Disclaimer: this is not meant as a defence of the behaviour in question, since I don’t exactly know what we’re talking about.
For my own part, I observe a small but significant number of people who seem to believe that LessWrong ought to be a community where it’s acceptable to differentially characterize women negatively
LessWrong characterizes outgroups negatively all the time. I cautiously suggest the whole premise of LW characterizes most people negatively, and it’s easier to talk about any outgroup irrationality, in this case women statistically, than look at our own flaws. If we talked about what men are like on average, we might not have many flattering things to say either.
Should negative characterizations of people be avoided in general, irrespective of how accurately we think they describe the average of the groups in question?
If you see characterizations that are wrong, you should obviously confront them.
I agree that there are also other groups of people who are differentially negatively characterized; I restricted myself to discussions of women because the original question was about sexism.
I cautiously suggest you could say the whole premise of lw characterizes most people negatively,
I would cautiously agree. There’s a reason I used the word “differentially.”
Should negative characterizations of people be avoided in general, irrespective of how accurately we think they describe the average of the groups in question?
Personally, I’m very cautions about characterizing groups by their averages, as I find I’m not very good about avoiding the temptation to then characterize individuals in that group by the group’s average, which is particularly problematic since I can assign each individual to a vast number of groups and then end up characterizing that individual differently based on the group I select, even though I haven’t actually gathered any new evidence. I find it’s a failure mode my mind is prone to, so I watch out for it.
If your mind isn’t as prone to that failure mode as mine, your mileage will of course vary.
I’m not sure how not being differential is supposed to work though. Different groups have different kinds of failure modes.
Suppose it’s actually true in the world that all people are irrational, that blue-eyed people (BEPs) are irrational in a blue way, green-eyed-people (GEPs) are irrational in a green way, and green and blue irrationality can be clearly and meaningfully distinguished from one another.
Now consider two groups, G1 and G2. G1 often discusses both blue and green irrationality. G2 often discusses blue irrationality and rarely discuss green irrationality. The groups are otherwise indistinguishable.
How would you talk about the difference between G1 and G2? (Or would you talk about it at all?)
For my own part, I’m comfortable saying that G2 differentially negatively characterizes BEPs more than G1 does. That said, I acknowledge that one could certainly argue that in fact G1 differentially negatively characterizes BEPs just as much as G2 does, because it discusses blue and green irrationality differently, so if you have a better suggestion for how to talk about it I’m listening.
What if G1=BEP and G2=GEP and discussing outgroup irrationality is much easier than discussing ingroup irrationality? Now suppose G1 is significantly larger than G2, and perhaps even that discussing G1 is more relevant to G2 winning* and discussing G2 is more relevant to G1 winning. How is the situation going to look like for a member of G2 who’s visiting G1? How about if you mix the groups a bit? Is it wrong?
if you have a better suggestion for how to talk about it I’m listening.
You connotationally implied the behaviour you described to be wrong. Can you denotationally do that?
How is the situation going to look like for a member of G2 who’s visiting G1?
I expect a typical G2/GEP visiting a G1/BEP community in the scenario you describe, listening to the BEPs differentially characterizing GEPs as irrational in negative-value-laden ways, will feel excluded and unwelcome and quite possibly end up considering the BEP majority a threat to their ongoing wellbeing.
How about if you mix the groups a bit?
I assume you mean, what if G1 is mostly BEPs but has some GEPs as well? I expect most of G1′s GEP minority to react like the G2/GEP visitors above, though it depends on how self-selecting they are. I also expect them to develop a more accurate understanding of the real differences between BEPs and GEPs than they obtained from a simple visit. I also expect some of G1′s BEP majority to develop a similarly more-accurate understanding.
Is it wrong?
I would prefer a scenario that causes less exclusion and hostility than the above. How about you?
You connotationally implied the behaviour you described to be wrong. Can you denotationally do that?
I’m not sure.
As I said, I’m cautious about characterizing groups by their averages, because it leads me to characterize individuals differently based on the groups I tend to think of them as part of, rather than based on actual evidence, which often leads me to false conclusions.
I suspect this is true of most people, so I endorse others being cautious about it as well.
I would prefer a scenario that causes less exclusion and hostility than the above. How about you?
I definitely want less exclusion and hostility, but I’m not sure the above scenario causes them for all values like GEP and BEP, nor for all kinds of examples of their irrationality. Perhaps we’re assuming different values for the moving parts in the scenario, although we’re pretending to be objective.
Many articles here are based on real life examples and this makes them more interesting. This often means picking an outgroup and demonstrating how they’re irrational. To make things personal, I’d say health care has gotten it’s fair share, especially in the OB days. I never thought the problem was that my ingroup was disproportionally targeted, but I was more concerned about strawmen and the fact I couldn’t do much to correct them.
Would it have been better if I had not seen those articles? I don’t think so, since they contained important information about the authors’ biases. They also told me that perhaps characterizations of other groups here are relatively inaccurate too. Secret opinions cannot be intentionally changed. Had their opinions been muted, I would have received information only through inexplicable downvotes when talking about certain topics.
I’m not sure the above scenario causes them for all values like GEP and BEP
I’m not exactly sure what reference class you’re referring to, but I certainly agree that there exist groups in the above scenario for whom negligible amounts of exclusion and hostility are being created.
Perhaps we’re assuming different values for the moving parts in the scenario, although we’re pretending to be objective.
I don’t know what you intend for this sentence to mean.
Would it have been better if I had not seen those articles? I don’t think so, [..] Had their opinions been muted, I would have received information only through inexplicable downvotes when talking about certain topics.
I share your preferences among the choices you lay out here.
Specific ones? Not especially. But it’s hard to know how to respond when someone concludes that C1 is superior to C2 and I agree, but I have no idea what makes the set (C1, C2) interesting compared to (C3, C4, .., Cn).
I mean, I suppose I could have asked you why you chose those two options to discuss, but to be honest, this whole thread has started to feel like I’m trying to nail Jell-O to a tree, and I don’t feel like doing the additional work to do it effectively.
So I settled for agreeing with the claim, which I do in fact agree with.
I have no idea what makes the set (C1, C2) interesting
I find that difficult to believe.
I’m trying to nail Jell-O to a tree,
I suggest this is because all we had was Jell-O and nails in the first place, but of course there are also explanations (E1, E2, .., En) you might find more plausible :)
If your mind isn’t prone to that failure mode, your mileage will of course vary.
Perhaps any such characterizations should be explicitly hedged against this failure mode, instead of being tabooed. I also think people should confront ambiguous statements, instead of just assuming they’re malicious.
Disclaimer: this is not meant as a defence of the behaviour in question, since I don’t exactly know what we’re talking about.
LessWrong characterizes outgroups negatively all the time. I cautiously suggest the whole premise of LW characterizes most people negatively, and it’s easier to talk about any outgroup irrationality, in this case women statistically, than look at our own flaws. If we talked about what men are like on average, we might not have many flattering things to say either.
Should negative characterizations of people be avoided in general, irrespective of how accurately we think they describe the average of the groups in question?
If you see characterizations that are wrong, you should obviously confront them.
I agree that there are also other groups of people who are differentially negatively characterized; I restricted myself to discussions of women because the original question was about sexism.
I would cautiously agree. There’s a reason I used the word “differentially.”
Personally, I’m very cautions about characterizing groups by their averages, as I find I’m not very good about avoiding the temptation to then characterize individuals in that group by the group’s average, which is particularly problematic since I can assign each individual to a vast number of groups and then end up characterizing that individual differently based on the group I select, even though I haven’t actually gathered any new evidence. I find it’s a failure mode my mind is prone to, so I watch out for it.
If your mind isn’t as prone to that failure mode as mine, your mileage will of course vary.
I don’t understand how not being differential is supposed to work though. Different groups are irrational in different ways.
I think the failure mode you mention is common enough that we should be concerned about it. I’m just not sure about the right way to handle it.
Suppose it’s actually true in the world that all people are irrational, that blue-eyed people (BEPs) are irrational in a blue way, green-eyed-people (GEPs) are irrational in a green way, and green and blue irrationality can be clearly and meaningfully distinguished from one another.
Now consider two groups, G1 and G2. G1 often discusses both blue and green irrationality. G2 often discusses blue irrationality and rarely discuss green irrationality. The groups are otherwise indistinguishable.
How would you talk about the difference between G1 and G2? (Or would you talk about it at all?)
For my own part, I’m comfortable saying that G2 differentially negatively characterizes BEPs more than G1 does. That said, I acknowledge that one could certainly argue that in fact G1 differentially negatively characterizes BEPs just as much as G2 does, because it discusses blue and green irrationality differently, so if you have a better suggestion for how to talk about it I’m listening.
What if G1=BEP and G2=GEP and discussing outgroup irrationality is much easier than discussing ingroup irrationality? Now suppose G1 is significantly larger than G2, and perhaps even that discussing G1 is more relevant to G2 winning* and discussing G2 is more relevant to G1 winning. How is the situation going to look like for a member of G2 who’s visiting G1? How about if you mix the groups a bit? Is it wrong?
You connotationally implied the behaviour you described to be wrong. Can you denotationally do that?
*rationality is winning
I expect a typical G2/GEP visiting a G1/BEP community in the scenario you describe, listening to the BEPs differentially characterizing GEPs as irrational in negative-value-laden ways, will feel excluded and unwelcome and quite possibly end up considering the BEP majority a threat to their ongoing wellbeing.
I assume you mean, what if G1 is mostly BEPs but has some GEPs as well? I expect most of G1′s GEP minority to react like the G2/GEP visitors above, though it depends on how self-selecting they are. I also expect them to develop a more accurate understanding of the real differences between BEPs and GEPs than they obtained from a simple visit. I also expect some of G1′s BEP majority to develop a similarly more-accurate understanding.
I would prefer a scenario that causes less exclusion and hostility than the above.
How about you?
I’m not sure.
As I said, I’m cautious about characterizing groups by their averages, because it leads me to characterize individuals differently based on the groups I tend to think of them as part of, rather than based on actual evidence, which often leads me to false conclusions.
I suspect this is true of most people, so I endorse others being cautious about it as well.
I definitely want less exclusion and hostility, but I’m not sure the above scenario causes them for all values like GEP and BEP, nor for all kinds of examples of their irrationality. Perhaps we’re assuming different values for the moving parts in the scenario, although we’re pretending to be objective.
Many articles here are based on real life examples and this makes them more interesting. This often means picking an outgroup and demonstrating how they’re irrational. To make things personal, I’d say health care has gotten it’s fair share, especially in the OB days. I never thought the problem was that my ingroup was disproportionally targeted, but I was more concerned about strawmen and the fact I couldn’t do much to correct them.
Would it have been better if I had not seen those articles? I don’t think so, since they contained important information about the authors’ biases. They also told me that perhaps characterizations of other groups here are relatively inaccurate too. Secret opinions cannot be intentionally changed. Had their opinions been muted, I would have received information only through inexplicable downvotes when talking about certain topics.
I’m not exactly sure what reference class you’re referring to, but I certainly agree that there exist groups in the above scenario for whom negligible amounts of exclusion and hostility are being created.
I don’t know what you intend for this sentence to mean.
I share your preferences among the choices you lay out here.
You understood me correctly.
I meant it’s tempting to replace “eye colour” with something less neutral and “irrationality” with something more or less reliably insulting.
I bet you have other choices in mind.
Specific ones? Not especially. But it’s hard to know how to respond when someone concludes that C1 is superior to C2 and I agree, but I have no idea what makes the set (C1, C2) interesting compared to (C3, C4, .., Cn).
I mean, I suppose I could have asked you why you chose those two options to discuss, but to be honest, this whole thread has started to feel like I’m trying to nail Jell-O to a tree, and I don’t feel like doing the additional work to do it effectively.
So I settled for agreeing with the claim, which I do in fact agree with.
I find that difficult to believe.
I suggest this is because all we had was Jell-O and nails in the first place, but of course there are also explanations (E1, E2, .., En) you might find more plausible :)
Perhaps any such characterizations should be explicitly hedged against this failure mode, instead of being tabooed. I also think people should confront ambiguous statements, instead of just assuming they’re malicious.