X-Message-Number: 7898
From:  (Thomas Donaldson)
Subject: Re: CryoNet #7883 - #7888
Date: Thu, 20 Mar 1997 01:10:16 -0800 (PST)

Hi again!

Here I am close to my bedtime, so I can't really do justice in my answers.
Some comments, however:

To Mr. Strout:

NO, we do NOT have good emulations of neurons right now. We have devices which
emulate only a few of their characteristics. We don't even have devices which
have as many connections. And the chemistry, at least for what it says about
how the neuron and memory works, IS important and cannot be dispensed with.
It is hardly an accident that we have many different nerve transmitters and
for each one more than one different receptor. All of that affects how the
neuron works --- in an important way. 

I also note that you are talking about analog devices now, and yes, they are
different. Whether we could really call the result "uploading" then gets to
be kinda vague. Reconstruction? Reconstruction with modified "neurons"? If
they make a copy of you with analog copies (or improvements) of neurons they
are doing something quite different from "uploading" you, at least as I 
understand the word. I would agree that it would be you if done well. Sure,
doing so is not at all simple.

Of course, perhaps you want to be uploaded to a computer only in the 
sense of storage, while these devices are created and combined together. 
I've already said that I have no objection to that.

To Mike Perry:

For all practical purposes, Tipler is wrong. We do not even know if the
universe is finite. A finite object, acted on by an infinite variety of 
inputs, will behave in an infinite variety of ways. Tipler has said some
highly percipient things about the universe, but that just isn't one of
them.

Nor are you a computer, or I. In my second posting, I explained the
essential difference: computers just move around bits in memory. You act
on the world. And ultimately to understand how your brain works (mine 
too) we must forget about any "symbols" we are using --- not that we
don't use them, but they come ultimately from the physical activity of
our neurons. (For that matter, most of your brain is not involved in 
symbolic activity --- it is reacting to events in the world or connections
it may have formed in the past. JUST LIKE THE MODEL T).

Nor does our thinking usually consist of manipulating symbols. Even as I
sit writing this, the words for what I want come to me, not because 
I am just manipulating symbols in my head but because I have learned to
describe something of what the physical state of my brain is. Since you
learned English (mostly, note, as a set of neural nets, not by using
a dictionary) you generally understand me, whether or not you agree.
Even when I do math, as a mathematician, manipulation of symbols is 
secondary to the thoughts I am trying to express -- or exploring in
my own mind for their merits or lack of merit.

Yes, we have found it very useful to set up systems to manipulate
symbols. When we humans invented arithmetic, and the rules for doing
it, we invented a nice tool. The same may be said of all the theories we
have developed. And with computers (as we use them now, I am not talking
about other kinds, or whatever a "computer" may be) we can do some of this
manipulation of symbols automatically, and that is even nicer. But all 
of that is just moving bits about in an electrical device of some kind.
We put those symbols into our computers, and we take them out again at
the end. All that is a subset of our human abilities to use language.

But behind that language there is meaning, which by its nature cannot
simply be a matter of symbols. You will never find out the meaning of 
a word by chasing it through the dictionary. Not only that, but without
our understanding of those symbols, the results we get from our computers
would mean nothing. There is no INFORMATION in such symbols unless we
know their meaning. All of that depends on us. 

We are no more processing information when we think than a Model T is
processing information. That is essential. We are doing things on a 
different level than the symbolic level. And yes, this is an easy
point to forget. But think how easy it is for us to MISunderstand one
another: where is the information when that happens?

If a physical device behaves as if it is you, and on scrutiny is not just
hooked up to a large computer, but has a physical structure resembling
you, I can believe it is aware. But if it is just moving symbols about,
no matter how complex, not only is it not aware but it simply will not
manage to emulate you at all --- for the same reasons as no finite theory
will ever describe your response to all the things you may encounter.
And yes, if I were turned into such a device (well, I am kinda one now,
though no one built me in any good sense of that word) then I would still
be aware. But not if I can only move symbols about according to various
rules in my "brain". That is not how we work, nor is it what gives us
our awareness.

This is probably not thorough enough to explain, but it's now past my
bedtime, and I have lots to do tomorrow. So goodnight.

		And long long life,

				Thomas Donaldson


Rate This Message: http://www.cryonet.org/cgi-bin/rate.cgi?msg=7898