Scott Alexander and Freddie DeBoer both published articles on Effective Altruism in the past few days. Scott’s piece gives a load of examples of Effective Altruists doing good things. EAs have saved 200,000 human lives (mostly by paying for malaria nets). They’ve convinced farms to move 400 million chickens from tiny cages to slightly-less-unpleasant barns or outdoor areas. They’ve also given quite a lot of cash to weirder causes that are probably good: the
Nice piece. But I disagree that it's purely a movement. I think the core idea of trying to do good effectively is also really distinctive. As you say, "it’s insanely weird to actually put [these principles] into practice." So there's plenty of room to defend the idea of effective altruism as *really obviously good and worth pursuing* even if one questions whether Big EA actually does a good job of realizing its ideals.
Though if it's helpful to have a different name to distinguish the core ideas from the actual movement, I quite like "beneficentrism":
Enjoyed this non-totalising reaction, essentially to FdB's churlish piece. His is typical of an approach that assumes the validity of a very subjective deontological morality and uses that as the basis for disparaging a much less subjective mode of thought. I'm getting tired of it.
As for the weird stuff, deontological ethics leads one into stupid-seeming conclusions too. But, unlike utilitarianism, it's rarely as transparent about how it gets there. Anyway, thanks for putting a thoughtful, reasonable case.
This whole piece seems based on the idea that effective altruism is not an idea but a particular group of people, but that goes against how the concept is used in practice by so many of us (e.g. Singer, MacAskill, Ord)
> Effective Altruists claim that the whole EA schtick is a commitment to doing as much good as possible, but basically everyone would agree we should do a lot of good!
But is basically everyone doing it? I would basically agree with the top comment on Freddie's post ( https://freddiedeboer.substack.com/p/the-effective-altruism-shell-game/comment/44394364 ) that « isn't EA basically just "evidence based philanthropy." Like in medicine, I agree that suggesting that we use evidence to determine what works seems obvious, but like in medicine, it is actually not that common. »
I wonder if there's any kind of evidence that would convince you that e.g. Bill Gates is doing "effective altruism", by how most people use the term.
Most EAs are probably fairly weird (the weirdest ones are very weird). Weird people are attracted to EA, and EA probably makes people weirder. EAs generally know they're weird, but justify their weirdness as rationalism/utilitarianism. In the most part their weirdness isn’t malicious, but does a lot of good.
This is great. I read both pieces, wrote a response, then read Scott's response which I mostly agreed with. Then I read this and it's made me think less negatively about Freddie's piece and I've accepted some of the weirdness and it's benefits.
Initially my reaction to FDB's piece was very negative but after talking with people outside the EA movement and reading some accounts of complaints I think I understand the issue a bit more.
It's not just that people feel that EA is weird. They do and that's never a plus but that alone wouldn't explain the reaction. It's that they get the vibe that they are being told "I'm so much smarter than you idiots the normal rules don't apply to me and I'm better than you"
Those can be literal rules like the feeling that SBF was able to use EA as an excuse for outlandish behavior but more often just rules about what kinds of claims require substantial evidence. Usually we demand extra proof before we upend settled ideas of what's beneficial/good.
And I think if people got the sense that EA was just saying we should be more open to counterintuitive results that wouldn't be an issue. But I think the sense they get (to some extent rightly imo) is that it's privleging a certain kind of elite... nontraditional to be sure. There is a very real way that claims that are appealing to the kind of people who are into EA about x-risk and AI aren't treated the same way similar claims coming from a religious background would be. (to be clear this doesn't justify the reaction...we've been fine privleging trad groups forever...just understandable)
That's not to say that there is no validity to these claims but it's the same problem with all tribalism. EAers worried about AI x-risk or longtermism or whatever parse as one of us so even when they make bad arguments they get the mistaken but well-intentioned treatment. That feels very different to people on the outside.
In fact, I'd suggest it's pretty similar to the situation with maga voters and the left. If you're on the left the environmentalist who advocates a useless policy or cites bad facts is misguided but on well-intentioned so you shrug while you jump on the person arguing we shouldn't reduce fossil fuels. From inside it feels like not being a dick but from outside it feels very different.
Isn't EA both an idea and a social movement? Conservatism as well.
I agree that it would be a weird defense for a conservative MP to make but totally reasonable for some conservative pundit especially if they didn't want to defend the actual party.
I dunno if MacAskill wants to defend EAs as a movement or even what exactly that might mean since there are a number of different groups that all might be described in that fashion. I'm sure he thinks that some of them are worthwhile but suppose he cares much more about the idea than any movement he may wish to focus on the idea.
Five billion dollars a year, tens of millions of lives saved. Not Effective Altruists, just effective and altruists. And not congratulating themselves on their distinctive weirdness.
This really is like the San Francisco sourdough baking fraternity persuading itself that it invented bread and introduced the world to it.
I think MacAskill's comparison is a good one, but only. because I think his attitude to science is inappropriately reductive. Science is a way of establishing truths; it is also a body of knowledge acquired through that method, and a community of practitioners. It's common to downplay that last one but it is a reasonable definition, and it definitely was when science was first being established as a practice! Arguably EA is in a similar position to science in the 17th century: primarily a community of practitioners, soon to have acquired a large body of evidence about how to do good and with a great track record of having done so.
> If you’re trying to decide how to do good, and you only consider options that sound reasonable and normal to most people, you probably won't end up with the conclusion that we should donate huge amounts of our income to people abroad.
Sorry what? That's exactly what I think you would conclude.
Really enjoyed a lot of this piece, but the final paragraph seemed to suddenly jump to a weird conclusion ("you can't think bednets are important unless you also think AI is important").
Nice piece. But I disagree that it's purely a movement. I think the core idea of trying to do good effectively is also really distinctive. As you say, "it’s insanely weird to actually put [these principles] into practice." So there's plenty of room to defend the idea of effective altruism as *really obviously good and worth pursuing* even if one questions whether Big EA actually does a good job of realizing its ideals.
Though if it's helpful to have a different name to distinguish the core ideas from the actual movement, I quite like "beneficentrism":
https://rychappell.substack.com/p/beneficentrism
Enjoyed this non-totalising reaction, essentially to FdB's churlish piece. His is typical of an approach that assumes the validity of a very subjective deontological morality and uses that as the basis for disparaging a much less subjective mode of thought. I'm getting tired of it.
As for the weird stuff, deontological ethics leads one into stupid-seeming conclusions too. But, unlike utilitarianism, it's rarely as transparent about how it gets there. Anyway, thanks for putting a thoughtful, reasonable case.
It's interesting that you write "there’s not really such a thing as a small-e small-a Effective Altruist", while so many others disagree, e.g. https://forum.effectivealtruism.org/posts/gBiQfCZd26mu4CSqa/shouldn-t-effective-altruism-be-capitalized?commentId=ywNz95Rcjy5KwdLbd
This whole piece seems based on the idea that effective altruism is not an idea but a particular group of people, but that goes against how the concept is used in practice by so many of us (e.g. Singer, MacAskill, Ord)
> Effective Altruists claim that the whole EA schtick is a commitment to doing as much good as possible, but basically everyone would agree we should do a lot of good!
But is basically everyone doing it? I would basically agree with the top comment on Freddie's post ( https://freddiedeboer.substack.com/p/the-effective-altruism-shell-game/comment/44394364 ) that « isn't EA basically just "evidence based philanthropy." Like in medicine, I agree that suggesting that we use evidence to determine what works seems obvious, but like in medicine, it is actually not that common. »
I wonder if there's any kind of evidence that would convince you that e.g. Bill Gates is doing "effective altruism", by how most people use the term.
Most EAs are probably fairly weird (the weirdest ones are very weird). Weird people are attracted to EA, and EA probably makes people weirder. EAs generally know they're weird, but justify their weirdness as rationalism/utilitarianism. In the most part their weirdness isn’t malicious, but does a lot of good.
This is great. I read both pieces, wrote a response, then read Scott's response which I mostly agreed with. Then I read this and it's made me think less negatively about Freddie's piece and I've accepted some of the weirdness and it's benefits.
My response: https://fourofalltrades.substack.com/p/effective-altruism-has-good-consequences
Initially my reaction to FDB's piece was very negative but after talking with people outside the EA movement and reading some accounts of complaints I think I understand the issue a bit more.
It's not just that people feel that EA is weird. They do and that's never a plus but that alone wouldn't explain the reaction. It's that they get the vibe that they are being told "I'm so much smarter than you idiots the normal rules don't apply to me and I'm better than you"
Those can be literal rules like the feeling that SBF was able to use EA as an excuse for outlandish behavior but more often just rules about what kinds of claims require substantial evidence. Usually we demand extra proof before we upend settled ideas of what's beneficial/good.
And I think if people got the sense that EA was just saying we should be more open to counterintuitive results that wouldn't be an issue. But I think the sense they get (to some extent rightly imo) is that it's privleging a certain kind of elite... nontraditional to be sure. There is a very real way that claims that are appealing to the kind of people who are into EA about x-risk and AI aren't treated the same way similar claims coming from a religious background would be. (to be clear this doesn't justify the reaction...we've been fine privleging trad groups forever...just understandable)
That's not to say that there is no validity to these claims but it's the same problem with all tribalism. EAers worried about AI x-risk or longtermism or whatever parse as one of us so even when they make bad arguments they get the mistaken but well-intentioned treatment. That feels very different to people on the outside.
In fact, I'd suggest it's pretty similar to the situation with maga voters and the left. If you're on the left the environmentalist who advocates a useless policy or cites bad facts is misguided but on well-intentioned so you shrug while you jump on the person arguing we shouldn't reduce fossil fuels. From inside it feels like not being a dick but from outside it feels very different.
Isn't EA both an idea and a social movement? Conservatism as well.
I agree that it would be a weird defense for a conservative MP to make but totally reasonable for some conservative pundit especially if they didn't want to defend the actual party.
I dunno if MacAskill wants to defend EAs as a movement or even what exactly that might mean since there are a number of different groups that all might be described in that fashion. I'm sure he thinks that some of them are worthwhile but suppose he cares much more about the idea than any movement he may wish to focus on the idea.
https://www.theglobalfund.org/en/
Five billion dollars a year, tens of millions of lives saved. Not Effective Altruists, just effective and altruists. And not congratulating themselves on their distinctive weirdness.
This really is like the San Francisco sourdough baking fraternity persuading itself that it invented bread and introduced the world to it.
I think MacAskill's comparison is a good one, but only. because I think his attitude to science is inappropriately reductive. Science is a way of establishing truths; it is also a body of knowledge acquired through that method, and a community of practitioners. It's common to downplay that last one but it is a reasonable definition, and it definitely was when science was first being established as a practice! Arguably EA is in a similar position to science in the 17th century: primarily a community of practitioners, soon to have acquired a large body of evidence about how to do good and with a great track record of having done so.
> If you’re trying to decide how to do good, and you only consider options that sound reasonable and normal to most people, you probably won't end up with the conclusion that we should donate huge amounts of our income to people abroad.
Sorry what? That's exactly what I think you would conclude.
Really enjoyed a lot of this piece, but the final paragraph seemed to suddenly jump to a weird conclusion ("you can't think bednets are important unless you also think AI is important").
Besides, private charity is about 1% as effective as sound government policies directed towards the same ends.