X-Message-Number: 11613
From: Thomas Donaldson <>
Subject: Sorry, Bob, but your arguments sound weak, though I do agree
Date: Mon, 26 Apr 1999 00:17:22 +1000 (EST)

To Bob Ettinger:

I personally have said in the past that I see no reason why a robot could
not be constructed so that it had consciousness and was really a person
--- though I can see many difficulties which others seem to ignore. My
chief reason for believing this is that we are constituted of constant
chemical changes at the synapses of our neurons.

To simply say that the notion of a person consisting of changes in the
writing on a tape comes awfully close to those changes in the chemistry
of synapses (and as with the tape, the changes also change behavior of
the neurons just as changes in its tape change the behavior of the Turing
machine).

I have also pointed out one major difference between brains and computers,
even those which are highly parallel: our neurons constantly grow new
synapses and do away with others, so that their connections change. Those
connections constitute the neural nets of which we are constituted, and
because of them our neural nets differ from any neural net so far 
made for computing. The possible changes are great enough that it would
be impractical (for instance) to construct such a neural net by allowing
all possible connections between even nearby neurons (axons can go for
relatively long distances, so a "nearby" neuron may be centimeters away).
Growth seems to me to be the only practical strategy for emulating such
neural nets.

I am saying, basically, that your argument against Mike actually seems
quite weak, even though I would agree with you on the point you wish to
argue. I do not see how the material of which we are constituted must
play any role in whether or not we have the minimum abilities needed to
be a person. Yes, right now if we want growth of connections (and their
destruction, too) we would immediately think of biological materials
and devices. Still, no one has shown that similar devices might not be
made by quite different means (no one has shown that they CAN be made by
different means, either, but that still doesn't make them impossible).

As for Turing machines, they fail completely to provide a system capable
of dealing with the real world. Even forgetting issues such as just how
fast tapes can be written on and erased, a person (or even a device of
the complexity of a frog) will need to act much faster than any sequential
computer could do --- regardless of its hardware. We don't see this mainly
because much of our brain processing is not conscious, and our conscious
processing is clearly slower than that of current computers.

And the Turing Test, specifically because it only tests the ability to
use language to answer questions given in the same language, falls short
of showing that the machine (?) on the other side of the curtain can
actually respond to the world itself. Our language is not the world.

			Best and long long life to all,

				Thomas Donaldson

Rate This Message: http://www.cryonet.org/cgi-bin/rate.cgi?msg=11613