You can see why the secularist might feel cheated. Every argument he makes against religious belief runs up against a great foggy X-factor called “God” and a useful hedge called “the Fall of Man” and an ace up the sleeve called “grace.”

He argues that people take to religion as a crutch, because they can’t get through life without help—or, as he thinks, the illusion of help—and the Christian and Jew and Muslim smile benignly and admit they can’t. They explain that we are crippled by sin and death, and God has graciously provided the aid we need. The lifesaving ring someone throws you when you are drowning remains real even though you want it desperately.

He may also argue that we believe in God because we have a “God gene” and are hard-wired to do so, and again the religious believer smiles benignly and admits that we might well have a God gene. He suggests that a loving creator might well arrange our wiring to make belief easier, knowing how hard it will be for us.

Or the secularist may argue that we believe in God because we want to claim Divine sanction for our worldly interests and desires, and points to the allied and German soldiers in World War I singing hymns as they tried to kill each other, and the religious believer shakes his head sadly and admits that many Christians have done this from the beginning. He shrugs and explains that God loves his creatures even though they make a mess of his gifts, and that some of them get it right anyway.

No evidence of the human origins of religious belief will upset the religious believer, because he can always appeal to a very convenient, and convently mysterious, relation between God and a defective humanity. It's all grace, he will say. What seems like good evidence that religion is a sham looks to the believer like yet more evidence that God loves us.

The secularist must feel cheated. He's playing cards with a rulebook that says, “Whatever hand you have, his is better, because all his cards are wild cards.”

I thought of this when running through my files looking for an illustration for a column I write for our diocesan newspaper, and finding a story from the bright, stylish, and very intelligent magazine New Scientist with the headline “Dear God, please confirm what I already believe.” Reporting on a study of religious belief, the news story began: “God may have created man in his image, but it seems we return the favour. Believers subconsciously endow God with their own beliefs on controversial issues.”

As it turns out, the writer was more certain of the conclusion than the scientists whose study he described. They kept saying “may,” although the few quotes in the article suggested that they really wanted to say “is.” The study, led by a professor from the University of Chicago, was published in the Proceedings of the National Academy of Sciences.

The study had found that people tend to think that they believe what God believes, or that God believes what they believe. (You may feel an Inspector Renault in the casino moment coming on.) The scientists involved then asked some of their subjects to do things that might change their beliefs, like write a paper arguing for the other side—the article mentions capital punishment—and found that people tend to change their beliefs and at the same time change what they thought were God's beliefs. They still thought that they believed what God believed, or God believed what they believed, even though they had changed their minds somewhat.

The scientists then scanned the brains of their subjects while they thought about God and found that they used “similar parts” of their brains when thinking about their own and about God's beliefs, and a different part when thinking about other people's. “This implies that people map God's beliefs onto their own,” the writer claimed.

The study gives what seems to be the obvious conclusion from this evidence:

"People may use religious agents as a moral compass, forming impressions and making decisions based on what they presume God as the ultimate moral authority would believe or want," the team write. "The central feature of a compass, however, is that it points north no matter what direction a person is facing. This research suggests that, unlike an actual compass, inferences about God's beliefs may instead point people further in whatever direction they are already facing."

But not really so obvious, even on purely scientific grounds. For one thing, judging from the story, the study doesn't seem to have measured whether the subjects' religion limited how much they changed their views when asked to do something designed to change them.

Maybe their religious beliefs kept them from changing their views as much as they would have if they didn’t believe what they believed. Perhaps it limited the effects of their new sympathy for the other point of view. That’s perfectly possible, and if true would show that they don’t use religion simply as an “echo chamber,” but that it binds them in some way.

The experience is one believers have all the time. You listen to a lonely homosexual man who desperately wants to be married, or a woman who feels she had to abort her child, and everything in you wants to affirm them and reduce their suffering. But you can't, because you know what God has said. You may push the limits of your belief and even push them too far, but you won't break them. You feel bound by an authority outside and above you.

The study also apparently didn’t investigate whether its subjects changed their views back later, when the effect of the exercise had worn off. This is also perfectly possible, and if true would also show that religion binds them and doesn't simply echo beliefs they hold for other reasons.

And again, this is an experience the believer has. You may believe in capital punishment, but find the story of a prisoner on death row—the child of an abusive home, perhaps—drives you to doubt it. Or you may reject capital punishment but find the story of a horrifically sadistic murder driving you to support it. In either case, when the emotional reaction wears off, you will remember that hard cases make bad law and go back to the position to which you'd come in more objective moments.

Nor does the study seem to have studied whether the extent to which its subjects changed their views varied with how authoritative they believed the teachings were. It may be that they let their sympathy affect them more the less they thought the issue mattered, and affect them less the more they thought it mattered. They might feel they have permission to change in some things and not others. This is also perfectly plausible, and would also suggest that they don’t use religion merely as an echo chamber.

Here too the believer knows the experience. You might change your position on the death penalty because you think it one on which believers can disagree, while refusing to change your position on abortion. There you are bound.

The secularist seems to assume that imperfection invalidates belief. If religion is a crutch, it can't be true. That, it seems to me, is itself a kind of cheating, because it requires asserting—unscientifically—the secularist reading of the evidence as if it were the objective, scientific, indisputable one.

That, at least, is what the author of the article and, though not so assertively, the scientists who produced the study, have done. “Believers subconsciously endow God with their own beliefs on controversial issues,” declares the first; “inferences about God's beliefs may instead point people further in whatever direction they are already facing,” says the second. The declaration is quite dubious, even on their own evidence, but delivered with certainty.

In these cases, the secularist and the believer will read the evidence differently, and both readings fit the data. The crutch may be an illusory crutch or a real one. The apparent convenience of religion may result from wishful thinking or from an encounter with the living God, who with infinite patience gives himself and his revelation to his stupid, rebellious children no matter how much they misunderstand and corrupt it.

In other words, everything the secularist says about the origins of religion may be true, and also irrelevant.

David Mills is Deputy Editor of First Things. His previous “On the Square” articles can be found here. The article he quotes is Andy Coghlan's Dear God, please confirm what I already believe.