You say you had 57 speakers (or i guess more that weren’t featured?). An attendee estimates that 8 speakers in lessonline and manifest had scientific racism controversies (with 2 more debatebly adjacent). Obviously this isn’t an exact estimate, but it looks like something on the order of 5-10% of the speakers had scientific racism ties.
What percentage of speakers were African American (or african anything else)? I did not see any of the 30 with pictures on the site, so i’d guess something on the order of 0-3%.
Do you see a problem with a conference that has something like twice or three times as many scientific racist speakers as it does black people speakers?
These speakers are not a representative slice of society. Scientific racists are much, much more rare, and black people are much, much more common. If your goal is a free exchange of ideas, the ideas you are recieving here are vastly skewed in one direction.
The actual effect of this type of speaker list is to push out anti-racists, and encourage more people sympathetic to scientific racism to join your community. I think this is bad!
Just to note, many unrelated communities underrepresent black people eg, to quote Scott Alexander,
“For Runners (3%). Bikers (6%). Furries (2%). Wall Street senior management (2%). Occupy Wall Street protesters (unknown but low, one source says 1.6% but likely an underestimate). BDSM (unknown but low) Tea Party members (1%). American Buddhists (~2%). Bird watchers (4%)...”
and manifest likely heavily overrepresented queer and neurodivergent people. It’s unclear to me that every single minority group should be represented perfectly in every single community (do we hold EA to this standard? what % of EA talks are given by black people?).
I think it’s pretty hard to have your community be about even 1 thing, let alone 1 thing + perfect representation of every group. The sectors that Manifest draws from (forecasting, crypto, heterodoxy, rationalism, EA, tech) probably all have low black representation, so it seems a lot to ask manifest alone to improve this.
To state the obvious, I don’t expect manifest ever to have 50⁄50 gender representation (though I think it would be better if it were, say, 80⁄20[1] than like 95⁄5). To give another example in the forecasting space Metaculus is very polite, online, and doesn’t allow controversial questions and for a long time had a female CEO. Even then I don’t think that more than 1 in 20 metaculus forecasters were women. That suggests to me that there is something going on more than purely “unwelcomeness”.
That said, I would appreciate if “strong racism not being common” was a pretty important goal, because if it isn’t, I imagine I’ll stop attending.
Strangely, of all the large events I saw over the LessOnline/Manifest 9 days the Scott Alexander meetup was the one that seemed to contain the most women. I have no idea what that implies.
I think we’re missing the dynamic though where there’s a very clear theory for why Manifest would be dramatically less appealing to black people or to women, when you have a platformed and promoted speaker(!!) who think we need mass surveillance of black people to reduce crime or think that intellectual debate is an inherently male activity that women are less well suited to. It’s not a mystery here. Why would it be fun to “balance out” that?
I am very confident that the variance explained by speaker choice is vastly less than the variance explained by the demographics of the forecasting community, compared to a US population baseline.
I would be happy to bet on this, since I am sure we will have events that have different speaker line up but drawing from the same community, and I am sure that from a US population baseline, they will differ relatively little (my best guess is there will be a small directional effect for gender for Manifest, since the event feels a bit more masculine than other community events, but no measurable directional effect for race).
I think it is likely that a perception of sympathy or ambivalence to racism reduces the number of black people interested in being part of the rationalism / Effective Altruism communities.
I doubt the signal from a single event like this is strong enough to be detectable, but I’d be surprised if media coverage like the Guardian’s write-up didn’t have an effect on demographic compositions of the community in the aggregate.
I’d grant that specific invite list of a given event are smaller drivers than the general popularity as important intellectuals of figures like Hanania who explicitly or implicitly endorse human biodiversity.
I would guess that even without Hanania the event would have had ~the same black attendance. I don’t think Hanania was at LessOnline and that event wasn’t notably more black. I don’t like Hanania either, but I don’t think your theory predicts what we see. Given the lack of black people in basically all the constituent communities of Manifest I wouldn’t expect much.
If we wanted manifest to have black attendees then probably the way to do it would be to invite some influencers with large black followings. If I recall correctly there are a number of debate streamers who are black.
But if Manifest really is wide-open as to subject matter, one would expect this kind of effect to be less pronounced than for narrower subject matter. For example, I’d be concerned if the speakers in a TED speaker series in the U.S. were only 2% were African-American (or African). My default level of concern would be much less in a bird-watching speaker series.
I’d note that the small percentage for some of those groups memberships are affected by systemic racism (Wall Street executives), by cultural factors (having parents who practice a religion is a major factor as to whether someone will do the same), by socioeconomic factors (running marathons is correlated with income, as is race).
Moreover, Manifest had control over the diversity of invitees, so “few African-Americans were interested in actually coming” wouldn’t be a great explanation of a very low percentage rate for invitees. This is particularly true where the organizers assert that they were trying to bring together diverse perspectives.
Right now, for me, removing speakers who say racist things is a much higher priority than seeking racial balance in attendees. And probably higher than either is having an event which pushes towards the truth, since I generally think this is so hard. But I agree that there are lots of black intellectuals and streamers I’d love to listen to.
I can imagine thinking that it would be good to push slightly on speakers in the representation direction. The problem is it can quite quickly shift away from having the speakers the community most wants to hear. Probably I’d have a big poll and then anyone can nominate speakers, anyone can vote. In that world, representative speakers can be bumped up if everyone wants that but that has a cost to attendees themselves.
You say you had 57 speakers (or i guess more that weren’t featured?). An attendee estimates that 8 speakers in lessonline and manifest had scientific racism controversies (with 2 more debatebly adjacent). Obviously this isn’t an exact estimate, but it looks like something on the order of 5-10% of the speakers had scientific racism ties.
What percentage of speakers were African American (or african anything else)? I did not see any of the 30 with pictures on the site, so i’d guess something on the order of 0-3%.
Do you see a problem with a conference that has something like twice or three times as many scientific racist speakers as it does black people speakers?
These speakers are not a representative slice of society. Scientific racists are much, much more rare, and black people are much, much more common. If your goal is a free exchange of ideas, the ideas you are recieving here are vastly skewed in one direction.
The actual effect of this type of speaker list is to push out anti-racists, and encourage more people sympathetic to scientific racism to join your community. I think this is bad!
Just to note, many unrelated communities underrepresent black people eg, to quote Scott Alexander,
and manifest likely heavily overrepresented queer and neurodivergent people. It’s unclear to me that every single minority group should be represented perfectly in every single community (do we hold EA to this standard? what % of EA talks are given by black people?).
I think it’s pretty hard to have your community be about even 1 thing, let alone 1 thing + perfect representation of every group. The sectors that Manifest draws from (forecasting, crypto, heterodoxy, rationalism, EA, tech) probably all have low black representation, so it seems a lot to ask manifest alone to improve this.
To state the obvious, I don’t expect manifest ever to have 50⁄50 gender representation (though I think it would be better if it were, say, 80⁄20[1] than like 95⁄5). To give another example in the forecasting space Metaculus is very polite, online, and doesn’t allow controversial questions and for a long time had a female CEO. Even then I don’t think that more than 1 in 20 metaculus forecasters were women. That suggests to me that there is something going on more than purely “unwelcomeness”.
That said, I would appreciate if “strong racism not being common” was a pretty important goal, because if it isn’t, I imagine I’ll stop attending.
Strangely, of all the large events I saw over the LessOnline/Manifest 9 days the Scott Alexander meetup was the one that seemed to contain the most women. I have no idea what that implies.
I think we’re missing the dynamic though where there’s a very clear theory for why Manifest would be dramatically less appealing to black people or to women, when you have a platformed and promoted speaker(!!) who think we need mass surveillance of black people to reduce crime or think that intellectual debate is an inherently male activity that women are less well suited to. It’s not a mystery here. Why would it be fun to “balance out” that?
I am very confident that the variance explained by speaker choice is vastly less than the variance explained by the demographics of the forecasting community, compared to a US population baseline.
I would be happy to bet on this, since I am sure we will have events that have different speaker line up but drawing from the same community, and I am sure that from a US population baseline, they will differ relatively little (my best guess is there will be a small directional effect for gender for Manifest, since the event feels a bit more masculine than other community events, but no measurable directional effect for race).
I think it is likely that a perception of sympathy or ambivalence to racism reduces the number of black people interested in being part of the rationalism / Effective Altruism communities.
I doubt the signal from a single event like this is strong enough to be detectable, but I’d be surprised if media coverage like the Guardian’s write-up didn’t have an effect on demographic compositions of the community in the aggregate.
I’d grant that specific invite list of a given event are smaller drivers than the general popularity as important intellectuals of figures like Hanania who explicitly or implicitly endorse human biodiversity.
I would guess that even without Hanania the event would have had ~the same black attendance. I don’t think Hanania was at LessOnline and that event wasn’t notably more black. I don’t like Hanania either, but I don’t think your theory predicts what we see. Given the lack of black people in basically all the constituent communities of Manifest I wouldn’t expect much.
If we wanted manifest to have black attendees then probably the way to do it would be to invite some influencers with large black followings. If I recall correctly there are a number of debate streamers who are black.
But if Manifest really is wide-open as to subject matter, one would expect this kind of effect to be less pronounced than for narrower subject matter. For example, I’d be concerned if the speakers in a TED speaker series in the U.S. were only 2% were African-American (or African). My default level of concern would be much less in a bird-watching speaker series.
I’d note that the small percentage for some of those groups memberships are affected by systemic racism (Wall Street executives), by cultural factors (having parents who practice a religion is a major factor as to whether someone will do the same), by socioeconomic factors (running marathons is correlated with income, as is race).
Moreover, Manifest had control over the diversity of invitees, so “few African-Americans were interested in actually coming” wouldn’t be a great explanation of a very low percentage rate for invitees. This is particularly true where the organizers assert that they were trying to bring together diverse perspectives.
Right now, for me, removing speakers who say racist things is a much higher priority than seeking racial balance in attendees. And probably higher than either is having an event which pushes towards the truth, since I generally think this is so hard. But I agree that there are lots of black intellectuals and streamers I’d love to listen to.
I can imagine thinking that it would be good to push slightly on speakers in the representation direction. The problem is it can quite quickly shift away from having the speakers the community most wants to hear. Probably I’d have a big poll and then anyone can nominate speakers, anyone can vote. In that world, representative speakers can be bumped up if everyone wants that but that has a cost to attendees themselves.