Base on my experience most halfers nowadays are actually double-halfers. However, not everyone agrees why. So I am just going to explain my approach.
The main point is to treat perspectives as fundamental and recognize in first-person reasoning indexicals such as “I, here, now” are primitive. They have no reference class other than themselves. So self-locating probabilities such as “the probability of now being Monday” is undefined. This is why there is no Bayesian update.
This also explains other problems of halferism. e.g. robust perspectivism: two parties sharing all information can give different answers to the same probability question. It is also immune to counter-arguments against double halfers as pointed out by Michael G. Titelbaum.
I don’t see why “the probability of now being Monday” is undefined. I agree that the information given in the problem isn’t enough to determine it, but that’s a very different objection.
Base on my experience most halfers nowadays are actually double-halfers. However, not everyone agrees why. So I am just going to explain my approach.
The main point is to treat perspectives as fundamental and recognize in first-person reasoning indexicals such as “I, here, now” are primitive. They have no reference class other than themselves. So self-locating probabilities such as “the probability of now being Monday” is undefined. This is why there is no Bayesian update.
This also explains other problems of halferism. e.g. robust perspectivism: two parties sharing all information can give different answers to the same probability question. It is also immune to counter-arguments against double halfers as pointed out by Michael G. Titelbaum.
I don’t see why “the probability of now being Monday” is undefined. I agree that the information given in the problem isn’t enough to determine it, but that’s a very different objection.