Here we face the tragedy of “reference class tennis”. When you don’t know how much to trust your own reasoning vs. someone else’s, you might hope to defer the historical record for some suitable reference class of analogous disputes. But if you and your interlocutor disagree on which reference class is appropriate, then you just have the same kind of problem again.
I really don’t think this is a reference class tennis problem, given that I’m criticizing a specific post for specific reasons, not making an argument that we should judge this on the basis of a specific reference class.
And given that, I’m still seeing amazingly little engagement of the object level question of whether the criticisms I noted are valid.
Thanks for the suggestion! I just re-skimmed the Bostrom et al. paper (it’s been a while) and wrote up my thoughts in a top-level post.
Here we face the tragedy of “reference class tennis”. When you don’t know how much to trust your own reasoning vs. someone else’s, you might hope to defer the historical record for some suitable reference class of analogous disputes. But if you and your interlocutor disagree on which reference class is appropriate, then you just have the same kind of problem again.
I really don’t think this is a reference class tennis problem, given that I’m criticizing a specific post for specific reasons, not making an argument that we should judge this on the basis of a specific reference class.
And given that, I’m still seeing amazingly little engagement of the object level question of whether the criticisms I noted are valid.