I take it this is a question for Jacob, right? I'll just chime in with one thought. - I think the comparison to wildlife suffering is relevant here too. Most wild animals live short lives and die of starvation, predation, disease of exposure. If the bar for net zero welfare is too high, it appears that one would be either pressed to drastically intervene and turn ecosystems upside down to avoid this suffering, or to eliminate all wildlife.
Thanks so much, this quasi-sociological perspective is quite helpful.
One thing that puzzles me is the role of intuition in this context. A few people have responded to the repugnant conclusion by saying that animals in CAFOs, even in cage-free poultry systems, have negative welfare. But that's not borne out by the empirical research on the topic. In my view, it's largely an unverified assumption, or intuition. That seems to run against the general project of "using reason and evidence to do the most good".
Similar tensions seemed apparent to me in what you write about stances of some effective altruists. You say that many EAs want to rely on reason rather than intuition, and don't consider their own moral intuitions trustworthy. But then you also say that they "consider consequentialism the strongest perspective to take — perhaps because they find it least counterintuitive." So, the acceptance of consequentialism itself is based on intuition.
The use of intuitions appears to be quite selective and arbitrary, when it serves prior commitments or helps to insulate parts of the worldview against objections.
That's a great question, Jacob, and I don't think I have a full answer!
I had funny conversations with my partner about this. He is (or was) an anti-natalist, and thought it is always wrong to "inflict existence" on someone. But I thought that I would always choose life over no-life, and would want to be born into virtually any context (provided the alternative is not being born at all). This is just to illustrate that people have widely different intuitions on this question.
I think the challenge for effective altruists is to find an answer to this question that is as much constrained by empirical evidence and good argument as possible.
Your concern was actually on my mind when writing a paper, and I made sure to address it head-on repeatedly. Let me just point to a few of the passages in the article that clearly refer to variations in the EA and effective animal advocacy movements:
"Effective animal advocates want to help animals as effectively as possible. I explore a popular way of spelling out this idea, according to which..." (second sentence in the abstract)
"Many effective animal advocates do not subscribe to a specific philosophical view but rather want to 'help animals as effectively as possible.' But if we try to add more specificity to this idea, the most popular interpretation is that when choosing between two actions to help animals, we should pick the one that maximizes the net aggregate welfare of animals." (first paragraph of the article)
"I call this more specific thesis effective animal altruism (EAA)." (Coining of the term "Effective Animal Altruism", in contrast to the more broadly used "Effective Animal Advocacy", to pick out a specific subgroup) (first paragraph).
"If the goal is to 'benefit animals as much as possible,' we need a different way of explaining what that means." (last line of the article)
My ascriptions of philosophical foundations to the movement was based on published articles by Will MacAskill, widely described as one of the founders or "originators" of the EA movement, with titles such as "What is Effective Altruism?" or "The definition of Effective Altruism", along with the book "Doing Good Better". Of course, as with any social movement, it is to be expected that there is a lot of variations in what individual members of the movement think and say.
He draws out welfarism and aggregationism as core commitments of the view. It's clear that there are many versions of utilitarianism, as well as many versions of welfarism (stronger an weaker) and many ways of aggregating welfare.
The main difference between utilitarianism and EA that MacAskill draws out is that utiliarians think you're doing something wrong when you're not maximizing utility, while EAs (on this presentation) don't think that. MacAskill describes EA as a "project", not a normative theory.
By "effective altruist" you can mean someone who is a member of the EA community. That's who you seem to have in mind. Members of the EA community of course are a heterogenous group, they hold many different views. Or you can mean someone who believes in the core commitments outlined above and is motivated by them. That's who I had in mind.
I agree with you that the phrasing "subscribing to a version of utilitarianism" slides over some nuances, since, as I said, effective altruism is not a normative theory. My text goes on to explain what I mean though.
Thanks so much for your response, Jacob. I really enjoy being in this conversation with you.
I'll take your questions point by point!
Regarding the logic of the larder: I do not endorse the claim that we do animals a favor by bringing them into existence for consumption, provided their lives have net positive welfare. My argument is instead meant as a reductio of the maximizing assumptions behind EAA. If those assumptions are accepted, they appear to push one toward something very close to the logic of the larder. So the point is not to defend that view, but to expose a troubling implication of the underlying framework.
On the Cumulative Pain Framework: I agree that whether a cage-free hen’s life is, all things considered, net positive is not a purely empirical finding. There is an irreducibly normative question here about how pleasures and pains should be weighed. I also agree that calling the threshold “a very low bar” is itself a normative claim. At the same time, I do not think “normative” means arbitrary or merely a matter of opinion. If effective altruists want to know how to do the most good, they should want these judgments to be constrained by empirical evidence as much as possible, even if the evidence does not by itself settle all evaluative questions.
On Frick: Utilitarians start with a definition of the good (=welfare), and then derive from this starting point an account of what you ought to do (do whatever maximizes welfare). Frick does not start with a definition of the good. Instead, he proposes that in a context c1, outcome o1 is better than outcome o2 if and only if you have overall most reason to bring about o1. So, he starts with a primitive notion of reasons and derives from that an account of outcome betterness (or the good). He discusses how he avoids the Mere Addition Paradox (and the repugnant conclusion) in this article: https://academic.oup.com/book/38952/chapter-abstract/338159303?redirectedFrom=fulltext My paper only gestures in that direction; it does not try to show that Frick’s view avoids every impossibility result or satisfies every desideratum in population ethics.
Relatedly, I do not mean in the paper to endorse the procreation asymmetry, or the claim that adding a happy life is simply neutral. In fact, I explicitly say that the neutrality view still ultimately faces the repugnant conclusion. The view I treat most sympathetically is Frick’s broader nonconsequentialist, reasons-first framework, which tries to avoid the repugnant conclusion not by saying flourishing lives add no value, but by rejecting the consequentialist assumption that value is always something to be promoted through maximization.
Finally, on “welfarism”: I agree that there is an important distinction between welfarism in the philosophical sense and “welfarism” in animal advocacy as a practical orientation toward on-farm welfare improvements. It is entirely possible to reject the former while still endorsing the latter as a tactic in pursuit of different goals. More generally, the same tactic can serve different ultimate aims. So I agree that my argument against maximizing net aggregate welfare does not by itself settle the strategic question of which interventions are most effective in practice. That is why I framed my conclusion relatively modestly: not that abolitionist or reductionist approaches are thereby shown to be superior, but that I hope the argument encourages greater interest in nonconsequentialist foundations and in strategies that are not guided by maximizing welfare alone. In practice, I have on several occasions actively supported welfare-oriented work, including work by organizations such as ACE, despite my doubts about the underlying philosophical framework.
Thanks, Jacob! From my perspective, the difference between CAFOs with negative welfare and those with slightly positive welfare is not very significant. Using metaphors, the zero point is often described by comparing it with suicidality. People with 0 welfare are indifferent wrt suicide. CAFOs with chickens whose welfare level is only a hair better than that do not appear like a good thing to me. As I say in the paper, a world with a small number of blissful chickens seems clearly better to me than a world with a large number of miserable chickens, even if their welfare is above 0. Since, from my perspective, the difference between CAFOs with positive and those with negative welfare is not very significant, this did not seem like a strong risk of misinterpretation to me. I can see that this would seem different, however, if for you the 0 welfare inflection point is crucially important.
What I'm understanding from what you're saying is this: some people might read my abstract and think that I argue that, if we ought to maximize net aggregate welfare, then we ought to build CAFOs of any kind, including ones in which animals have net negative welfare. Then they read my article and find out that, actually, my argument shows that if we ought to maximize net aggregate welfare, then we ought to build CAFOs of certain kinds, in which animals have net positive welfare. And they might feel disappointed or mislead by that.
To which my response is: Fair! Explaining the difference between CAFOs in which animals have net negative welfare and CAFOs in which they have net positive welfare in the abstract could potentially have forestalled certain misunderstandings.
I take it this is a question for Jacob, right? I'll just chime in with one thought. - I think the comparison to wildlife suffering is relevant here too. Most wild animals live short lives and die of starvation, predation, disease of exposure. If the bar for net zero welfare is too high, it appears that one would be either pressed to drastically intervene and turn ecosystems upside down to avoid this suffering, or to eliminate all wildlife.
Hi Kevin,
Thanks so much, this quasi-sociological perspective is quite helpful.
One thing that puzzles me is the role of intuition in this context. A few people have responded to the repugnant conclusion by saying that animals in CAFOs, even in cage-free poultry systems, have negative welfare. But that's not borne out by the empirical research on the topic. In my view, it's largely an unverified assumption, or intuition. That seems to run against the general project of "using reason and evidence to do the most good".
Similar tensions seemed apparent to me in what you write about stances of some effective altruists. You say that many EAs want to rely on reason rather than intuition, and don't consider their own moral intuitions trustworthy. But then you also say that they "consider consequentialism the strongest perspective to take — perhaps because they find it least counterintuitive." So, the acceptance of consequentialism itself is based on intuition.
The use of intuitions appears to be quite selective and arbitrary, when it serves prior commitments or helps to insulate parts of the worldview against objections.
Vera
That's a great question, Jacob, and I don't think I have a full answer!
I had funny conversations with my partner about this. He is (or was) an anti-natalist, and thought it is always wrong to "inflict existence" on someone. But I thought that I would always choose life over no-life, and would want to be born into virtually any context (provided the alternative is not being born at all). This is just to illustrate that people have widely different intuitions on this question.
I think the challenge for effective altruists is to find an answer to this question that is as much constrained by empirical evidence and good argument as possible.
Hi Leo,
Your concern was actually on my mind when writing a paper, and I made sure to address it head-on repeatedly. Let me just point to a few of the passages in the article that clearly refer to variations in the EA and effective animal advocacy movements:
My ascriptions of philosophical foundations to the movement was based on published articles by Will MacAskill, widely described as one of the founders or "originators" of the EA movement, with titles such as "What is Effective Altruism?" or "The definition of Effective Altruism", along with the book "Doing Good Better". Of course, as with any social movement, it is to be expected that there is a lot of variations in what individual members of the movement think and say.
Vera
Hi Joe,
My understanding of the relationship betwee effective altruism and utilitarianism is informed by this article by Will MacAskill on "What Is Effective Altruism": https://static1.squarespace.com/static/5506078de4b02d88372eee4e/t/68ac5735dad92b460f97786d/1756124981213/What+Is+Effective+Altruism+International+Encyclopedia+of+Ethics.pdf
He draws out welfarism and aggregationism as core commitments of the view. It's clear that there are many versions of utilitarianism, as well as many versions of welfarism (stronger an weaker) and many ways of aggregating welfare.
The main difference between utilitarianism and EA that MacAskill draws out is that utiliarians think you're doing something wrong when you're not maximizing utility, while EAs (on this presentation) don't think that. MacAskill describes EA as a "project", not a normative theory.
By "effective altruist" you can mean someone who is a member of the EA community. That's who you seem to have in mind. Members of the EA community of course are a heterogenous group, they hold many different views. Or you can mean someone who believes in the core commitments outlined above and is motivated by them. That's who I had in mind.
I agree with you that the phrasing "subscribing to a version of utilitarianism" slides over some nuances, since, as I said, effective altruism is not a normative theory. My text goes on to explain what I mean though.
Vera
Thanks so much for your response, Jacob. I really enjoy being in this conversation with you.
I'll take your questions point by point!
Finally, on “welfarism”: I agree that there is an important distinction between welfarism in the philosophical sense and “welfarism” in animal advocacy as a practical orientation toward on-farm welfare improvements. It is entirely possible to reject the former while still endorsing the latter as a tactic in pursuit of different goals. More generally, the same tactic can serve different ultimate aims. So I agree that my argument against maximizing net aggregate welfare does not by itself settle the strategic question of which interventions are most effective in practice. That is why I framed my conclusion relatively modestly: not that abolitionist or reductionist approaches are thereby shown to be superior, but that I hope the argument encourages greater interest in nonconsequentialist foundations and in strategies that are not guided by maximizing welfare alone. In practice, I have on several occasions actively supported welfare-oriented work, including work by organizations such as ACE, despite my doubts about the underlying philosophical framework.
Thanks again for the careful engagement, Jacob!
Thank you! I'll think it through and get back to you in 1-2 days.
I'm still curious, apart from how I worded the abstract, what's your take on the substance of the argument? If you're willing to share!
Thanks, Jacob! From my perspective, the difference between CAFOs with negative welfare and those with slightly positive welfare is not very significant. Using metaphors, the zero point is often described by comparing it with suicidality. People with 0 welfare are indifferent wrt suicide. CAFOs with chickens whose welfare level is only a hair better than that do not appear like a good thing to me. As I say in the paper, a world with a small number of blissful chickens seems clearly better to me than a world with a large number of miserable chickens, even if their welfare is above 0. Since, from my perspective, the difference between CAFOs with positive and those with negative welfare is not very significant, this did not seem like a strong risk of misinterpretation to me. I can see that this would seem different, however, if for you the 0 welfare inflection point is crucially important.
Thanks, Clara.
What I'm understanding from what you're saying is this: some people might read my abstract and think that I argue that, if we ought to maximize net aggregate welfare, then we ought to build CAFOs of any kind, including ones in which animals have net negative welfare. Then they read my article and find out that, actually, my argument shows that if we ought to maximize net aggregate welfare, then we ought to build CAFOs of certain kinds, in which animals have net positive welfare. And they might feel disappointed or mislead by that.
To which my response is: Fair! Explaining the difference between CAFOs in which animals have net negative welfare and CAFOs in which they have net positive welfare in the abstract could potentially have forestalled certain misunderstandings.