in culturally distinctive rituals, we reinforce the values and identities that make our tribe unique.

Another foundation for our ability to construct niches is language. Unlike animal signals, verbal language allows human beings to communicate about imperceptible domains: the past, the future, and imaginary things. Only language, then, enables us to envision social roles as opposed to concrete individuals. A bus driver is both a physical person and an abstract role, but only the individual is visible. Species without verbal language can have friends, mates, and rivals. But they can’t have transposable roles like presidents, lawyers, or bus drivers.

Recently, a colleague and I put forward an argument that gods and spirits are, cognitively speaking, just templated roles that have no embodiment. Despite their invisibility, we represent these divine roles cognitively in a way that matches how we represent other roles, like “president.” The difference is that all the responsibility for our templated interactions with gods rests on our shoulders.

Bowing to chairs

Imagine that everyone who comes into a room bows, one after another, to an empty chair in the middle of the floor. Then they join the others in sitting around the room’s perimeter, taking care not to sit too near the chair. Watching this display, you’d soon start to feel like there was something in that chair after all. The careful, repeated behavior directed toward it makes it seem special, marked off.

In the same way, people “act out” their relationships with religious figures such as Jesus or the Hindu god Shiva by directing repeated actions and prayers toward them, diligently treating these divine beings as roles that require particular forms of deference and attention. Thus, people create invisible boundaries and identities—even disembodied ones—by treating them ritualistically as if they were real.

We mark off normal roles similarly—treating our spouse one way, our coworker another. The difference is that everyday roles are inhabited by real human beings, so our interactions with them are two-way and rich with improvisation. By contrast, interactions with gods and spirits are behaviorally one-sided, which caps variation. So even among charismatic Christians—who value spontaneity—prayers to God tend to take reliable, templated forms.

In our paper, my colleague and I argued that such invisible roles (whether gods, spirits, or ancestors) are common precisely because interactions with them are one-sided. This ensures that at least some interactions—those between humans and gods—are largely free of transactional or utilitarian incentives. We might ask gods for things. But we can’t manipulate them into giving us what we want, as we might with a spouse or coworker.

The more we treat gods appropriately—despite the fact that they aren’t normal transactional players—the more natural it becomes to conscientiously treat other people according to their roles, too. Patterns of deference and dominance become more closely linked to people’s positions in the abstract social structure, not, say, their brute physical size or attractiveness.

In other words, gods and spirits increase the predictability of people’s behavior as a function of their normative social roles, which increases the adaptive value of the cultural niche. (After all, a niche is an environment an organism knows how to predict.) And so these invisible, imagined beings are a recurrent core feature of all religious systems.

Wrestling with naturalism

It’s an interesting idea. It might even be true. But if our hypothesis is correct, then it seems Ludwig Feuerbach was right. Feuerbach, a 19th-century German critic of religion, argued that God is nothing but a projection of our human capacities and wishes onto the cosmic canvas—at its worst, a way of relieving humanity of the responsibility of living up to our own potential. Like Marx, Feuerbach saw God as a natural product of the social human mind.

The theory also reflects the work of French sociologist Émile Durkheim, who conceived of religious beliefs as emerging from and reflecting social conditions. For Feuerbach, religious beliefs were external projections of our mental lives. For Durkheim, they were internal impressions of our social environment. Either way, naturalistic explanations for religion seem to contradict the possibility of sincerely held religious belief by reducing it to mechanical, inbuilt processes.

But the anthropologist Mary Douglas, a groundbreaking pioneer in charting the social influences on cognition and a lifelong Catholic, saw things differently. To religious believers who feared that scientific anthropology would undermine religious faith, she confidently replied:

“It does not help our understanding of religion to protect it from profane scrutiny by drawing a deferential boundary around it. Religion should not be exempted at all.”

Douglas saw the construction of religious beliefs as a natural function of the way the evolved human mind works. Rather than debunking religion, though, this insight simply furnished researchers with a better model of how people arrive at religious beliefs: through a complex interplay of social and cognitive links, embedded in cultures and constrained by the brain’s limits. For canny believers, Douglas believed this meant that “God, having made man a social being, allows His Face to be seen only through a distorted lens, through the medium of the society which men themselves create.”

Douglas’s steady-eyed realism meant that she wasn’t intimidated by the cultural and cognitive conditions that revealed themselves when she carried out a “profane scrutiny” of religious beliefs. Daily Catholic practice also ensured that she had a steady base to stand on. She could stare into the whirling void of reductionism, yet not blink.

A reflection of ultimate reality

Her ability to see both the reductionistic and the holistic at the same time illustrates fellow social scientist Peter Berger’s principle that, although religious belief is indeed a cognitive projection, “What is projected is, however, itself a reflection, an imitation, of ultimate reality.”

Nonbelievers don’t have to buy this comforting, synthetic vision (and by definition, they don’t). But good naturalistic accounts of religion and culture stand on their own evidence. Although Douglas was a Catholic and Berger a Lutheran, their social scientific theories are used today by countless researchers of all faith backgrounds (and no religion at all).

Similarly, if my and my colleague’s theory of religion and the cognitive basis of social roles holds any water, then it should eventually be formalizable and testable in a way that carries weight with knowledgeable believers and nonbelievers alike.

Evolutionary and cognitive approaches to religion can be off-putting to believers because they’re explicitly reductionistic. (Ironically, they’re also off-putting to many secular humanities scholars for the same reason.) But if people of faith are to take seriously our condition as animals with language, they need confidence like Mary Douglas’s, not fearful protectionism.

It’s not an easy ask. But it does no good to protect religion from objective inquiry, to declare “Up to here science can approach, but no further!” In this case as in so many others, faith requires courage.

Research Associate, Center for Mind and Culture