A friend pointed to this massive collation of statistics on atheism across the world. I myself keep track of this literature and most of the values are pretty plausible, or I’ve seen them before (you can find The World Values Survey publications in any college library). This section caught my attention:
Justin Barret (2004) has argued that belief in God is a result of the “way our minds are structured” (p.viii) and “the way human minds operate” (p.30). He argues that belief in God is “greatly supported by intuitive mental tools”(p.17) and is “an inevitable consequence of the sorts of minds we are born with” (p.91). Belief in God is “natural,” resulting from the “natural workings of the human mind,” and atheism is thus unnatural (p.108). David Wilson (2002) suggests that religion is part of humanity’s naturally evolving adaptive strategy, and that religious belief represents “the healthy functioning of the biologically and culturally well-adapted human mind” (p.228). Michael Persinger (1987) has stressed the role of the hippocampus, the amygdala, temporal lobes, and hormonal processes, in explaining religious belief in God. Ashbrook and Albright (1997) focus on the neural underpinnings and workings of the brain in explaining belief in God. Newberg and D-Aquili (2001) argue that the religious impulse lies in an evolved “neurological process” (p.9), that the roots of belief in God are to be found in “the wiring of the human brain” (p.129), and that “as long as our brains are arranged the way they are,” belief in God will remain (p.172).
The data presented in this chapter delivers a heavy blow to this new explanation of theism. First of all, the sheer numbers must be contended with. With between 500,000,000 and 750,000,000 non-theists living on this planet today, any suggestion that belief in God is natural, inborn, or a result of how our brains are wired becomes manifestly untenable. Secondly, anyone who argues that theism has neural roots and is a direct result of the natural way human minds work must then explain the dramatically different rates of belief among similar countries. Consider Britain (31-44% atheist) compared to nearby Ireland (4-5% atheist), the Czech Republic (54-61% atheist) compared to nearby Poland (3-6% atheist), and South Korea (30-52% atheist) compared the Philippines (less than 1% atheist). It is simply unsustainable to argue that these glaring differences in rates of atheism among these nations is due to different biological, neurological or other such brain-related properties. Rather, the differences are better explained by taking into account historical, cultural, economic, political, and sociological factors (Norris and Inglehart, 2004; Grontenhuis and Scheepers, 2001; Verweij, Ester, and Natua, 1997; Zuckerman, 2003).
I object to this characterization of the idea that theism is ‘natural’ and atheism is ‘unnatural.’ A minor point is that Justin L. Barrett believes that the growth of atheism is predominantly due to social forces (I asked him). I don’t think the above quote would give that impression. But a bigger issue vague and imprecise way that the author above uses the term “innate.”
What does “innate” mean? Consider the number of fingers on your hand. You have 5, unless you are a non-human intelligence. Not having 5 fingers is abnormal, pathological even. The number of fingers, and how they develop, are specified purely by a particular genotypic configuration, and that configuration is fixed across our species (there isn’t any genetic variation). This is a genetic trait, there is no intrapopulational variance in the character aside from pathologies (e.g., mutations or developmental abnormalities). Now, consider your height. This is a quantitative trait, the population is distributed in a normal fashion, there is a mean value, and a deviation from the mean. Being short or tall is not a pathology, but just part of the range which we expect to see within the population.
So how does the normal distribution of height occur? Via the combined action of a multitude of genetic and non-genetic factors. If all the genes contribute a large “average effect” to the trait, then the individual has a high quantitative value, and if all of the genes contribute a small average effect, then the individual has a low quantitative value. The majority of humans though tend to have quantitative values in the middle because of the average effects…average out. In the United States nearly 90% of the population level variation in height is due to genes. In other words, 10% of the variation is due to environmental factors. In contrast, among Jamaicans the genetically controlled variance is about 75%, and among Nigerians about 60%. Why is this? Are Nigerians and Jamaicans genetically different from each other in regards to the alleles which result in height? And from Americans? No! Actually, the height distributions for black and white males in the United States are about the same (Jamaicans and African Americans derive from West African populations). The two populations differ in many ways, but not so much on height (there are small differences which can be non-trivial at the “tails,” but I’m talking overall here). So what’s going on? Simple: the way a quantitative trait unfolds in an individual is contingent upon their environmental context. I won’t make this complicated, it is simply the fact that even if genes have relevance for a trait, so does environment. In fact, if you change the environment uniformly across the population the proportion of variance due to genes can remain the same even as the distribution shifts considerably. In the United States nutrition is not a limiting factor in how tall you become, the environmental parameter is saturated so the only source of variance is genetic. In contrast, in Jamaica there is environmental variance, the poor eat less than the rich, and in Nigeria this is even more true. This is obviously different from the genetic trait above, which is invariant across these populations and exhibits no intrapopulational variance either.
Why is this relevant to what we are speaking about? A year ago there was a paper out about heritability of political beliefs, which came up with a value of 50% of the intrapopulation variance being due to genes. The important point here is that the variance is being explained, the distribution, not any absolute value. Swedish Americans in Minnesota are more politically conservative than Swedes in Sweden, but this is not due to any genetic difference, likely, Swedish American political attitudes are shaped by the American political environment. Likely the variance explained by genes is going to be about 50% of the total variance in both groups, but the distribution from “Left” to “Right” will be framed by the local environment. In relation to atheism, it then makes sense that religiosity, or belief in supernatural agents, can vary over time and space due to local contingencies even when there are innate cognitive biases. The psychobiological ones are not the only parameters here, so if you weight the values enough you can reshape the mental orientation appropriately. The author above pretty much implied that those those who favor that theism is innate believe it is a genetic trait, like 5 fingers on your hand, or the ability to learn language. So the data above clearly falsifies that hypothesis, but that’s not what the proponents of the innate school believe!
No, religious predisposition is conceived of as a tendency which emerges out of a multitude of mental skills and aptitudes.1 This synthesis results in a “byproduct” orientation, the tendency to see supernatural agents in the universe. This tendency is powerful, and combined with particular cultural inputs it leads to a universal bias toward religious orientation, but like a quantitative trait if the various innate parameters are shifted in a particular direction that bias may not exist. Some scholars have proposed that individuals with a weak theory of mind, or attenuated social intelligence, are far more likely to be irreligious or lack the basic “religious sense.” And, many quantitative traits have environmental components, when Communism was ascendent many more individuals were professed atheists in what is now Russia. Just because environmental variables are non-trivial does not mean that biological variables are irrelevant, we need to get beyond simplistic models.
Finally, let me offer that the author above really does try to “high ball” the numbers of “atheists and agnostics” a bit in my opinion (I’ve looked at some of the data, and the phrasing of the questions and the penetration of exactly what “God” means in someone’s mind is highly relevant to their response). And yet nonetheless they show that around 90% of humans believe in God. This is a threshold trait with different norms of reaction.
I know from long experience that it is almost impossible to make heritability a basal background assumption, so I’m going to illustrate what I mean….
The three curves represent three normal frequency distributions. That is, the peak is where most people are, and the distribution tails off toward the ends. The x-axis depicts religiosity, ranging from atheism to intense religiosness. Above, I have illustrate three “environments” which the population is placed within,
1) Totalitarian atheism
2) Neutrality in regards to religion
3) Totalitarian theism
The distributions show what I suspect are the normal population level reaction. When there is positive inducement for atheism, there are more atheists, and when there is positive inducement for religion, there are more religionists. When you shift the parameters in such a manner the “die is loaded.” But, the variation is remains, and the cognitive innate camp would assert that that underlying variation is due in part to genetic/biological factors. Twin studies and what not tend to show about a 50% heritability for “religious zeal,” which suggests that there are some psychologically hard-wired factors which influence susceptibility to religious orientation. Societies which are forced in either direction often enter in a requilibration mode when the social pressures are relaxed. The former Soviet Union is characterized by far more religiosity than during the Communist era, while Spain is far more secular than during the Franco period.
1 – Imagine that each ‘mental module’ or cognitive aptitude, is a random variable, or the equivalent of a genetic locus with a particular average effect.