Science Saves Religion

By driving the nail into the coffin of 2010s macho atheism

Back in the heady 2000s, the heyday of macho internet atheism, a lot of smart people were betting that Christianity would soon blink out forever, like the lights in an abandoned motel.

The 9/11 attacks by Islamist fanatics seemed to torch the credibility of institutional religion, convincing millions of Redditors that Bronze Age ideologies just aren’t compatible with modern progress. Science, with its aura of cool competence, seemed the only way forward. Sniffing opportunity, in 2006 a gang of white guys led by Richard Dawkins launched what they hoped would be a decapitation strike at Christianity itself, publishing a set of wildly popular manifestos to unmask Biblical religion as intellectually indefensible and morally bankrupt.

---

These New Atheists became the popular faces not only of Science™, but of an entire era of ascendant liberal triumphalism. The legitimation apparatus for the Obama era, if you will.

One of their biggest pet projects, though, wasn’t ideological, but scientific.

This was to explain — or explain away — belief in gods, spirits, and the afterlife. In particular, Daniel Dennett’s book Breaking the Spell: Religion as a Natural Phenomenon called for a rigorous investigation into the naturalistic reasons for religious belief. The book argued that we should subject religion to the same ruthless scientific investigation as any other phenomenon.

Dennett, who died earlier this year, seemed pretty confident that the rigorous scientific study of religion (SSR) would debunk faith forever, unmasking it as a bunch of quaint cognitive errors and parasitic cultural memes.

That’s not quite what happened, but he was right in that SSR has absolutely exploded in the 21st century. For two decades, droves of psychologists, anthropologists, cognitive scientists, and neuroscientists have increasingly trained their scientific lenses on religion, ritual, and belief. Papers on the psychological nature of gods now regularly appear in Nature and Science. Books on the topic have become bestsellers. (I’ve been a minor contributor, receiving my PhD in the scientific study of religion and publishing some papers on the psychology and cognitive science of ritual.)

Debunking Fail

But there’s a hitch: SSR hasn’t exactly done a stellar job debunking religion. Instead of quantitatively proving that “religion poisons everything,” as New Atheist Christopher Hitchens put it, research studies kept turning up puzzling evidence that religious practice is actually pretty helpful. Studies showed that gods and spirits were somehow linked with encouraging cooperation. Rituals boosted social trust and improved mental health. Religious believers had far more children — the core measure of biological fitness — than nonbelievers.

Of course, SSR can’t prove the existence of God. But this work does seem to hint that religion quietly improves a lot of human life, even in the most heavily secularized countries.

Now, in 2024, I think these revelations about the utility of religious belief and practice are one small part of the “vibe shift” that’s underway in culture and values. After decades of  progressive liberalism, many young people in America (particularly men) seem to be taking a conservative turn. The Catholic Church, that most un-modern of institutions, is enjoying a sort-of renaissance, with many dioceses reporting record numbers of adult baptisms in recent years. Increasing numbers of intellectuals and celebrities are taking an interest in Christianity.

The cultural dominance of left-coded worldviews such as atheism and wokism, unchallengeable only a couple of summers ago, is fast fading.

Culture usually moves in pendulum swings, and a serious backlash to the impossible-to-please, holier-than-thou secular progressivism of the 2010s — and its jaundiced attitudes toward Christianity — was always likely. But I don’t think that fully explains what we’re seeing.

Consider the change in how scientists have been writing about religion. A decade and a half ago, Breaking the Spell and The God Delusion were on the bestseller charts.

But more recently, philosopher Stephan Asma’s 2018 book Why We Need Religion surveyed the psychological and social needs that religion elegantly fulfills, even though Asma is a skeptical nonbeliever. How God Works, by psychologist David DeSteno, similarly highlights the many ways that religious rituals, beliefs, and practices help real people navigate the vicissitudes of life. And anthropologist Dimitris Xygalatas’s bestselling Ritual takes the reader on a tour of research (much of it from his lab) showing the benefits and functions of its eponymous subject.

Books such as these show that the very research that New Atheists were confident would undermine faith for good — the empirical study of religion as an object of naturalistic inquiry — has instead ignited a newfound appreciation for it among straight-talking intellectuals.

Over the coming posts, I’ll show what I mean by surveying four areas where the research demonstrates that religion has adaptive functions or benefits: cooperation, mental health, the psychology of ritual, and romance/mating.

Again, none of this research says anything about the existence of God or the truth of any religious tradition. (Although, full disclosure, working in this field did influence my own conversion to Christianity; that’s a story for a different post. Or series.) But it does push back on the previous narrative that religion is a dangerous programming bug of the human mind, something we need to overcome to march forward in progress.

It also helps illustrate one of this newsletter’s main themes: the liberal-modern conception of human nature is just plain false. Scholars in the humanities like Charles Taylor have been arguing this point cogently for decades, but the 21st-century human sciences are showing empirically how correct their arguments are. We’re not actually enlightened, rational, autonomous individuals who thrive best when we throw off the heavy shackles of blind custom and tradition. Instead, we’re ineluctably social beings, biologically and cognitively designed for tradition and ritual, for the kinds of institutions and practices that SSR investigates. As a species, we’re adapted for these things the way that fish are adapted for water. Without them, we’re gasping for air.

Religion and Cooperation

Let’s start today with cooperation. You can’t have human life unless people help each other out. Not only that, but they have to do so even when it costs them in the short term. August Comte called this “altruism,” but researchers in psychology, game theory, and evolutionary biology usually just call it cooperation. It’s the sine qua non of human society, from the smallest families to the biggest, globe-spanning civilizations.

One of religion’s most important functions, from a purely secular or naturalistic perspective, is that it builds the foundation for cooperation. This is an important function because cooperation is hard.

Take an example. Imagine a college housing co-op where all the idealistic young housemates have agreed to take turns doing dishes. Cooperating, in this case, means scrubbing everyone’s plates on your assigned day, sacrificing your comfort for one evening to benefit your housemates, and trusting that someone else will do your dishes tomorrow.

But this is an anarchist co-op, the kind where a couple of the roommates hide out permanently in their shade-darkened rooms, never going to class, emerging only to sell pot to sketchy characters who turn up in the living room. Soon after the semester starts, these deadbeat housemates are already shrugging off their kitchen responsibilities, leaving extra-large piles of dishes caked with questionable leftovers for everyone else to deal with.

These guys are defectors, or free riders. They take advantage of others’ goodwill to get benefits for themselves without providing much in return.

Free riding is poison for communities.

A minority of remaining housemates, earnest girls from nice homes, initially pick up everyone else’s slack, but they soon (understandably) get discouraged. As they wipe down yet another quinoa-caked bowl they didn’t use, they entertain colorfully resentful thoughts about their deadbeat housemates. Soon, in exasperation, they stop doing their share, too. Dishes stack up in the sink and people begin avoiding the kitchen.

At this point, the house isn’t a community anymore. It’s devolved into an agglomeration of passive-aggressive, mutually hostile individuals. A Hobbesian state of nature with wicker furniture.

This is the story of countless group projects throughout history, from global empires to startups to hippie co-ops in Madison or Berkeley. Free riders sidle in, like what they see, take advantage of others, and ruin everything.

But at another co-op down the street, things turn out a bit differently. To apply for membership in this co-op, you have to be actively enrolled in a Christian church or parish that you attend weekly. A mandatory Thursday night house dinner and prayer night anchors the school week.

Drugs and overnight sleepovers are big no-nos, with breaches of the rules following a two-strike rule: first time, you get a group intervention and request to change your ways. Second time, you’re packing your stuff and finding another house.

The Christian co-op also has a rule for sharing dish duty, but people actually follow it — for the most part. People being people, some are more committed and conscientious than others. Conflicts periodically arise when someone lets things slip for too long, and the housemates inevitably contend with petty resentments, insecurities, and the occasional passive-aggressive comment.

But for the most part, everyone actually holds up their end of the collective bargain, trusting each other to do their parts. The house behaves like a community throughout the school year and beyond.

This co-op succeeds partly because the housemates all know that they’ll face real consequences if they break the house rules. If you know that you’ll soon be homeless should you let dishes rot in the sink too often, you’ll probably be incentivized to scrub when it’s your turn to scrub.

But more than that, the Christian housemates also share at least some level of commitment to common ideals, symbols, and values. These commitments encourage everyone intrinsically to live up to the high standards those ideals set. It tacitly encourages trust — with everyone believing the same far-out story about how the universe works, the housemates see each other as fellow inhabitants of a meaningful cosmos. This makes everyone feel better about relying on each other.

Perhaps most importantly, there are also costs to belonging to the Christian co-op.

By definition, free riders don’t like to pay costs or make sacrifices. By requiring Thursday night prayer dinners and weekly church attendance, the Christian co-op is already screening out potential deadbeats from the get-go. Would-be free riders get turned off by the high expectations and search instead for communities with lower standards. Voilà — the Christian co-op fields applicants who are already somewhat inclined to shoulder burdens.

Costly Signaling

Research strongly backs up the idea that costly barriers to entry help shore up cooperation in religious communities. One team of researchers, led by anthropologist Rich Sosis, retrospectively studied 19th-century American communes and found that, based on their curiously excellent records, only the religious communes had any real chance of lasting for more than a couple of years.

In fact, religious communes lasted for a median of 25 years, whereas the median secular commune fell apart after only 5 years.

It’s not hard to see why most communes would flame out quickly. Communes are risky, deeply cooperative endeavors, but they also often attract lazy, idealistic people who dream of free-love utopias without work. In Sosis’s 19th-century sample, religiosity seemed to somehow help communes avoid those lazy idealists while attracting serious, hard-working contributors.

The difference seemed to come down to a combination of religion and costliness. Sosis and his co-author also looked at the number of costly signals the different communes required.¹ Costly signals were coded as practices such as mandatory fasts, restrictions on the consumption of staples like coffee or meat, or obligatory celibacy. Religious communes that imposed 12 or more costly requirements lasted for an average of over 60 years, whereas those that imposed only one or two fell apart in less than a decade.

Yet, a bit surprisingly, costs made no real difference to the longevity of secular communes. In other words, it didn’t matter whether a nonreligious commune asked potential members to give up a lot or only a little — such communes fell apart in short order either way.

As in our hypothetical college co-op, the researchers reasoned that costly signals made group membership less appealing to people who weren’t truly committed to the commune’s spiritual worldview. The more committed people were, the more likely they were to sacrifice time, effort, and treasure in support of the collective project. In other words, they were more likely to cooperate. Religious costs screened out potential free riders.

But for this signaling process to work, it seems there actually had to be a religious worldview to commit to. The secular communes had plenty of ideologies, from anarchism to primitive communism. But somehow secular worldviews don’t quite pair up with costly rituals and signals to screen out free riders.

Your Perceived Costs May Vary

One key difference is that ritual costs just don’t seem as costly for true believers. A nonbeliever faces an opportunity cost for going to church on a Sunday morning. That’s two hours she’d rather use for going to brunch, or playing pickleball, or scrolling YouTube, or whatever. But for someone truly committed to the Christian worldview, going to church on Sunday doesn’t feel like a cost. It’s what he already wanted to do anyway.

This difference in perceived costliness for committed versus uncommitted people amplifies the costly signaling effect. Communities that expect people to pour time and energy into rituals, fast from certain foods, or give up certain conveniences seem extra unappealing to people who aren’t invested in their worldviews. But for those who are invested, the supposed “costs” seem like freewill offerings, or even like fun.

Costly Signaling Today

The costly signaling effect can help explain religious dynamics in our society today. Liberal Protestant denominations in America — which tend to emphasize openness and inclusion and downplay the costly aspects of Christianity — are rapidly collapsing, but more conservative denominations that impose more costs are either shrinking more slowly, holding steady, or even growing.

Within Catholicism, parishes that set high expectations — weekly Mass, monthly confession, no living together before marriage — are often growing and bringing in new converts, whereas parishes that strive to make things easy and convenient tend to be shrinking.

The economist Lawrence Iannaccone argued that, by selecting for people who are strongly committed to the church’s teachings, a “strict church” winds up with a highly intrinsically motivated congregation, one that actually wants to do hard things and sacrifice for the church and for each other.

Being surrounded by other highly committed people leads each individual member to relax and trust in the community, further encouraging investment and cooperation. These positive vibes (and benefits, like having a lot of people around willing to help with childcare or moving) attract yet more people from outside. It’s a virtuous cycle.

By contrast, a lax church, one that tries to be too welcoming and low-cost, has no real way to screen out the kind of person who will take advantage of others’ goodwill. The general level of investment is inevitably lower, so people find themselves trusting others less, as in the anarchist co-op example.

This half-committed vibe isn’t very attractive, and many of the “secular benefits” of religious belonging are weaker here: there are fewer young moms willing to help each other out, fewer teens available to babysit, fewer congregants willing to show up and help you move. So the community draws in fewer converts and loses more members over time.

Happy to Sacrifice

I’ve mostly used Christianity as an example in this post, but faith traditions of all kinds seem to offer a suspiciously well-matched cluster of tools that help screen out free riders from and encourage cooperation within their communities.

I didn’t mention the idea that God or the gods can punish uncooperative behavior, which is found everywhere in world religions, from Islam to Christianity to — yes — Buddhism (especially its bodhisattva- and demon-haunted Mahayana varieties), as well as smaller-scale traditions, too. Most studies find that belief in supernatural punishment encourages people to act more cooperatively, both in the lab and in real life. For example, belief in hell is correlated with less crime at the national level.

But costly signaling is a leading reason why religions seem to be so good at encouraging cooperation and driving away free riders. The combination of unfalsifiable beliefs and symbols with costly displays, rituals, and sacrifices pulls in committed people and drives away uncommitted people. It also naturally screens out those who would rather not do anything difficult or challenging.

In other words, it automatically selects for people who are happy to sacrifice. The more of that kind of person you have in your community, the more benefits everyone will reap from belonging to it, and the stronger it will be.

If religion were really just a tool for elites to oppress the masses or a random cognitive error, it probably wouldn’t bundle these functions together so effectively.

Of course, sometimes (often?) elites do use religion for oppressive purposes, and our innate cognitive tendencies do affect our religious beliefs (how could they not?). But SSR research challenges the idea, implicit in a lot of older New Atheist writings, that religion is basically maladaptive, something we’d be better off without. Instead, we’re the cooperative species par excellance, and religion seems to be a critical part of the reason why.

1. In biology, a costly signal honestly conveys information about invisible qualities such as physical robustness or fitness. On the African savanna, Thompson’s gazelles jump up and down in place when first spotted by a predator. Scientists see this quirky behavior, which they call stotting, as a costly signal of vigor. Stotting essentially tells the predators “don’t chase me, I’ll just wear you out.” Taking the hint, predators chase the stotting-challenged old and sick gazelles instead. Everyone wins! (Except the non-stotting gazelles.) The key in this example is that weak or sick gazelles either can’t stot or don’t stot, because doing so would be too risky. If you’re old or injured, it’s best to save whatever remaining energy you have for trying to outrun the lion. Maybe you’ll get lucky and escape this time. But if you try to stot, you’ll probably be exhausted by the time the chase begins. The cost is what makes the signal reliable.