I think there was perhaps some miscommunication around your use and my interpretation of âcollapseâ. To me it implies that something is at an unrecoverable stage, like a âcollapsedâ building or support for a presidential candidate âcollapsingâ in a primary race. In your pinned quick take you posit that Effective Altruism as a brand may be damaged to an unrecoverable extent which makes me feel this is the right reading of your post, or at least it was a justified interpretation at least.
***
I actually agree with a lot of your claims in your reply. For example, I think that the last 12 months has been the worst the movement has ever faced. I think many inside the EA movement have lost a lot of faith and trust in both the movement as a whole and its leadership. I donât think anything I said in my comment would support the view that âthereâs no big problemâ, I guess I donât things have to be so binary as âEAâs reputation has collapsedâ or âitâs not that bad weâre back to normalâ.
Looking into the future, I think you do raise a very good point about risks from media that reflects on the FTX collapse and the role EA played. I think thatâs something that the movement does need to prepare for. I donât have specific suggestions myself, apart from believing that we need a more positive strategic direction to counter mistaken criticisms of us instead of leaving them unchallenged, and accepting reforms where they have merit, but I think thatâs a whole other discussion.
***
On the EA movement itself, I guess I find it harder to divorce it from the ideas and values of the movement. Here I think Scottâs already said what I believe in a better way:
âa lot of media is predicting the death of EA, or a major blow to EA, or something in that category. Not going to happen. The media isnât good at understanding people who do things for reasons other than PR. But most EAs really believe. Like, really believe. If every single other effective altruist in the world were completely discredited, I would just shrug and do effective altruism on my own. If they instituted the death penalty for effective altruism, I would do it under cover of night using ZCash. And Iâm nowhere near the most committed effective altruist; honestly Iâm probably below average. âSaint gets eaten by lions in Colosseum, can early Christianity possibly survive this setback?â Update your model or prepare to be constantly surprised.â
Ideas have importance all on their own. The ideas that make up the philosophy of âEffective Altruismâ exist, and cannot be destroyed. People would believe them, want to co-ordinate on it. Then theyâd want to organise to help make their own ideas more efficient and boom, weâre just back to an EA movement all over again.
I guess where we most disagree given this would would be the tone/âimplications of your section 12.
a) I think most EAs are not yet ready to let go of the existence of an EA movement.
I donât really know, given above, that this is an option.
b) Still, itâs hard to see how tweaking EA can lead to a product that we should be excited about growing.
Iâm still excited about EA? I donât know how broad the âweâ is meant to be. I still want concern about reducing the suffering on non-human animals to grow, I still want humanity to expand its moral circle beyond the parochial, I still want us to find the actions individually and collectively that will lead to humanity flourishing. Apologies if Iâm misinterpreting, but this sentance really seems to come out of left-field from me given the rest of your post.
c) In this picture, EA would grow more slowly or shrink for a while, and maybe ultimately be overtaken by cause-specific communities.
I think again, given the ideas of EA exist, these cause-specific communities would find themselves connected again over time.
Yes, I had this exact quote in mind when I said in Sect 5 that âReligions can withstand persecution by totalitarian governments, and some feel just about as strongly about EA.â
People would believe them, want to co-ordinate on it. Then theyâd want to organise to help make their own ideas more efficient and boom, weâre just back to an EA movement all over again.
One of my main theses is supposed to be that people can and should coordinate their activities without acting like a movement.
I still want concern about reducing the suffering on non-human animals to grow, I still want humanity to expand its moral circle beyond the parochial, I still want us to find the actions individually and collectively that will lead to humanity flourishing. Apologies if Iâm misinterpreting, but this sentance really seems to come out of left-field from me given the rest of your post.
This feels like the same misunderstanding. Spreading EA ideas and values seems fine and good to me. Itâs the collectivism, branding, identity-based reasoning, and other âmovement-likeâ characteristics that concern me.
I think again, given the ideas of EA exist, these cause-specific communities would find themselves connected
This seems like black and white thinking to me. Of course these people will connect over their shared interests in consequentialism, RCTs, and so on. But this is different from branding and recruiting together, regulating this area as one community, hosting student chapters, etc.
Thanks for explaining your viewpoints Ryan. I think I have a better understanding, but Iâm still not sure I grok it intuitively. Let me try to repeat what I think is your view here (with the help of looking at some of your other quick takes)
note for readers, this is my understanding of Ryanâs thoughts, not what heâs said
1 > The EA movement was directly/âsignificantly causally responsible for the FTX disaster, despite being at a small scale (e.g. âthere are only ~10k effective altruistsâ)
2 > We should believe that without reform, similar catastrophes will contain to happen as the movement grows, which would lead to many more such FTXs catastrophes (i.e. if we get to a movement of 1 million EAs in the current movement, we should expect ~100 FTX-scale disasters)
3 > This outcome is morally unacceptable, so the EA movement shouldnât grow in its current form
4 > An alternative way to grow would be to continue to grow as a set of different, interconnected movements focused around direct action (e.g. The xRisk community, The animal welfare community, the Global Development community, The Longtermist community etc...)
5 > This would allow EA values to spread without the harms that we see occurring with EA as a movement
Iâm think I follow along. Iâm not sure about the extrapolation of FTX (would it scale linearly or logarithmically? Does it actually make any sense to extrapolate as if EA will continue the same way at all?) But that aside I think my main disagreement is to think why a set of separate fields/âcommunities that co-ordinate would be better at avoiding the failure modes you see in EA than the current one. I feel like âcollectivism, branding, identity-based reasoning, and other âmovement-likeâ characteristicsâ are going to occur whenever humans organise themselves into groups.
I think perhaps an underlying disagreement we have is about the power of ideas. I just donât think you can cleanly separate the EA movement from EA values. Ideas are powerful things which have logical and empirical consequences. The EA movement has grown so much so quickly, in my view, because its ideas and values are true[1] and convincing. That causes movements and not the other way around. I guess Iâm finding it difficult to picture what a movement-less EA would look like?
As an intuition pump, itâd be like a reformer saying Christians so just go to church on Sunday and listen to sermon, follow the commandments, read the bible, tithe, and do good works, but not bother with all of the Father/âSon/âHoly Ghost stuff. But that belief is the reason why theyâre doing the former. In a world where that was attempted to be removed, I think people would either stop doing the activities or reinvent them.
Iâm also worried about Leverage and various other cults and disasters, not just FTX.
I wouldnât think of the separate communities as âmovementsâ per se. Rather, each cause area would have a professional network of nonprofits and companies.
Basically, why do mid-sized companies usually not spawn cults and socially harm their members like movements like EA and the animal welfare community sometimes do? I think itâs because movements by their nature try to motivate members towards their goals, using social pressures. This attracts young idealists, some of whom will be impressionable. People will try radical stuff like traveling to locations where theyâre unsupported, going on intensive retreats, circling, drugs, polyamory, etc. These things benefit some people in some situations, but in they also can put people in vulnerable situations. My hypothesis is that predators detect this vulnerability and then start even crazier and more cultish projects, arguably including Leverage and FTX, under the guise of advancing the movementâs goals.
Companies rarely put junior staff in such vulnerable positions. People generally know not to sleep with subordinates, and better manage conflicts of interest. They donât usually give staff a pass for misbehaviour due to being value-aligned.
We donât need to lose our goals, or our social network, but we could strip away a lot of risk-increasing behaviour that âmovementsâ do, and take on some risk-reducing âprofessionalisingâ measures thatâs more typical of companies..
I agree that ideas are powerful things, and that people will continue to want to follow those ideas to their conclusions, in collaboration with others. But Iâm suggesting to be faithful to those ideas might be to shape up a little bit and practice them somewhat differently. For the case of Christianity, itâs not like telling Christians to disavow the holy Trinity. Itâs more like noticing abuse in a branch of Christianity, and thinking âweâve got to do some things differentlyâ. Except that EA is smaller and thousands of years younger, so can be more ambitious in the ways we try to reform.
JWS, do you think EA could work as a professional network of âimpact analystsâ or âimpact engineersâ rather than as a âmovementâ?
Ryan, do you have a sense of what that would concretely look like?
If we look at other professionals, for example, engineers have in common some key ideas, values, and broad goals (like âbuild things that workâ). Senior engineers recruit young engineers and go to professional conferences to advance their engineering skills and ideas. Some engineers work in policy or politics, but they clearly arenât a political movement. They donât assume engineering is a complete ethos for all major life decisions, and they donât assume that other engineers are trustworthy just because they are engineers.
I share your appreciation for EA ideas and think theyâll have longevity. I donât know that there is a way to push back against the pitfallls of being a social movement instead of just being a collection of professionals. But I agree with Ryan that if there were a way to just be a group of skilled colleagues rather than âbrethrenâ, it would be better. Social movements have the pitfalls of religions, tribes, and cults that most professions do not and fall prey to more demagogues as a result.
JWS, do you think EA could work as a professional network of âimpact analystsâ or âimpact engineersâ rather than as a âmovementâ?
I guess I still donât have a clear idea of what Ryanâs ânetwork of networksâ approach would look like without the âmovementâ aspect broadly defined. How definitely would that be practically from current EA but with more decentralisation of money and power, and more professional norms?
But would this be a set of rigid internal norms that prevent people from the philanthropy space connecting with those in specific cause areas? Are we going to split AI technical and governance fields strictly? Is nobody meant to notice the common philosophical ideas which underline the similar approaches to all these cause areas? Itâs especially the latter Iâm having trouble getting my head around.
Some engineers work in policy or politics, but they clearly arenât a political movement. They donât assume engineering is a complete ethos for all major life decisions, and they donât assume that other engineers are trustworthy just because they are engineers.
I donât think that âfield of engineeringâ is the right level of analogy here. I think the best analogies for EA are other movements, like âEnvironmentalismâ or âFeminismâ or âThe Enlightenmentâ.
Social movements have the pitfalls of religions, tribes, and cults that most professions do not and fall prey to more demagogues as a result.
Social movements have had a lot of consequences in the human history, some of them very positive and some very negative. It seems to me that you and Ryan think that thereâs a way to structure EA so that we can cleanly excise the negative parts of a movement and keep the positive parts without being a movement, and Iâm not sure thatâs really possible or even a coherent idea.
***
[to @RyanCarey I think you updated your other comment as I was thinking of my response, so folding in my thoughts on that here]
We donât need to lose our goals, or our social network, but we could strip away a lot of risk-increasing behaviour that âmovementsâ do, and take on some risk-reducing âprofessionalisingâ measures thatâs more typical of companies.
Iâm completely with you here, but to me this is something that ends up miles away from âwinding down EAâ, or EA being ânot a movementâ.
But Iâm suggesting to be faithful to those ideas might be to shape up a little bit and practice them somewhat differently. For the case of Christianity, itâs not like telling Christians to disavow the holy Trinity. Itâs more like noticing abuse in a branch of Christianity, and thinking âweâve got to do some things differentlyâ.
I think abuse might be a bit strong as an analogy but directionally I think this is correct, and Iâd agree we need to do things differently. But in this analogy I donât think the answer is end âChristianityâ as a movement and set up an overlapping network of tithing, volunteering, Sunday schools etc, which is what I take you to be suggesting. I feel like weâre closer to agreement here, but on reflection the details of your plan here donât sum up to âend EA as a movementâ at all.
JWS, do you think EA could work as a professional network of âimpact analystsâ or âimpact engineersâ rather than as a âmovementâ? Ryan, do you have a sense of what that would concretely look like?
Well Iâm not sure it makes sense to try to fit all EAs into one professional community that is labelled as such, since we often have quite different jobs and work in quite different fields. My model would be a patchwork of overlapping fields, and a professional network that often extends between them.
It could make sense for there to be a community focused on âeffective philanthropyâ, which would include OpenPhil, Longview, philanthropists, and grant evaluators. That would be as close to âimpact analysisâ as you would get, in my proposal.
There would be an effective policymaking community too.
And then a bevy of cause-specific research communities: evidence-based policy, AI safety research, AI governance research, global priorities research, in vitro meat, global catastrophic biorisk research, global catastrophic risk analysis, global health and development, and so on.
Lab heads and organisation leaders in these research communities would still know that they ought to apply to the âeffective philanthropyâ orgs to fund their activities. And they would still give talks at universities to try to attract top talent. But there wouldnât be a common brand or cultural identity, and we would frown upon the risk-increasing factors that come from the social movement aspect.
I think there was perhaps some miscommunication around your use and my interpretation of âcollapseâ. To me it implies that something is at an unrecoverable stage, like a âcollapsedâ building or support for a presidential candidate âcollapsingâ in a primary race. In your pinned quick take you posit that Effective Altruism as a brand may be damaged to an unrecoverable extent which makes me feel this is the right reading of your post, or at least it was a justified interpretation at least.
***
I actually agree with a lot of your claims in your reply. For example, I think that the last 12 months has been the worst the movement has ever faced. I think many inside the EA movement have lost a lot of faith and trust in both the movement as a whole and its leadership. I donât think anything I said in my comment would support the view that âthereâs no big problemâ, I guess I donât things have to be so binary as âEAâs reputation has collapsedâ or âitâs not that bad weâre back to normalâ.
Looking into the future, I think you do raise a very good point about risks from media that reflects on the FTX collapse and the role EA played. I think thatâs something that the movement does need to prepare for. I donât have specific suggestions myself, apart from believing that we need a more positive strategic direction to counter mistaken criticisms of us instead of leaving them unchallenged, and accepting reforms where they have merit, but I think thatâs a whole other discussion.
***
On the EA movement itself, I guess I find it harder to divorce it from the ideas and values of the movement. Here I think Scottâs already said what I believe in a better way:
Ideas have importance all on their own. The ideas that make up the philosophy of âEffective Altruismâ exist, and cannot be destroyed. People would believe them, want to co-ordinate on it. Then theyâd want to organise to help make their own ideas more efficient and boom, weâre just back to an EA movement all over again.
I guess where we most disagree given this would would be the tone/âimplications of your section 12.
I donât really know, given above, that this is an option.
Iâm still excited about EA? I donât know how broad the âweâ is meant to be. I still want concern about reducing the suffering on non-human animals to grow, I still want humanity to expand its moral circle beyond the parochial, I still want us to find the actions individually and collectively that will lead to humanity flourishing. Apologies if Iâm misinterpreting, but this sentance really seems to come out of left-field from me given the rest of your post.
I think again, given the ideas of EA exist, these cause-specific communities would find themselves connected again over time.
Yes, I had this exact quote in mind when I said in Sect 5 that âReligions can withstand persecution by totalitarian governments, and some feel just about as strongly about EA.â
One of my main theses is supposed to be that people can and should coordinate their activities without acting like a movement.
This feels like the same misunderstanding. Spreading EA ideas and values seems fine and good to me. Itâs the collectivism, branding, identity-based reasoning, and other âmovement-likeâ characteristics that concern me.
This seems like black and white thinking to me. Of course these people will connect over their shared interests in consequentialism, RCTs, and so on. But this is different from branding and recruiting together, regulating this area as one community, hosting student chapters, etc.
Thanks for explaining your viewpoints Ryan. I think I have a better understanding, but Iâm still not sure I grok it intuitively. Let me try to repeat what I think is your view here (with the help of looking at some of your other quick takes)
Iâm think I follow along. Iâm not sure about the extrapolation of FTX (would it scale linearly or logarithmically? Does it actually make any sense to extrapolate as if EA will continue the same way at all?) But that aside I think my main disagreement is to think why a set of separate fields/âcommunities that co-ordinate would be better at avoiding the failure modes you see in EA than the current one. I feel like âcollectivism, branding, identity-based reasoning, and other âmovement-likeâ characteristicsâ are going to occur whenever humans organise themselves into groups.
I think perhaps an underlying disagreement we have is about the power of ideas. I just donât think you can cleanly separate the EA movement from EA values. Ideas are powerful things which have logical and empirical consequences. The EA movement has grown so much so quickly, in my view, because its ideas and values are true[1] and convincing. That causes movements and not the other way around. I guess Iâm finding it difficult to picture what a movement-less EA would look like?
As an intuition pump, itâd be like a reformer saying Christians so just go to church on Sunday and listen to sermon, follow the commandments, read the bible, tithe, and do good works, but not bother with all of the Father/âSon/âHoly Ghost stuff. But that belief is the reason why theyâre doing the former. In a world where that was attempted to be removed, I think people would either stop doing the activities or reinvent them.
or true-enough, or seem true enough. Not claiming EA has anywhere near âultimate truthâ on any issue
Roughly yes, with some differences:
I think the disasters would scale sublinearly
Iâm also worried about Leverage and various other cults and disasters, not just FTX.
I wouldnât think of the separate communities as âmovementsâ per se. Rather, each cause area would have a professional network of nonprofits and companies.
Basically, why do mid-sized companies usually not spawn cults and socially harm their members like movements like EA and the animal welfare community sometimes do? I think itâs because movements by their nature try to motivate members towards their goals, using social pressures. This attracts young idealists, some of whom will be impressionable. People will try radical stuff like traveling to locations where theyâre unsupported, going on intensive retreats, circling, drugs, polyamory, etc. These things benefit some people in some situations, but in they also can put people in vulnerable situations. My hypothesis is that predators detect this vulnerability and then start even crazier and more cultish projects, arguably including Leverage and FTX, under the guise of advancing the movementâs goals.
Companies rarely put junior staff in such vulnerable positions. People generally know not to sleep with subordinates, and better manage conflicts of interest. They donât usually give staff a pass for misbehaviour due to being value-aligned.
We donât need to lose our goals, or our social network, but we could strip away a lot of risk-increasing behaviour that âmovementsâ do, and take on some risk-reducing âprofessionalisingâ measures thatâs more typical of companies..
I agree that ideas are powerful things, and that people will continue to want to follow those ideas to their conclusions, in collaboration with others. But Iâm suggesting to be faithful to those ideas might be to shape up a little bit and practice them somewhat differently. For the case of Christianity, itâs not like telling Christians to disavow the holy Trinity. Itâs more like noticing abuse in a branch of Christianity, and thinking âweâve got to do some things differentlyâ. Except that EA is smaller and thousands of years younger, so can be more ambitious in the ways we try to reform.
JWS, do you think EA could work as a professional network of âimpact analystsâ or âimpact engineersâ rather than as a âmovementâ?
Ryan, do you have a sense of what that would concretely look like?
If we look at other professionals, for example, engineers have in common some key ideas, values, and broad goals (like âbuild things that workâ). Senior engineers recruit young engineers and go to professional conferences to advance their engineering skills and ideas. Some engineers work in policy or politics, but they clearly arenât a political movement. They donât assume engineering is a complete ethos for all major life decisions, and they donât assume that other engineers are trustworthy just because they are engineers.
I share your appreciation for EA ideas and think theyâll have longevity. I donât know that there is a way to push back against the pitfallls of being a social movement instead of just being a collection of professionals. But I agree with Ryan that if there were a way to just be a group of skilled colleagues rather than âbrethrenâ, it would be better. Social movements have the pitfalls of religions, tribes, and cults that most professions do not and fall prey to more demagogues as a result.
I guess I still donât have a clear idea of what Ryanâs ânetwork of networksâ approach would look like without the âmovementâ aspect broadly defined. How definitely would that be practically from current EA but with more decentralisation of money and power, and more professional norms?
But would this be a set of rigid internal norms that prevent people from the philanthropy space connecting with those in specific cause areas? Are we going to split AI technical and governance fields strictly? Is nobody meant to notice the common philosophical ideas which underline the similar approaches to all these cause areas? Itâs especially the latter Iâm having trouble getting my head around.
I donât think that âfield of engineeringâ is the right level of analogy here. I think the best analogies for EA are other movements, like âEnvironmentalismâ or âFeminismâ or âThe Enlightenmentâ.
Social movements have had a lot of consequences in the human history, some of them very positive and some very negative. It seems to me that you and Ryan think that thereâs a way to structure EA so that we can cleanly excise the negative parts of a movement and keep the positive parts without being a movement, and Iâm not sure thatâs really possible or even a coherent idea.
***
[to @RyanCarey I think you updated your other comment as I was thinking of my response, so folding in my thoughts on that here]
Iâm completely with you here, but to me this is something that ends up miles away from âwinding down EAâ, or EA being ânot a movementâ.
I think abuse might be a bit strong as an analogy but directionally I think this is correct, and Iâd agree we need to do things differently. But in this analogy I donât think the answer is end âChristianityâ as a movement and set up an overlapping network of tithing, volunteering, Sunday schools etc, which is what I take you to be suggesting. I feel like weâre closer to agreement here, but on reflection the details of your plan here donât sum up to âend EA as a movementâ at all.
To be clear, winding down EA is something I was arguing we shouldnât be doing.
At a certain point it becomes semantic, but I guess readers can decide, when you put together:
the changes in sec 11 of the main post
ideas about splitting into profession-oriented subgroups, and
shifting whether we âmotivate members re social pressuresâ and expose junior members to risk
whether or not it counts as changing from being a âmovementâ to something else.
Fair.
Having run through the analogy, EA becoming more like an academic field or a profession rather than a movement seems very improbable.
I agree that âtry to reduce abuses common within the churchâ seems a better analogy.
Well Iâm not sure it makes sense to try to fit all EAs into one professional community that is labelled as such, since we often have quite different jobs and work in quite different fields. My model would be a patchwork of overlapping fields, and a professional network that often extends between them.
It could make sense for there to be a community focused on âeffective philanthropyâ, which would include OpenPhil, Longview, philanthropists, and grant evaluators. That would be as close to âimpact analysisâ as you would get, in my proposal.
There would be an effective policymaking community too.
And then a bevy of cause-specific research communities: evidence-based policy, AI safety research, AI governance research, global priorities research, in vitro meat, global catastrophic biorisk research, global catastrophic risk analysis, global health and development, and so on.
Lab heads and organisation leaders in these research communities would still know that they ought to apply to the âeffective philanthropyâ orgs to fund their activities. And they would still give talks at universities to try to attract top talent. But there wouldnât be a common brand or cultural identity, and we would frown upon the risk-increasing factors that come from the social movement aspect.