X-Message-Number: 9250
From: Thomas Donaldson <>
Subject: Re: CryoNet #9244 - #9245
Date: Fri, 6 Mar 1998 22:39:55 -0800 (PST)

Hi Mike!

The first thing I thought when I read about the nonTuring neural net was 
that it was a counterexample to the notion that not all machines must be
Turing machines. The particular features of the counterexample aren't so 
important: what it tells us, more than anything else, is that we cannot make
the ASSUMPTION that Turing machines can emulate everything we find in the 
world. We have to use much more argument to get there.

Now frankly for this particular argument I don't care what Dennett had to
say. I don't pay attention to arguments from authority as if they were
arguments based on good grounds (No, this doesn't mean that I think poorly
of Dennett, who has come us with lots of good stuff; it just means that 
you would need to USE his actual arguments when discussing this issue with
me, you can't just refer to them). Moreover it seems to me, simply from 
looking around, that 1). there are lots of things which are not digital
in any obvious way 2). my own understanding of quantum mechanics is that
it does not claim the world is digital, but that under some circumstances
things behave as quanta rather than waves, and other circumstances as
waves. What's the difference? Suppose that everything was quantal; nowhere
do we know that these quanta are COMMENSURATE. Jumps of sqrt(2) or pi 
are not disallowed. Considering that objects move, relativity itself
suggests that their mass need not be commensurate, even if they are
the same kind of particle. Light may sometimes appear as quanta, but the 
spread of light in a spectrum goes through the whole range.

As you might guess, I am dubious of the notion that we will return
automatically in any reasonable sense. Sure, if we really were both digital 
and finite, that's easy to argue, but I suspect we are neither digital nor
finite, seen long term. 

This is not an argument against finite models at all. At any given time,
no matter how large and elaborate they may be, we will find it useful to use
models of people, the universe, objects in the universe, etc etc. And such
models have proven very useful for the last 2000 years. Yet models are
not the same as the world, which always has the feature that one day it 
does not behave as our models suggest it should.  
 
And when we really get down to it, I'd be quite disappointed to find that
we had made a digital model of the Universe which always worked (not that 
I think that's possible --- it would probably be possible, though, to have
almost everyone BELIEVE it is perfect for centuries, just like the Ptolemaic
System. And if I awakened in such a world, such a belief would make me feel
that people had lost their curiosity, not that they fully understood the 
world).

If you wish to argue that the world is digital, please do so. You can use
arguments from wherever and whoever you wish. And sure, if the world is
finite and digital we will have assurance of our eventual return: so 
that we can argue this question, with these very words, on Cryonet an
infinite number of times. What a horrible, boring kind of resurrection 
that would be.

			Best and long long life to everyone,

				Thomas Donaldson

Rate This Message: http://www.cryonet.org/cgi-bin/rate.cgi?msg=9250