Cross-posted to my website.
Like the last time I wrote something like this, my suggestions here could apply to any large foundation. But most large foundations don’t care at all about what I say, and the Open Philanthropy Project cares at least a tiny bit about what I say, so I’m going to focus on Open Phil.
The Open Philanthropy Project ought to prioritize wild animal suffering (WAS). Here’s why:
- WAS is important and neglected.
- WAS is not tractable for most actors, but it’s tractable for Open Phil.
(Previously I discussed some of my issues with the importance/neglectedness/tractability framework, but I believe it works reasonably well for our purposes here.)
Why wild animal suffering matters
The problem of wild animal suffering has enormous scale. There exist far more sentient wild animals than there do humans or factory-farmed animals. Wild animal suffering dwarfs all other problems that currently exist. Some other problems (such as existential risk) may matter more, but WAS is certainly the biggest problem that’s happening right now.
Additionally, wild animal suffering is neglected: hardly anyone cares about this problem, and of the people who care, hardly any of them are trying to do anything about it. Animal Ethics is the only organization spending non-trivial time on the problem of wild animal suffering, and it’s a small organization with limited staff time and narrow focus–I see room for much, much more work on reducing suffering in the wild than what Animal Ethics does currently.
Why Open Phil should prioritize wild animal suffering
For people who care about animals, their biggest objection to reducing wild animal suffering is that it’s intractable. But this is mistaken: we can do lots of things right now to work toward reducing wild animal suffering. (If you doubt that we can do anything about wild animal suffering, please, please read my essay on this subject, and if you disagree, leave a comment explaining why.)
Even given the sad state of WAS research, we already have some concrete proposals for how to reduce wild animal suffering without risking big negative side effects. For example, Brian Tomasik has suggested paying farmers to use humane insecticides. Calculations suggest that this could prevent 250,000 painful deaths per dollar. This intervention alone looks much more cost-effective than GiveDirectly even if we heavily discount insects’ capacity for suffering. And this is just an initial idea; surely there exist much more effective interventions than this, and we could find them if we spent more time looking.
Reducing suffering in the wild is probably much more tractable than most people tend to think. That said, if you want to work on wild animal suffering, you either need specific relevant skills (which are rare and hard to develop) or you need to fund an organization doing relevant work; and right now Animal Ethics is the only such organization. We have something of a coordination problem here where people won’t work on wild animal suffering because they can’t get funding, and people don’t want to fund it because so few people are working on it.
What we need is a large, committed source of funding to jump-start the cause. If the Open Philanthropy Project began funding work on wild animal suffering, it could stimulate new research efforts or small-scale interventions by offering grants. Specifically, Open Phil should probably create a new focus area for wild animal suffering and possibly hire dedicated staff. This problem has such large scale, and so many possible interventions, that it absolutely deserves to be a dedicated focus area. Open Phil might consider lumping WAS under its farm animal welfare program, but this would excessively constrain its budget and limit the amount of staff time that it could receive. Wild animal suffering is a massive problem, and easily deserves as much attention as most of Open Phil’s other focus areas.
How many painful mosquito deaths would you have to be offered to prevent to choose that over causing one new human life (of quality equal to that of a typical person today) to be lived (all instrumental effects / consequences aside)?[1][2][3] (For my answer see [2].)
What would the distribution of EAs' answers look like? College graduates' answers? Everyone's answers?
What range of answers does the OP assume?
Or more broadly, for what range of moral theories can a case be made that WAS should be prioritized?
I ask these questions because, while I find the OP argument intriguing, my current values (or my current beliefs about my values, depending on how you want to think about it) are such that preventing mosquito suffering is very insignificant relative to many other things (e.g. there being more humans that live good lives, or humans living better lives) and is therefore far from being a high priority for me.
While I haven't dived deeply into arguments for negative utilitarianism or other arguments that could conceivably change my view significantly, I think it's unlikely (~10%, reported in [2]) that doing so would lead me to change my view significantly.[4]
It seems to me that the most probable way that my view could be changed to believe that (e.g.) OPP ought to prioritize WAS would be to persuade me that I should adopt a certain view on how to deal with moral uncertainty that would, if adopted, imply that OPP ought to prioritize WAS even given my current beliefs about how much I value the suffering of mosquitoes relative to other things (e.g. the lives of humans).
Is there a case to be made for prioritizing WAS if one assigns even a small probability (e.g. 1%) to a negative utilitarian-like view being correct given that they also subscribe to certain plausible views on moral uncertainty?
My views on how to deal with moral uncertainty are very underdeveloped. I think I currently have a tendency to evaluate situations or decide on actions on the basis of the moral view I deem most probable, however as the linked LessWrong wiki article points out, this has potential problems. (I'm also not aware of a less problematic view, so I will probably continue to do this until I encounter something else that appeals to me more. Bostrom's parliamentary model seems like a reasonable candidate, although I'm unsure how this negotiation process works exactly or would play out. Would have to think about it more.
Lastly, let me just note that I don't challenge the non-normative factual claims of the OP. Rather, I'm simply stating that my hesitation to take the view that OPP should prioritize WAS comes from my belief that I value things significantly differently than I would have to in order for WAS to be something that OPP should prioritize.
{1] A similar question was asked in the Effective Altruism Facebook group. My version gets at how much one values the life of a typical person today relative to the life of a typical mosquito rather than how much one values extreme pleasure relative to extreme suffering.
[2] Since I'm asking for others' answers, I should estimate my own answer. Hmm. If I had to make the decision right now I would choose to create the new human life, even if the number of painful mosquito deaths I was offered to prevent was infinite. Although note that I am not completely confident in this view, perhaps only ~60%. Then maybe ~30% to 10^10-infinity and ~10% to <10^10 mosquitoes, where practically all of that 10% uncertainty comes from the possibility that a more enlightened version of myself would undergo a paradigm shift or significant change in my fundamental values / moral views. In other words, I'm pretty uncertain (~40/60) about whether mosquitoes are net negative or not, but I'm pretty certain (~75%=30%/40%) that if I do value them negatively that the magnitude of their negative value is quite small (e.g. relative to the positive value I place on (the conscious experience of) human life).
[3] Knowing that my view is controversial among EAs (see the link at [1]), perhaps I should meta-update significantly towards the consensus view that not only is the existence of suffering inherently bad, but it's also a much greater magnitude bad than I think in the ~30% scenario that it is. I'll refrain from doing this for now, or figuring out how much I should update if I only think there's an X% that it's proper to update. (I'm also not sure how much my intuitions / current reported estimates already take into account others estimates or not.)
[4] The basis of my view that the goodness of a human life is much greater than the possible (~40% in my view) badness of a mosquito's suffering or painful death (and the basis of more general versions of this view) is my intuition. Thinking about the question from different angles I have been unable to shift my view significantly towards placing substantially more value on mosquitoes' significance or preventing mosquito suffering.
Even if you discount insects that heavily (which I believe is wrong), there's still a strong case to be made for trying to prevent wild vertebrates from suffering.