In a research paper published earlier this year, a team of psychologists reported a study in which 39 children (from Belfast, as it happens) were each given a challenging game to play. They had been split into three groups: in one, each child played with an adult present; in the second, they played alone; and in the third, they were told that they were being watched by the invisible “Princess Alice”.
In order to do well at the game, it was actually necessary to cheat, to break the rules. As you might expect, the children who thought that nobody was watching cheated most, and those watched by an adult cheated least. Those who had been told they were in the presence of Princess Alice averaged in between; but on further investigation the researchers found that this group was composed of two sets: the children who believed that Princess Alice was real, and those who didn’t.
The behaviour of the believers was exactly the same as that of the children who were supervised by a real, non-invisible adult.
The conclusion is almost too obvious to require pointing out. If you believe that an invisible being is observing your every move, you won’t cheat; your behavior will be more moral.
The senior researcher in this experiment was Jesse Bering, author of The God Instinct (The Belief Instinct in the USA). The main theme of the book is Bering’s hypothesis that religious belief is “adaptive”, that is, it’s something which gives an evolutionary advantage, and thus tends to spread in a human population. In this, he is directly in conflict with Richard Dawkins, who suggests in The God Delusion that instinct for religion is a mere by-product of the evolution of the human mind.
I’ve argued the latter point myself. It seems to me that there are two obvious features of human cognition that might be enough to create religion. First, as social animals, humans need to understand “agency” — who did that? why did he do it? what was in his mind? — and have a tendency to over-apply the concept to, say, natural phemomena — God did it. Secondly, the human brain is fantastically tuned to detecting patterns, both in time and space. Even, sometimes, patterns that aren’t really there. “Every time I wear my lucky socks, we win the game.”
I have to admit, though, that Bering’s point of view is better-argued and supported by research and experiment, rather than Dawkins’ vague appeal to things that “make sense” (as I did in the last paragraph). The idea that religious early cultures out-competed non-religious ones through enhanced cooperation and lower numbers of “cheaters” has not been proven, but it’s certainly a possibility.
But atheists like myself (and Dawkins, and Bering) have an instinct to be moral too. Even if a moral culture really is more successful than a purely selfish one, there’s no actual need for the religious aspects to enter into it. The way evolution works though, is that some pre-existing feature always undergoes a mutation or modification that gives a reproductive advantage, or an environmental change makes it so.
So I think they both could be right. Religion might have been an inherent quirk of human cognition which only became evolutionarily adaptive once societies grew beyond a certain size or complexity. But after that, as communication, education, democracy and justice developed, religion lost its evolutionary advantage again. Now it’s just useless.