Read our COVID-19 research and news.

Sistine Chapel (Cappella Sistina) by Michelangelo Buonarroti, 16th century, fresco (post restoration), Buonarroti, Michelangelo (1475-1564)/Musei e Gallerie Pontificie, Musei Vaticani, Vatican City/Mondadori Portfolio/Bridgeman Images

Did judgmental gods help societies grow?

*Update, 7 July 2021, 11 a.m.: The authors have retracted the study described in this story because of a data coding issue. They misleadingly coded societies for which belief in moralizing gods is unknown as having an absence of moralizing gods. However, the authors say they have since "thoroughly refined our data and analyses, and have found that our original conclusions are still strongly supported." They have submitted a new analysis to another journal. 

Today’s most popular religions have one thing in common: gods or supernatural laws (such as karma) that dictate moral behavior and punish transgressions. Act morally and these supernatural forces will reward you; break the rules and you’ll be punished.

But moralizing gods seem to be quite rare in human history. Researchers know from ethnographies that the gods of hunter-gatherer societies, for example, don’t much concern themselves with humans, much less their moral behavior. (Many of them focus on nature instead.) Now, a new study tests a popular hypothesis about why moralizing gods eventually took over.

Many scholars argue that moralizing gods were needed to build large-scale societies, an idea sometimes known as the “big gods” hypothesis, although it applies to impersonal supernatural moral laws like karma as well. Hunter-gatherers live in small bands in which everybody knows everybody else, so immoral behavior is virtually guaranteed to be discovered and punished. But in larger, more anonymous societies—from networks of interconnected villages to the first cities—people can break the rules without anyone noticing. If everyone did that, society would fall apart, so moralizing gods were needed to keep an eye on everyone and encourage cooperation instead of cheating. The more people cooperate, the more the society can grow.

To test this idea, a team of researchers used a new historical database called Seshat (named for the ancient Egyptian goddess of wisdom). Seshat contains information about the sizes, governments, militaries, religions, economies, and more of hundreds of societies spanning the past 10,000 years, making it possible for researchers to quantitatively compare them.

The scientists analyzed 414 societies from 30 regions around the world, from the deep past until the Industrial Revolution. They classified each society according to 51 measures of social complexity, such as how many people belonged to it and whether its government had hierarchical leadership. They also attempted to determine whether each society believed in a moralizing god (or gods) or a supernatural law that enforced values such as fairness and loyalty. It’s “very ambitious,” says Carol Ember, a cross-cultural anthropologist at Yale University, who wasn’t involved in the new research.

Large-scale societies did tend to have moralizing gods, whereas small-scale societies didn’t, the team reports today in Nature. But when the researchers zeroed in on the 12 regions for which they could examine societies before and after the emergence of moralizing gods, they found that moralizing gods consistently appeared after a society had already grown large and complex.

That means these deities couldn’t have helped a society with its initial growth spurt, says Patrick Savage, an anthropologist and ethnomusicologist at Keio University in Fujisawa, Japan, and an author of the new study. He suggests participating in religious rituals—which do tend to appear as social complexity is increasing—may be more important than belief in moralizing gods for first promoting cooperation. Once societies reach 1 million members or so, he says, moralizing gods seem to come in to stabilize cooperation between people who may have different languages, ethnicities, or cultural backgrounds.

That’s “an interesting alternative hypothesis” that deserves to be investigated, says Edward Slingerland, a historian at the University of British Columbia in Vancouver, Canada, who helped develop the “big gods” hypothesis. But he worries about the reliability of Seshat’s data, because the majority of them were collected and classified by research assistants and not expert historians.

“People are really going to be scrutinizing the data,” and rightfully so, adds Quentin Atkinson, an evolutionary psychologist at the University of Auckland in New Zealand who wasn’t involved in the new research. He points out, for example, that written or archaeological evidence for moralizing gods likely appeared well after belief in them begins, a lag that can skew the timing of their emergence in a database such as Seshat. “A lot rests on the quality of that information.”