This is another of those posts I thought up a while ago -- it arose from a couple of comments elsewhere and a lot of stuff I was thinking about last year -- and didn't get around to actually writing. It's about bellringers.
Actually it's not. It's about prejudices and stuff. But with bells on.
Thing is, humans are optimised for spotting patterns in things. We're very good at noticing minor correlations. As a consequence of this, we're also very good at spotting correlations that don't exist. An example is random play on most music players -- when multiple songs by the same artist come up in a row, most people will start complaining, humourously or otherwise, that random play isn't random enough. Itunes even apparently corrects for this in their smart shuffle function. In fact it's perfectly normal for clusters like that to arise in random data, and filtering them out makes the data less random.
Once we've noticed a (real or not) correlation, we're also rather better at noticing and assimilating data which supports out idea than that which disproves it. This can end up being a problem.
Another thing is that assuming things about people based on the limited data that you have about them is part of the social glue that holds everyday interaction together. It's frequently used in order not to offend -- one might be more wary of telling a recently-met woman that her arse is large than a recently-met man, for example. Our social glue is also peppered through with safety instructions: be wary of new things, they're frequently harmful; if you have a suspicion about something, best to go with your instincts; people who are most like you are more likely to be safe; don't interact with people who might be dangerous. Of course, if you never interact with someone, it's quite hard to find out they're not dangerous.
And yet another thing is that there is often a grain of truth in some prejudices, if (and this is a very big if indeed) you are talking in terms of probability distributions and not individual people. So, say, on average, men are slightly better at logic puzzles than women. This doesn't mean that any given man will be better at logic puzzles than any given woman. In fact (given that the correlation is pretty slight) there may be roughly a fifty-fifty chance that the woman is better. It certainly doesn't mean that only men should do logic puzzles, or that only men should be considered for cushy logic-puzzle-solving jobs. However the misapprehension that 'average of A slightly higher than average of B' is exactly equivalent to 'all A at least slightly better than all B' is widespread.
That's not the main point of this entry, though. The main point is that there are numerous ways that prejudices can take hold and spread even if there is no truth in them at all, because of the way the brain works.
So. Suppose we have a group and a property, and any correlation between the group and the property is essentially negligible. The only link between them is one person has once told you (in this case incorrectly) that they are linked. I thought long and hard about suitable examples, before realising that they were in my midst all along.
I refer, of course, to bellringers. They lurk, hidden, in the heart of every social gathering. Next time you're in a group of more than four people, ask yourself: which one is the secret bellringer?. I have tried this, you see, and after a small pause there's always a sheepish that'd be me, then from someone whom you previously wouldn't have expected of raving campanology.
Look, see for yourself:
Are you a bellringer?
The scurrilous rumour which I shall make up is this: All bellringers are filthy perverts. Following are just some of the ways I can think of off-hand that having a vague idea that someone thinks this can make you believe it, even if it's completely untrue. There's a fairly comprehensive (and interesting) wikipedia page on cognitive biases in general here.
The 'women drivers' effect
You are aware that there is a rumour going round that bellringers are filthy perverts. Occasionally in your daily life you see acts of filthy pervertism. Whenever you see an act of filthy pervertism committed by a bellringer, you think oh look, there's a bellringer being a filthy pervert. When you see one committed by a member of any other group, you think oh look, there's a person being a filthy pervert. Thus unconsciously you end up with the impression that you've seen way more bellringers being perverts than other groups. Example, of the campanologically-unconnected sort: I realised I was doing this when I saw a (male) driver hit a bollard in Cambridge. It took me a while to work out why 'there's a bloke driving badly' seemed a nigglingly odd thought, when (casting my mind back properly) it wasn't that uncommon an event.
The 'why smart people believe stupid things' effect
I'm wary of invoking this, since it's cleary misusable as the why smart people don't agree with me theory. But I think there is a point within it, if it is not taken too far. The basic gist is given here.
So, someone tells you that bellringers are filthy perverts. That's interesting, you think: why on earth could that be? Perhaps to be inducted into bellringing requires a very traditional church upbringing, leading to repression and subsequent rebellion in private life. It must be true! It's like the sort of debating where you're not told beforehand which side of the debate you're going to be speaking on. If you surmise that bellringers are very straight-edge, law-abiding folk then that would obviously also be because of their assumed traditional church upbringing.
But people rarely get as far as surmising the opposite, because the oh, I've discovered a connection! clever me! mental payoff has already been obtained by then.
The 'stop and search' effect
In order to find out if the scurrilous rumour is true, I go to some churches which have bells at bell-ringing practice time and ask around. Do you ring bells?, I enquire, and Are you a pervert? (which we will assume the good folk of the parish answer truthfully). At the end of the day I count up my results, and lo and behold: Nearly all the perverts are also bellringers!. Mission accomplished.
Of course, this was because nearly all the people I talked to were bellringers. But why quibble over such matters?
'Everyone else is saying it, so it must be right (and I'll say it)'
Firstly: you hear one or two people in your peer group expressing the idea that bellringing sessions, really just huge great sticky bucket-of-soapy frogs felchathons, yadda yadda. You assume from this that this must be what everyone in your peer group thinks, since the others haven't expressed an opinion either way (another well-known bias is that people consistently overestimate the number of people who agree with them). You further assume that some of these people who think this must have thought long and hard about it and come to the conclusion that it's right. This suggests to you that they must have a point, so you don't need to think too long and too hard yourself about whether it's true.
Secondly: once (nearly) everyone believes something it becomes very hard to shift the idea, because everyone knows it's right because everyone else knows it's right: a sort of 'castle built on blancmange' situation. An interesting side issue is that of societal taboos: in present-day England, for example, it's disgusting to have sex with a twelve-year-old. You'd be hard-pressed to find anyone who says otherwise. In Spain, twelve is (or was at least until recently: it may have gone up a year now) the age of consent. And of Juliet, famous lover and thirteen-year-old: younger than she are happy mothers made. Most of the bitter moral debates around at the moment centre on things which are legal and normal in some places, times or cultures, and considered utterly beyond the pale, locking-up-for-life or lynch-mob offences elsewhere.
It's odd that the offence which is engendered by things which are legal elsewhere should be often greater than that engendered by things which are illegal everywhere. But I digress.
The MMR effect
Someone says that pervertism and bellringingness are linked. Bellringers take umbrage: of course that's not the case! So they make a point of vigourously dispelling this rumour whenever possible, e.g. in the form of newspaper articles entitled Study shows bellringing not related to incidents of rogue frottage and the like (the analogy may be wearing a bit thin here: work with me, OK?). Result: in the public mind, a stronger association is created between the concepts 'bellringing' and 'pervert'. OK, so there's a 'not' in there. But the situation that bellringers have been campaigning for is that the concepts should become unlinked, and their campaigning is in fact producing a different state; one in which there will always be at least suspicion. It is also likely that in order to seem fair, their discussions of how bellringers are not perverts will be obliged to have someone putting the case for bellringers indeed being perverts.
'doing what you're told'
Or to put it another way, three misconceptions:
a) Members of a discriminated-against group cannot themselves hold prejuices (in general; cf. the black vs. asian riots in Birmingham).
b) Members of a discriminated-against group cannot hold prejudices against other members of their group (see: women who believe, whether openly or no, that generally women are worse than men at most things, but they themselves are an exception)
c) Members of a discriminated-against group cannot be prejudiced against themselves (it's a lot easier to fail when you've been told repeatedly that you're going to).
You can probably fit those darned campanologists in there somewhere.
The Us and Them effect
Neither you nor anyone you know closely is a bellringer. Nor is anyone you know closely a filthy pervert (though some of them may be 'kinky', 'different', 'non-conventional' or 'open-minded'). Extrapolating from the people you know to everyone who's a bit like you, you decide that nobody who shares various characteristics with you could possibly be a pervert. All the perverts must belong to other groups. Those shifty strange bellringing-types, probably. I wouldn't put it past them.
In summary: your brain doesn't work like you think it works. No, you are not an exception. I'm certainly not an exception; I'm guilty of several of the above. Be nice to your brain and the brains of other people and think about your assumptions from time to time.
 Some also come from the twisting of things that were true in a way many years ago. For example, there were at one time laws in various parts of Europe which effectively prohibited people who were Jewish from participating in most trades (see e.g. here). One of the very few professions which they were allowed to pursue was the lending of money, primarily because Catholics held it to be a sin so didn't want to do it themselves. You can see where this is leading.
 My son started bellringing at age 13, and shortly afterwards I discovered that he had begun to masturbate! I took him out of bellringing classes, but it was too late: he'd clearly been scarred for life. Soon he got into pornography, staying out late at night and swearing. Now he's gay!