> For the most part, animal welfare scientists aren’t interested in scoring welfare on a cardinal scale, so it’s an oddity when they try.
Just to confirm, you and Rethink Priorities are using a cardinal scale for your welfare ranges, right? So when you say that a cow has a welfare range of 0.5, you implicitly mean that there is some universal scale where a cow’s minimal welfare is −0.25 and maximum is +0.25 (or shifted if we don’t assume symmetry).
I guess I’m confused on why there isn’t more work on estimating the average realized values of welfare, both from Rethink and from other animal welfare scientists. Those values are necessary for foundational claims like “eating 1000 calories of beef creates demand for X units of suffering”, or “moving cows to a pasture will increase welfare by Y units”.
Yes, Chris: we’re using a cardinal scale. To your point about estimating the average realized values of welfare, I agree that this would be highly valuable. Animal welfare scientists don’t do it because they don’t face decisions that require it. If you’re primarily responsible for studying broiler welfare, you don’t need to know how to compare broiler welfare with pig welfare. You just need to know what to recommend to improve broiler welfare. As for RP, we’d love to work on this and I’ve proposed such projects many times. However, this work has never been of sufficient interest to funders. If that changes, you can bet I’ll devote a lot of time to it!
Thanks Bob, much appreciated.
> For the most part, animal welfare scientists aren’t interested in scoring welfare on a cardinal scale, so it’s an oddity when they try.
Just to confirm, you and Rethink Priorities are using a cardinal scale for your welfare ranges, right? So when you say that a cow has a welfare range of 0.5, you implicitly mean that there is some universal scale where a cow’s minimal welfare is −0.25 and maximum is +0.25 (or shifted if we don’t assume symmetry).
I guess I’m confused on why there isn’t more work on estimating the average realized values of welfare, both from Rethink and from other animal welfare scientists. Those values are necessary for foundational claims like “eating 1000 calories of beef creates demand for X units of suffering”, or “moving cows to a pasture will increase welfare by Y units”.
Yes, Chris: we’re using a cardinal scale. To your point about estimating the average realized values of welfare, I agree that this would be highly valuable. Animal welfare scientists don’t do it because they don’t face decisions that require it. If you’re primarily responsible for studying broiler welfare, you don’t need to know how to compare broiler welfare with pig welfare. You just need to know what to recommend to improve broiler welfare. As for RP, we’d love to work on this and I’ve proposed such projects many times. However, this work has never been of sufficient interest to funders. If that changes, you can bet I’ll devote a lot of time to it!