From: Dan Fabulich (daniel.fabulich@yale.edu)
Date: Sun Mar 05 2000 - 11:47:59 MST
> This was just an example to introduce the notation. Use the mugging
> fact instead if that suits you better.
The mugging fact won't do. The mugging fact is a claim which is true iff
a certain ethical claim is right and if a certain empirical claim is true.
You can settle the empirical claim, I'm arguing, but not the ethical one.
> You keep restating claims like this as if they were obvious;
> they aren't. This is the issue we are arguing about.
I'll fall back on my Omega Point argument, because I think I can force
this one through more easily than my computability argument (though I
actually like that one a bit better).
Allow me to restate the claim in a more general manner. Assume, for the
moment, that the state of my beliefs can be completely determined by some
numbering scheme. (Think Godel numbers here.) Each number follows from a
very complicated rule of inference from the previous number. Suppose you
knew this rule of inference completely.
Now you want to ask which number I'll "end up believing," or at least tell
me some number theoretic properties about that number. Suppose you're
curious as to whether that number will be even or odd.
Now notice that if I live forever and have an infinite number of thoughts
(that is, if there is no end to the list of numbers which represent my
beliefs), there's no way for you to tell whether my "last" number will be
even or odd, in exactly the same respect as it would be impossible for you
to determine whether the last natural number will be even or odd. And
notice as well that it doesn't matter how many finite odd or even examples
you provide: there's STILL no way for you to tell me that the last natural
number is odd. There IS no last natural number on this picture.
You might step in and say "Well, number theory is very young, so I can't
tell you what sort of examples would provide evidence that the last number
is even..." but no one should believe you because the number *doesn't
exist*, and, if I have an infinite number of thoughts, we can see this
NOW.
What's more, if I DON'T have an infinite number of thoughts, I can't be
sure that I've figured out ethics yet. So if I WERE going to die, and you
could somehow tell me what number I'll believe when I die, I'd still have
no reason to believe it. Who's saying that I'll necessarily have it right
then? Or that I'll have figured it out best? Or that this is even the
best I'm capable of? Hal's parallel to physics is apropos here. Just
because I'll die with a certain conception of physics doesn't mean that
it's right. Lots of dead people got physics very wrong indeed; the same
holds of ethics.
> The empirical science of ethics is very young, so I can't tell you now
> what sortsof empirical facts will turn out to be very relevant. But
> in some possible states of the universe you agree with S, and in
> others you disagree, and *anything* that helps me exclude some of
> those possible states can inform me about whether you will agree with
> S. If this isn't obvious to you, you haven't had a sufficient
> introduction to information theory (e.g., decision theory) for me to
> proceed.
I understand the information theory well enough; what I'm contesting is
that there *are* facts that help you exclude those "possible states." I
put "possible states" in scare quotes because none of those states are
possible on this Omega point picture. It can neither turn out that the
last number is odd and I'm related to bonobos, nor that the last number is
odd and I'm related to baboons, nor that the last number is even and I'm
related to bonobos nor that the last number is even and I'm related to
baboons. None of these are possible because there *is* no last number.
> In my example, the brain does *not* remember which atom it picked;that
> info is *only* encoded in that light signal. So the analogy is
> closer. We will later be able to verify the atom mass by getting the
> light signal, and we will later be able to verify your ethical
> opinions, after the universe influences you.
After the universe influences me? On this Omega point picture, you'll
have to wait a very long time indeed. This is akin to saying "We can
verify that the last number is even just by counting all the way up to the
last number and checking out its ones digit." My response? *No we can't.*
-Dan
-unless you love someone-
-nothing else makes any sense-
e.e. cummings
This archive was generated by hypermail 2.1.5 : Fri Nov 01 2002 - 15:27:11 MST