The shrimp welfare project

Apologies for focusing on just one sentence of this article, but I feel like it's crucial to the overall argument:

> ... if [shrimp] suffer only 3% as intensely as we do ...

Does this proposition make sense? It's not obvious to me that we can assign percentage values to suffering, or compare it to human suffering, or treat the values in a linear fashion.

It reminds me of that vaguely absurd thought experiment where you compare one person undergoing a lifetime of intense torture vs billions upon billions of humans getting a fleck of dust in their eyes. I just cannot square choosing the former with my conscience. Maybe I'm too unimaginative to comprehend so many billions of bits of dust.

Yeah this (along with the "billion headaches" inanity) rests on a fallacy: insisting an abstraction can be measured as a quantity when it clearly cannot. This trick is usually done by blindly averaging together some concrete quantities and claiming it represents the abstraction. The illusion is fostered by "local continuity" of these abstractions - if pulling your earlobe causes suffering, pulling harder causes more suffering. And of course the "mathiness" gives an aura of rigor and rationality. But a terrible error in quantitative reasoning occurs when you break locality: going from pulled earlobe to emotional loss, or pulled earlobe to pulled antennae, etc. The very nature of the abstraction - "suffering," "badness," - changes between entities and situations, so the one formula cannot possibly apply.

ETA: see also the McNamara fallacy https://en.wikipedia.org/wiki/McNamara_fallacy