I think we have good reason to believe the article is broadly right, even if some of the specific anecdotes don’t do a good job of proving this.
If someone invests a lot of effort into searching for good evidence and comes up empty that’s a signal for the availability of good evidence.
But it’s just hard to present evidence that conclusively proves
That leaves the question of why it’s hard. In plenty of communities, it’s easy to find a lot of women who were sexually touched without their consent.
The fact that the article suggests that this is very hard to find that in the EA community, suggests that something goes right.
“#MeToo urged society to ‘believe women;’ EAs tend to be a bit more skeptical.” (also seems true)
This seems to make it sound like society on average updated due to #MeToo to ‘believe women;’ and EAs didn’t. In reality, most of society didn’t update. I would expect that on average EA might even lean more toward ‘believe women;’ than the average person.
partly because there are many polyamorous people in the EA/rationalist communities; this creates an environment in which sexual misconduct may be addressed suboptimally.” This strikes me as totally plausible.
One thing that distinguishes polyamorous people is that they are more willing to talk about sexual conduct of other people than most other groups. That results in some people feeling bad because they hear about sexual conduct, but it also helps to police bad conduct.
Thanks for your comment, which has been helpful in clarifying my own thinking. Particularly this:
If someone invests a lot of effort into searching for good evidence and comes up empty that’s a signal for the availability of good evidence.
I take the article’s thesis to be:
(1) The culture of EA is characterized by a skewed gender ratio, gendered power imbalances, mixing of professional/personal relationships, etc; (2) this (Increases the risk of? Leads to more of? Undermines reporting of?) sexual misconduct
I think the article does a pretty good job of proving (1), which is what I meant when I said the article is “broadly right.” Perhaps the crux of our disagreement is that it’s not exactly clear what claim (2) is.
I think it’s pretty intuitive that (1) would increase the risk of a weak version of (2): i.e., the cultural dynamics in EA lead to women having encounters of a sexual nature that they don’t want to be having (e.g., getting hit on at professional events, feeling like they’ll suffer professional repercussions for rejecting people, etc). I also think this kind of claim is hard to prove—it’s just difficult to establish causation between one amorphous cultural phenomenon and another. Given that, I think the article does a good job of showing that women EAs in the Bay Area were repeatedly made uncomfortable by men’s behavior towards them. Without more context, it’s hard to know how much to condemn the men here (e.g., were people making innocent mistakes, or was this more malicious?), but clearly a lot of women felt bad about the encounters they had. I don’t think we have reason to doubt this, and I think this is something the EA community should reflect on and investigate further.
There’s also a separate, stronger claim that the article seems to be gesturing at, which is maybe closer to: “rates of sexual harassment and assault are higher in the EA community than elsewhere, at least in the Bay Area.” And I agree with you that the article hasn’t really provided much evidence for this (beyond that one extremely troubling anecdote), although I think we have some independent reasons to worry about this. I’ll be curious to see what conclusions the Community Health Team reaches on this point.
“Given that, I think the article does a good job of showing that women EAs in the Bay Area were repeatedly made uncomfortable by men’s behavior towards them.”
Even this isn’t really established in an interesting way. If out of thousands of women in a group dozens feel this way, it is probably actually a really safe place, while if half of them feel that way there is an issue that probably should be addressed.
And I seriously don’t know which is going on (and sampling women who stay in the EA community about how they feel creates a survivorship bias, because they are the ones who aren’t offended enough to leave).
But the TIME article is weak evidence in either direction, because the reporter is simply not trying to establish base rates.
Given you aren’t commenting on the truth value of the claims, just thought I’d nitpick that I don’t think establishing base rates is required for updating (or perhaps more accurately, some stories are sufficiently bad that even 1 credible case in the EA movement may be seen as bad enough). For example, what do you think is the acceptable base rate of this allegation? Can you think of stories that “if true”, would be sufficiently bad that it would make you think EA wasn’t a safe space for women, even without interviewing thousands of women? How do you feel about reading a report about sexual abuse of young children by Catholic priests which were not taken seriously by the Catholic Church—do you also suggest that this is “weak evidence in either direction, because the reporter is simply not trying to establish base rates?”
So first, the story in time is stripped of context—it is impossible to judge from that story being told by a journalist how serious it is, because I know the journalist stripped out any information that would make it seem less bad.
Second, that one individual in a group of thousands did something is never, ever, ever enough to judge the whole group—though the official reaction to that individual of course might be. I think that you are referring to Owen not being banned from all community positions immediately after this was reported to the community team by the anonymous woman. If you judge that as a reason to say the whole movement yas a problem, and I see it as totally reasonable, we disagree.
But yes: That some priests sexually abuse children is irrelevant. There are tens of thousands. That a reporter claims that covering this up is a systematic issue is something that I will only believe if the reporter does the work to prove it. My extremely uninformed impression was that this had been proven in that case, but you correctly have pointed out to me that I should lower my confidence in the view that the catholic church pre 2010 or so had a real problem with sexual abuse.
Second, that one individual in a group of thousands did something is never, ever, ever enough to judge the whole group—though the official reaction to that individual of course might be.
Never, ever, ever? Are you saying you’d put a 0% chance that applying any kind of judgement on the group based on the actions of one individual is appropriate? This just seems clearly misguided.
In any case, I never made the claim that “Owen not being banned from all community positions immediately after this was reported to the community team by the anonymous woman” means that the EA movement as a whole has a problem, that’s a strawman. I just expressed skepticism around your view that base rates are a necessary condition for you to consider something good evidence, by asking you whether there was anything at all that was qualitatively bad enough such that even 1 case would make you think it would be reasonable for a woman to conclude that EA in the Bay area, or even EA more broadly, isn’t a safe space for them. but sounds like you think the answer is a firm and absolute no, which is pretty surprising to me.
I think we might be attaching different concepts to the same words here.
When you say that one incident could indicate there is a problem, are you including the way the surrounding community reacts in what you mean by the ‘incident’, or does the ‘incident’ only include the fact that one individual who was part of the community acted that way?
The pressing question for EA is not whether or not EA has a base rate of harassment that is worse than average. The question is “is there reasonable room for improvement?”. I think the answer is yes, and I think the time article provides helpful evidence of that. It’s not impossible that the answer is no and EA is already perfect on this issue, but from I’ve read it seems unlikely.
If someone invests a lot of effort into searching for good evidence and comes up empty that’s a signal for the availability of good evidence.
That leaves the question of why it’s hard. In plenty of communities, it’s easy to find a lot of women who were sexually touched without their consent.
The fact that the article suggests that this is very hard to find that in the EA community, suggests that something goes right.
This seems to make it sound like society on average updated due to #MeToo to ‘believe women;’ and EAs didn’t. In reality, most of society didn’t update. I would expect that on average EA might even lean more toward ‘believe women;’ than the average person.
One thing that distinguishes polyamorous people is that they are more willing to talk about sexual conduct of other people than most other groups. That results in some people feeling bad because they hear about sexual conduct, but it also helps to police bad conduct.
Thanks for your comment, which has been helpful in clarifying my own thinking. Particularly this:
I take the article’s thesis to be:
(1) The culture of EA is characterized by a skewed gender ratio, gendered power imbalances, mixing of professional/personal relationships, etc; (2) this (Increases the risk of? Leads to more of? Undermines reporting of?) sexual misconduct
I think the article does a pretty good job of proving (1), which is what I meant when I said the article is “broadly right.” Perhaps the crux of our disagreement is that it’s not exactly clear what claim (2) is.
I think it’s pretty intuitive that (1) would increase the risk of a weak version of (2): i.e., the cultural dynamics in EA lead to women having encounters of a sexual nature that they don’t want to be having (e.g., getting hit on at professional events, feeling like they’ll suffer professional repercussions for rejecting people, etc). I also think this kind of claim is hard to prove—it’s just difficult to establish causation between one amorphous cultural phenomenon and another. Given that, I think the article does a good job of showing that women EAs in the Bay Area were repeatedly made uncomfortable by men’s behavior towards them. Without more context, it’s hard to know how much to condemn the men here (e.g., were people making innocent mistakes, or was this more malicious?), but clearly a lot of women felt bad about the encounters they had. I don’t think we have reason to doubt this, and I think this is something the EA community should reflect on and investigate further.
There’s also a separate, stronger claim that the article seems to be gesturing at, which is maybe closer to: “rates of sexual harassment and assault are higher in the EA community than elsewhere, at least in the Bay Area.” And I agree with you that the article hasn’t really provided much evidence for this (beyond that one extremely troubling anecdote), although I think we have some independent reasons to worry about this. I’ll be curious to see what conclusions the Community Health Team reaches on this point.
“Given that, I think the article does a good job of showing that women EAs in the Bay Area were repeatedly made uncomfortable by men’s behavior towards them.”
Even this isn’t really established in an interesting way. If out of thousands of women in a group dozens feel this way, it is probably actually a really safe place, while if half of them feel that way there is an issue that probably should be addressed.
And I seriously don’t know which is going on (and sampling women who stay in the EA community about how they feel creates a survivorship bias, because they are the ones who aren’t offended enough to leave).
But the TIME article is weak evidence in either direction, because the reporter is simply not trying to establish base rates.
Given you aren’t commenting on the truth value of the claims, just thought I’d nitpick that I don’t think establishing base rates is required for updating (or perhaps more accurately, some stories are sufficiently bad that even 1 credible case in the EA movement may be seen as bad enough). For example, what do you think is the acceptable base rate of this allegation? Can you think of stories that “if true”, would be sufficiently bad that it would make you think EA wasn’t a safe space for women, even without interviewing thousands of women? How do you feel about reading a report about sexual abuse of young children by Catholic priests which were not taken seriously by the Catholic Church—do you also suggest that this is “weak evidence in either direction, because the reporter is simply not trying to establish base rates?”
So first, the story in time is stripped of context—it is impossible to judge from that story being told by a journalist how serious it is, because I know the journalist stripped out any information that would make it seem less bad.
Second, that one individual in a group of thousands did something is never, ever, ever enough to judge the whole group—though the official reaction to that individual of course might be. I think that you are referring to Owen not being banned from all community positions immediately after this was reported to the community team by the anonymous woman. If you judge that as a reason to say the whole movement yas a problem, and I see it as totally reasonable, we disagree.
But yes: That some priests sexually abuse children is irrelevant. There are tens of thousands. That a reporter claims that covering this up is a systematic issue is something that I will only believe if the reporter does the work to prove it. My extremely uninformed impression was that this had been proven in that case, but you correctly have pointed out to me that I should lower my confidence in the view that the catholic church pre 2010 or so had a real problem with sexual abuse.
Thank you
Never, ever, ever? Are you saying you’d put a 0% chance that applying any kind of judgement on the group based on the actions of one individual is appropriate? This just seems clearly misguided.
In any case, I never made the claim that “Owen not being banned from all community positions immediately after this was reported to the community team by the anonymous woman” means that the EA movement as a whole has a problem, that’s a strawman. I just expressed skepticism around your view that base rates are a necessary condition for you to consider something good evidence, by asking you whether there was anything at all that was qualitatively bad enough such that even 1 case would make you think it would be reasonable for a woman to conclude that EA in the Bay area, or even EA more broadly, isn’t a safe space for them. but sounds like you think the answer is a firm and absolute no, which is pretty surprising to me.
I think we might be attaching different concepts to the same words here.
When you say that one incident could indicate there is a problem, are you including the way the surrounding community reacts in what you mean by the ‘incident’, or does the ‘incident’ only include the fact that one individual who was part of the community acted that way?
The pressing question for EA is not whether or not EA has a base rate of harassment that is worse than average. The question is “is there reasonable room for improvement?”. I think the answer is yes, and I think the time article provides helpful evidence of that. It’s not impossible that the answer is no and EA is already perfect on this issue, but from I’ve read it seems unlikely.