3 ms·

My main critique to effective altruism is that, because it is based on utilitarianism, it is a perfectly good framework with 0 ways to apply it practically. For

My main critique to effective altruism is that, because it is based on utilitarianism, it is a perfectly good framework with 0 ways to apply it practically. For example, show me the calculation where they take into account the probability of climate solving technology comes from supporting schools in Detroit. The amount of ALL small likelihood events with large impacts need to be taken into account. I emphasize ALL because otherwise you can just pick and choose the events you care about. Then, the next problem is quantifying the probability. Again, it is mostly assumptions and these assumptions dictate what you believe. If you believe "AI is the highest risk to humanity", utilitarianism doesn't matter as you are already deciding the probabilities. At that point, you are just doing performative art to make your belief feel legitimate. I do think that people should be encouraged to donate more (and should be progressive by income, not a fixed 10%) but I also think the organization misses the whole side of volunteering or personal actions, although they have "altruistic jobs". For example, converting to veganism would greatly outweigh 10% of a median salary to the most effective climate charity.

colinmhayes 4y ago
As effective altruists love to claim, the core tenants aren't really based on utilitarianism. Yes, pretty much all the commentators are utilitarians, but the core stance of "you should help others, and you should do so in an effective way" doesn't actually say anything about what effective means. You can effectively seperate the church from the religion here. Yes, givewell is an inherently utilitarian organization, but being EA doesn't mean just doing whatever givewell tells you. It means you're putting some level of thought into the best way to help people and you're acting on your conclusions. Converting to veganism is something EA advocates for constantly by the way, it's pretty much the first thing you'll see when looking into animal welfare. I think the biggest issue is that people really like meat, and being told they should stop eating it is likely to turn them off of animal welfare all together, so organizations bury the lede sometimes.
- skybrian 4y ago
  Just doing what GiveWell says seems fine? Why not?
lrobinovitch 4y ago
> converting to veganism would greatly outweigh 10% of a median salary to the most effective climate charity This claim itself is based on probabilities assigned to the potential impacts of veganism and donations. I would argue that this is a form of practical utilitarianism - assigning probabilities of event occurrence, weighted by their negative impacts, and attempting to do the most good despite the uncertainties and evolving information landscapes.
- april_22 4y ago
  I also think that people shouldn't overthink this. You shouldn't be spending hours trying to calculate which charity could have most impact in 25 years. Donating is good. Not eating meat is good. Doing both is even better. Just try to help as much as you are willing to help.
- solveit 4y ago
  The word "utilitarianism" means so many different things that it usually hurts discussions more than it helps. For some people, utilitarianism is a maximally permissive framework that allows you to discuss any set of preferences following the VNM axioms (I won't get into technicalities, but these axioms basically just say that your preferences are minimally coherent. This is not a real restriction on your ethics) by couching it in the language of a "utility function". This does not say anything about your ethics!. In this point of view, practically everyone who has ever lived can be cast as a utilitarian with some appropriate utility function! For some people, "utilitarianism" implies much more stringent constraints on your utility function and also come with moral imperatives about how to act (notice the previous interpretation tells you what is good with respect to some utility function, but doesn't tell you that you should do anything). For instance, it might imply that you view utilities as additive, so two happy people are twice as valuable as one. And it might also command you to maximize expected utility, or worst-case utility, or it might be a straw-man of sane utilitarianism and command you to complete the impossible task of precisely computing the utility of every possible world-state before being able to ethically take any action. And of course, there are people who think utilitarianism is anything between these two examples. For example, most flavours of consequentialism can be sensibly described as a form of utilitarianism and some people say utilitarianism when they mean consequentialism. Anyway, the word "utilitarianism" encompasses a ridiculously wide range of views, and when two people with wildly differing interpretations of the word talk to each other, they do not get a productive discussion.
ouid 4y ago
modern american broiler chickens can turn feed corn into chicken meat calories at a rate of around 50%, and feed corn is our most efficient crop per area by about a factor of two, after potatoes, I think, but isnt really human food. thats just back of the envelope stuff, but i think you might be doing some of that working backwards yourself.
- drekk 4y ago
  1.7kg of feed gets you 1kg of chicken[0]. for a crop that humans can't even eat. and then you eat the american broiler chicken for calories, only getting 10% of them. if you get 500 calories from eating some chicken, that took 5,000 calories of chicken which took almost 10,000 calories of feed. this is assuming you eat all the chicken, including organ meat, feet, etc. trophic levels are fun! this is why you can get "funny numbers" like the napkin math that lets you know the US produces enough food each year to feed 10-20% more than the entire world population. it mostly goes to animal feed or landfills though. aside from the obvious animal cruelty it's wasteful. and the corporate capture from animal agribusiness means fruits and vegetables are artificially more expensive than meat and animal products. wonder how that ties into the obesity epidemic and poverty? I don't get it. I really don't. From any angle I choose to look at it (ethical, health, environmental) meat seems like a terrible choice. Why is it that during perma-drought I can buy a burger every day for a few dollars, and each one of those patties took over a thousand gallons of water to produce? [0] https://seekingalpha.com/article/4476215-its-about-chicken-feed-investing-poultry-industry https://seekingalpha.com/article/4476215-its-about-chicken-f...
  - ouid 4y ago
    uh, you just threw in a bullshit factor of 10 on the caloric content of chicken. a kg of assorted skeletal chicken meat is about 2500 kcal. A broiler is about 70% meat. Most crops are less than a quarter as land efficient as feed corn. Chicken is as environmentally friendly as those crops.
skybrian 4y ago
I think that’s generally true of utilitarianism, but utilitarianism being very difficult to put into practice isn’t an excuse to avoid making any attempt to determine effectiveness. For example, if someone goes from “I don’t believe in utilitarianism” to “therefore there is no reason to do trials to determine which medicines are effective because any medicine could indirectly do something good, who knows” then I think something has gone wrong? Similarly, it doesn’t mean we can’t make rough guesses about effectiveness and avoid all the charities that, while well-intended, are unlikely to move the needle. Somehow, though, some EA people convince themselves to donate to very speculative causes anyway. I think the answer is just: don’t do that? You can stick with things that are likely to work. And there are not zero ways to put it into practice. Donating through GiveWell is something you can do.
joshuamorton 4y ago
> For example, converting to veganism I know multiple people (like, a dozen maybe?) who were convinced to become vegan due to EA. One specifically did so because at the time they didn't feel they could donate a significant amount to charity, and so felt veganism would be higher value. Of course the reason here was animal cruelty, not climate (veganism is likely worse than vegitarianism or careful omnivorism from a strictly climate perspective).
- ZeroGravitas 4y ago
  Any links on that? I don't think veganism would be much better than vegetarianism or something like the EAT-lancet diet, but can't quite see how it could be worse from a climate perspective? Meat and dairy are usually the big ones. (Though personal diet choices are probably meaningless, unless you can start a movement, which veganism seems to have done)
tlb 4y ago
Some benefits can be measured, some can't. Mosquito nets can be measured. Interventions to reduce AI risk can never be measured, because we'll only ever get one data point (many years from now) and evaluating an intervention needs at least two data points. I don't think anyone working on, say, AI risk is claiming that their work is 1.37x more valuable than mosquito nets. Rather, they claim that there are plausible assumptions under which it's highly valuable. If I also find their assumptions plausible, I'm now persuaded they're working on something worthwhile. Great! It distinguishes them from lots of people working on stuff that definitely isn't. Utilitarianism doesn't need all the numbers to be precise. You really only need to get the signs right to improve your decision making. Often, you can get different signs for overall utility from two different assumptions that are both plausible. That's an important signal for which assumptions to worry about. The assumptions that only change the magnitude by a digit or two aren't that important.

colinmhayes 4y ago

As effective altruists love to claim, the core tenants aren't really based on utilitarianism. Yes, pretty much all the commentators are utilitarians, but the core stance of "you should help others, and you should do so in an effective way" doesn't actually say anything about what effective means. You can effectively seperate the church from the religion here. Yes, givewell is an inherently utilitarian organization, but being EA doesn't mean just doing whatever givewell tells you. It means you're putting some level of thought into the best way to help people and you're acting on your conclusions. Converting to veganism is something EA advocates for constantly by the way, it's pretty much the first thing you'll see when looking into animal welfare. I think the biggest issue is that people really like meat, and being told they should stop eating it is likely to turn them off of animal welfare all together, so organizations bury the lede sometimes.

skybrian 4y ago
Just doing what GiveWell says seems fine? Why not?

lrobinovitch 4y ago

> converting to veganism would greatly outweigh 10% of a median salary to the most effective climate charity This claim itself is based on probabilities assigned to the potential impacts of veganism and donations. I would argue that this is a form of practical utilitarianism - assigning probabilities of event occurrence, weighted by their negative impacts, and attempting to do the most good despite the uncertainties and evolving information landscapes.

april_22 4y ago
I also think that people shouldn't overthink this. You shouldn't be spending hours trying to calculate which charity could have most impact in 25 years. Donating is good. Not eating meat is good. Doing both is even better. Just try to help as much as you are willing to help.
solveit 4y ago
The word "utilitarianism" means so many different things that it usually hurts discussions more than it helps. For some people, utilitarianism is a maximally permissive framework that allows you to discuss any set of preferences following the VNM axioms (I won't get into technicalities, but these axioms basically just say that your preferences are minimally coherent. This is not a real restriction on your ethics) by couching it in the language of a "utility function". This does not say anything about your ethics!. In this point of view, practically everyone who has ever lived can be cast as a utilitarian with some appropriate utility function! For some people, "utilitarianism" implies much more stringent constraints on your utility function and also come with moral imperatives about how to act (notice the previous interpretation tells you what is good with respect to some utility function, but doesn't tell you that you should do anything). For instance, it might imply that you view utilities as additive, so two happy people are twice as valuable as one. And it might also command you to maximize expected utility, or worst-case utility, or it might be a straw-man of sane utilitarianism and command you to complete the impossible task of precisely computing the utility of every possible world-state before being able to ethically take any action. And of course, there are people who think utilitarianism is anything between these two examples. For example, most flavours of consequentialism can be sensibly described as a form of utilitarianism and some people say utilitarianism when they mean consequentialism. Anyway, the word "utilitarianism" encompasses a ridiculously wide range of views, and when two people with wildly differing interpretations of the word talk to each other, they do not get a productive discussion.

ouid 4y ago

modern american broiler chickens can turn feed corn into chicken meat calories at a rate of around 50%, and feed corn is our most efficient crop per area by about a factor of two, after potatoes, I think, but isnt really human food. thats just back of the envelope stuff, but i think you might be doing some of that working backwards yourself.

drekk 4y ago
1.7kg of feed gets you 1kg of chicken[0]. for a crop that humans can't even eat. and then you eat the american broiler chicken for calories, only getting 10% of them. if you get 500 calories from eating some chicken, that took 5,000 calories of chicken which took almost 10,000 calories of feed. this is assuming you eat all the chicken, including organ meat, feet, etc. trophic levels are fun! this is why you can get "funny numbers" like the napkin math that lets you know the US produces enough food each year to feed 10-20% more than the entire world population. it mostly goes to animal feed or landfills though. aside from the obvious animal cruelty it's wasteful. and the corporate capture from animal agribusiness means fruits and vegetables are artificially more expensive than meat and animal products. wonder how that ties into the obesity epidemic and poverty? I don't get it. I really don't. From any angle I choose to look at it (ethical, health, environmental) meat seems like a terrible choice. Why is it that during perma-drought I can buy a burger every day for a few dollars, and each one of those patties took over a thousand gallons of water to produce? [0] https://seekingalpha.com/article/4476215-its-about-chicken-feed-investing-poultry-industry https://seekingalpha.com/article/4476215-its-about-chicken-f...
- ouid 4y ago
  uh, you just threw in a bullshit factor of 10 on the caloric content of chicken. a kg of assorted skeletal chicken meat is about 2500 kcal. A broiler is about 70% meat. Most crops are less than a quarter as land efficient as feed corn. Chicken is as environmentally friendly as those crops.

skybrian 4y ago

I think that’s generally true of utilitarianism, but utilitarianism being very difficult to put into practice isn’t an excuse to avoid making any attempt to determine effectiveness. For example, if someone goes from “I don’t believe in utilitarianism” to “therefore there is no reason to do trials to determine which medicines are effective because any medicine could indirectly do something good, who knows” then I think something has gone wrong? Similarly, it doesn’t mean we can’t make rough guesses about effectiveness and avoid all the charities that, while well-intended, are unlikely to move the needle. Somehow, though, some EA people convince themselves to donate to very speculative causes anyway. I think the answer is just: don’t do that? You can stick with things that are likely to work. And there are not zero ways to put it into practice. Donating through GiveWell is something you can do.

joshuamorton 4y ago

> For example, converting to veganism I know multiple people (like, a dozen maybe?) who were convinced to become vegan due to EA. One specifically did so because at the time they didn't feel they could donate a significant amount to charity, and so felt veganism would be higher value. Of course the reason here was animal cruelty, not climate (veganism is likely worse than vegitarianism or careful omnivorism from a strictly climate perspective).

ZeroGravitas 4y ago
Any links on that? I don't think veganism would be much better than vegetarianism or something like the EAT-lancet diet, but can't quite see how it could be worse from a climate perspective? Meat and dairy are usually the big ones. (Though personal diet choices are probably meaningless, unless you can start a movement, which veganism seems to have done)

tlb 4y ago

Some benefits can be measured, some can't. Mosquito nets can be measured. Interventions to reduce AI risk can never be measured, because we'll only ever get one data point (many years from now) and evaluating an intervention needs at least two data points. I don't think anyone working on, say, AI risk is claiming that their work is 1.37x more valuable than mosquito nets. Rather, they claim that there are plausible assumptions under which it's highly valuable. If I also find their assumptions plausible, I'm now persuaded they're working on something worthwhile. Great! It distinguishes them from lots of people working on stuff that definitely isn't. Utilitarianism doesn't need all the numbers to be precise. You really only need to get the signs right to improve your decision making. Often, you can get different signs for overall utility from two different assumptions that are both plausible. That's an important signal for which assumptions to worry about. The assumptions that only change the magnitude by a digit or two aren't that important.