Your independent source for Harvard news since 1898

Your independent source for Harvard news since 1898

Right Now | Biology and Belief

Foundations of Faith?

September-October 2007

Illustration by Keith A. Negley


Illustration by Keith A. Negley

Public debates about evolution frequently pit science against religion. But work by professor of mathematics and of biology Martin Nowak adds a surprising twist to that age-old argument. Evolution, he says, might actually lead to religious beliefs.

Nowak, the director of Harvard’s program for evolutionary dynamics and author of a recent, prizewinning book on the subject, has worked for the last decade on questions of cooperation in biology. Although evolution is commonly understood as competition played out in mutation and natural selection, “You don’t really get very far in biology without cooperation,” he says. “The evolution of multicellularity and the evolution of eukaryotic cells which have a distinct nucleus—the big steps—seem to be based on cooperation” and lead to “the emergence of something completely new in biology. You get cooperation on a lower level and then you get the emergence of a higher level: single cells cooperating to form multicellular organisms.”

Researchers have long studied cooperative strategies for evolutionary success among their fellow human beings by using game theory. The classic “prisoner’s dilemma” involves the hypothetical case of two suspects arrested by police and held in separate cells. Each is offered the same deal: confess and reduce your own sentence, but in doing so, implicate the other prisoner. The best outcome for both prisoners is for neither one to say anything (a form of cooperation), but when neither prisoner knows what the other will do, the incentive to defect (or plea bargain, in this case) is strong.

Nowak runs a variant of this game using money, and has discovered that the most successful players start by cooperating with each other—paying a dollar, for example, so the other player receives two dollars—and then continue to cooperate until someone defects. This tit-for-tat strategy, in which players cooperate when others cooperate with them, but defect when their opposite number has defected in a previous round, proved the most successful in a well-known computer tournament that pitted various strategies against each other. But in the real world, Nowak points out, mistakes or misunderstandings (which weren’t part of the computer simulations) can destroy cooperation just as surely as true defection: “If you have two tit-for-tat players playing against each other and one defects by mistake, the other will retaliate, and then the first will retaliate, and so on.”

Nowak has shown mathematically that in an error-filled world, a more generous variant of tit for tat—in which if one player defects, the other “forgives” and still cooperates a certain amount of the time—is more successful. (Players who are too forgiving, or always cooperate, however, are exploited by others.) In other words, says Nowak, “To win the prisoner’s dilemma in a world of errors, you must know how to forgive. Natural selection leads to forgiveness.”

This simple form of interaction, known as direct reciprocity, relies on repeated encounters between two people. But direct reciprocity fails to explain a whole range of observed “cooperative” or altruistic human behaviors in which individuals perform selfless acts for others without apparent expectation of a reciprocal act. Nowak wanted to explain such ad hoc actions, and in 1998, he formulated a much more powerful mechanism, “indirect reciprocity,” by which evolution can lead to cooperation.

Indirect reciprocity relies on reputation. Observers watch the game and tell others what they’ve seen. “Indirect reciprocity is powerful because I gain experience even without playing,” Nowak explains. Humans “are obsessed with finding out about each other and...monitoring the social network of the group. There are empirical studies of what people actually talk about on British trains, for example, or in small-scale societies, sitting around fireplaces. Sixty percent of the conversations” deal with interactions with others.

When indirect reciprocity is incorporated into Nowak’s game, cooperative players tend to find each other and form clusters, because a single defector can exploit and “pop” a bubble of cooperation. The ability to communicate about reputation thus becomes extremely important to a cooperative player’s success.

Nowak, in fact, has demonstrated mathematically how indirect reciprocity provides selective pressure for the development of human language, which he calls the most important invention of the last 500 million years. “Prior to the invention of language, evolution was based on genetics only,” he explains. “With human language, we have the machinery for unlimited cultural evolution.”

Religion is a form of social behavior that Nowak studies as part of a joint-research project with the Divinity School on the “Evolution and Theology of Cooperation.” Religion, like language, is a human universal, he points out: every civilization has discovered it. And he finds it remarkable that the winning cooperative strategies in both direct and indirect reciprocity require players to be hopeful, generous, and forgiving, traits often encouraged by religions. “Hopeful, because if we haven’t met prior to the first time we interact, the winning strategy must actually assume the best of the other person. Generous, because the winning strategy is happy with a somewhat smaller piece of a pie”: the strategy wins not by getting a bigger slice of any one transaction, but by cooperating successfully in many transactions. “Forgiving, because if I’m in a relationship and some one defects, and I always hit back—that’s a losing strategy. I have to have a mechanism to forgive a failure to cooperate.

“Religions want to understand what causes suffering and they also want to help people. When you study their prescriptions for how to live one’s life, what is formulated is how to live the best possible life right now. The work on cooperation shows which kinds of behavior are important for human evolution,” he explains, and it “seems that religion actually helps people to behave accordingly: to cooperate with each other.”

~Jonathan Shaw