Re: Opinions as Evidence: Should Rational Bayesian Agents Commonize Priors

From: Robin Hanson (rhanson@gmu.edu)
Date: Wed May 09 2001 - 19:45:45 MDT


hal@finney.org wrote:

> Counterfactuals ... get even trickier when you try to imagine who "you"
> are in a world with a different history. What if I were that poor orphan
> I see in Bombay? Who was I in ancient Egypt? ...
> In the context of this paper, what does it mean to say that I might
> have had different priors, or that your priors and mine might have been
> reversed? To some extent, my priors *are* me. ... If you swap his
> priors with someone else, you don't thereby give him different priors.
> Identity goes with the priors, plus the experiences. ...
> No doubt other people have different philosophical views about the
> nature of identity where swapping priors makes sense. But I think the
> paper needs to address this issue head-on rather than talk of swapping
> priors as though it were a philsophically unproblematic counterfactual.

There are indeed conceptual difficulties with counterfactuals, and with
identity. But counterfactuals are ubiquitous in language and science. So
I'm not sure I should discuss it unless they are more difficult than usual
in my context.

Folks on this list should be familiar with a wide range of identity issues,
such as for example whether an upload copy of you is you. I think we're
used to thinking of this as a matter of choice, and of definition, rather
than of reality. Each of us gets to decide what variations on us are close
enough to still be "us." Even if it isn't "you" you should still be able
to make sense of the counterfactual - just call it the "you-substitute."

It is hard to make much sense of most physical laws without counterfactuals.
They tend to say what would happen if you did things you usually don't do.
And prior probabilities themselves don't make sense without counterfactuals.
So I don't consider it a great leap to consider counterfactuals about which
prior you have. And I don't see how you can talk about the causal processes
that produced your priors without counterfactuals. Perhaps Nick Bostrom,
who is better trained than I in philosphy, could offer his opinion on this
topic.



This archive was generated by hypermail 2b30 : Mon May 28 2001 - 10:00:04 MDT