There are 9% fewer atheists in foxholes

According to an old military saying, “There are no atheists in foxholes.” The idea is that when you live with death close by, you think a lot about your mortality, its possible abrupt end, and the meaning and value of it all. Winston Churchill said “Nothing in life is so exhilarating as to be shot at without result” (though when he wrote that, while reporting from Afghanistan in 1897, he was commenting about the reaction of members of a British patrol he hadn’t actually been on). Of course, another possible reaction to combat is despair. What the “Lost Generation” of the 1920s had lost was hope. How could a merciful God have allowed the mass slaughter of 1914-18?

It’s amazing what economists study these days, but “Have data, will travel.” A new research study in the working papers series of the National Bureau of Economic Research in Cambridge, Massachusetts, carries the profound and provocative title: “Death, Trauma and God: The Effect of Military Deployments on Religiosity.”

Never again accuse economists of dealing only in petty materialism.

The paper’s authors are the economists Resul Cesur, Travis Freidman and Joseph J. Sabia of, respectively, the University of Connecticut, the Peter T. Paul College of Business and Economics in Durham, Hew Hampshire, and San Diego State University.

At one time, discussion of how people respond to the proximity of death might have been reserved for philosophers. But now we have data—lots and lots of data, terabytes more every day, it seems—so social scientists can have their say, too.

In this case, Cesur, Freidman and Sabia use two sets of data. One is a national survey studying “adolescent to adult health” that has been following a cohort of high school students since 1994-95 and which now includes a “military module” of people who served in the armed forces. The other is a 2008 survey conducted by the U.S. Department of Defense on “health and related behaviors” that also asks questions about combat experience, including being shot at and witnessing death. As always, each data set has shortcomings that make it less than perfectly ideal for the problem under study, but on balance the two sets allow for an investigation of the effects of combat, either or actual or potential, on people’s religious feelings and behaviour.

Let’s cut to the chase. The researchers find that “servicemen assigned to combat zones are 8.9 percentage points more likely to attend weekly religious services than their counterparts in non-combat zones.” As it happens, they’re also 8.9 percentage points “more likely to engage in private prayer… and 4.3 percentage points more likely to report religion as important to them.” How big are 8.9- and a 4.3-percentage point changes? Both pretty big and not so big. The relevant baselines are that “15.4 per cent of active duty deployed servicemen attend religious services weekly” so an 8.9 percentage-point jump is actually pretty big. At the same time, 51.1 per cent say religion “is an important aspect of their lives” so a combat-related 8.9-point jump in that share isn’t quite so big, while fully 75.1 per cent “report praying outside their house of worship” so a 4.3-point increase in that number isn’t very big at all.

Of course, these numbers could be spurious correlation. It might be that for whatever reason only the more religious combat units get assigned to combat zones. So before calculating these effects the researchers had to see whether the data they had on individuals predicted whether or not they got sent into combat. After a careful review, they conclude that assignment was random.   

The results I’ve quoted are from the first data set described above. The second data set allows the researchers to get at the separate effects of having been in a firefight, seen death or been wounded. In general, though significant, these effects are smaller, which the researchers attribute to the fact that everyone in a combat zone exhibits higher religiosity. Actually being in combat, as opposed to being forced to anticipate it more or less constantly, apparently doesn’t add much.

What’s causing this increase in religiosity among soldiers assigned to combat zones?

As good economists, Cesur, Freidman and Sabia consider several alternatives and, as is often the case, have trouble deciding definitively what exactly is going on. One possibility is that, because of an active chaplaincy service in the U.S. military, there are more opportunities to worship and, frankly, to be proselytized. Another is that it’s part of peer group behaviour. And a third is that, yes, looking his or her own death in the face does turn a person toward religion.

And, oh yes, let’s not forget more mundane economic possibilities. As the researchers write, “A final pathway through which combat deployments could affect religiosity is via income effects. Combat deployments are accompanied by increased income in the form of hostile fire pay (HFP) or imminent danger pay (IDP). Religious engagement has been found to be positively related to income, consistent with religious organizations serving as club goods.” In the end, however, their research doesn’t say much about whether these income effects do finally play a role.