I don’t see why they should. Merely holding beliefs to be revisable, doesn’t tell you how to revise them....and they are going to be revised by other meta-level beliefs (ie. epistemology), which will vary (and also reviseable!)
So if Alice and Bob start off with a shared object level belief and feel the need to revise it, they are only going to concur if they are revising it the same way.
Aumanns theorem is more or less wrong, at least extremely impractical, for similar reasons.
So if Alice and Bob start off with a shared object level belief and feel the need to revise it, they are only going to concur if they are revising it the same way.
Or if they hit the disagreement with an experiment and concur on the resolution.
That’s just a way of assuming they agree on epistemology. In real life there is no firm agreement on what an experiment is, or how to draw a conclusion from one.
In general, what you say may be true. In specific cases, it may still work. (For example, two parties may be able to come up with an experiment they agree on. This may be easier if both parties are more similar to each other.)
Yes, but my claim was not that convergence would never occur, it was that it would not necessarily occur. Of course convergence can occur between parties that have similar epistemology, but in general that’s not a given. It’s not a refutation to say that it’s only true in general.
I don’t see why they should. Merely holding beliefs to be revisable, doesn’t tell you how to revise them....and they are going to be revised by other meta-level beliefs (ie. epistemology), which will vary (and also reviseable!)
So if Alice and Bob start off with a shared object level belief and feel the need to revise it, they are only going to concur if they are revising it the same way.
Aumanns theorem is more or less wrong, at least extremely impractical, for similar reasons.
Or if they hit the disagreement with an experiment and concur on the resolution.
That’s just a way of assuming they agree on epistemology. In real life there is no firm agreement on what an experiment is, or how to draw a conclusion from one.
In general, what you say may be true. In specific cases, it may still work. (For example, two parties may be able to come up with an experiment they agree on. This may be easier if both parties are more similar to each other.)
Yes, but my claim was not that convergence would never occur, it was that it would not necessarily occur. Of course convergence can occur between parties that have similar epistemology, but in general that’s not a given. It’s not a refutation to say that it’s only true in general.