I think there is a lot of detail and complexity here and I don’t think that this comment is going to do it justice, but I want to signal that I’m open to dialog about these things.
For example, allowing introductory EA spaces like the EA Facebook group or local public EA group meetups to disallow certain forms of divisive speech, while continuing to encourage serious open discussion in more advanced EA spaces, like on this EA forum.
On the face of it, this seems like a bad idea to me. I don’t want “introductory” EA spaces to have different norms than advanced EA spaces, because I only want people to join the EA movement to the extent that they have a very high epistemic standards. If people wouldn’t like the discourse norms in the central EA spaces, I don’t want them to feel comfortable in the more peripheral EA spaces. I would prefer that they bounce off.
To say it another way, I think it is a mistake to have “advanced” and “introductory” EA spaces, at all.
I am intending to make a pretty strong claim here.
[One operationalization I generated, but want to think more about before I fully endorse it: “I would turn away billions of dollars of funding to EA causes, if that was purchased at the cost of ‘EA’s discourse norms are as good as those in academia.’”]
Some cruxes:
I think what is valuable about the EA movement is the quality of the epistemic discourse in the EA movement, and almost nothing else matters (and to the extent that other factors matter, the indifference curve heavily favors better epistemology). If I changed my mind about that, it would change my view about a lot of things, including the answer to this question.
I think a model by which people gradually “warm up” to “more advanced” discourse norms is false. I predict that people will mostly stay in their comfort zone, and people who like discussion at the “less advanced” level will prefer to stay at that level. If I were wrong about that, I would substantially reconsider my view.
Large number of people at the fringes of a movement tend to influence the direction of the movement, and significantly shape the flow of talent to the core of the movement. If I thought that you could have 90% of the people identifying as EAs have somewhat worse discourse norms than we have on this forum without meaningfully impacting the discourse or action of the people at the core of the movement, I think I might change my mind about this.
Surely there exists a line at which we agree on principle. Imagine that, for example, our EA spaces were littered with people making cogent arguments that steel manned holocaust denial, and we were approached by a group of Jewish people saying “We want to become effective altruists because we believe in the stated ideals, but we don’t feel safe participating in a space where so many people commonly and openly argue that the holocaust did not happen.”
In this scenario, I hope that we’d both agree that it would be appropriate for us to tell our fellow EAs to cut it out. While it may be a useful thing to discuss (if only to show how absurd it is), we can (I argue) push future discussion of it into a smaller space so that the general EA space doesn’t have to be peppered with such arguments. This is the case even if none of the EAs talking about it actually believe it. Even if they are just steel-manning devil’s advocates, surely it is more effective for us to clean the space up so that our Jewish EA friends feel safe to come here and interact with us, at the cost of moving specific types of discussion to a smaller area.
I agree that one of the things that makes EA great is the quality of its epistemic discourse. I don’t want my words here to be construed that I think we should lower it unthinkingly. But I do think that a counterbalancing force does exist: being so open to discussion of any kind that we completely alienate a section of people who otherwise would be participating in this space.
I strongly believe that representation, equity, and inclusiveness is important in the EA movement. I believe it so strongly that I try to look at what people are saying in the safe spaces where they feel comfortable talking about EA norms that scare them away. I will report here that a large number of people I see talking in private Facebook groups, on private slack channels, in PMs, emails, and even phone calls behind closed doors are continuously saying that they do not feel safe in EA spaces. I am not merely saying that they are “worried” about where EA is heading; I’m saying that right here, right now, they feel uncomfortable fully participating in generalized EA spaces.
You say that “If people wouldn’t like the discourse norms in the central EA spaces…I would prefer that they bounce off.” In principle, I think we agree on this. Casual demands that we are being alienating should not faze us. But there does exist a point at which I think we might agree that those demands are sufficiently strong, like the holocaust denial example. The question, then, is not one of kind, but of degree. The question turns on whether the harm that is caused by certain forms of speech outweighs the benefits accrued by discussing those things.
Q1: Do you agree that this is a question of degree, not kind? If not, then the rest of this comment doesn’t really apply.
Q2: You mentioned having similar standards to academia. If it became standard for undergraduate colleges to disallow certain forms of racist speech to protect students, would you be okay with copying those norms over to EA? Or do you mean only having similar standards to what academics discuss amongst each other, setting aside completely how universities deal with undergraduate students’ spaces.
I have significant cognitive dissonance here. I’m not at all certain about what I personally feel. But I do want to report that there are large numbers of people, in several disparate places, many of which I doubt interact between themselves in any significant way, who all keep saying in private that they do not feel safe here. I have seen people actively go through harm from EAs casually making the case for systemic racism not being real and I can report that it is not a minor harm.
I’m extremely privileged, so it’s hard for me to empathize here. I cannot imagine being harmed by mere speech in this way. But I can report from direct experience watching private Facebook chats and slack threads of EAs who aren’t willing to publicly talk about this stuff that these speech acts are causing real harm.
Is the harm small enough to warrant just having these potential EAs bounce off? Or would we benefit from pushing such speech acts to smaller portions of EA so that newer, more diverse EAs can come in and contribute to our movement? I hope that you’ll agree that these are questions of degree, not of kind. After seeing the level of harm that these kinds of speech acts cause, I think my position of moving that discourse away from introductory spaces is warranted. But I also strongly agree with traditional enlightenment ideals of open discussion, free speech, and that the best way to show an idea is wrong is to seriously discuss it. So I definitely don’t want to ban such speech everywhere. I just want there to be some way for us to have good epistemic standards and also benefit from EAs who don’t feel safe in the main EA Facebook groups.
To borrow a phrase from Nora Caplan-Bricker, they’re not demanding that EA spaces be happy places where they never have to read another word of dissent. Instead, they’re asking for a level of acceptance and ownership that other EAs already have. They just want to feel safe.
Surely there exists a line at which we agree on principle. Imagine that, for example, our EA spaces were littered with people making cogent arguments that steel manned holocaust denial, and we were approached by a group of Jewish people saying “We want to become effective altruists because we believe in the stated ideals, but we don’t feel safe participating in a space where so many people commonly and openly argue that the holocaust did not happen.”
In this scenario, I hope that we’d both agree that it would be appropriate for us to tell our fellow EAs to cut it out.
I agree with your conclusion about this instance, but for very different reasons, and I don’t think it supports your wider point of view. It would be bad if EAs spent all the time discussing the holocaust, because the holocaust happened in the past, and so there is nothing we can possible do to prevent it. As such the discussion is likely to be a purely academic exercise that does not help improve the world.
It would be very different to discuss a currently occurring genocide. If EAs were considering investing resources in fighting the Uighur genocide, for example, it would be very valuable to hear contrary evidence. If, for example, we learnt that far fewer people were being killed than we thought, or that the CCP’s explanations about terrorism were correct, this would be useful information that would help us prioritize our work. Equally, it would be valuable to hear if we had actually under-estimated the death toll, for exactly the same reasons.
Similarly, Animal Rights EAs consider our use of factory farming to be a modern holocaust, far larger than any prior. But debate about this is a perfectly acceptable EA topic—even on debate around subjects like ‘but do the victims (animals) have moral value?’
Or again, pro-life activists consider our use of abortion to be a modern holocaust, far larger than any prior. But debate about this is a perfectly acceptable EA topic—even on debate around subjects like ‘but do the victims (fetuses) have moral value?’
It might be the case that people make a dedicated ‘Effective Liberation for Xinjiang’ group, and intend to discuss only methods there, not the fundamental premise. But if they started posting about the Uighurs in other EA groups, criticism of their project, including its fundamental premises, would be entirely legitimate.
I think this is true even if it made some hypothetical Uighur diaspora members of the group feel ‘unsafe’. People have a right to actual safety—clearly no-one should be beating each other up at EA events. But an unlimited right to ‘feel safe’, even when this can only be achieved by imposing strict (and contrary to EA) restrictions on others is clearly tyrannical. If you feel literally unsafe when someone makes an argument on the internet you have a serious problem and it is not our responsibility (or even within our power) to accommodate this. You should feel unsafe while near cliff edges, or around strange men in dark allys—not in a debate. Indeed, if feeling ‘unsafe’ is a trump card then I will simply claim that I feel unsafe when people discuss BLM positively, due to the (from my perspective) implied threat of riots.
The analogy here I think is clear. I think it is legitimate to say we will not discuss the Uighur genocide (or animal rights, or racism) in a given group because they are off-topic. What is not at all legitimate is to say that one side, but not the other, is forbidden.
Finally, I also think your strategy is potentially a bit dishonest. We should not hide the true nature of EA, whatever that is, from newcomers in an attempt to seduce them into the movement.
If you’re correct that the harms that come from open debate are only minor harms, then I think I’d agree with most of what you’ve said here (excepting your final paragraph). But the position of bipgms I’ve spoken to is that allowing some types of debate really does do serious harm, and from watching them talk about and experience it, I believe them. My initial intuition was closer to your point of view — it’s just so hard to imagine how open debate on an issue could cause such harm — but, in watching how they deal with some of these issues, I cannot deny that the harm from something like a casual denial of systemic racism caused them significant harm.
On a different point, I think I disagree with your final paragraph’s premise. To me, having different moderation rules is a matter of appropriateness, not a fundamental difference. I think that it would not be difficult to say to new EAs that “moderation in one space has different appropriateness rules than in some other space” without hiding the true nature of EA and/or being dishonest about it. This is relevant because one of the main EA Facebook groups is currently deciding how to implement moderation rules with regard to this stuff right now.
Improving signaling seems like a positive-sum change. Continuing to have open debate despite people self-reporting harm is consistent with both caring a lot about the truth and also with not caring about harm. People often assume the latter, and given the low base rate of communities that actually care about truth they aren’t obviously wrong to do so. So signaling the former would be nice.
Note: you talked about systemic racism but a similar phenomenon seems to happen anywhere laymen profess expertise they don’t have. E.g. if someone tells you that they think eating animals is morally acceptable, you should probably just ignore them because most people who say that haven’t thought about the issue very much. But there are a small number of people who do make that statement and are still worth listening to, and they often intentionally signal it by saying “I think factory farming is terrible but XYZ” instead of just “XYZ”.
First of all, I took this comment to be sincere and in the spirit of dialog. Thank you and salutations.
[Everything that I say in this comment is tentative, and I may change my mind.]
Surely there exists a line at which we agree on principle. Imagine that, for example, our EA spaces were littered with people making cogent arguments that steel manned holocaust denial, and we were approached by a group of Jewish people saying “We want to become effective altruists because we believe in the stated ideals, but we don’t feel safe participating in a space where so many people commonly and openly argue that the holocaust did not happen.”
In this scenario, I hope that we’d both agree that it would be appropriate for us to tell our fellow EAs to cut it out.
If that were actually happening, I would want to think more about the specific case (and talk directly to the people involved), but I’m inclined to bite the bullet of allowing that sort of conversation.
The main reason is that, (I would guess, though you can say more about your state of mind), that there is an implicit premise underlying the stance that we shouldn’t allow that kind of talk. Namely, that “the Holocaust happened, and Holocaust denial is false”.
Now, my understanding is that there is an overwhelming historical consensus that the Holocaust happened. But the more I learn about the world, the more I discover that claims that I would have thought were absurd are basically correct, especially in politicized areas.
I am not so confident that the Holocaust happened, and especially that the holocaust happened the way it is said to have happened, that I am willing to sweep out any discussion to the contrary.
If they are making strong arguments for a false conclusion, then they should be countered with arguments, not social censure.
This is the case even if none of the EAs talking about it actually believe it. Even if they are just steel-manning devil’s advocates...
In the situation where EAs are making such arguments not out of honest truth-seeking, but as playing edge-lord / trying to get attention / etc., then I feel a lot less sympathetic. I would be more inclined to just tell them to cut it out in that case. (Basically, I would make the argument that they are doing damage for no gain.)
But mostly, I would say if any people in an EA group were threatening violence, racially-motivated or otherwise, we should have a zero-tolerance policy. That is where I draw the line. (I agree that there is a bit of a grey area in the cases where someone is politely advocating for violent-action down the line, eg the Marxist who has never personally threatened anyone, but is advocating for a violent revolution.)
...
Q1: Do you agree that this is a question of degree, not kind? If not, then the rest of this comment doesn’t really apply.
I think so. I expect that any rigid rule is going to have edge cases, that are bad enough that you should treat them differently. But I don’t think we’re on the same page about what the relevant scalar is.
If it became standard for undergraduate colleges to disallow certain forms of racist speech to protect students, would you be okay with copying those norms over to EA?
It depends entirely on what is meant by “certain forms”, but on the face of it, I would not be okay with that. I expect that a lot of ideas and behaviors would get marked as “racist”, because that is a convenient and unarguable way to attack those ideas.
I would again draw the line at the threat of violence: if a student group got together to discuss how to harass some racial minority, even just as a hypothetical (they weren’t actually going to do anything), Eli-University would come down on them hard.
If a student group came together to discuss the idea a white ethno-state, and the benefits of racial and cultural homogeneity, Eli-University would consider this acceptable behavior, especially if the epistemic norms of such a group are set high. (However if I had past experience that such reading groups tended to lead to violence, I might watch them extra carefully.)
The ethno-state reading group is racist, and is certainly going to make some people feel uncomfortable, and maybe make them feel unsafe. But I don’t know enough about the world to rule out discussion of that line of thinking entirely.
...
I will report here that a large number of people I see talking in private Facebook groups, on private slack channels, in PMs, emails, and even phone calls behind closed doors are continuously saying that they do not feel safe in EA spaces.
I would love to hear more about the details there. In what ways do people not feel safe?
(Is it things like this comment?)
I’m extremely privileged, so it’s hard for me to empathize here. I cannot imagine being harmed by mere speech in this way. But I can report from direct experience watching private Facebook chats and slack threads of EAs who aren’t willing to publicly talk about this stuff that these speech acts are causing real harm.
Yeah. I want to know more about this. What kind of harm?
My default stance is something like, “look, we’re here to make intellectual progress, and we gotta be able to discuss all kinds of things to do that. If people are ‘harmed’ by speech-acts, I’m sorry for you, but tough nuggets. I guess you shouldn’t participate in this discourse. ”
That said, if I had a better sense of what kind of harms are resulting, I might have a different view, or it might be more obvious where there are cheep tradeoffs to be made.
Is the harm small enough to warrant just having these potential EAs bounce off? Or would we benefit from pushing such speech acts to smaller portions of EA so that newer, more diverse EAs can come in and contribute to our movement? I hope that you’ll agree that these are questions of degree, not of kind.
Yep. I think I do, though I think that the indifference curve is extremely lopsided, for EA in particular.
...
I agree that one of the things that makes EA great is the quality of its epistemic discourse. I don’t want my words here to be construed that I think we should lower it unthinkingly. But I do think that a counterbalancing force does exist: being so open to discussion of any kind that we completely alienate a section of people who otherwise would be participating in this space.
I’m tentatively suggesting that we should pay close to no attention to possibility of alienating people, and just try to do our best to actually make progress on the intellectual project.
It is a (perhaps unfortunate) fact that many true conclusions alienate a lot of people. And it is much more important that we are able to identify those conclusions than that we find more people to join our ranks, or that our ranks are more ethnically / culturally / etc. diverse.
It is a (perhaps unfortunate) fact that many true conclusions alienate a lot of people. And it is much more important that we are able to identify those conclusions than that we find more people to join our ranks, or that our ranks are more ethnically / culturally / etc. diverse.
We are agreed that truth is of paramount importance here. If a true conclusion alienates someone, I endorse not letting that alienation sway us. But I think we disagree on two points:
I believe diversity is a serious benefit. Not just in terms of movement building, but in terms of arriving at truth. Homogeneity breeds blind spots in our thinking. If a supposed truth is arrived at, but only one group recognizes it as truth, doesn’t that make us suspect whether we are correct? To me, good truth-seeking almost requires diversity in several different forms. Not just philosophical diversity, but diversity in how we’ve come up in the world, in how we’ve experienced things. Specifically including BIPGM seems to me to very important in ensuring that we arrive at true conclusions.
I believe the methods of how we arrive at true conclusions doesn’t need to be Alastair Moody-levels of constant vigilance. We don’t have to rigidly enforce norms of full open debate all the time.
I think the latter disagreement we have is pretty strong, given your willingness to bite the bullet on holocaust denial. Sure, we never know anything for sure, but when you get to a certain point, I feel like it’s okay to restrict debate on a topic to specialized places. I want to say something like “we have enough evidence that racism is real that we don’t need to discuss it here; if you want to debate that, go to this other space”, and I want to say it because discussing racism as though it doesn’t exist causes a level of harm that may rise to the equivalent to physical harm in some people. I’m not saying we have to coddle anyone, but if we can reduce that harm for almost no cost, I’m willing to. To me, restricting debate in a limited way on a specific Facebook thread is almost no cost. We already restrict debate in other, similar ways: no name calling, no doxxing, no brigading. In the EAA FB group, we take as a given that animals are harmed and we should help them. We restrict debate on that there because it’s inappropriate to debate that point there. That doesn’t mean it can’t be debated elsewhere. To me, restricting the denial of racism (or the denial of genocide) is just an additional rule of this type. It doesn’t mean it can’t be discussed elsewhere. It just isn’t appropriate there.
In what ways do people not feel safe? (Is it things like this comment?) … I want to know more about this. What kind of harm?
No, it’s not things like this comment. We are in a forum where discussing this kind of thing is expected and appropriate.
I don’t feel like I should say anything that might inadvertently out some of the people that I have seen in private groups talking about these harms. Many of these EAs are not willing to speak out about this issue because they fear being berated for having these feelings. It’s not exactly what you’re asking for, but a few such people are already public about the effects from those harms. Maybe their words will help: https://sentientmedia.org/racism-in-animal-advocacy-and-effective-altruism-hinders-our-mission
“[T]aking action to eliminate racism is critical for improving the world, regardless of the ramifications for animal advocacy. But if the EA and animal advocacy communities fail to stand for (and not simply passively against) antiracism, we will also lose valuable perspectives that can only come from having different lived experiences—not just the perspectives of people of the global majority who are excluded, but the perspective of any talented person who wants to accomplish good for animals without supporting racist systems.
I know this is true because I have almost walked away from these communities myself, disquieted by the attitudes toward racism I found within them.”
“I think a model by which people gradually “warm up” to “more advanced” discourse norms is false.”
I don’t think that’s the main benefit of disallowing certain forms of speech at certain events. I’d imagine it’d be to avoid making EA events attractive and easily accessible for, say, white supremacists. I’d like to make it pretty costly for a white supremacist to be able to share their ideas at an EA event.
We’ve already seen white nationalists congregate in some EA-adjacent spaces. My impression is that (especially online) spaces that don’t moderate away or at least discourage such views will tend to attract them—it’s not the pattern of activity you’d see if white nationalists randomly bounce around places or people organically arrive at those views. I think this is quite dangerous for epistemic norms, because white nationalist/supremacist views are very incorrect and deter large swaths of potential participants and also people with those views routinely argue in bad faith by hiding how extreme their actual opinions are while surreptitiously promoting the extreme version. It’s also in my view a fairly clear and present danger to EA given that there are other communities with some white nationalist presence that are quite socially close to EA.
I don’t know anything about Leverage but I can think of another situation where someone involved in the rationalist community was exposed as having misogynistic and white supremacist anonymous online accounts. (They only had loose ties to the rationalist community, it came up another way, but it concerned me.)
I just upvoted this comment as I strongly agree with it, but also, it had −1 karma with 2 votes on it when I did so. I think it would be extremely helpful for folks who disagree with this, or otherwise want to downvote it, to talk about why they disagree or downvoted it.
I didn’t downvote it, though probably I should have. But it seems a stretch to say ‘one guy who works for a weird organization that is supposedly EA’ implies ‘congregation’. I think that would have to imply a large number of people. I would be very disappointed if I had a congregation of less than ten people.
JoshYou also ignores important hedging in the linked comment:
Bennett denies this connection; he says he was trying to make friends with these white nationalists in order to get information on them and white nationalism. I think it’s plausible that this is somewhat true.
So instead of saying
We’ve already seen white nationalists congregate in some EA-adjacent spaces.
It would be more fair to say
We’ve already seen one guy with some evidence he is a white nationalist (though he somewhat plausibly denies it) work for a weird organization that has some EA links.
Which is clearly much less worrying. There are lots of weird ideologies and a lot of weird people in California, who believe a lot of very incorrect things. I would be surprised if ‘white nationalists’ were really high up on the list of threats to EA, especially given how extremely left wing EA is and how low status they are. We probably have a lot more communists! Rather, I think the highlighting of ‘White Nationalists’ is being done for ideological reasons—i.e. to cast shade on more moderate right wing people by using a term that is practically a slur. I think the grandparent would not have made such a sloppy comment had it not been about the hated outgroup.
I also agree that it’s ridiculous when left-wingers smear everyone on the right as Nazis, white nationalists, whatever. I’m not talking about conservatives, or the “IDW”, or people who don’t like the BLM movement or think racism is no big deal. I’d be quite happy for more right-of-center folks to join EA. I do mean literal white nationalists (like on par with the views in Jonah Bennett’s leaked emails. I don’t think his defense is credible at all, by the way).
I don’t think it’s accurate to see white nationalists in online communities as just the right tail that develops organically from a wide distribution of political views. White nationalists are more organized than that and have their own social networks (precisely because they’re not just really conservative conservatives). Regular conservatives outnumber white nationalists by orders of magnitude in the general public, but I don’t think that implies that white nationalists will be virtually non-existent in a space just because the majority are left of center.
Describing members of Leverage as “white nationalists” strikes me as pretty extreme, to the level of dishonesty, and is not even backed up by the comment that was linked. I thought Buck’s initial comment was also pretty bad, and he did indeed correct his comment, which is a correction that I appreciate, and I feel like any comment that links to it should obviously also take into account the correction.
I have interfaced a lot with people at Leverage, and while I have many issues with the organization, saying that many white nationalists congregate there, and have congregated in the past, just strikes me as really unlikely.
Buck’s comment also says at the bottom:
Edited to add (Oct 08 2019): I wrote “which makes me think that it’s likely that Leverage at least for a while had a whole lot of really racist employees.” I think this was mistaken and I’m confused by why I wrote it. I endorse the claim “I think it’s plausible Leverage had like five really racist employees”. I feel pretty bad about this mistake and apologize to anyone harmed by it.
I also want us to separate “really racist” from “white nationalist” which are just really not the same term, and which appear to me to be conflated via the link above.
I also have other issues with the rest of the comment (namely being constantly worried about communists or nazis hiding everywhere, and generally bringing up nazi comparisons in these discussions, tends to reliably derail things and make it harder to discuss these things well, since there are few conversational moves as mindkilling as accusing the other side to be nazis or communists. It’s not that there are never nazis or communists, but if you want to have a good conversation, it’s better to avoid nazi or communist comparisons until you really have no other choice, or you can really really commit to handling the topic in an open-minded way.)
My description was based on Buck’s correction (I don’t have any first-hand knowledge). I think a few white nationalists congregated at Leverage, not that most Leverage employees are white nationalists, which I don’t believe. I don’t mean to imply anything stronger than what Buck claimed about Leverage.
I invoked white nationalists not as a hypothetical representative of ideologies I don’t like but quite deliberately, because they literally exist in substantial numbers in EA-adjacent online spaces and they could view EA as fertile ground if the EA community had different moderation and discursive norms. (Edited to avoid potential collateral reputational damage) I think the neo-reactionary community and their adjacency to rationalist networks are a clear example.
Just to be clear, I don’t think even most neoreactionaries would classify as white nationalists? Though maybe now we are arguing over the definition of white nationalism, which is definitely a vague term and could be interpreted many ways. I was thinking about it from the perspective of racism, though I can imagine a much broader definition that includes something more like “advocating for nations based on values historically associated with whiteness”, which would obviously include neoreaction, but would also presumably be a much more tenable position in discourse. So for now I am going to assume you mean something much more straightforwardly based on racial superiority, which also appears to be the Wikipedia definition.
I’ve debated with a number of neoreactionaries, and I’ve never seen them bring up much stuff about racial superiority. Usually just arguing against democracy and in favor of centralized control and various arguments derived from that, though I also don’t have a ton of datapoints. There is definitely a focus on the superiority of western culture in their writing and rhetoric, much of which is flawed and I am deeply opposed to many of the things I’ve seen at least some neoreactionaries propose, but my sense is that I wouldn’t characterize the philosophy fundamentally as white nationalist in the racist sense of the term. Though of course the few neoreactionaries that I have debated are probably selected in various ways that reduces the likelihood of having extreme opinions on these dimensions (though they are also the ones that are most likely to engage with EA, so I do think the sample should carry substantial weight).
Of course, some neoreactionaries are also going to be white nationalists, and being a neoreactionary will probably correlate with white nationalism at least a bit, but my guess is that at least the people adjacent to EA and Rationality that I’ve seen engage with that philosophy haven’t been very focused on white nationalism, and I’ve frequently seen them actively argue against it.
I think that it seems like accusations of EA associations with white supremacy of various sorts come up enough to be pretty concerning.
I also think the claims would be equally concerning if JoshYou had said “white supremacists” or “really racist people” instead of “white nationalists” in the original post, so I feel uncertain that Buck stepping back the original post actually lessens the degree we ought to be concerned?
I also have other issues with the rest of the comment (namely being constantly worried about communists or nazis hiding everywhere, and generally bringing up nazi comparisons in these discussions, tends to reliably derail things and make it harder to discuss these things well, since there are few conversational moves as mindkilling as accusing the other side to be nazis or communists. It’s not that there are never nazis or communists, but if you want to have a good conversation, it’s better to avoid nazi or communist comparisons until you really have no other choice, or you can really really commit to handling the topic in an open-minded way.)
I didn’t really see the Nazi comparisons (I guess saying white nationalist is sort of one, but I personally associate white nationalism as a phrase much more with individuals in the US than Nazis, though that may be biased by being American).
I guess broadly a trend I feel like I’ve seen lately is occasionally people writing about witnessing racism in the EA community, and having what seem like really genuine concerns, and then those basically not being discussed (at least on the EA Forum) or being framed as shutting down conversation.
I don’t follow how what you’re saying is a response to what I was saying.
I think a model by which people gradually “warm up” to “more advanced” discourse norms is false.
I wasn’t saying “the point of different discourse norms in different EA spaces is that it will gradually train people into more advanced discourse norms.” I was saying if that I was mistaken about that “warming up effect”, it would cause me to reconsider my view here.
In the comment above, I am only saying that I think it is a mistake to have different discourse norms at the core vs. the periphery of the movement.
I think there is a lot of detail and complexity here and I don’t think that this comment is going to do it justice, but I want to signal that I’m open to dialog about these things.
On the face of it, this seems like a bad idea to me. I don’t want “introductory” EA spaces to have different norms than advanced EA spaces, because I only want people to join the EA movement to the extent that they have a very high epistemic standards. If people wouldn’t like the discourse norms in the central EA spaces, I don’t want them to feel comfortable in the more peripheral EA spaces. I would prefer that they bounce off.
To say it another way, I think it is a mistake to have “advanced” and “introductory” EA spaces, at all.
I am intending to make a pretty strong claim here.
[One operationalization I generated, but want to think more about before I fully endorse it: “I would turn away billions of dollars of funding to EA causes, if that was purchased at the cost of ‘EA’s discourse norms are as good as those in academia.’”]
Some cruxes:
I think what is valuable about the EA movement is the quality of the epistemic discourse in the EA movement, and almost nothing else matters (and to the extent that other factors matter, the indifference curve heavily favors better epistemology). If I changed my mind about that, it would change my view about a lot of things, including the answer to this question.
I think a model by which people gradually “warm up” to “more advanced” discourse norms is false. I predict that people will mostly stay in their comfort zone, and people who like discussion at the “less advanced” level will prefer to stay at that level. If I were wrong about that, I would substantially reconsider my view.
Large number of people at the fringes of a movement tend to influence the direction of the movement, and significantly shape the flow of talent to the core of the movement. If I thought that you could have 90% of the people identifying as EAs have somewhat worse discourse norms than we have on this forum without meaningfully impacting the discourse or action of the people at the core of the movement, I think I might change my mind about this.
Surely there exists a line at which we agree on principle. Imagine that, for example, our EA spaces were littered with people making cogent arguments that steel manned holocaust denial, and we were approached by a group of Jewish people saying “We want to become effective altruists because we believe in the stated ideals, but we don’t feel safe participating in a space where so many people commonly and openly argue that the holocaust did not happen.”
In this scenario, I hope that we’d both agree that it would be appropriate for us to tell our fellow EAs to cut it out. While it may be a useful thing to discuss (if only to show how absurd it is), we can (I argue) push future discussion of it into a smaller space so that the general EA space doesn’t have to be peppered with such arguments. This is the case even if none of the EAs talking about it actually believe it. Even if they are just steel-manning devil’s advocates, surely it is more effective for us to clean the space up so that our Jewish EA friends feel safe to come here and interact with us, at the cost of moving specific types of discussion to a smaller area.
I agree that one of the things that makes EA great is the quality of its epistemic discourse. I don’t want my words here to be construed that I think we should lower it unthinkingly. But I do think that a counterbalancing force does exist: being so open to discussion of any kind that we completely alienate a section of people who otherwise would be participating in this space.
I strongly believe that representation, equity, and inclusiveness is important in the EA movement. I believe it so strongly that I try to look at what people are saying in the safe spaces where they feel comfortable talking about EA norms that scare them away. I will report here that a large number of people I see talking in private Facebook groups, on private slack channels, in PMs, emails, and even phone calls behind closed doors are continuously saying that they do not feel safe in EA spaces. I am not merely saying that they are “worried” about where EA is heading; I’m saying that right here, right now, they feel uncomfortable fully participating in generalized EA spaces.
You say that “If people wouldn’t like the discourse norms in the central EA spaces…I would prefer that they bounce off.” In principle, I think we agree on this. Casual demands that we are being alienating should not faze us. But there does exist a point at which I think we might agree that those demands are sufficiently strong, like the holocaust denial example. The question, then, is not one of kind, but of degree. The question turns on whether the harm that is caused by certain forms of speech outweighs the benefits accrued by discussing those things.
Q1: Do you agree that this is a question of degree, not kind? If not, then the rest of this comment doesn’t really apply.
Q2: You mentioned having similar standards to academia. If it became standard for undergraduate colleges to disallow certain forms of racist speech to protect students, would you be okay with copying those norms over to EA? Or do you mean only having similar standards to what academics discuss amongst each other, setting aside completely how universities deal with undergraduate students’ spaces.
I have significant cognitive dissonance here. I’m not at all certain about what I personally feel. But I do want to report that there are large numbers of people, in several disparate places, many of which I doubt interact between themselves in any significant way, who all keep saying in private that they do not feel safe here. I have seen people actively go through harm from EAs casually making the case for systemic racism not being real and I can report that it is not a minor harm.
I’m extremely privileged, so it’s hard for me to empathize here. I cannot imagine being harmed by mere speech in this way. But I can report from direct experience watching private Facebook chats and slack threads of EAs who aren’t willing to publicly talk about this stuff that these speech acts are causing real harm.
Is the harm small enough to warrant just having these potential EAs bounce off? Or would we benefit from pushing such speech acts to smaller portions of EA so that newer, more diverse EAs can come in and contribute to our movement? I hope that you’ll agree that these are questions of degree, not of kind. After seeing the level of harm that these kinds of speech acts cause, I think my position of moving that discourse away from introductory spaces is warranted. But I also strongly agree with traditional enlightenment ideals of open discussion, free speech, and that the best way to show an idea is wrong is to seriously discuss it. So I definitely don’t want to ban such speech everywhere. I just want there to be some way for us to have good epistemic standards and also benefit from EAs who don’t feel safe in the main EA Facebook groups.
To borrow a phrase from Nora Caplan-Bricker, they’re not demanding that EA spaces be happy places where they never have to read another word of dissent. Instead, they’re asking for a level of acceptance and ownership that other EAs already have. They just want to feel safe.
I agree with your conclusion about this instance, but for very different reasons, and I don’t think it supports your wider point of view. It would be bad if EAs spent all the time discussing the holocaust, because the holocaust happened in the past, and so there is nothing we can possible do to prevent it. As such the discussion is likely to be a purely academic exercise that does not help improve the world.
It would be very different to discuss a currently occurring genocide. If EAs were considering investing resources in fighting the Uighur genocide, for example, it would be very valuable to hear contrary evidence. If, for example, we learnt that far fewer people were being killed than we thought, or that the CCP’s explanations about terrorism were correct, this would be useful information that would help us prioritize our work. Equally, it would be valuable to hear if we had actually under-estimated the death toll, for exactly the same reasons.
Similarly, Animal Rights EAs consider our use of factory farming to be a modern holocaust, far larger than any prior. But debate about this is a perfectly acceptable EA topic—even on debate around subjects like ‘but do the victims (animals) have moral value?’
Or again, pro-life activists consider our use of abortion to be a modern holocaust, far larger than any prior. But debate about this is a perfectly acceptable EA topic—even on debate around subjects like ‘but do the victims (fetuses) have moral value?’
It might be the case that people make a dedicated ‘Effective Liberation for Xinjiang’ group, and intend to discuss only methods there, not the fundamental premise. But if they started posting about the Uighurs in other EA groups, criticism of their project, including its fundamental premises, would be entirely legitimate.
I think this is true even if it made some hypothetical Uighur diaspora members of the group feel ‘unsafe’. People have a right to actual safety—clearly no-one should be beating each other up at EA events. But an unlimited right to ‘feel safe’, even when this can only be achieved by imposing strict (and contrary to EA) restrictions on others is clearly tyrannical. If you feel literally unsafe when someone makes an argument on the internet you have a serious problem and it is not our responsibility (or even within our power) to accommodate this. You should feel unsafe while near cliff edges, or around strange men in dark allys—not in a debate. Indeed, if feeling ‘unsafe’ is a trump card then I will simply claim that I feel unsafe when people discuss BLM positively, due to the (from my perspective) implied threat of riots.
The analogy here I think is clear. I think it is legitimate to say we will not discuss the Uighur genocide (or animal rights, or racism) in a given group because they are off-topic. What is not at all legitimate is to say that one side, but not the other, is forbidden.
Finally, I also think your strategy is potentially a bit dishonest. We should not hide the true nature of EA, whatever that is, from newcomers in an attempt to seduce them into the movement.
I think this comment says what I was getting at in my own reply, though more strongly.
If you’re correct that the harms that come from open debate are only minor harms, then I think I’d agree with most of what you’ve said here (excepting your final paragraph). But the position of bipgms I’ve spoken to is that allowing some types of debate really does do serious harm, and from watching them talk about and experience it, I believe them. My initial intuition was closer to your point of view — it’s just so hard to imagine how open debate on an issue could cause such harm — but, in watching how they deal with some of these issues, I cannot deny that the harm from something like a casual denial of systemic racism caused them significant harm.
On a different point, I think I disagree with your final paragraph’s premise. To me, having different moderation rules is a matter of appropriateness, not a fundamental difference. I think that it would not be difficult to say to new EAs that “moderation in one space has different appropriateness rules than in some other space” without hiding the true nature of EA and/or being dishonest about it. This is relevant because one of the main EA Facebook groups is currently deciding how to implement moderation rules with regard to this stuff right now.
Improving signaling seems like a positive-sum change. Continuing to have open debate despite people self-reporting harm is consistent with both caring a lot about the truth and also with not caring about harm. People often assume the latter, and given the low base rate of communities that actually care about truth they aren’t obviously wrong to do so. So signaling the former would be nice.
Note: you talked about systemic racism but a similar phenomenon seems to happen anywhere laymen profess expertise they don’t have. E.g. if someone tells you that they think eating animals is morally acceptable, you should probably just ignore them because most people who say that haven’t thought about the issue very much. But there are a small number of people who do make that statement and are still worth listening to, and they often intentionally signal it by saying “I think factory farming is terrible but XYZ” instead of just “XYZ”.
First of all, I took this comment to be sincere and in the spirit of dialog. Thank you and salutations.
[Everything that I say in this comment is tentative, and I may change my mind.]
If that were actually happening, I would want to think more about the specific case (and talk directly to the people involved), but I’m inclined to bite the bullet of allowing that sort of conversation.
The main reason is that, (I would guess, though you can say more about your state of mind), that there is an implicit premise underlying the stance that we shouldn’t allow that kind of talk. Namely, that “the Holocaust happened, and Holocaust denial is false”.
Now, my understanding is that there is an overwhelming historical consensus that the Holocaust happened. But the more I learn about the world, the more I discover that claims that I would have thought were absurd are basically correct, especially in politicized areas.
I am not so confident that the Holocaust happened, and especially that the holocaust happened the way it is said to have happened, that I am willing to sweep out any discussion to the contrary.
If they are making strong arguments for a false conclusion, then they should be countered with arguments, not social censure.
In the situation where EAs are making such arguments not out of honest truth-seeking, but as playing edge-lord / trying to get attention / etc., then I feel a lot less sympathetic. I would be more inclined to just tell them to cut it out in that case. (Basically, I would make the argument that they are doing damage for no gain.)
But mostly, I would say if any people in an EA group were threatening violence, racially-motivated or otherwise, we should have a zero-tolerance policy. That is where I draw the line. (I agree that there is a bit of a grey area in the cases where someone is politely advocating for violent-action down the line, eg the Marxist who has never personally threatened anyone, but is advocating for a violent revolution.)
...
I think so. I expect that any rigid rule is going to have edge cases, that are bad enough that you should treat them differently. But I don’t think we’re on the same page about what the relevant scalar is.
It depends entirely on what is meant by “certain forms”, but on the face of it, I would not be okay with that. I expect that a lot of ideas and behaviors would get marked as “racist”, because that is a convenient and unarguable way to attack those ideas.
I would again draw the line at the threat of violence: if a student group got together to discuss how to harass some racial minority, even just as a hypothetical (they weren’t actually going to do anything), Eli-University would come down on them hard.
If a student group came together to discuss the idea a white ethno-state, and the benefits of racial and cultural homogeneity, Eli-University would consider this acceptable behavior, especially if the epistemic norms of such a group are set high. (However if I had past experience that such reading groups tended to lead to violence, I might watch them extra carefully.)
The ethno-state reading group is racist, and is certainly going to make some people feel uncomfortable, and maybe make them feel unsafe. But I don’t know enough about the world to rule out discussion of that line of thinking entirely.
...
I would love to hear more about the details there. In what ways do people not feel safe?
(Is it things like this comment?)
Yeah. I want to know more about this. What kind of harm?
My default stance is something like, “look, we’re here to make intellectual progress, and we gotta be able to discuss all kinds of things to do that. If people are ‘harmed’ by speech-acts, I’m sorry for you, but tough nuggets. I guess you shouldn’t participate in this discourse. ”
That said, if I had a better sense of what kind of harms are resulting, I might have a different view, or it might be more obvious where there are cheep tradeoffs to be made.
Yep. I think I do, though I think that the indifference curve is extremely lopsided, for EA in particular.
...
I’m tentatively suggesting that we should pay close to no attention to possibility of alienating people, and just try to do our best to actually make progress on the intellectual project.
It is a (perhaps unfortunate) fact that many true conclusions alienate a lot of people. And it is much more important that we are able to identify those conclusions than that we find more people to join our ranks, or that our ranks are more ethnically / culturally / etc. diverse.
We are agreed that truth is of paramount importance here. If a true conclusion alienates someone, I endorse not letting that alienation sway us. But I think we disagree on two points:
I believe diversity is a serious benefit. Not just in terms of movement building, but in terms of arriving at truth. Homogeneity breeds blind spots in our thinking. If a supposed truth is arrived at, but only one group recognizes it as truth, doesn’t that make us suspect whether we are correct? To me, good truth-seeking almost requires diversity in several different forms. Not just philosophical diversity, but diversity in how we’ve come up in the world, in how we’ve experienced things. Specifically including BIPGM seems to me to very important in ensuring that we arrive at true conclusions.
I believe the methods of how we arrive at true conclusions doesn’t need to be Alastair Moody-levels of constant vigilance. We don’t have to rigidly enforce norms of full open debate all the time.
I think the latter disagreement we have is pretty strong, given your willingness to bite the bullet on holocaust denial. Sure, we never know anything for sure, but when you get to a certain point, I feel like it’s okay to restrict debate on a topic to specialized places. I want to say something like “we have enough evidence that racism is real that we don’t need to discuss it here; if you want to debate that, go to this other space”, and I want to say it because discussing racism as though it doesn’t exist causes a level of harm that may rise to the equivalent to physical harm in some people. I’m not saying we have to coddle anyone, but if we can reduce that harm for almost no cost, I’m willing to. To me, restricting debate in a limited way on a specific Facebook thread is almost no cost. We already restrict debate in other, similar ways: no name calling, no doxxing, no brigading. In the EAA FB group, we take as a given that animals are harmed and we should help them. We restrict debate on that there because it’s inappropriate to debate that point there. That doesn’t mean it can’t be debated elsewhere. To me, restricting the denial of racism (or the denial of genocide) is just an additional rule of this type. It doesn’t mean it can’t be discussed elsewhere. It just isn’t appropriate there.
No, it’s not things like this comment. We are in a forum where discussing this kind of thing is expected and appropriate.
I don’t feel like I should say anything that might inadvertently out some of the people that I have seen in private groups talking about these harms. Many of these EAs are not willing to speak out about this issue because they fear being berated for having these feelings. It’s not exactly what you’re asking for, but a few such people are already public about the effects from those harms. Maybe their words will help: https://sentientmedia.org/racism-in-animal-advocacy-and-effective-altruism-hinders-our-mission
“I think a model by which people gradually “warm up” to “more advanced” discourse norms is false.”
I don’t think that’s the main benefit of disallowing certain forms of speech at certain events. I’d imagine it’d be to avoid making EA events attractive and easily accessible for, say, white supremacists. I’d like to make it pretty costly for a white supremacist to be able to share their ideas at an EA event.
We’ve already seen white nationalists congregate in some EA-adjacent spaces. My impression is that (especially online) spaces that don’t moderate away or at least discourage such views will tend to attract them—it’s not the pattern of activity you’d see if white nationalists randomly bounce around places or people organically arrive at those views. I think this is quite dangerous for epistemic norms, because white nationalist/supremacist views are very incorrect and deter large swaths of potential participants and also people with those views routinely argue in bad faith by hiding how extreme their actual opinions are while surreptitiously promoting the extreme version. It’s also in my view a fairly clear and present danger to EA given that there are other communities with some white nationalist presence that are quite socially close to EA.
I don’t know anything about Leverage but I can think of another situation where someone involved in the rationalist community was exposed as having misogynistic and white supremacist anonymous online accounts. (They only had loose ties to the rationalist community, it came up another way, but it concerned me.)
I just upvoted this comment as I strongly agree with it, but also, it had −1 karma with 2 votes on it when I did so. I think it would be extremely helpful for folks who disagree with this, or otherwise want to downvote it, to talk about why they disagree or downvoted it.
I didn’t downvote it, though probably I should have. But it seems a stretch to say ‘one guy who works for a weird organization that is supposedly EA’ implies ‘congregation’. I think that would have to imply a large number of people. I would be very disappointed if I had a congregation of less than ten people.
JoshYou also ignores important hedging in the linked comment:
So instead of saying
It would be more fair to say
Which is clearly much less worrying. There are lots of weird ideologies and a lot of weird people in California, who believe a lot of very incorrect things. I would be surprised if ‘white nationalists’ were really high up on the list of threats to EA, especially given how extremely left wing EA is and how low status they are. We probably have a lot more communists! Rather, I think the highlighting of ‘White Nationalists’ is being done for ideological reasons—i.e. to cast shade on more moderate right wing people by using a term that is practically a slur. I think the grandparent would not have made such a sloppy comment had it not been about the hated outgroup.
I also agree that it’s ridiculous when left-wingers smear everyone on the right as Nazis, white nationalists, whatever. I’m not talking about conservatives, or the “IDW”, or people who don’t like the BLM movement or think racism is no big deal. I’d be quite happy for more right-of-center folks to join EA. I do mean literal white nationalists (like on par with the views in Jonah Bennett’s leaked emails. I don’t think his defense is credible at all, by the way).
I don’t think it’s accurate to see white nationalists in online communities as just the right tail that develops organically from a wide distribution of political views. White nationalists are more organized than that and have their own social networks (precisely because they’re not just really conservative conservatives). Regular conservatives outnumber white nationalists by orders of magnitude in the general public, but I don’t think that implies that white nationalists will be virtually non-existent in a space just because the majority are left of center.
Describing members of Leverage as “white nationalists” strikes me as pretty extreme, to the level of dishonesty, and is not even backed up by the comment that was linked. I thought Buck’s initial comment was also pretty bad, and he did indeed correct his comment, which is a correction that I appreciate, and I feel like any comment that links to it should obviously also take into account the correction.
I have interfaced a lot with people at Leverage, and while I have many issues with the organization, saying that many white nationalists congregate there, and have congregated in the past, just strikes me as really unlikely.
Buck’s comment also says at the bottom:
I also want us to separate “really racist” from “white nationalist” which are just really not the same term, and which appear to me to be conflated via the link above.
I also have other issues with the rest of the comment (namely being constantly worried about communists or nazis hiding everywhere, and generally bringing up nazi comparisons in these discussions, tends to reliably derail things and make it harder to discuss these things well, since there are few conversational moves as mindkilling as accusing the other side to be nazis or communists. It’s not that there are never nazis or communists, but if you want to have a good conversation, it’s better to avoid nazi or communist comparisons until you really have no other choice, or you can really really commit to handling the topic in an open-minded way.)
My description was based on Buck’s correction (I don’t have any first-hand knowledge). I think a few white nationalists congregated at Leverage, not that most Leverage employees are white nationalists, which I don’t believe. I don’t mean to imply anything stronger than what Buck claimed about Leverage.
I invoked white nationalists not as a hypothetical representative of ideologies I don’t like but quite deliberately, because they literally exist in substantial numbers in EA-adjacent online spaces and they could view EA as fertile ground if the EA community had different moderation and discursive norms. (Edited to avoid potential collateral reputational damage) I think the neo-reactionary community and their adjacency to rationalist networks are a clear example.
Just to be clear, I don’t think even most neoreactionaries would classify as white nationalists? Though maybe now we are arguing over the definition of white nationalism, which is definitely a vague term and could be interpreted many ways. I was thinking about it from the perspective of racism, though I can imagine a much broader definition that includes something more like “advocating for nations based on values historically associated with whiteness”, which would obviously include neoreaction, but would also presumably be a much more tenable position in discourse. So for now I am going to assume you mean something much more straightforwardly based on racial superiority, which also appears to be the Wikipedia definition.
I’ve debated with a number of neoreactionaries, and I’ve never seen them bring up much stuff about racial superiority. Usually just arguing against democracy and in favor of centralized control and various arguments derived from that, though I also don’t have a ton of datapoints. There is definitely a focus on the superiority of western culture in their writing and rhetoric, much of which is flawed and I am deeply opposed to many of the things I’ve seen at least some neoreactionaries propose, but my sense is that I wouldn’t characterize the philosophy fundamentally as white nationalist in the racist sense of the term. Though of course the few neoreactionaries that I have debated are probably selected in various ways that reduces the likelihood of having extreme opinions on these dimensions (though they are also the ones that are most likely to engage with EA, so I do think the sample should carry substantial weight).
Of course, some neoreactionaries are also going to be white nationalists, and being a neoreactionary will probably correlate with white nationalism at least a bit, but my guess is that at least the people adjacent to EA and Rationality that I’ve seen engage with that philosophy haven’t been very focused on white nationalism, and I’ve frequently seen them actively argue against it.
Thanks for elaborating!
I think that it seems like accusations of EA associations with white supremacy of various sorts come up enough to be pretty concerning.
I also think the claims would be equally concerning if JoshYou had said “white supremacists” or “really racist people” instead of “white nationalists” in the original post, so I feel uncertain that Buck stepping back the original post actually lessens the degree we ought to be concerned?
I didn’t really see the Nazi comparisons (I guess saying white nationalist is sort of one, but I personally associate white nationalism as a phrase much more with individuals in the US than Nazis, though that may be biased by being American).
I guess broadly a trend I feel like I’ve seen lately is occasionally people writing about witnessing racism in the EA community, and having what seem like really genuine concerns, and then those basically not being discussed (at least on the EA Forum) or being framed as shutting down conversation.
I don’t follow how what you’re saying is a response to what I was saying.
I wasn’t saying “the point of different discourse norms in different EA spaces is that it will gradually train people into more advanced discourse norms.” I was saying if that I was mistaken about that “warming up effect”, it would cause me to reconsider my view here.
In the comment above, I am only saying that I think it is a mistake to have different discourse norms at the core vs. the periphery of the movement.