Newsgroups: comp.ai,comp.ai.philosophy,sci.logic,sci.cognitive
Path: cantaloupe.srv.cs.cmu.edu!bb3.andrew.cmu.edu!nntp.sei.cmu.edu!news.psc.edu!hudson.lm.com!hookup!usenet.eel.ufl.edu!news.mathworks.com!gatech!swrinde!cs.utexas.edu!utnut!utgpu!pindor
From: pindor@gpu.utcc.utoronto.ca (Andrzej Pindor)
Subject: Re: Zeleny on predictability (was FIRST order?)
Message-ID: <DCFyxI.62w@gpu.utcc.utoronto.ca>
Organization: UTCC Public Access
References: <jqbDBu09v.G9H@netcom.com> <GUDEMAN.95Jul27091026@baskerville.cs.arizona.edu> <DCE209.LqL@gpu.utcc.utoronto.ca> <GUDEMAN.95Jul27195705@baskerville.cs.arizona.edu>
Date: Fri, 28 Jul 1995 19:46:29 GMT
Lines: 242
Xref: glinda.oz.cs.cmu.edu comp.ai:31954 comp.ai.philosophy:31017 sci.logic:13256 sci.cognitive:8639

In article <GUDEMAN.95Jul27195705@baskerville.cs.arizona.edu>,
David Gudeman <gudeman@cs.arizona.edu> wrote:
>In article <DCE209.LqL@gpu.utcc.utoronto.ca> pindor@gpu.utcc.utoronto.ca (Andrzej Pindor) writes:
>
>]In article <GUDEMAN.95Jul27091026@baskerville.cs.arizona.edu>,
>]David Gudeman <gudeman@cs.arizona.edu> wrote:
>]................
>]>My own reason for rejecting mechanism might be considered irrational
>]>in a sense: I cannot imagine what sort of theory would even count as
>]>an explanation of cognition in terms of physical events.  One can
>]>imagine an account of brain physiology which successfully predicts
>]>cognitive events in terms of physical events, but this would not count
>]>as an explanation.  How does one get from a proposition involving
>]>mass, charge, position and other physical properties to a proposition
>]>involving intent, belief, reference and other mental properties?  The
>]>gulf seems to be inherently uncrossable.
>]>
>]I am a bit puzzled - if the account of brain physiology predicted cognitive
>]events, it would mean that it mapped a way from physical properties to 
>]cognitive events, wouldn't it?
>
>I don't know.  What do you mean by "map a way from physical properties
>to cognitive events"?  

See above what you have said : "brain physiology which successfuly predicts
cognitive events...". In other words it would (hypotheticaly) indicate how 
physical properties give rise to cognitive events, isn't it what you meant?
This is what I refered to as "map a way from ... to...".

>................Certainly it does not explain anything.  

If this would not count as explanation, what would? I feel that (and you seem 
to have said this below) word "explain" has an empty domain for you.

>................................................................There
>are well-known behavioral theories now that are fairly accurate in
>judging emotional events.  They are collectively called "looking at
>the person's expression".  Does an angry face explain the anger just
>because it can be used as an indicator of angor?  It can also be used
>to predict with fair reliability that any vocaliations will be angry
>also.  Does it explain angry vocalizations?  The point is that a
>correlation does not indicate an explanatory connection.
>
Can't answer you queries until I know what counts as an explanation for you.
At the moment I am at loss concerning this. Can you give an example of
"an explanation" if not a definition?

>]If these cognitive events (as predicted from brain physiology) had any 
>]empirically verifiable consequences (discussion about Popper's falsification 
>]is relevent here) then such a theory could count as their explanation,
>]assuming observations would not falsify predictions of such consequences. If 
>]they did not, then their description would have a value of a Harlequin novel.
>
>How would they do so?  All I'm looking for is a vague scheme here, it
>does not have to be at all plausible.  Just give me a theory such that
>_if_ it were true, then it would explain cognitive events in terms of
>physical events.
>
You mean "how would they have any empirically verifiable consequences"?
Hmm, I do not know although I could speculate. However, it might be a waste
of time if you are saying that that they cannot have such consequences "in
principle". If this is the case then their "existence" is on par with
existence of angels, fairies, etc. in my view.

>]>People like to point to the growth of scientific theory as evidence
>]>that things can be explained that were thought impossible of
>]>explanation previously, but the cases are highly divergent.  Consider
>]>the mechanical explanation of heat.  Heat, if construed as a physical
>]>event (rather than a perceptual one) has the property of location, and
>]>has apparent physical sources and effects.  Everyone always viewed
>]>heat as essentially physical, even if not as mechanical in the usual
>]>sense.  All of what is known about heat is known either physically or
>]>by touch, and the mechanical explanation of heat only explains the
>]>physical facts (including the physical effects on human skin), not the
>]>perceptual facts.
>]>
>]I am puzzled again - "by touch" is the same as "physically", but what
>]perceptual facts about heat you have in mind which are not physical facts??
>
>The perceptual facts are the cognitive events associated with heat.
>Saying that the molecules of a surface have a high kinetic energy does
>not explain why I feel warmth when I touch it.  Why could I not feel
>coldness instead?  Or vibration?  Or why could the touch of a warm
>surface not make me see green stars flashing before my eyes?  What is
>the causal connection between perceptual warmth and physical heat?
>You might say that you just can't answer this because science has not
>progressed that far.  I say that I cannot even imagine what would
>count as an answer.
>
Sorry, but I have problems understanding your questions. Are you asking
why do we see red as red and green as green and not green as red and red as 
green?
Isn't such a question completely meaningless? You are presumably not surprised
that we have different perceptions for different stimuli, are you? How we
call them is totally irrelevant, isn't it? And don't you think that we could
find out that different stimuli lead to different processes (excitations,
transitions or like) in the brain? If this is not an explanation for you, can 
you suggest what possible _shape_ could the answer have? If you cannot, then
I do not see how the question can have any meaning as to "why".

>].................
>]>A reasonable line of research, as long as one does not make undue
>]>suppositions about the nature of the taxonomy.  In particular, just
>]>because a particular model of the mind is esthetic, well-organized,
>]>and capable of useful predictions, that does not mean the model is any
>]>truer than any other model capable of making a similar class of
>]>predictions.
>]>
>]What criteria of being "truer" or "less true" do you use?
>
>I don't have any such criterion in mind.  The point is that one theory
>is _not_ less true than another, implying that they are both equally
>true (or both equally false).
>
As long as both theories explain all known relevant facts, I agree with you.
The choice between them is then a matter of pragmatics - which one is
simpler, which one requires fewer assumptions, etc. Some people like also
to use esthetic criteria, but I think that these are strongly connected to
the pragmatic ones.
>]..................
>]>The presence of the mechanism is superfluous except as a proof of
>]>correctness.  What is being investigated is how human behaviors may be
>]>described by rules.  Many (perhaps all) human behaviors are subject to
>]>such description, including language, social interactions and
>]>architecture (I may have a reference for this if you are interested).
>]>While it may be interesting to see how well behaviors may be described
>]>by rule sets, the production of such rules tells us more about human
>]>ingenuity and ability to organize data than it tells us about the
>]>sources of the behavior.
>]>
>]If deducing rules which could describe behavior would not satisfy you,
>]what do you mean by "sources of behavior"?
>
>I'm not "unsatisfied" by sets of rules.  I find them quite interesting
>at times.  My point is that just because it is possible to give a set
>of rules that humans follow in certain endeavers, with some
>consistency, does not prove, or even suggest, that the behaviors are
>caused exclusively by physical events.  It _does_ show that it is in
>principle possible to simulate the same behaviors with physical
>mechanisms.  I'll go further, and say that even if you could give a
>complete set of rules that perfectly predicted the behavior of
>someone, that would not prove that behavior has exclusively physical
>causes.  There is no reason to suppose that cognitive causes cannot
>have predictable effects also.
>
If you invoke "non-physical" events, you are rise a spectre of dualism. For
me dualism is a literature rather then science, and a bad literature at that.
In this forum I am prepared to discuss science only.

>Frankly, it is obvious (a posteriori) that there are patterns, not
>only in the physical universe, but in animal and human behavior as
>well.  It is equally obvious that the better one knows these patterns,

I find the above distinction surprising - animal and human behavior is 
a part of physical universe for me.

>the better one can predict events, or even mimic them.  But if one
>gives a description of the patterns by way of a theory that contains
>references to unobservable entities and events, this is no reason at
>all to suppose that the occult entities and events exist.
>
If these "unobservable entities and events" are unobservable "in principle"
then such a description is not a scientific one (rather what I'd call a 'bad
literature'). And you are right, saying that these exist is an unwarranted use
of the word "exist" (as by Wittgenstein).

>Consider that I can mimic a game of ping pong on a computer screen by
>understanding the patterns followed by the ball (that is, the
>mechanics of collision and trajectory).  Suppose my implementation of
>the game treats the universe as an two-dimensional array of integers,
>where the integer represents the height the ball will have when it
>hits that cell (the integers are calculated at a collision).  Does
>this suggest that space is really just an array of cells, where each
>cell contains a number saying what will happen at that point in the
>future?  No, it is just my way of representing a pattern.
>
Shouldn't you rather say "implementing a pattern"? And I can plausibly say
that for this "ball" on the computer screen the space in which it 'lives'
IS an array of cells. The problem arises only if you start mixing space of
a real ping-pong ball with the space of the 'ball' pattern on the screen.

>]...........
>]>I don't know why you should be distressed by this, since everything
>]>else also works by some kind of magic that can never be understood or
>]>explained.
>
>]If so, then what is the point of debating any science question?
>
>Well, there still plenty of work to do once you give up explanations.

What might this be? Could you give an example? Would this 'work' be equivalent
to doing 'science'?

>But on the other hand, much scientific work of the past and present
>would then come to seem pretty pointless.
>
I do not understand what you mean by this. Had it not been for this scientific 
work of the past and present you migh be living in a cave and communicating 
with others by voice and body language and not by Internet, for instance.
Misgivings about certain features of our civilization notwithstanding, I'd
rather have it the way it is. And it seems to be the case with you too,
since presumably you participate in these discussions by choice.

>]It seems to me that your notions of "understanding" and "explanation" are so
>]idealized as to make these terms totally useless. Reminds me of people who
>]look for "ideal love" (or "ideal woman/man") and end up their lives
>]empty-handed :-).
>
>I think you are missing my point.  To use your analogy, I'm not
>looking for an ideal woman since I've decided they don't exist.  Lady
>science is a mythical creature, ephemeral and unapproachable.  She has
>never genuinely put out for anyone, and she never will.  Plenty of
>deluded young men have had wet dreams over the illusion that they have
>touched her, but all they had was the same rheumy old whore Aristotle
>slept with.  The only thing they changed was the style of poetry used
>to describe her scarce virtues, and they covered her with the finery
>and the makeup of a more advanced technology.
>
I assume you are speaking for yourself :-). How about "if you don't have what
you like, you like what you have"? "Science" is what people do to be more
effective in dealing with the nature, where does anything mythical come in?
It is true that in the past people belived that they are in the center of 
the universe, and even today some think that the nature is limited to those
aspects which participated in forming us the way we are. I'd rather say that
we suffer the lack of imagination and not the overabundance of it, as you 
seem to be suggesting. 
Specifically, please give what you consider to be a legitimate example of
"an explanation".

>Gee, I didn't start out to get that depressing...

Try to like what you have instead of regretting what you don't :-).

>					David Gudeman
>gudeman@cs.arizona.edu

Andrzej
-- 
Andrzej Pindor                        The foolish reject what they see and 
University of Toronto                 not what they think; the wise reject
Instructional and Research Computing  what they think and not what they see.
pindor@gpu.utcc.utoronto.ca                           Huang Po
