Newsgroups: comp.ai.philosophy
Path: cantaloupe.srv.cs.cmu.edu!bb3.andrew.cmu.edu!news.sei.cmu.edu!cis.ohio-state.edu!math.ohio-state.edu!howland.reston.ans.net!ix.netcom.com!netcom.com!kovsky
From: kovsky@netcom.com (Bob Kovsky)
Subject: Re: Is the mind/brain deterministic?
Message-ID: <kovskyCyxKCo.Hv5@netcom.com>
Organization: NETCOM On-line Communication Services (408 261-4700 guest)
References: <39ig6i$joq@seralph9.essex.ac.uk>
Date: Tue, 8 Nov 1994 03:58:00 GMT
Lines: 58

In article <39ig6i$joq@seralph9.essex.ac.uk>,
Butler J M <butljw@essex.ac.uk> wrote:
>I would like to poll readers of this group on the following question:
>
>(You may or may not assume that the mind and brain are one and the same
>but say so and why in your posting.)
>
>Please discuss on 2 levels, 1) the level of single neurons or small groups
>of neurons and 2) the mind/brain as a whole. Are these 2 levels deterministic
>or non-deterministic with regards to their behaviour?
>
>My view:
>
>If mind/brain behaviour is non-deterministic then why are we AIers wasting our
>time?

	1)  In my view, neurons are so intensively interconnected 
synaptically and so responsive to chemical environments that discussion 
of "single neurons or small groups of neurons" is meaningless;

	2)  "Determinism" (cause-and-effect) is inusufficent to account 
for "behavior" or experience.  In particular, consciousness and freedom 
are beyond the reach of explanation in terms of cause-and-effect.  (Those 
who believe that there is no freedom are welcome to visit the ftp site 
indicated below.)

	In my view, cause-and-effect relations are generated by the
processes of experience and do not necessarily correspond to reality. 
Cause-and-effect relations are, however, often susceptible to being
assembled into structures that can approximate reality.  In some regimes,
such as that established in the laboratory, structures of cause-and-effect
relations can be crafted so as to become an approximation that can be made
to appear faultless in the limit.  (Given any epsilon greater than 0, it
is possible to construct a laboratory environment such that the difference
between the predictions of the approximating causal structures -- the laws
of physics -- and measured reality is less than epsilon.) In other
regimes, however, such as the one obtaining in the brain, cause-and-effect
relations cannot provide even a good approximation.  (Freedom is not
susceptible to description in terms of causal relations.)

	AI researchers are not wasting time.  We humans are stuck with
causal relations as the best we can do and AI researchers are
investigating those areas where mechanistic (cause-and-effect) approaches
are useful.  Moreover, I believe that it is possible to develop a device
that has both consciousness and freedom; and AI is paving the way.  The
device will not, however, be a machine.  My prediction is that such a
device will be developed from neural networks operating "on the edge of
chaos." 

	Obviously the "mind" (a notion with many difficulties) and the
brain are not the same.  (Neither is a book the same as the paper that
goes into its construction.)

-- 

*   *    *    *    *    *    *    *    *    *    *    *    *    *    *    *   * 
    Bob Kovsky          |  A Natural Science of Freedom 
    kovsky@netcom.com   |  Materials available by anonymous ftp
                        |  At ftp.netcom.com/pub/freeedom
*   *    *    *    *    *    *    *    *    *    *    *    *    *    *    *   * 
