Socially Influenced Beliefs

The discussion of the atheistic tendencies of professors leads me to posit the following.

1.  Suppose that we say that beliefs are primarily influenced by social considerations.  You believe X because you want to earn the friendship/respect of people around you.  So, if you are around God-fearers, your social instinct is to believe in God.  If you are around atheists, your social instinct is to be atheist.

2.  Suppose that you are inclined to believe x.  If your reference group does not believe x, then you will pay attention to evidence against x and reconsider your position.  However, if the group also believes x, you will want to search for evidence in favor of x and to be skeptical about evidence against x.  That is, we try pretty hard to align our thinking to conform to that of our reference group.

3.  Even our belief in mathematical and scientific propositions has a social component to it.   

4.  Academic intellectuals learn something of how to question beliefs in a rational way.  This makes them a bit less inclined to fall for popular superstitions.

5.  However, even academic intellectuals are leery of questioning beliefs within their own reference group.  So it is possible for a group of academics to get stuck in an equilibrium in which they believe a dubious proposition.  One hopes that eventually someone comes along and questions the conventional wisdom in such a way as to disturb that equilibrium.

6.  The atheism of academics looks like an equilibrium.  I think it is a sound one.  However, other equilibrium beliefs among academics strike me as more problematic.  That is, a huge majority of academics may hold some political views, and I do not share those views.

GD Star Rating
Tagged as:
Trackback URL:
  • It is striking how people seem to think that the possibility of social pressure on beliefs justifies their disagreement with some group. The possibility of social pressure undermines everyone’s beliefs. It might make deviations from average beliefs less bad than otherwise, but it still doesn’t make them good. It’s another variation on the argument “No one knows anything about this topic, therefore I’m right.”

  • I’d disagree with 4. I don’t think it’s true, mainly because I deal with academics as part of Chemistry, Physics, and Material Science over the years. I’ve found them to be able to do this to some degree in their discipline, but not in general. I think Hogan was right in his Sacred Cows, that people do not generalize this ability.

    As to the atheism, I’d disagree there too as often I find it is not a result of any real consideration (unlike some agnostic chemists I have known) but a dual response to both class issues (those danged rednecks townies believe in God so I won’t) and also a social issue (my colleagues would mock me if I admitted belief and call me a fundamentalist. Unless I was Moslem…then it would be un-PC to do so). I’d not discount how strong that first choice was. My limited experince is that academics have an elitism as strong as any.

  • I don’t doubt there is a certain degree of socialization going on, but I still think selection explains most of it, as per my blog: If I’m recalling correctly, this is supported by Wuthnow’s article “Science and the sacred” found in “The sacred in a secular age” edited by Phillip Hammond.

  • Stuart Armstrong

    My personal conversion to atheism came as a result of socialisation with a particular group of Christians who encouraged me to examine the evidence – which I then did, and my personal conclusion was that they were wrong.

    However the social atmosphere of academia in general has definetly allowed me to maintain my atheism, without constantly requestionning my core beliefs.

    Since the socialisation is going on, we should give more weight to the opinions of those who differ strongly with their own social group – and the higher the penalty for disagreement, the more weight we should give to their opinions.

    The penalty for religiousity in academia is not very high (some embarassement, some loneliness, and the consequences of these – correct me if people have encountred much worse). The penalty for atheism in some religious communities can be very high indeed.

    So on this subject we should initially give greater weight to the arguments of religious academics that to atheist ones. But the arguments of, say, an atheist ex-moromon, voluntarily removed from his community, should be given greater weight than either. Similar thoughts for various converts in extreme situations (christian converts in afganistan, white musilms in some US states, etc…)

  • Ned

    I see two problems with arguments in the main article:

    1) if everyone’s beliefs are determined by group socializing, what then determines group’s bliefs? I.e. there may be some other factor, such as IQ, that determines individual beliefs, and further:

    2) could it actually be reverse, that we select groups we want to belong to, partly, according to our agreement with group members’ beliefs?

    It seems to me, on the topic of atheism, (I’m an agnostic) that since high-IQ people are more likely to be atheists, their beliefs are probably evidence based.

  • Stuart Armstrong

    2) could it actually be reverse, that we select groups we want to belong to, partly, according to our agreement with group members’ beliefs?

    Unfortunately that would be even more worrying, implying that people who would disagree are premptively turned away from academia!

    And can IQ be taken as a guide for establishing religious truth? What biases does having a high IQ bring to people? A few spring to mind:
    1) That their own opinions are better than most
    2) That they can solve most problems they turn their minds to

    1) doesn’t seem to be too much of a problem here, but 2) is closely related to “2′) Every problem has a solution”, which is problematic. If you believe 2′) and 2), then since there are many contradictory religious positions, most with comparable amounts of evidence, you are practically forced to atheism, or, if your are more modest, to agnosticism.

    In Bayesian terms, (2′) and 2)) would be a bias that would give a higher weight to evidence that would make the postiori distribution have a sharp, high peak. Many academics I know suffer from this bias.

    The best argument for respecting the positions of academics is their use of the scientific method. If they use it, even approximately, in this issue, then we should respect their opinions above the average.

  • Carl Shulman

    “The possibility of social pressure undermines everyone’s beliefs.”

    People with relatively low Agreeableness, or whose track records show relatively little difference in prediction accuracy (or prediction market returns) between socially salient and non-salient questions, could justifiably increase confidence on balance by asymetrically discounting the views of those without said markers. Of course such a rule will be widely misapplied due to standard self-serving and confirmation biases, but the appropriate level of imperfect prophylactic heuristics will generally have to be adjusted for the level of bias being addressed.

  • Carl, I don’t see why you think your strategy works.

  • Regarding getting stuck in unwise equilibria for social reasons among academics, if the equilibrium involves deeply held views at the “paradigmatic” level as analyzed by Kuhn, then I think we are dealing with what Paul Samuelson took from Max Planck, the old line about scientific progress being made “funeral by funeral.”

  • I’ve always been quite amused by the idea that scientific progress occurs when a previous generation dies out and a new generation of students grows up familiar with the ideas. If this were true, the entire power of science would rest on the ability of grad students to distinguish truth from falsehood.

  • Carl Shulman

    “Carl, I don’t see why you think your strategy works.”
    We should aggregate across analytical processes, not individuals as such. If an advocate of view X has a million transporter duplicates concur with her, we should count that as one viewpoint for aggregation. If we are fortunate enough to have objective measures of susceptibility to social influence, as I postulated, we should discount the effective number of advocates for a particular conclusion accordingly. A dispute that seems to divide opinion in a ratio of 50:50 may reveal a clear majority of independent viewpoints after discounting if the sides differ on the indicia of social malleability.

  • “It seems to me, on the topic of atheism, (I’m an agnostic) that since high-IQ people are more likely to be atheists, their beliefs are probably evidence based.”

    Ned, where’s the evidence that IQ and atheism are correlated?

  • Carl Shulman