However, unless we assign only an extremely (and arguably, implausibly) low weighting to animals, their sheer numbers mean that they are still likely to dominate humans by several orders of magnitude.3
[...]
3Current attempts to weight based on neuron count (e.g. MacAskill 2022) are unconvincing (see Shriver 2022 [report by Adam Shriver which was part of Rethink Priorities (RPâs) moral weight project]). [As a side note, you could add the footnotes to the EA Forum post such that people do not have to check the end of the substack post.]
I think the total welfare of humans may be larger than the absolute value of the total welfare of animals more easily than you suggest above. Adamâs report concludes âthere is no straightforward empirical evidence or compelling conceptual arguments indicating that relative differences in neuron counts within or between species reliably predicts welfare relevant functional capacitiesâ. However, as illustrated in the graph below, the estimates for the (expected) welfare ranges in Bob Fischerâs book about comparing welfare across species, which contains what RP stands behind now, are pretty well explained by âindividual number of neurons as a fraction of that of humansâ^0.188.
Assuming individual welfare per fully-healthy-animal-year is proportional to âindividual number of neuronsâ^âexponent of the number of neuronsâ, I estimate the total welfare of humans is larger than the absolute value of the total welfare of:
Cattle, hens, broilers, and farmed black soldier fly (BSF) larvae and mealworms, finfishes, and shrimps for an exponent of at least 0.61.
Below is the graph illustrating this. I believe the exponent may easily be higher than 1, and therefore would not be surprised if the total welfare of humans was much larger than the absolute value of the total welfare of animals.
Thanks Vasco! Please feel free to post your entire response in the comment section on Substack, as this is where most of our readers are. I will write more on the Welfare Ranges project later, but I will say that I am quite skeptical of neural counts as a proxy for welfare ranges.
You are welcome! I only posted the comment here to avoid having to keep 2 threads updated. You may consider linking to your EA Forum crossposts at the start of your substack posts.
I will write more on the Welfare Ranges project later
Good to know!
I am quite skeptical of neural counts as a proxy for welfare ranges
Are you sceptical of welfare ranges being proportional to âindividual number of neuronsâ^âexponentâ? I think it makes a lot of sense that one is spectical of any particular exponent. For example, of welfare ranges being proportional to the individual number of neurons, which corresponds to an exponent of 1. However, being sceptical about any exponent requires much more confidence. For instance, it would require being sceptical of the welfare ranges presented in Bobâs book, which are pretty well explained by an exponent of 0.188 (see 1st graphs in my last comment).
Thanks for the post, Walter and Heather!
I think the total welfare of humans may be larger than the absolute value of the total welfare of animals more easily than you suggest above. Adamâs report concludes âthere is no straightforward empirical evidence or compelling conceptual arguments indicating that relative differences in neuron counts within or between species reliably predicts welfare relevant functional capacitiesâ. However, as illustrated in the graph below, the estimates for the (expected) welfare ranges in Bob Fischerâs book about comparing welfare across species, which contains what RP stands behind now, are pretty well explained by âindividual number of neurons as a fraction of that of humansâ^0.188.
Assuming individual welfare per fully-healthy-animal-year is proportional to âindividual number of neuronsâ^âexponent of the number of neuronsâ, I estimate the total welfare of humans is larger than the absolute value of the total welfare of:
Cattle, hens, broilers, and farmed black soldier fly (BSF) larvae and mealworms, finfishes, and shrimps for an exponent of at least 0.61.
Soil ants, termites, springtails, mites, and nematodes for an exponent of at least 1.17.
Below is the graph illustrating this. I believe the exponent may easily be higher than 1, and therefore would not be surprised if the total welfare of humans was much larger than the absolute value of the total welfare of animals.
Thanks Vasco! Please feel free to post your entire response in the comment section on Substack, as this is where most of our readers are. I will write more on the Welfare Ranges project later, but I will say that I am quite skeptical of neural counts as a proxy for welfare ranges.
You are welcome! I only posted the comment here to avoid having to keep 2 threads updated. You may consider linking to your EA Forum crossposts at the start of your substack posts.
Good to know!
Are you sceptical of welfare ranges being proportional to âindividual number of neuronsâ^âexponentâ? I think it makes a lot of sense that one is spectical of any particular exponent. For example, of welfare ranges being proportional to the individual number of neurons, which corresponds to an exponent of 1. However, being sceptical about any exponent requires much more confidence. For instance, it would require being sceptical of the welfare ranges presented in Bobâs book, which are pretty well explained by an exponent of 0.188 (see 1st graphs in my last comment).