Atheism and intelligence

There’s some research out there concerning correlations between intelligence as psychometricians understand it and atheism. I’ve come across (thanks to Prem Dhanesh) another example: “Average intelligence predicts atheism rates across 137 nations,” by Richard Lynn, John Harvey and Helmuth Nyborg. The abstract:

Evidence is reviewed pointing to a negative relationship between intelligence and religious belief in the United States and Europe. It is shown that intelligence measured as psychometric g is negatively related to religious belief. We also examine whether this negative relationship between intelligence and religious belief is present between nations. We find that in a sample of 137 countries the correlation between national IQ and disbelief in God is 0.60.


One way to think of this sort of research is as confirmation of the prejudice of nonbelievers that being smart works against believing in fairy tales. Still, some caveats apply, aside from the obvious one that this is just one study that I’ve happened upon, and that as a non-psychologist, I am not in any position to tell whether this research is good or not. One important point, for example, is that, as Keith Stanovich points out in his excellent What Intelligence Tests Miss: The Psychology of Rational Thought, intelligence and rationality can be associated with very different skills. Psychometric intelligence is, I am inclined to think, oversold and overrated.

So I think I’ll let research like this increase my SQ (smugness quotient) a bit. But not too much, especially if I remember what a complete idiot I’m also capable of making myself.

About Taner Edis

Professor of physics at Truman State University

  • PersonalFailure

    I tend to think, if the research is valid, that it’s not that stupid=religious or religious=stupid, but that the more intelligent you are, the more likely you are to questions things and the more able you are to think up those questions.

    Unfortunately for religion, questioning fairy tales tends to lead in one direction.

  • Keith

    So I think I’ll let research like this increase my SQ (smugness quotient) a bit.Careful. A spiritually inclined quantum physicist has developed a “spiritual quotient” which has already acquired the SQ acronym:

    I suspect that credulously believing in at least some sort of spiritual reality is a prerequisite for being “spiritually intelligent” (whatever the hell that means) :)

  • Dianelos Georgoudis

    Who was it who said that there are lies, damned lies, and statistics?

    This paper shows that in some countries, namely the poorer ones, there is a negative correlation between IQ and religious belief. So what’s the best explanation of this fact? Well, as should be more or less obvious, poorer people tend to need religion more and therefore tend to be more religious. Poorer people also tend to have lower IQ (for many reasons such as bad nutrition, bad education, etc). So it’s not like the paper’s implication, namely that intelligence causally leads to non-belief, holds any water. (Also see: . And perhaps atheists should know that this paper’s main author, Richard Lynn, is a rather unsavory character who also believes in race and sex differences in intelligence.)

    Now this may be a good place to think about what “intelligence” means. One obvious thing is that there are many kinds of intelligence: So the intelligent scientist is good in discovering the hidden order in physical phenomena. The intelligent cook is one who makes tasty meals. The intelligent criminal is one who is successful at crimes. So intelligence is a means to something and it’s observable success which defines the respective intelligence. In the case of metaphysics though we do not know who’s been successful in forming true beliefs about how reality is, so the whole issue is mute. Certainly to use scientific prowess (as in “scientists tend to be non-religious”) doesn’t work at all. After all ontological naturalism is usually the result of reifying scientific models, and one would expect scientists to be especially inclined to do this. I am not implying that reifying scientific models is necessarily a fallacy; reality may be such that you get a good description of it if you reify scientific models (and believe, say, that mass bends spacetime around it just because gravitational phenomena are best described by the respective model)[1]. My point is that as long as one doesn’t know that the reification of scientific models is a good description of reality the fact that scientists tend to be naturalists is entirely irrelevant. (Another common argument is that the greater the scientist the more probably it is that they are not religious. On the other hand at least in modern times in order to be a great scientist you must be completely absorbed in science and spend little time thinking about other matters, so, again this evidence is mute. I predict that, similarly, the greater the pianist the more probable it is that they are not religious.)

    Perhaps there is a way to use statistics about intelligence as evidence for the truth of metaphysical beliefs. I submit that the most relevant kind of intelligence is the one that leads us to be good persons and have a good life. Let’s call that intelligence “intelligence of wellbeing”. There are several statistical studies which show that all other factors being the same (people living in the same country, having the same educational and socioeconomic level, etc) there is a positive correlation between personal wellbeing in this sense and being religious. So I submit that “intelligence of wellbeing” correlates positively with religion, which can be construed as evidence that religious worldviews come closer to describing reality.

    On the other hand, frankly, I find there are so many good arguments for religion when compared to naturalism, that I don’t see why a religious person should use arguments based on intelligence in the first place. I think it’s evidence of the epistemic poverty of the naturalistic worldview that naturalists are reduced to so often try to use arguments from intelligence or from scientific prowess.

    [1] That the stepwise refinement of scientific theories tends to require completely different models, as well as the fact that quantum mechanics resists naturalistic modeling are perhaps an indication that the epistemology of reifying scientific models is shaky, but never mind.