Newsgroups: comp.ai.philosophy
Path: utzoo!utgpu!watserv1!watdragon!mpdevine
From: mpdevine@watdragon.waterloo.edu (Michel P. Devine)
Subject: Re: Introspection
Message-ID: <1991Feb13.185046.25750@watdragon.waterloo.edu>
Organization: University of Waterloo
References: <1991Feb4.224101.18846@watdragon.waterloo.edu> <1574@pdxgate.UUCP>
Date: Wed, 13 Feb 1991 18:50:46 GMT
Lines: 152

In article <1574@pdxgate.UUCP> erich@eecs.cs.pdx.edu (Erich Stefan Boleyn) writes:
>Why must a mechanistic universe have no good explanation of emotions 
>and intuition?  

A very good question.  I think the problem lies in assigning meanings.  I 
don't want to give a full expose on this complex matter, but I will briefly
outline my point.  A "sign" is something that has no meaning in and of
itself, but that refers to something else.  For example, our mathematical
language consists entirely of signs; every part is arbitrary and ordered
according to agreed upon rules.  So, it is not possible to refer to something
which has no sign, or for which no sign can be constructed.  In that sense,
our modeling tools, based on math, are necessarily a "closed system".  
The assignment of the signs to something outside the system (for example,
the intuitive idea of the number 2) is a mysterious process happening
within the mind of the observer, and therefore hard to analyse, let alone
detect.  If the universe is mechanistic, then there must be some description
constructed out of signs that captures it, in all its glory.

On the other hand, there are "symbols" which are both signs (pointers to)
and are themselves greater than signs.  In psychological terms, the symbols 
have meaning but the signs do not.  Symbols are things like "father", "water", 
"pain", "mathematician" which are not purely conceptual but much richer 
in content and association.  The problem with the mechanistic view, as I see it,
is that it precludes anything but an arbitrary assignment to the axioms
of nature, i.e, there are no intrinsic meanings since everything is
a denotation.  But what does it denote?  That is where the importance of
establishing a science of symbols comes in.

>   When I hear explanations that there must be "something more" to the
>universe than a mechanism in operation, many seem to also hold an
>undercurrent of revulsion to attributing all of the beauty of the world to
>"just a machine".  

I agree.  I too consider machines beautiful, but perhaps the perception
of beauty is due to something in me that is inherently non-mechanical. 

>   My view of a mechanistic world is
>extremely *rich*, full of sensuality and imagery of incredible complexity,
>with feelings, life, and all the other amenities...  It could even allow
>for a God if done right.

I don't understand that last statement.  Any God would still have to be
around and if the universe is mechanical, why do we need it/Him? and
where is he hiding?  I like your statement "if done right", because it
is familiar to me...although not satisfying since I prefer experience 
to argument.  

>   *My* point is that, although you could be correct, of what value is
>it for science as a whole (or even a subdiscipline like Cognitive Science)
>to disregard its models totally...  science's tenet is to discover
>empirical relationships.  Now, admittedly, there is a strong debate about
>whether cognitive science fits the bill, so to speak, but can it hurt to
>try?
>

Right. One cannot ditch all models because they are necessary.  However,
there *is* a scientific discipline that can answer our questions, but it
isn't CS.  According to Wilber (author of many  fascinating books) a
scientific fact is established by the following procedure:

1. Injunction.     A recipe or procedure is provided.  Following the 
		recipe should result in 
2. Experience.     The person looks at the result of following the 
		instructions, interprets, analyzes and forms a new
		personal world view.
3. Validation.     One consults a group of peers who have acquired 
		the knowledge to verify the truth.  Facts are based
		on agreement with established authority.

So, a mathematical theorem is a procedure by which I can acquire the same
mind-state as the author and thereby verify his claim; similarly for
the physical sciences.  With this very broad definition of the scientific
paradigm, one can attack the consciousness problem.  

Note that all three steps are essential.  To paraphrase Wilber, a person who 
has not acquired the skills necessary to make an informed statement ought to
be ignored.  This may sound harsh, and even non-conformist in an age where
the nation gobbles up the word of the inexpert and uninformed on a daily
basis (I'm thinking of actors and "personalities" on talk shows), but it seems
only reasonable.  By this definition of science, there are many sciences of
the mind that are extremely sophisticated, have a large body of theoretical
knowledge *and* groups of experts for validation.  It is entirely possible
that the Tibetans, the Indian Gurus and Zen Masters (not to mention the
Christian "mystics") have more to say about the mind than CS.  

I admit readily that I don't "buy" the mystic's viewpoint.  However, if I
insist on being scientific, then I *must* carefully consider their arguments,
try out their exercises and find out for myself, whether this is the prevalent
"scientific" (read, conventional) tactic or not.  Besides, I never understood
why a nation like Tibet never developed technology; perhaps they were
concentrating on something more "organic"?

>
>>The levels represent qualitative differences in thinking ability or 
>>"intelligence".  For example, the ability of plants to orient their 
>>photoreceptors with the sun seems to be the action of some kind of
>>thought, although not necessarily consciousness.
>
>   Some extremely simple algorithms for self-adaptive and self-configuring
>systems seem to be able to do this quite well.
>

Yes, but the pattern (algorithm) does not do justice to the plant.  We still
can't build plants, even though we can build systems that behave somewhat
like plants.

>The right kind of shock to one's system can even kill us.  One could then
>think of a godel string as a "shock to the system", since how do we know
>what form they would take?
>

Sorry, I don't understand these statements relating godel strings to 
our systems at all.

>
>   Degrading?  This sounds a bit like elitism...  why are we better than
>the rocks and stars?  We're more interesting to ourselves, I'll grant you,
>and we can do neat things, but they are beautiful too.
>

Absolutely. We are *both* better (I can things they can't) and not better
(we're all the same).

  I am going out on a limb here, but
>in trying to be what I feel as an honest scientist, I feel one should
>think critically about everything, and am therefore an agnostic.  (Atheism
>in my view would be an assumption too)

I disagree; agnosticism is a way of saying that I'm not willing to take 
a stand.  Most agnostics I know are actually atheists who, for social
reasons, prefer not to be singled out.  (I am not criticising since I
have spent a long time as an agnostic, and I may change yet, but I'm not
sure :-).  

>
>   Yes, but so many people have fundamentally disagreed about how the
>human mind operates from introspection...  can we rely on their
>interpretations for real information that is not just useable in a social
>context?  

No, let us not rely on anyone else.  Introspection is taken up for *personal*
discovery.  If something generic surfaces, we shall celebrate but chances
are that the result of introspection is to increase the disparity between
us and the people around us, as we gain more understanding and clarity.
Then, it is almost impossible to communicate directly; but that is not new.
 
Michel
-- 
--------------------------------------------------------------------------------
mpdevine@watdragon.waterloo.{edu|csn}            (519) 884-7123 Michel P. Devine
mpdevine@watdragon.uwaterloo.ca				   CS Dept., U. Waterloo
{uunet|utzoo|decvax|utai}!watmath!watdragon!mpdevine       Waterloo, Ont. N2L3G1