Right, but there really aren’t any good arguments for adopting P(n) for any n—none worth considering, at least. And that’s a countably infinite number of policy debates that we don’t need to have!
No cheaper than leaving out the sack and the potatoes. Do you really think that there are any benefits of P(n) for any n that would justify having a debate over it? I think all the arguments go the same way for sufficiently small values of “all”—that is, it’s “one-sided” enough that it shouldn’t even be brought up.
One reason to bring up argument X against policy P when policy P is clearly better is that there might be a slight modification of P that retains the advantages of P while addressing argument X.
I meant that as saying all the considerations for deciding any given issue go the same way, not all issues to be decided go the same way.
Right, but there really aren’t any good arguments for adopting P(n) for any n—none worth considering, at least. And that’s a countably infinite number of policy debates that we don’t need to have!
But that’s an example of “wins by a comfortable margin”, not “all the arguments go the same way”. For example, P(n) is cheap to implement for low n.
No cheaper than leaving out the sack and the potatoes. Do you really think that there are any benefits of P(n) for any n that would justify having a debate over it? I think all the arguments go the same way for sufficiently small values of “all”—that is, it’s “one-sided” enough that it shouldn’t even be brought up.
One reason to bring up argument X against policy P when policy P is clearly better is that there might be a slight modification of P that retains the advantages of P while addressing argument X.