Not a happy message: >the aggregate sentience losses of animals in the past half century outstrip the sentience gains to humans from population growth by at least 10,000-fold. Breaking even in welfare terms requires assigning a correspondingly small welfare weight to animals.

In a way this paper is silly. It's a more serious attempt of the shrimp argument: "Yes, you should save 10^100 shrimp instead of one human"https://morallawwithin.substack.com/p/yes-you-should-save-10...

But, since I think of myself as mostly utilitarian, I do need to take such arguments seriously. Either I need a good counterargument to this paper, or I should question my moral philosophy

In addition, I devour biology anecdotes, and everything I hear increases my belief that animals are more sentient & intelligent than our common sense suggests. E.g. consider how sophisticated hermit crabs are in trading off pain vs. a valuable home: https://www.sciencedirect.com/science/article/abs/pii/S00033...

With over 40 pages, I only barely skimmed this paper so far. My reactions are:

Tactical:

1) Are they really measuring biodiversity (vs. biomass)? I'm skeptical that weighted biomass has decreased so much.

2) The effect is bigger the further you get from humans taxonomically. e.g. it's all about the shrimp ;-P3) The assumptions highlight how little we actually understand about welfare and sentience, in any animal, including us. Neuron counts and cognitive tests are poor measures.

Philosophical:

1) This highlights our uncertainty about sentience and welfare in nature. This uncertainty provides some justification for the status quo. But it also argues we should work to reduce it.

2) Arguments like this highlight that one should be skeptical of utilitarian arguments that are scaled so far.

3) But overall, the utilitarian approach is better than others, and thus this question is important to ask. The lack of a good answer doesn't invalidate the question. It should drive us to improve the answer.