Improve your self control… with religion?

The ability to control short term urges in order to achieve long-term goals makes a big difference to what you get out of life. People with high levels of self control forgo cream cakes for healthy food, and opt to study rather than succumb to the temptations of all night raves. As a result, they tend to be healthier, earn more, and live longer.

Religious people tend to have higher self control – after adjusting for all the other factors that make religious people different. So here’s the big question: which comes first? Does religion increase your powers of self control, or do people with high levels of self control tend to gravitate towards religion?

Michael McCullough (with colleague Brian Willoughby), at the University of Miami, thinks that religion has a direct effect on increasing self control, and he’s put down the evidence in a rather lengthy paper to be published this month in the Bulletin of the American Psychological Association (you can get a preprint from McCullough’s homepage). The NY Times has a nice summary, but misses a couple of crucial insights.

First off, it isn’t religious belief that has the effect. It’s religious institutions. People who score higher on ‘Self-Transcendence’ (i.e. they answer ‘yes’ to questions like, ‘“Sometimes I have felt my life was being directed by a spiritual force greater than any human being’, and ‘“I sometimes
feel a spiritual connection to other people that I cannot explain in words’) in fact have lower than average levels of conscientiousness (a key aspect of personality linked to self control).”

Second, there’s actually precious little direct evidence that religion really does increase self-control. For example, there’s one study that indicates that religiousness as an adolescent is linked to changes in only one of two personality factors linked to self-control, and only in women, not men. On the other hand, there are five studies showing quite clearly that high levels of self control precede changes in religiousness. McCullough & Willoughby acknowledge this rather fundamental weakness in their arguments:

Except for one study suggesting that individual differences in religiousness precede longitudinal changes in Agreeableness (at least for women) and a single experiment showing that religious cognition is automatically recruited for self-control (Fishbach et al., 2003; Wink et al., 2007), however, the available evidence for evaluating whether religion causes self-regulation or self-control is rather meager.

But I would go further, and suggest that what the actual evidence (rather than the theory) seems to show is that people with high levels of spiritual belief who also have high self control tend to turn to organized religion – and that this is the primary driver of the link between religiousness and self control.

Now this doesn’t mean that religion doesn’t increase self control. It’s just that the most important effect is probably in the reverse direction. And this has important implications for the conclusion. It’s not enough to show that religion could, in theory, have an effect on self control. If you want to draw the sorts of conclusions drawn by the NY Times reporter, you have to show that the magnitude of the effect is meaningful in the real world.

And here’s the other thought. McCullough & Willoughby think that this provides a rationale for an evolutionary basis to religion. In other words, they argue that religion enhances reproductive fitness by increasing self control, and that this means that religiousness is selected for by Darwinian evolution.

But to argue this, you would have to show that there is something specific about religion, as opposed to mystical beliefs or secular institutions, that enhances self control. And studies conducted mostly in the US, where the major institutional route to civic participation is religion, are not going to be able to tease these things apart.

I think that McCullough would agree with the idea that secular institutions could fulfil the role of religious ones in the modern world, according to the NY Times report:

Religious people, he said, are self-controlled not simply because they fear God’s wrath, but because they’ve absorbed the ideals of their religion into their own system of values, and have thereby given their personal goals an aura of sacredness. He suggested that nonbelievers try a secular version of that strategy.

“People can have sacred values that aren’t religious values,” he said. “Self-reliance might be a sacred value to you that’s relevant to saving money. Concern for others might be a sacred value that’s relevant to taking time to do volunteer work. You can spend time thinking about what values are sacred to you and making New Year’s resolutions that are consistent with them.”

The last bone of contention I have with the conclusions being drawn is the extrapolation from the general to the personal. Religious people are not the same as non-religious people. They tend to be of lower educational achievement, and may well start off with lower levels of self control. Even supposing that religion increases self control to a meaningful extent in these people, that is not the same as saying that it will work for you!

Michael E. McCullough, Brian L. B. Willoughby (2009). Religion, Self-Regulation, and Self-Control: Associations, Explanations, and Implications Psychological Bulletin. In press.