Re: Rationality, Miracles and ETI

Dan Fabulich (
Tue, 2 Nov 1999 20:59:25 -0500 (EST)

'What is your name?' 'Eliezer S. Yudkowsky.' 'IT DOESN'T MATTER WHAT YOUR...'

> The problem with Putnam's theory is that you can't get humans, capable
> of *forming* models, without a pre-existing reality to create them.

And? Putnam would agree with that, he'd simply point out that we have POSITED the existence of a real world with real people in it in order to explain/predict our experiences.

With that having been said, I (as well as Putnam) believe that it is the case that we have posited correctly, that is, there certainly IS a real world, with real people in it, etc. By that I mean that the above claims are certainly all rationally acceptable under ideal epistemic conditions.

Putnam is NOT saying that there is no real world; that would be rationally unacceptable, after all! He is, however, saying that truth does not consist in corresponding to the "real world," but in idealized rational acceptability.

> So internalism doesn't strike me as being very self-consistent. In my
> philosophy, which is called "Externalism" for a very good reason,
> external reality predated humans, who evolved to model that reality.

Speaking as a self-affirmed internalist, your "externalism" seems like an extremely reasonable model of the world; so reasonable, in fact, that I ALSO believe that it is the case!

Your position in no way disagrees with Putnam's internalism. Internalism simply specifies the conditions under which externalism would turn out to be true.

> You can't "prove" reference, of course, because if that were true our
> thoughts would define reality. (And of course, I don't believe in
> instantiation, especially of Turing-computable behaviors, but that's
> not the point.)

I'm not sure what you mean by this... I don't recall any mention of this particular snippet online. What DO you mean by it?

> To me, this sounds like another version of that old dilemna of "How do
> you prove that rational thought works without assuming the rational
> processes you use for the proof?"

Not really, though that's certainly an interesting question. I'll probably go take a look at your TMOL page in a few minutes.


-unless you love someone-
-nothing else makes any sense-

e.e. cummings