Field of Science

Why are atheists so disliked?

Bruce Hood has a post up about the atheist bus ad controversy in the US state of Iowa (OK, it was a couple of weeks ago, but I've been away...). What caught my eye was a comment by Konrad:

The thing that got me was the governor of the state saying that he found the ad disturbing. Clearly, people seem to treat religious adherence as symbolic of group identity so that they find the idea of atheists in their midst as threatening as that of enemy spies.

The hostile reaction to what was a pretty innocuous ad certainly is extraordinary. But is group identity - and the distrust of non-group members, really the cause of it?

Some intriguing hints come in a masters thesis by Will Gervais, a student at the University of British Columbia (I took it with me on vacation for some pool-side reading!). In it he describes a series of three experiments in anti-atheist prejudice among fellow students.

The first was an implicit association task. What this found is that religious people have a fairly deep-seated conception of atheists as unpleasant and untrustworthy - but it was the lack of trust that came through strongest.

The second explored the idea of trust further, by exploring how religiosity affects willingness to hire atheists.

It turns out that it depends on the kind of job. Religious people were quite prepared to hire atheists for jobs that don't require require particularly trustworthy people. But they weren't prepared to hire atheists for high trust jobs.

Religious people didn't show this bias when the jobs were split into those that do or do not require pleasant people, or when the jobs were split according to the required degree of intelligence.

These two experiments show that the primary driver for religious hostility to atheists is specifically a lack of trust, rather than a belief that they're more generally unpleasant. But it doesn't explain why they have this level of distrust.

It could be in-group favouritism. Trust is the classic victim of group divisions, and so if the religious see atheists as an alien group then you would expect them to be distrustful.

However, Gervais argues that this might not be the whole story, for several reasons. Firstly, there was no evidence that atheists distrusted the religious, which you would expect if this were a standard case of distrust between groups.

Also, it's not at all clear that 'atheists' are seen to be a group. Although the 'religious' are also highly diverse, by and large they all subscribe to some doctrine that defines them as group members (of one religion or another). Atheists, by definition, have no such common ground that make them an identifiable group.

What's more, open atheists are a tiny minority in North America. Normally, between-group hostility is proportional to the size of the group. The hostility towards atheists seems to be, quite literally, out of all proportion.

It might be that distrust of atheists is driven, or at least augmented, by fears that non-belief in a punishing god will lead atheists to behave dishonestly. That's certainly what a lot of evangelical Christians believe (and cognitive psychologists, for that matter).

But what about the third experiment? Here's where it gets rather interesting. In the third experiment, Gervais gave the subjects one of three passages to read and react to - one on food, an excerpt from The God Delusion in which Dawkins argues that belief is nonsensical, and a passage detailing the increasing numbers of atheists in the USA in recent decades. This last passage included the crucial fact that at least 20% of Americans aged 18-25 are atheists.

For the religious, reading that atheism was rather more common than they previously believed had a remarkable effect. It effectively abolished their distrust of atheists.

To me, this result strongly suggests that distrust of atheists is mostly due to fear of 'others'. It suggests that the main reason for the distrust is that the subjects had not realised that many of their fellow students were, in fact, atheists.

Once they learned that atheists were not a weird, alien group, but rather people just like them, they felt able to trust them. And I think this conclusion is supported by the experience of atheists in places like the UK, where overt atheism is much more prevalent and distrust of atheists is correspondingly lower.

There are two lessons here. First, it suggest that theories that religion evolved as a tool to enforce in-group trust may be wrong.

Second, it suggests that all those bus ads may well be serving a useful function, even if they're unlikely to convert anyone. If they normalise atheism, then they should also help to change the lot of atheists in the USA from social pariahs to trusted community members.


By the way, if you're interested in group cohesion, you might be interested in an earlier post on The Hand Grenade Experiment.

_____________________________________________________________________________________
Creative Commons LicenseThis work by Tom Rees is licensed under a Creative Commons Attribution-Share Alike 2.0 UK: England & Wales License.

11 comments:

  1. Interesting post but I do not see how the conclusion follows from the evidence presented.

    First of all, the issue isn't 'distrust between groups' but how religion creates a strong in-group/out-group distinction. The fact that atheists don't distrust theists isn't counterevidence for this - it is actually exactly what you'd expect since they lack the beliefs which would lead to them making that distinction.

    A similar point can be made about the atheists not being a group. Of course they are not - they are viewed as the out-group by the theists. The out-group is not necessarily a group, however. Indeed, it usually isn't. The out-group for a stone-age tribe member was everyone except those who were in his tribe. This included the guys across the creek, the bunch of hunters busily making their way across what would become Indonesia, as well as the lone guy sitting at an ice hole on river in Siberia - i.e. a collection that has nothing in common other than that they are not members of the original tribe member's in-group. Kind of like 'everything that is not a pencil' is not a natural kind.

    Finally, the results that distrust drops when the number of atheists in America is mentioned suggest a relatively rational overcoming of the basic instinct when faced with the evidence that religious adherence is a lousy way to define the in-group in the modern world. I bet the effect would disappear if the subjects were under cognitive load and, thus, could not process this information with their critical faculties.

    The in-group/out-group distinction is the ultimate explanation for the fear of 'others' you mention. The particular fear of atheists in the US is not just due to this, of course - cultural factors play a role in blowing it out of all proportion. One that comes to mind is the way that American society is fed a constant, carcinogenic diet of fear that superstimulates the mind to overproduce religious and superstitious behaviour. Another, is the way this culture constantly pushes religious imagery to the foreground. Both kinds of behaviour are known to have effects that will lead to the distrust of atheists. So, culture is not a helpless victim but a co-conspirator in this. I do agree with the final conclusion, however, that bus advertising is going to have the effect of getting people used to the idea that there are atheists around.

    I am wondering, how did you come across the masters' thesis the studies come from? Also, it is not clear from the post how much of the analysis is your own and how much it is due to Gervais.

    ReplyDelete
  2. These two experiments show that the primary driver for religious hostility to atheists is specifically a lack of trust, rather than a belief that they're more generally unpleasant. But it doesn't explain why they have this level of distrust.

    I've often wondered what this trust is that we're talking about. Trust in regards to what? That they will think like them, do like them, not cheat them...? What?

    Second, it suggests that all those bus ads may well be serving a useful function, even if they're unlikely to convert anyone. If they normalise atheism, then they should also help to change the lot of atheists in the USA from social pariahs to trusted community members.

    If this is true, then I would argue that almost no matter what is said, talking about atheism would have the same effect. The more talk, the more normal it becomes.

    Konrad said it is actually exactly what you'd expect since they lack the beliefs which would lead to them making that distinction.

    I don't think it is a religious belief that creates in-group/out-group distinctions. It is just one way that defines groups. Atheists could still very well have strong group feelings, though not necessarily with other atheists.

    ReplyDelete
  3. Trust in regards to what? That they will think like them, do like them, not cheat them...? What?

    See Locke's Letter Concerning Toleration:

    Lastly, those are not at all to be tolerated who deny the being of a God. Promises, covenants, and oaths, which are the bonds of human society, can have no hold upon an atheist. The taking away of God, though but even in thought, dissolves all; besides also, those that by their atheism undermine and destroy all religion, can have no pretence of religion whereupon to challenge the privilege of a toleration.

    ReplyDelete
  4. To John: say hello to the Melbourne HPS people when you see them next, especially Howard.

    To Bjorn: how about saying that trust is the willingness to initiate reciprocal altruism? As for atheism and in-group/out-group feelings, you are right, of course, to say that religion does not create these feelings but is just one thing around which they can crystalise. If what I said suggested otherwise, I apologise.

    ReplyDelete
  5. This post has been linked for the HOT5 Daily 9/3/2009, at The Unreligious Right

    ReplyDelete
  6. Atheists are not an "identifiable" group and, for the most part, have nothing in common except their lack of belief in a God. That is not enough to define a group and certainly not enough to predict their individual behavior. For the most part its a case of "better safe than sorry!"

    In today's world, identifying "atheists" as a group has as much validity is saying that "those who eat mashed potatoes are an identifiable group".

    ReplyDelete
  7. Enjoyed the analysis. But can you provide a reference for the statement that cognitive scientists believe that atheists are more likely to act dishonestly?

    ReplyDelete
  8. Konrad, I pretty much agree with everything you put there (the post is my interpretation of Gervais' arguments, except for the last two paragraphs). There are some counterarguments - some atheists do consider themselves to be a distinct group. And they regularly meet religious people, and so you might expect distrust to be lower.

    I think that the result in the last experiment is surprising, whichever way you look at it. I wouldn't expect that people would revise their opinion that non-belief is linked to untrustworthiness simply because they learn they know a few of them. After all, they probably know quite a few untrustworthy people! If you believed that untrustworthy people are closet atheists, then this wouldn't shake that belief.

    So I think it's more likely that it's a simple case of familiarity reducing fear. But of course it's disputable.

    What would have been interesting would be to have a passage explaining why atheists are trustworthy, and see if that had any effect (rather than the Dawkins one).

    ReplyDelete
  9. Bjørn, well the basic idea is that the religious won't cheat because they fear supernatural punishment. In Gervais' study, trust of atheists was assessed by word association - so what the participants understood it to mean is open to question.

    I'd agree that the more atheists, the less distrusted. Question is, why!

    ReplyDelete
  10. Anonymous, for the standard view that religion is associated with trustworthiness, try the review by Azim Shariff. I blogged it last year.

    ReplyDelete
  11. I'm a Christian who doesn't distrust atheists at all (except perhaps to teach French phenomenology :) ). Indeed it may surprise you to know that I myself was discriminated against by an atheist who eventually squeezed me out of my job.

    But I certainly don't make statements that infer that all or most atheists must act in such a manner.

    ReplyDelete

Markup Key:
- <b>bold</b> = bold
- <i>italic</i> = italic
- <a href="http://www.fieldofscience.com/">FoS</a> = FoS