Field of Science

Religion and support for torture

So, what do you think? Are the religious in America more or less likely than average to support the use of torture?

To find out, Arial Malka (Yeshiva University, New York) and Christopher Soto (Colby College, Waterville, Maine) used data from a couple of opinion polls, one conducted in 2004 for ABC News/Washington Post and a larger one (nearly 2,000 people) conducted in 2008 (the 2008 American National Election Studies time series).

Here's an example of what they asked (this is from the 2004 poll):

“Some people say it’s acceptable to torture people suspected of terrorism, in cases where other methods have failed and the authorities believe the suspect has information that could prevent terrorist attacks and save lives. Other people say the use of torture is never acceptable because it’s cruel, it may violate international law, it may not work, and it could be used unnecessarily or by mistake on innocent people.”

Well, it turns out that, in both surveys, religious Americans were actually slightly less likely than the less religious to condone torture (they measured religion using a composite of attendance, prayer, and subjective ratings of importance). But that's only half the story.

They were also interested in the interaction with political persuasion (liberal or conservative), and they tested this using a statistical technique that allows you to check if one variable (in this case, religiosity) might be influencing another variable (in this case, attitudes to torture) only indirectly - via its effects on a third variable (political persuasion).

What they found was consistent with a set up where religion makes people conservative, and that in turn makes them support torture. In other words, religion has a direct and an indirect effect. Basic religion (in their model) opposes torture, but it also religion increases support for conservative politics. As a result, it indirectly increases support for torture.

What's more, this indirect effect was much stronger in in educated people. In educated people, religion is more likely to be linked to conservative views, and conservative views are more likely to be linked to support for torture.

In my view, the real interest in these results is that they underscore once again just how complex religion is. I think that the motives for educated people to embrace religion differ from the motives of the less educated.As a result, the kind of religion they have, and the purposes they put it too, are different.

They make religion in their own image.
Malka A, & Soto CJ (2011). The Conflicting Influences of Religiosity on Attitude Toward Torture. Personality & Social Psychology Bulletin PMID: 21525330

Creative Commons License This article by Tom Rees was first published on Epiphenom. It is licensed under Creative Commons.


  1. ==I think that the motives for educated people to embrace religion differ from the motives of the less educated.==

    Yeah, probably educated people have to be twice as stupid to embrace religion. Like in the expression: "I think twice before I say something stupid."

  2. My thoughts about the interaction with education are somewhat similar to Mihai's. My thought is this: Many people -- including myself -- oppose torture only partially on ethical grounds, but also because it doesn't work. Furthermore, the ethical argument can be a bit subtle for some people, e.g. like it's easy to just say, "he bad man so it not matter what we do to him." Bottom line, *if everything goes as planned*, becoming more educated should help a person to better understand the problems with torture, and so it should increase opposition (and I believe it does).

    But if you throw in some screwed-up ideology that doesn't allow the person to reject torture... there ya go.

  3. I am suspicious of this report; I would like to see more of the specifics, but the study is behind a paywall. It seems unclear--that religion represents both less and more support for torture. Not to mention that this conclusion directly contradicts a 2005 Pew survey which found that religious people, especially Catholics, supported torture more, and that secularists opposed it most.

  4. Broadly speaking, more religious were slightly less likely to support torture on average. But the real point here is that there is no such thing as 'religion'. There's a multitude of religions (even within Christianity), and the one you choose is linked to the type of person you are (and whether you support torture).

    But what this study doesn't do is give any information on non-believers. Or rather, there are so few of them in the USA that they would have been swamped in the statistical noise.

    Non-believers are more likely to be liberals, and liberals are less likely to support torture. So I can imagine that non-believers are less likely to support torture, despite the headline results of this study.

    In fact, you could develop a model that says non-belief -> liberalism -> less support for torture. That would probably fit the data just as well as the model they created.


Markup Key:
- <b>bold</b> = bold
- <i>italic</i> = italic
- <a href="">FoS</a> = FoS