Friday, February 16, 2007

Do Atheists Commit Less Vicious Sexual Crimes?

This article suggests so. From the abstract:

This article examines associations between self-reported religious affiliations and official offense histories among 111 incarcerated adult male sexual offenders. Four categories of religiosity were devised according to self-reported continuities and discontinuities in life-course religious affiliations: atheists, dropouts, converts, and stayers. ANCOVAs indicated that stayers (those who maintained religious involvement from childhood to adulthood) had more sexual offense convictions, more victims, and younger victims, than other groups. Results challenge assumptions that religious involvement should, as with other crimes, serve to deter sexual offending behavior....

One is reminded, of course, of the Catholic church's troubles over the last several years. The average age of the youngest victim of the "stayers" was 9.5 years.

A few caveats: (1.) "Atheists" just means comparatively low reported religiosity, both childhood and present, and may not correspond to full-blown atheism. (2.) One cannot draw the conclusion that atheists are less likely, in general, to commit sex crimes, since rates of atheism and religiosity were not measured in the general population. (3.) The researchers (Donna Eshuys and Stephen Smallbone) seem to overlook the possibility that the "stayers" may come from a sub-population in which sex crimes are less likely to be detected and/or lead to jail time, in which case one might expect only the worst offenders to be incarcerated; or that maybe the more vicious sexual offenders are more likely to exaggerate their religiosity.

The general literature on the relationship between religiosity and crime is mixed and difficult to interpret. If only we could randomly assign religiosity! ;) Unfortunately, the social factors affecting religiosity are entangled with those affecting criminality. Furthermore, the literature may be distorted by the fact that null results are hard to publish and by the prior commitments of the researchers (many are from religiously affiliated institutions, for example).

Why do I care about this issue? Well, if academic moral philosophy has no great positive effect on one's moral character (as I'm inclined to think), then one might hope for religion as a better source. I'm not especially optimistic about that either (though I have some hope that there may be particular religious practices with positive moral effect), but I haven't yet explored the empirical literature as much as I'd like.


Sam said...

Hi Eric,

What a great post! I think part of the problem is defining what religion is...a difficult undertaking! If we take religion in terms of those who profess to belong to a particular religious tradition, I do not believe we would see much difference between the religious and the non-religious person in terms of behavior. From what you report, I'm actually not surprised that it appears to be worse in some areas (sex crimes). I would attribute that to the extensive amount of socially enforced silence and repression that goes on in many religious subcultures.

On a related topic, my friend and mentor Dallas Willard (philosophy professor at USC) is working hard on a new book (or essay - I'm not sure yet) on moral knowledge. You might get an better idea about his work here:; there you will also find a lot of his writings on religion/spirituality. I am honestly giving a plug for my mentor and his work; he really is an amazing person.


Lester Hunt said...


Interesting post! I had not heard of these results. Perhaps you have heard of the study that shows a surprisingly high correlation between atheism and marital stability, with born again Christians getting rather low scores in that regard. See:

The truth of the matter may not be that religion has no effect on behavior, but that other things have more. We atheists tend to be more prosperous and better educated than other people. Education-level and solvency might be what really matter.

Everybody knows, roughly, what rules we have to follow in order to live together in peace and cooperate with each other. You don't need a pipeline to God, or philosophical talent, to figure that out. The question is, who is most likely to actually comply those rules? This might be a matter of incentives more than anything else. Better educated, more solvent people might just have a stronger stake in society than others, a stronger interest in keeping it going, etc.

Eric Schwitzgebel said...

Thanks for the kind words, Sam. I met Dallas a few years ago when I gave a talk as USC, but I don't know much about his work. I will definitely follow up your lead though! I'm also sympathetic with your suggestion that maybe self-described religiosity isn't the best measure of genuine religiosity. (I had a post on this a while back, actually, regarding the question of whether people really believe in God and Heaven.)

I hadn't heard of the results you mention, Lester, and I appreciate the tip. I'm inclined to agree that there's a complex entanglement of social factors here, and that one should be vary wary of drawing causal conclusions!

I'd hate to be forced to the conclusion that moral thinking and moral behavior are completely unrelated; but if my readings and research are right, the relationship is considerably more tenuous than one might hope or expect -- at least when the matter is viewed from afar, without looking at particular ethical views or religious teachings.

Eric Schwitzgebel said...

Oh and P.S.: You may be right, Lester, that part of the issue is that the basic moral commands aren't really news -- atheists, non-ethicists, and everyone else already knows the basics of moral behavior.

(But that can only be part of the story, I think; since even what isn't news we can be helpfully reminded of and strengthened in, in the right conditions.)

Anonymous said...


This is a great, if disturbing, discussion.

I share your your sentiment: "I have some hope that there may be particular religious practices with positive moral effect." My suspicion is that those who engage in religious practices such as contemplative prayer and certain ascetic pratices, like fasting (for the purpose of training one's will, not for the purpose of bartering with God) would be much less likely to commit a variety of crimes. Consequently, I'm inclined to think that religious traditions that emphasize such practices -- e.g., Eastern Orthodox Christianity, Buddhism, strands of Roman Catholicism that are committed to the monastic practices of the early church, etc. -- would have substantially fewer members committing such crimes. Currently, however, that is just my suspicion. Are you aware of any interesting research that has been done on the topic?


Eric Schwitzgebel said...

Nice to hear from you, Rico! I'm hoping to spend some more time with the empirical literature on the question of the relationship between particular religious practices and moral behavior. When I have something interesting to report, I'll definitely post on it!

Part of the problem is that psychologists studying this question tend to go for easy measures such as self-reported religiosity or church attendance.

On contemplative prayer and asceticism in particular: If there are particular salutary religious practices, I agree that those might be among them. (I would also add mindfulness meditation.) Unfortunately, though, I don't think a simple measure of religious affiliation (e.g., Eastern Orthodox, Buddhist) is going to have a tight relationship to engagement with such practices, especially among people born into households with those religious affiliations.

Anonymous said...

Right. A simple measure is not likely to be particularly helpful. (The forms of measuring one's 'religiosity' seem deeply problematic in studies such as this one.) Although I would expect the people who engage in such virtue developing practices (assuming there are such practices) to be disproportionately among those who are members of religious traditions that emphasize the practices.

BTW: One implication of this discussion that I find fascinating is that certain religions, which seem to have strikingly different doctrines, might be very much kindred spirits insofar as they have strikingly similar prayer, meditative, and ascetic practices.

Eric Schwitzgebel said...

Yes, I agree with both your points, Rico -- thanks!

Anonymous said...

TO answer to points above...

... The study that showed a "Surprisingly high" level of Ateist Fidelity in Marriage, and a signifigantly higher DIvorce rate in Born-Againt CHristians, doen by Barna, has made its roudns in the usual Ahteist Venues, but isn't accurate.

Many even think it means all CHristaisn have a signifigantky higher Divorce rate than Atiests.

Acutlaly, Mainline Protestants did better than Ateists, and Cahtolics and Lutherans both scroed far superior numbers. At leats if "Far superior" is measure dby 2 or 3 percent.

The rate for Atheists was 21% and for "Born-Again Christians" it was 24%. This is a 3% margion of diffrence, nto a Signifigant gap. In facft, its such a small difference that if the sma epoll as done only a week later the figures may be reversed withotu raising an eyebrow, with the Born Again CHristains rankign at 21% and the Ahtists at 24%.

A 3 Percent difference in the survey sample is not relaly a signifigant increase.

I'm surprised tha many have lape donto it, without criticlaly examinign it.

Besides, the same survey reported a lower divorc rate for the Mainline Protestant Chruches and for Catholic and Lutherans, as noted before. Why is htis always lef tout? (Well, other han it doesn't help with the idea of Atheists beign somehow mor emoral.)

In order.

Denomination (in order of decreasing divorce rate) % who have been divorced

Non-denominational ** 34%
Baptists 29%
Mainline Protestants 25%
Mormons 24%
Atheists, Agnostics 21%
Catholics 21%
Lutherans 21%

So, no, the "Atheists divorce less oftenthan CHristians' Claim isnt relaly true. Not only is the margion of difference acualy negligable, but many Christain gorups score evenly witht he Agnostic and Atheists and on further pollign less.

Alo, the old Myth of Athiest beign more intelelgent and better educated isn't quiet true either.

Psycological studies show no correlation ebtween Intellegence and Religion. Neither is it true that Atheists tend to be better educated. Many Ahteists in Eurioe ony have a basic education, and in the United States this is also becomign a trend.

Intellegence and education ar enot relaly gurentees for loosing oens rleigiou convicitons.


Eric Schwitzgebel said...

Thanks for the data, Zar!

I'm inclined to think that overall the literature shows no relationship between religiosity and morality, as the study you cite also suggests. (This study is an interesting exception.)

I don't see the lack of such a relationship as particularly good news, though, for those who think that religion is effective promoting morality.

Bad Horse said...

I keep searching this post for where you wrote the study results, but... you didn't.

Eric Schwitzgebel said...

Brad, click the first link.