« Seek Superstar Slavery | Main | Talk Is Not About Info »

April 08, 2009


Well, an alternative explanation: folks will cheat more when told that insiders cheat too...

Is the control more inline with the cheating level if insiders cheat or outsiders cheat...

Without knowing if insiders cheating makes people cheat more or outsiders cheating makes them cheat less, we can't be sure about which message will be more effective (everyone else are immoral chattered damaged people, or we are better than everyone else, one of which provokes outright hostility, and the other arrogance)

No wonder Nazi Germany was so moral and upstanding.....

So folks will cheat less when told that outsiders cheat more. When can this justify preaching that other nations, religions, races, genders, etc. are evil or immoral?

My knee-jerk answer: When said outsiders are fictional and don't actually exist.

'When can this justify preaching that other nations, religions, races, genders, etc. are evil or immoral?'

Worked for Ted Haggard. What? It didn't? Oh.

Some how I think that in groups and out groups being part of a larger group (students) with well established behavioral norms that are similar enough across schools to be comparable is very important.

I'm not saying it wouldn't have some effect, but have we observed anything to suggest it holds up across more distinct groups? Do mosques which rage against the violence of the Jews produce less violent Muslims than more moderate mosques? I would think the answer's no, but I suppose we'd have to call Scott Atran.

"Without knowing if insiders cheating makes people cheat more or outsiders cheating makes them cheat less"

My understanding of the research is that both effects occur, relative to a baseline of receiving no information about anybody else cheating.

"When can this justify preaching that other nations, religions, races, genders, etc. are evil or immoral?"

The vacuous answer is when such preaching (a) will actually be effective in promoting better behavior, (b) the benefits of better behavior outweigh the costs of demonizing outgroups; and (c) there aren't any better ways of improving behavior.

(a) When might we actually expect improvements in in-group behavior as a function of bad out-group behavior?

Here, it seems important that the out-group behavior should trigger thoughts about one's own self-identity, and that one has a relatively *immediate* opportunity to engage in bad behavior that would negatively affect that self-identity. Both of these seem to militate against broad, and vague demonizations of others, and suggest that we may be most likely to reap benefits by focusing on specific behaviors that others engage in ("I love you, but I don't love your behavior"?) at times when we are contemplating similar behaviors.

It also seems important that the bad behavior be both (i) something that we all recognize as bad, but (ii) that we are nonetheless be tempted to engage in. Do you (Robin) have any such behaviors in mind as potential applications?

We should also be aware that there may also be situations in which bad outgroup behavior allows people to more easily justify their own bad behavior. This could occur especially if we are in competition with that outgroup, and are in danger of losing unless we "cheat" too.

(b) How much damage will it do?

Generic, and false demonization of others seems highly problematic for reasons that others have noted. It strikes me that the damage done by preaching that others are evil or immoral could be substantially reduced if we instead chose to focus on specific behaviors that particular outgroups actually do engage in. But again, I'm struggling to think of applications (Treatment of women in some societies? But maybe even that's too controversial to gain traction.)

(c) Are there other, potentially less costly ways to improve behavior? Other research along these lines has found that priming people with other, positive moral cues also decreases cheating. Apparently getting people to think about the ten commandments has this effect (even for athiests). This seems likely to be a lower cost strategy when it's available and relevant.

FWIW Robin, I think that both your title and the way you've phrased this question may have been unnecessarily sensationalistic/emotive. ("Haters cheat less" is pretty inaccurate as a description of the research, and the sort of thing I would expect this site to ridicule if a newspaper did it.) You may successfully provoke people, but I'm not sure it's conducive to getting good answers to your question.

You do have a way with titles, Robin, ;)

FWIW - I think this only illustrates how dependent our moral behavior is on social context/reinforcement/how we think we will be judged by others who are important to us. If our significant in-group respects cunning and the ability to deceive, then guess what we'll soon find ourselves doing?

How the outgroup is perceived is probably not as significant of a contributing factor as how the ingroup is perceived. (I suspect.)

conchis point A: "Do you (Robin) have any such behaviors in mind as potential applications?"

This community identifies its enemy as using inappropriate mental shortcuts: the irrational, the comforting fiction, the recursive buck. "They," the soto, are trying to cognitively cheat. To avoid being like Them, "We," the uchi, must be constantly aware of our biases and tendency to use cognitive cheats, with an ironic sense that this cognitive cheating does not really help one win.

Or you could see this as failing: because cognitive cheating is something They do, We must be the ones who are successfully thinking past our biases.

Doug S.,

my knee-jerk reaction was also "when the enemy is fictional", but I wondered if this might not be sufficient: Preaching even non-existent enemies as being immoral may cause intolerance towards unrelated attributes which were given to this fictional enemy (perhaps to make the lies more convincing).

E.g. "Blubworbs are evil. They steal, murder, rape and have blue eyes." -> "Let's hate people with blue eyes."

The comments to this entry are closed.

Less Wrong (sister site)

May 2009

Sun Mon Tue Wed Thu Fri Sat
          1 2
3 4 5 6 7 8 9
10 11 12 13 14 15 16
17 18 19 20 21 22 23
24 25 26 27 28 29 30