Evolution is stingy. Evolutionary biologists tell us that natural selection tends to weed out any trait or behavior that’s inefficient, reduces fitness, causes drag.

But natural selection produced human beings, and we do a lot of things that seem pretty inefficient. Take religion. We humans have spent thousands of years sacrificing otherwise perfectly edible animals, wasting time and energy on rituals, and believing in empirically unprovable beings. Yet evolutionary logic would seem to prefer that we believe only true things and not waste time and resources. So how has evolution allowed religious beliefs and practices to spread so widely?

In a new paper with evolutionary anthropologist John Shaver, I argue that religion plays a previously overlooked role in anchoring cultural norms—because you can’t bargain with gods.

We’re not the first scientists to argue that religion is important for stabilizing norms. Thinkers from Émile Durkheim to Clifford Geertz have made similar claims. Both John and I share a mentor in Rich Sosis, an anthropologist who has argued that religious rituals can serve as “costly signals” of commitment to group identity, and can also motivate obedience to social norms, which vary widely between cultures.

This cultural variability is critical—human beings have been so biologically successful in large part because our unprecedented behavioral flexibility helps us adapt to different contexts. But we have to find a way of coordinating that flexible behavior, since genes don’t do it for us. Religion, many have argued, is one way of accomplishing this goal.

Religion vs. culture

But “religion” is notoriously hard to define. Does it have centrally to do with worshiping gods and spirits, or with rituals, or what? After all, many secular events, from college football games to presidential inaugurations, display religion-like features. So where’s the boundary between religion and culture?

In our paper, John and I argue that there isn’t any such boundary. But instead of concluding that  religion is a mere social construct, we acknowledge that culture itself is, in key ways, essentially religious. In this reading, football games and patriotism often look like religion because they really share core features with it.

We draw on the anthropologist Maurice Bloch’s argument that human beings live in two social worlds: the transactional and the transcendental. The transactional social world is where people treat each other according to their inherent abilities and capacities, as concrete individuals. The big kids at school beat up on the shrimpy kid just because he’s weaker. All the guys swoon for the new girl in class for her looks. These interactions are transactional because physical strength or attractiveness are just basic characteristics of concrete, individual human beings—characteristics that they carry with them between their different roles, at home, work, and school.

By contrast, the transcendental world is made up of abstract social roles and norms, which exist as mental constructs or templates that we can enter, occupy, and then leave behind for others. Personal traits aren’t as relevant for these roles. For example, if Carolyn is a judge, her unique personality and preferences don’t define her professional identity. Instead, that identity is mostly defined by external norms and legal responsibilities—wearing robes when on the bench, making decisions according to the law, and so forth. These rules and norms preceded Carolyn, and will persist after her when she retires and someone else takes her place.

Bloch illustrates the transcendental social world with a powerful example from his fieldwork in Madagascar. An esteemed elder he’d known for a long time was treated with deference and respect even after he was senile—particularly in ritual or religious settings:

“By now, he is old, physically weak and a little bit senile. He has difficulty in recognizing people. He spends most of his days in a foetal position wrapped up in a blanket. Yet he is treated with continual deference, consideration, respect and even fear. Whenever there is a ritual to be performed, he has to be put in charge so that he can bless the participants.”

In other words, other members of the tribe treat this old man as his role dictates they should, rather than in the way his weakness and senility might suggest. Similarly, juries and trial attendees rise when the judge enters the courtroom, regardless of whether that judge is healthy, sick, tall, short, black, white, old, or young. Who the person is doesn’t matter. We stand up because the judge has entered, not because somebody named Carolyn has.

Importantly, Bloch claims that “the transcendental social (world) and phenomena that we have ethnocentrically called religion are part and parcel of a single unity.” For Americans and residents of other individualistic Western societies, this claim sounds far-fetched. But it makes more sense when we remember that transcendental social roles of any kind are templates for behavior, rooted in the surrounding social environment rather than our own preferences. And in many cultures, the templates for “wife,” “husband,” or “elder” are indeed deeply enmeshed with religious norms and rules.

Another way of thinking about this is that social roles and norms are examples of institutions: socially grounded behavioral patterns that don’t reduce to objective facts. A U.S. circuit court is an institutional entity—unlike rocks or trees, it only exists because we say that it exists, and we behave accordingly. Similarly, marriage isn’t a natural, objective entity, but an agreement between people that imposes role expectations and norms.

Acts of shared imagination

Religious entities and ideas fit this description, too. Take the Catholic doctrine of the transubstantiation. Catholic catechism teaches that the consecrated elements of the eucharist are literally transformed into the body and blood of Christ, but of course they still taste and look like wafers and wine. So even the most devout Catholics have to perform an act of shared imagination, adding cognitive content atop their sensory perceptions, agreeing to treat the eucharist as if it were transubstantiated.

Bloch argues that the transcendental social world is populated with all the social templates that depend on imagination in this way. This includes basic roles like wife, elder, or judge. But it also includes roles like ancestor, ghost, spirit, and god, which don’t require embodiment. We have to use our imaginations to perceive or interact with them.

Yet a real judge is both a role and a unique person. The visible and invisible components of her identity are entwined. This means people typically treat her as her role requires and in ways that reflect her intrinsic traits. We might dutifully stand when an elderly judge hobbles into the courtroom, but then rudely surf the web on our phones during the proceedings, assuming that the judge is too blind to see us. So interacting with real people is always a mixture of social propriety and self-interest.

But in everyday experience, gods and spirits don’t generally have embodied representatives. Yes, Christians believe that Christ was God incarnate. But very few Christians alive today report ever actually seeing Christ physically. This means that, cognitively speaking, gods and spirits are pure behavioral templates for our own behavior toward them. They have only the transcendental component of a social identity. When Catholics make the sign of the cross, the invisible role of Jesus determines their actions, just as courtroom attendees must stand to acknowledge the entrance of a judge.

Gods and spirits, then, have no transactional component. Although people from all cultures plea, cajole, or try to strike bargains with them, it’s generally understood that gods aren’t the least bit under our control; their decision to answer prayer or offer blessings is understood as a pure matter of grace or caprice. Yet in culture after culture, they do demand that we obey certain rules, which often take the form of taboos or restrictions: fasting for Ramadan, abstaining from certain food during the rainy season, staying home from the hunt after seeing a woman give birth. Our behavior toward gods is thus either fully deontic, or based on normative expectations and templates, or it isn’t anything—even when we ask them for things. Yes, Christians reciting the Lord’s Prayer are asking God for help. But they’re also behaving in a normative way that acknowledges the “role” of God in the interaction.

Can’t bribe gods and spirits

So if we try to manipulate or get away with things with gods or spirits, where does that get us? Pure transcendental roles don’t respond to bribes, bargains, or negotiations the way real, embodied roles do. Compare this to, say, a border security agent, who is both a transcendental role and a real person. You can offer him a bribe. If he takes it, you both benefit as people by breaking the prescriptive rules that apply to your respective roles. But gods and spirits don’t reward rule-breaking. If you cheat in your responsibilities to them, you don’t get anything except the time or resources you would have spent in worship. And if you’re not willing to expend those resources, then you’re not acknowledging the spiritual entities in the first place, since they’re pure transcendental roles.

John and I argue that this, then, is the function of gods and spirits. By ensuring that a portion of the transcendental social structure is immune to transactional or utilitarian motives, they increase the hold that the normative system has on us in general. From our paper:

“because these agents, lacking embodiment, cannot provide the same actual strategic benefits to interactants that embodied persons can, the various obligations that people owe to them tend to slip out of the reach of utilitarian cost-benefit calculations. The resulting sanctification of core social obligations renders the entire moral system more stable.”

This doesn’t mean that gods and spirits are unequivocally good. For example, the Aztecs believed that the sun god Huitzilopochtli required human sacrifice. Cutting the hearts out of living victims was the normative behavior that recognized the god, just like standing up recognizes the judge. All transcendental roles are defined by normative behaviors. But gods and spirits cognitively anchor our transcendental norms by putting some of them beyond the reach of transactional concerns—whether the results are benign or not.

Religious worship of unfalsifiable spiritual beings, then, is perhaps something we should expect from any creature with language and high levels of behavioral flexibility. Such creatures will always need to coordinate norms that differ between varying groups. Something is needed to cohere each unique and different sets of norms, getting it to “stick” with its group’s inhabitants. Behavioral obligations toward beings who can’t be seen and who don’t reward bargaining, pleading, or cheating may be exactly what the transcendental doctor ordered.

Icon-O
Research Associate, Center for Mind and Culture