NAS is a severe outlier, so look to something particular there. For example, to become a member of NAS you must be sponsored by an existing member. A “birds of a feather” effect surely applies.
In support of this, some survey I read indicated that the vast majority of atheist scientists were atheists before they chose science. … Those with a heart may choose other professions… Just pokin’ y’all! 
(and yes, I know I’m now open to comments about those with a brain!)
Do you think that the existing member considers the religious views of the one being sponsored, or even knows what they are?
Not like that, you’re right. But to sponsor someone, you’d want to know them, which means it’s more than just a “professional” relationship. And people who become friends tend to share similar views on many things. So I don’t think it’s a conscious exclusion or inclusion, but it’s just the way we humans work.
I have doubts that it works like that. I would think that a solely professional relationship would be sufficient for sponsorship. I doubt that many of the sponsors even know the religious views of those they sponsor. Unfortunately, I don’t think many NAS members are commenting here, so there’s no way to confirm these speculations.
OK. Not a “scientific” analysis by me, for sure! But there have to be reasons why NAS is such a severe outlier. It’s probably a combination of things, and I’m just expecting “birds of a feather” has some impact.
It’s not that severe an outlier. Scientists in general are less religious than the public, and other developed countries are less religious than the U.S. It’s no surprise that the most prominent U.S. scientists should be less religious than other U.S. scientists.
This portrayal of NAS membership is inaccurate. New members are elected by the existing membership, following nomination by an existing member. There is no such thing as “sponsorship” and there is no need or expectation of a relationship that is more than professional. You can read the basics on the NAS site.
Let me see, 50% of scientists are devout, NAS is claimed that 93% of them are not, and that’s NOT a severe outlier. Oh I see, that’s because it is “prominent” scientists. Hey, believe whatever you want.
And “nomination” vs “sponsorship”, the word they use on their site is not relevant to my point.
C’mon guys. It is impossible that my point about birds of a feather has zero merit. It’s a human thing. Happens everywhere. We can argue at what level it is true, but it would be absurd to argue that NAS has none of it.
Where do you get that statistic?
Nobody said it was. You’re being strangely defensive.
It’s certainly possible. You’re claiming a bias in favor of atheists for membership in the NAS. I claim that the religious views of NAS members, absent that poll, would probably not have been known to other members as a rule, including when they were candidates. I wouldn’t claim there’s none of; I would claim that it’s not significant.
Instead of all this unscientific speculation, a more rational way to proceed on this question would be to break down the membership of the NAS by different markers (geographic, education, ethnic, race, gender, institutions, etc.), and similarly for whichever survey detected 50% of scientists in general believe in God. It could be that there are more mundane reasons for this. Some example hypotheses:
It is possible that Christian scientists are more likely to take teaching-intensive jobs which will not lend them well to getting scientific accomplishments that get them into the NAS.
It is possible that Christian scientists tend to be educated in non-Ivy League (or top-tier) universities (due to geographical origin) and so have less chance of getting a faculty spot in a top-tier university which will increase their chances of getting into the NAS.
Did anyone say that? [Narrator: no.]
There are two pretty big problems with your claims. One is that you did mischaracterize the process by which NAS members are nominated and elected. Once we clear away the chaff about “sponsorship” and “more than a professional relationship,” we just have a standard process by which people are chosen for an honor in a supposed meritocratic fashion.
The second problem is a lot worse. You are suggesting that unbelievers are more likely to honor fellow unbelievers due to bias (implicit or explicit). By suggesting this, you are claiming that NAS members know the religious positions and affiliations of candidate scientists. And that means that you are claiming that Christian members of the NAS are biased toward fellow Christians. Birds of a feather and all.
It would be nice to know the survey. But he didn’t say “believe in God”; he said “devout”, which seems much stronger. Even if this survey is correct, scientists are still enriched in non-believers compared to the general population, which suggests that eminent scientists might be even more enriched for similar reasons.
We don’t know whether scientists are more likely to be non-believers because of their expertise in science compared to the general public. Thus if the NAS scientists are better in science than the average scientist we don’t know if that has anything to do with their belief in God or lack thereof.
I’m not sure if anyone has commented on this, but a form of selection bias that I should have included in the original post is that Christian parents (and churches and youth groups and Christian peer pressure) often direct kids away from science. And Christian colleges often have an abysmal science curriculum.
What culture would that be?
That’s one type of selection bias I had in mind. And scientists can’t really be blamed for this.
A good point. However, note that, while this would explain why people from fundamentalist, anti-science backgrounds tend not to go into science, and while it would help to explain why many scientists and leading scientists do not hold to certain narrow formulations of religious faith, it would not answer the question why so few scientists believe in God. After all, the survey questions ask only about belief in God generally, not about belief in six 24-hour days, a global Flood, etc. One would think that there would be plenty of non-fundamentalist religious believers in America, and therefore plenty of them in science as well.
No one would expect the top scientists of America to hold to an anti-scientific or anti-intellectual form of religion, any more than one would expect the top literary critics or philosophers to do so. But it’s still a question why the top philosophers, literary critics, scientists, etc., seem less inclined to believe in God (even a God with no fundamentalist associations) than the general population. And another question would be whether the lack of belief among the scientists is any greater, percentage-wise, than among philosophers, literary critics, etc. If the numbers are about the same, that would suggest that it is not the contents or methods of science per se that are the cause of the gap, but some more general cultural condition connected with modern education.
Since we’ve been talking about American scientists, I presumed everyone would understand the culture out of which America grew, i.e., Western European culture, from about the time of Constantine or a bit later onward. Of course if one wants to quibble, one can say that until about 1,000 or so there were pagans within European borders whose religion was not Christian, but even still, both pagans and Christians believed in divine beings, whereas what we are talking about today is the difference between an elite which does not believe in such beings and the majority of the population which does. And certainly by 1,000 AD all of central and Western Europe was Christianized, so adjust the number downward from 1500 to 900 years if you like. The point is the same. There is still something to be explained. Why should someone who studies the laws of nature be any more disinclined to believe in God than a mailman or a florist? We don’t see that in the Middle Ages, or in the 17th century when science began. What has changed? That’s what needs to be explained.
You said “a culture that has for 1500 years or so shown a rough religious harmony between the beliefs of its common folks and the beliefs of its intelligentsia”. If all you meant was that everyone believed in god(s), you could have said that directly and foregone the pretentious floweriness that implied something deeper.
But the answer is simple. What has changed is that we now have much more information about the history of our planet and the life on it than was available before. Information that contradicts a 6-literal-day-global-flood conclusion.
As a selection bias, if real, it would partially explain the effect. Taken to the extreme it would indicate that scientists are not unbelievers because their science education causes them to reject superstition, but rather because high ability believing students are directed to different careers.
After all, the very same effect explains why there is a shortage of women in the sciences.
In short, we produce far too many pastors and seminarians.