The “chinese robber fallacy” is being overstretched, in my opinion. All it says is that having many examples of X behaviour within a group doesn’t necessarily prove that X is worse than average within that group. But that doesn’t mean it isn’t worse than average. I could easily imagine the catholic church throwing this type of link out in response to the first bombshell articles about abuse.
Most importantly, we shouldn’t be aiming for average, we should be aiming for excellence. And I think the poor response to a lot of the incidents described is pretty strong evidence that excellence is not being achieved on this matter.
If having many examples of behavior X within a group doesn’t show that the group is worse at that or better at that than average—if you expect to see the same thing in either case—then being presented with such a list has given you zero evidence on which to update.
They would have written the same article whether behavior X was half as common or twice as common or vanishingly rare. They would have written the same article whether things were handled well or poorly, as shown by their framing things misleadingly and their lies of omission. They had an ax to grind and they’ve ground it. We should be aiming for excellence but when we get there (or if we’ve gotten there) it will do absolutely nothing to prevent people from writing these articles.
When someone goes looking for examples of X behavior, knowing that they’ll find a list, with the goal of damaging your reputation among third parties, being presented with the list does not seem to me a good impetus for paroxysms of soul-searching and finger pointing.
In the absence of evidence that rationalism is uniquely good at dealing with sexual harrasment (it isn’t), then the prior assumption about the level of misconduct should be “average”, not “excellent”. Which means that there is room for improvement.
Even if these stories do not update your beliefs about the level of misconduct in the communities, they do give you information about how misconduct is happening, and point to areas that can be improved. I must admit I am baffled as to why the immediate response seems to be mostly about attacking the media, instead of trying to use this new information to figure out how to protect your community.
I wonder if you both might just believe the same thing here? titotal, do you not think it possible that lumpyproletariat was offering that as one option out of many as a sort of insurance plan that witchhunts against EA not begin? Witchhunts are very easy to start, especially once you know more details and you know you need them, but they are very hard to stop. So I guess I think they agree with you but just want to drop in that reminder for people of that possibility to try to help things go well? Rather than making a claim that base rates is necessarily the case? After all, you both want the community to do better and be “aiming for excellence”?
[I agree it would have been better framed as one reason out of various though. I’ve been liking the allegory of the blind men and the elephant more for this myself]
baffled as to why the immediate response seems to be mostly about attacking the media, instead of trying to use this new information to figure out how to protect your community.
I don’t really want to get involved in this thread other than saying “I think you guys agree” so it’s okay if you consider it a tangent but… I’ll just flag that I think this bit isn’t accurate in character. What if the lessons were learned back then and the “immediate response” has actually passed? What about this response from the Community Health Team about an ongoing project to help clarify problems and reveal avenues for making the community safer, which probably implies that the rest of us don’t have much to do quite yet except maybe help people be patient for that? Another option is, if we want to go our own way, actually to try to figure out the veracity of the media and other sources of info, because questioning the importance of various pieces of info would also lilely be the first step in helping determine which potential interventions might do nothing or do amazingly or do net harm. I wouldnt recommend getting stuck on using such selective data, when there should be better to some soon, but you are certainly welcome to try to use this information to protect the community, and let us know if you think of something for us to do!
I think it’s good to both address sexual misconduct and to correct misleading context in media pieces. But if you only mention the latter, it gives the impression that the former doesn’t matter. I would highly encourage people who care about both to at least mention that you care about reducing the level of misconduct. It may sound like stating the obvious, but it really does matter.
While I certainly hope everyone cares about both, I can’t honestly say I believe that. Going through the lesswrong thread, it honestly looks to me like a lot of people genuinely don’t want to think about the issue at all, and I find this concerning. For example, downvoting the thread to 0 seems completely unwarranted.
None of this was news to the people who use LessWrong.
The time to have a conversation about what went wrong and what a community can do better, is immediately after you learned that the thing happened. If you search for the names of the people involved, you’ll see that LessWrong did that at length.
The worst possible time to bring the topic up again, is when someone writes a misleading article for the express purpose of hurting you, which was not written to be helpful and purposefully lacks the context that it would require in order to be helpful. Why would you give someone a button they can press to make your forum talk for weeks about nothing?
It was a low-quality article and was downvoted so fewer people saw it. I wish the same had happened here.
Someone on the LessWrong crosspost linked this relevant thing: https://slatestarcodex.com/2015/09/16/cardiologists-and-chinese-robbers/
The “chinese robber fallacy” is being overstretched, in my opinion. All it says is that having many examples of X behaviour within a group doesn’t necessarily prove that X is worse than average within that group. But that doesn’t mean it isn’t worse than average. I could easily imagine the catholic church throwing this type of link out in response to the first bombshell articles about abuse.
Most importantly, we shouldn’t be aiming for average, we should be aiming for excellence. And I think the poor response to a lot of the incidents described is pretty strong evidence that excellence is not being achieved on this matter.
I agree that we should be aiming for excellence.
If having many examples of behavior X within a group doesn’t show that the group is worse at that or better at that than average—if you expect to see the same thing in either case—then being presented with such a list has given you zero evidence on which to update.
They would have written the same article whether behavior X was half as common or twice as common or vanishingly rare. They would have written the same article whether things were handled well or poorly, as shown by their framing things misleadingly and their lies of omission. They had an ax to grind and they’ve ground it. We should be aiming for excellence but when we get there (or if we’ve gotten there) it will do absolutely nothing to prevent people from writing these articles.
When someone goes looking for examples of X behavior, knowing that they’ll find a list, with the goal of damaging your reputation among third parties, being presented with the list does not seem to me a good impetus for paroxysms of soul-searching and finger pointing.
In the absence of evidence that rationalism is uniquely good at dealing with sexual harrasment (it isn’t), then the prior assumption about the level of misconduct should be “average”, not “excellent”. Which means that there is room for improvement.
Even if these stories do not update your beliefs about the level of misconduct in the communities, they do give you information about how misconduct is happening, and point to areas that can be improved. I must admit I am baffled as to why the immediate response seems to be mostly about attacking the media, instead of trying to use this new information to figure out how to protect your community.
I wonder if you both might just believe the same thing here? titotal, do you not think it possible that lumpyproletariat was offering that as one option out of many as a sort of insurance plan that witchhunts against EA not begin? Witchhunts are very easy to start, especially once you know more details and you know you need them, but they are very hard to stop. So I guess I think they agree with you but just want to drop in that reminder for people of that possibility to try to help things go well? Rather than making a claim that base rates is necessarily the case? After all, you both want the community to do better and be “aiming for excellence”?
[I agree it would have been better framed as one reason out of various though. I’ve been liking the allegory of the blind men and the elephant more for this myself]
I don’t really want to get involved in this thread other than saying “I think you guys agree” so it’s okay if you consider it a tangent but… I’ll just flag that I think this bit isn’t accurate in character. What if the lessons were learned back then and the “immediate response” has actually passed? What about this response from the Community Health Team about an ongoing project to help clarify problems and reveal avenues for making the community safer, which probably implies that the rest of us don’t have much to do quite yet except maybe help people be patient for that? Another option is, if we want to go our own way, actually to try to figure out the veracity of the media and other sources of info, because questioning the importance of various pieces of info would also lilely be the first step in helping determine which potential interventions might do nothing or do amazingly or do net harm. I wouldnt recommend getting stuck on using such selective data, when there should be better to some soon, but you are certainly welcome to try to use this information to protect the community, and let us know if you think of something for us to do!
I think it’s good to both address sexual misconduct and to correct misleading context in media pieces. But if you only mention the latter, it gives the impression that the former doesn’t matter. I would highly encourage people who care about both to at least mention that you care about reducing the level of misconduct. It may sound like stating the obvious, but it really does matter.
While I certainly hope everyone cares about both, I can’t honestly say I believe that. Going through the lesswrong thread, it honestly looks to me like a lot of people genuinely don’t want to think about the issue at all, and I find this concerning. For example, downvoting the thread to 0 seems completely unwarranted.
None of this was news to the people who use LessWrong.
The time to have a conversation about what went wrong and what a community can do better, is immediately after you learned that the thing happened. If you search for the names of the people involved, you’ll see that LessWrong did that at length.
The worst possible time to bring the topic up again, is when someone writes a misleading article for the express purpose of hurting you, which was not written to be helpful and purposefully lacks the context that it would require in order to be helpful. Why would you give someone a button they can press to make your forum talk for weeks about nothing?
It was a low-quality article and was downvoted so fewer people saw it. I wish the same had happened here.
The Bloomberg piece was not an update on how misconduct has happened in EA to anyone who has been previously paying attention.