The Science of Religion Part III: Ritual
How a seemingless arbitrary behavior builds self-control, develops executive function, and helps us grow up
From a certain point of view, the very existence of religion seems like a puzzle. Picture it: every moment of every day, Earth is coated with vast hordes of otherwise sensible people sitting alone in their rooms talking animatedly to invisible spirits; gathering in lavish buildings to perform elaborate sequences of liturgy; and making sacrifices of perfectly good money, flowers, milk, and more at shrines. If you’re the kind of person who wants to believe that people make sense — that our behavior links useful knowledge with practical agendas in linear ways — you probably find this picture baffling.
But as I’ve been surveying for the past couple of posts, a large contingent of 21st-century scientists have been trying to clear up the confusion by applying the tools of their trade to the psychological and cognitive roots of religion. After all, science can investigate anything that’s observable, and many religious phenomena — cathedrals, temples, groups of pilgrims wandering along mountain roads — certainly fit that description.
Many thought leaders, such as the New Atheists, expected the scientific study of religion (SSR) to show once and for all that religiosity was really just a maladaptive cognitive error, a bug in the brain’s code. Maybe SSR would show us how to edit out that bug, freeing us from religion forever.
But the science hasn’t really met those expectations.
Instead, religion keeps turning out to be adaptive, on the whole. It supports and stabilizes in-group cooperation. It’s very good at dissuading people from suicide and other self-destructive behaviors. And it’s mildly but consistently predictive of better mental health. If religion is maladaptive, I’d love to see what’s adaptive.
In the last couple of posts, I examined these topics mostly at the sociological and epidemiological levels. But today, let’s drill down. How does religion have the positive effects that it does? What are the cognitive and psychological mechanisms?
Let’s focus on mental health. One common suggestion is that religion improves mental health by providing social support. Going to church, temple, or mosque regularly is a great way to build community, and having lots of close relationships is one of the best things you can do for your mental well-being. Accordingly, statistical analyses show that social support does mediate, or explain, some of the relationship between religiosity and mental health.
But it doesn’t explain all of it. Even if you control for stronger social networks, you still find that religious believers still enjoy better mental (and physical) health than nonbelievers on average.
One possibility is that religious practices are beneficial in themselves. I mentioned in my previous post that rituals may actually help build self-control. Self-control, in turn, robustly predicts all kinds of good mental health outcomes, including protecting against suicide. And there’s good longitudinal research showing that religiosity — which usually involves taking part in public rituals such as attending services — predicts future gains in self-control.
Rituals are usually social — one famous anthropologist, Roy Rappaport, called ritual humanity’s “basic social act.” So it’s clear that rituals can help us build community. But this doesn’t really explain how they help us develop self-regulation. To answer this question, we’ll have to describe what rituals are.
Ritual Doesn’t Do Anything
Ritual is hard to define, but, like pornography, most of us know it when we see it.
For one thing, rituals seem formalized compared with everyday behaviors. Performers go through basically the same steps each time they perform a ritual. Their motions and words are precise, repetitive, even stereotyped.
Rituals are also impractical. The way someone performs a ritual seems somehow more important than what the ritual actually accomplishes, at least from an everyday point of view. For example, a Jewish seder is sort of a meal, but its main goal isn’t satisfying participants’ hunger. Instead, the focus is on a prescribed sequence of actions, readings, and recitations that recall the long-ago exodus from Egypt. The ritual is a story, not a means to an end.
And many ritual actions seem characteristically exaggerated. After consecrating the communion host (the wafer) during a Mass, a Catholic priest usually holds it aloft for a long pause, displaying it above his head for the congregation to see. Often, an altar server rings a bell simultaneously. These things mark out Mass as a ritualized practice, not an everyday affair. If you did this to your muffin in a food court, you’d get some strange looks.1
In the words of American sociologist George C. Homans, then, “ritual actions do not produce a practical result on the external world—that is one of the reasons we call them ritual.” Most psychologists, cognitive scientists, and social scientists today agree with Homans. Some use the term “goal demotion” to talk about how rituals redirect our attention away from final goals and toward the steps of the actions themselves.
In short, rituals are the opposite of efficient.
So how do these inefficient, impractical, exaggerated behaviors affect us psychologically?
The Ritual Stance
The first thing to know is that rituals grab our attention.
As social animals, we’re already somewhat primed to hone into each other’s words and deeds. Other humans are the most important things in our environment. Paying attention to them is how we learn vital survival skills like technical know-how and languages, as well as the social norms that hold society together. So we’re always paying attention to what others are doing.
But exaggerated, goal-demoted, and formal or stereotyped actions seem to be particularly attention-grabbing.
One research team reported that people found conventional action sequences more salient and interesting than practical, goal-directed actions. That is, impractical but ritualized actions seemed more “special” than actions that were aimed at accomplishing a goal. Objects that had been treated ritualistically also came across as more desirable than those that had been treated instrumentally.
Observing a ritual thus activates what some psychologists call the ritual stance. Tufts philosopher (and New Atheist horseman) Daniel Dennett coined the idea of the “intentional stance” to describe how we interpret others’ actions as springing from their intentions. When we take the intentional stance, we see other people as persons rather than physical objects.
Analogously, the so-called ritual stance encourages us to see others’ actions as conventional rather than practical, which in turn makes those actions seem particularly noteworthy and interesting.
Overimitation: Apes Don’t Ape
Next, ritualized actions encourage imitation. In one famous study, a team of researchers showed both human children and chimpanzees (separately, of course) how to open a seemingly complicated box that contained a delicious treat. The box had two compartments, a top one gated by a bolt and a bottom one with a small door. The treat awaited in the bottom compartment, which was inaccessible through the top. The top compartment was thus totally irrelevant for getting the treat.
But the researchers demonstrated a method that involved tapping the top bolt three times with a tool, pushing the bolt aside, and inserting the tool into the top compartment. Only after these steps were complete did they open the door to the lower compartment and finally retrieve the treat with the tool.
That is, they included an invariable, formalized, yet causally irrelevant action sequence in their demonstration.
When the box was opaque, both human children and chimpanzees faithfully copied all the steps, even the irrelevant ones. For all they knew, sticking the tool into the top compartment was a necessary step.
But when the treat box was transparent, it became obvious that fiddling around with the top bolt and compartment was useless.
As soon as this became clear (pun intended), it was the smart little chimps who skipped all the ritualized tapping, rushing ahead to retrieve the treat in the most efficient way possible. They didn’t waste time slavishly following what others had done. Immanuel Kant, who implored his Enlightenment-era readers to sapere aude! (“Dare to use your own reason!”), would have cheered the apes’ cleverness and initiative.
The human subjects, by contrast, kept fastidiously imitating the entire action sequence. They carefully tapped the bolt three times and sticking the tool in the top compartment as if it were a sacred duty — even though they now knew it didn’t accomplish anything.
Kant would have rolled his eyes.
Subsequent research has shown that humans exhibit this propensity for overimitation all over the world, in all kinds of cultures. The ritual stance seems to particularly evoke the overimitation response. When we see a conventional sequence of gestures or actions— one whose form is highly structured but whose practical goal isn’t obvious — we assume that every part of the sequence must be important, which inspires us to imitate the whole thing. Evidence suggests that overimitation helps people fit into cultural groups and demonstrate affiliation.2 For instance, in one lab experiment, children showed an even greater tendency to overimitate after they’d been made to feel socially excluded. We use imitation of purposeless actios to show that we belong — or that we want to belong.
The Ritual Marshmallow
Finally, rituals seem linked with self-regulation and self-control at a basic cognitive level.
Inhibitory self-control is one of the three core executive functions that develop over the course of childhood. It helps us suppress impulses and keep focused on a longer-term goal. Self-control is often measured with a version of the famous marshmallow test. In this task, you sit a young child down at a table with a single marshmallow on a plate. You let her know that she’s free to eat the marshmallow once you step out of the room, but if you come back in 15 minutes and the marshmallow is still there, you’ll give her another marshmallow. It’s an acute test of delayed gratification.
There are lots of other versions of the marshmallow test, some involving toys or other treats such as animal crackers. But regardless what treat is used, if you can resist the temptation that’s staring you in the face for a quarter-hour (a long time for a preschooler), you’ve got pretty good self-control. Some research has suggested that people who perform better on the marshmallow test as children are statistically more likely to enjoy better health, self-regulation, education, and relationships as adults, in keeping with the overwhelmingly beneficial effects of self-control on life outcomes.3
One Oxford SSR researcher, Veronika Rynanska, used a version of the marshmallow test to examine how arbitrary action sequences — essentially artificial rituals — affected kids’ self-control and executive function. At two separate sites in Slovakia and the Pacific island nation of Vanuatu, her team had groups of elementary school–aged children perform a marshmallow test with candy to test self-control. Another test measured general executive function by challenging kids’ ability to follow instructions whose rules conflicted with verbal statements — i.e., when the instructor says “touch your toes,” you’re actually supposed to touch your head.
Then, over a period of three months, some groups in both locales took part in exercises where the children learned and precisely repeated complex sequences of actions. The researchers told some groups that the exercises had a practical rationale, such as learning how to dance. But for others, they suggested that the exercises were purely conventional — for example, by cryptically telling the children that “it has always been done this way.” The researchers hoped that such cues would activate the ritual stance, encouraging the kids to interpret the actions as conventional and meaningful, and thus evoking overimitation.
After the three-month intervention period, the researchers had the children at both sites perform the tests of executive function and delayed gratification for a second time. As expected, they found that practicing the arbitrary action sequences improved scores on both the marshmallow/candy and verbal instruction tasks. This finding reiterated that self-control and executive functions can be trained.
However, the kids in the ritual stance condition showed much better improvement on these tests than the kids in the practical/instrumental condition. That is, copying random sequences of actions that served no practical purpose built self-control and executive function faster than copying purposeful actions.
An obvious implication is that, although learning skills for a trade or profession is probably good for self-control, actively participating in the rituals of a religious community or cultural group may be even better.4
Other Benefits for Self-Control
Rybanska’s findings dovetail with the writings of American psychologist and philosopher William James, who long ago argued that self-control can either atrophy from lack of use or grow with exercise:
(I)f we often flinch from making an effort, before we know it the effort-making capacity will be gone…do every day or two something for no other reason than that you would rather not do it, so that when the hour of dire need draws nigh, it may find you not unnerved and untrained to stand the test.
Putting James’s principle into practice in real life reveals wisdom. Take the stairs in the morning instead of the escalator, and it might seem easier to face a difficult client or a boring task later in the day. Shirk the little difficult things, and more serious challenges afterwards seem insurmountable.
The American James was individualistic and post-Protestant, though. He saw strengthening self-control as an individual challenge, something for each person to strive for independently. But that’s not really how it works. To riff on Aristotle, the real gymnasiums for virtue are communities. The obligations and responsibilities of community life help us to learn how to defer gratification, accept discomfort, do things we’d rather not do. In other words, to build self-regulation.
Religious communities, then, are particularly effective “gymnasiums.” Many call for difficult sacrifices such as fasting or abstinence, such as Muslim Ramadan or Christian Lent. Religious rituals are often physically challenging or taxing, such as the Hindu Thaipusam festival, which obligates participants to carry heavy displays on their heads after days of fasting and celibacy. (Many participants outside India also pierce themselves with hooks to represent the divine spear of the Hindu god Murugan.)
Then there are the simple time commitments of regular service attendance and prayers. Catholics are supposed to attend Mass every Sunday as well as multiple times during the week throughout the year, such as on the Feast of the Assumption. Orthodox Christians often stand through their entire divine liturgy, which lasts for hours. (And hours.) Even evangelical Protestants, not known for their love of ritual, often spend two to three hours in church each Sunday.
The point here is that religions automatically provide regular opportunities to develop the “effort-making capacity” that William James thought we had to cobble together out of independent initiatives.
Ritual, Prepackaged
A growing body of research shows that ritual is an important ingredient in well-being, even outside religious contexts. For instance, one recent study found that regular family rituals — such as sitting down to dinner together at the same time — protected toddlers against distress in low-income African-American families. Performing rituals before stressful experiences reduces stress, and athletes who perform personal rituals before a match or event often enjoy a boost in performance.
So you don’t need religion to benefit from rituals. But religion does bundle all these functions into a one-stop shop. By requiring believers to partake in prescribed cycles of liturgies, prayers, and other practices, religions deliver a preprepared suite of tools for building self-control, stabilizing mood, reducing anxiety, and even enhancing performance on certain types of tricky tasks.
Religion isn’t a panacea, of course. Rituals can and do go awry. Obsessive-compulsive disorder often features personal rituals that become destructive. Nor are all religious rituals benign. Human sacrifice has a deep history in many different cultures, for example. It probably results from the same logic as other forms of ritual — giving up time, treasure, or comfort for the sake of some social or religious ideal — but takes this logic in a brutal direction.
Yet on the whole, it seems clear that rituals are important for mental health, thanks to their ability to catalyze and develop self-control and other executive functions and to reduce anxiety and stress. Even the most supposedly under-ritualized traditions, such as evangelical Protestantism, are actually rife with practices, disciplines, and habits that serve these same functions.
The evangelical philosopher James K.A. Smith has aptly observed that all of life is liturgical. We’re always taking part in conventional actions whose means are decoupled from practical ends, displaying our identities, commitments, and values rather than trying to get something done. What religions do is unify and cohere these liturgical habits, bringing them in line with a single overarching set of values.
I’ve argued in my own scholarly work that this is probably one of the most important reasons why religion benefits mental health. William James was 100% right that everyone should practice and exercise self-control, but religion provides a way to automate this. The tradeoff is that we have to surrender some of our individualism and autonomy — you can’t benefit from external rituals if you insist on always setting your own standards or being your own authority. For this reason, the mode of self-regulation that religions offer may not fit well with our individualistic consumer culture. But this doesn’t make it any less powerful for those willing to try it.
Edit: changed one bad link. Thanks to
for the tip.Although you should definitely give this a try and let me know the results. For science.
It also helps us learn complex technological skills like carving out dugout canoes or making bows and arrows. These skills usually include so many steps that we’d never be able to tell at first glance which were necessary and which weren’t. As human evolutionary biologist Joe Henrich has pointed out, the best strategy in this sort of context is to just copy all the actions you see the expert doing. But in contrast to ritual, people usually start skipping steps for technological skills once they’ve mastered the sequence and know the difference between the necessary steps and the unnecessary ones. For ritual, though, over-imitation persists indefinitely.
Recent research has raised questions about how good the marshmallow task on its own actually is at predicting outcomes later in life, but this may be because performance on the task reflects the social environment kids inhabit. Namely, kids who perform better on the marshmallow test tend to come from families that are stable and provide a predictable home environment. In general, self-control is scaffolded by cultural expectations — when we inhabit communities that expect and reward delay of gratification, we’re better at those things. But of course this describes most religious communities, so my points seem to hold.
One recent study found that performing rituals helped participants increase self-control directly after “depleting” their self-control resources. This study relied on the so-called “resource model” of self-control, which construes self-control as a general cognitive resource that we can temporarily deplete by using it. This was the theory behind those studies you were probably reading about six or seven years ago in the Harvard Business Review supposedly proving that resisting the temptation to eat chocolate made it harder to do math problems afterward, or whatever. However, the resource model of self-control has come under a lot of scrutiny recently, including failed replication attempts for the core experiments that established it. So it’s probably best to interpret any findings about ritual and self-control resources with several grains of salt.