Newsgroups: comp.ai,comp.robotics,comp.ai.philosophy
Path: cantaloupe.srv.cs.cmu.edu!rochester!udel!gatech!howland.reston.ans.net!ix.netcom.com!netcom.com!jqb
From: jqb@netcom.com (Jim Balter)
Subject: Re: Minsky's new article
Message-ID: <jqbCzJ9I4.F9v@netcom.com>
Organization: NETCOM On-line Communication Services (408 261-4700 guest)
References: <3agf03$qi5@mp.cs.niu.edu> <3ak0va$n6e@mp.cs.niu.edu> <jqbCzI9IE.Cqt@netcom.com> <3al5f7$4lu@mp.cs.niu.edu>
Date: Sat, 19 Nov 1994 21:10:52 GMT
Lines: 80
Xref: glinda.oz.cs.cmu.edu comp.ai:25334 comp.robotics:15490 comp.ai.philosophy:22324

In article <3al5f7$4lu@mp.cs.niu.edu>, Neil Rickert <rickert@cs.niu.edu> wrote:
>In <jqbCzI9IE.Cqt@netcom.com> jqb@netcom.com (Jim Balter) writes:
>>You said "this may seem implausible but then there's CI".  Just what was your
>>point?
>
>Perhaps I should have said "this may seem implausible, but then you
>need to stretch the bounds of your implausibility tolerance." Special
>relativity was also widely thought implausible when it was
>introduced.  Some still consider it implausible.  My point was that
>lack of plausibility is not, by itself, a convincing counter
>argument.

There's well-founded implausibility based upon known facts and then there's
implausibility based merely upon emotional response and cognitive dissonance.
Perhaps I missed it; where did someone argue against you based upon the latter
sort of implausibility?  Your "argument" here applies just as well to any
theory, regardless of how hare-brained.  There are specific complaints about
your use of the term "free will" and your claim that we necessarily have it.
Suggesting that people reject it based merely upon some narrow implausibility
tolerance seems unfair.

>Consider the following thought experiment.
>
>We live in a universe where tossing a coin will lead to heads 90% of
>the time.  There is also an evil demon in this universe, who has
>subtle control over people.  The evil demon ensures that
>experimenters think that coin tosses will be heads 50% of the time.
>This is easy for him to do.  After all, the people don't really have
>free will, for he controls their every choice.  Any time any human
>has a choice which could allow that human to gain any information at
>all about the frequency of coin tosses, the evil demon controls that
>choice in such a way that the the aggregate of human observations
>actually made will show a 50% probability for heads.  Lacking free
>will, people simply will not look at the result of the coin toss on
>occasions when that would lead to contrary evidence.
>
>The result:  People draw a bogus scientific conclusion about
>	whether coin tossing is fair.
>
>Comment:  What difference does it make?  In all human affairs, coin
>	tossing will be fair.  For all practical purposes we might as
>	well say that coin tossing is fair.  If it does not match the
>	way the actual universe is, then we should not care about the
>	way the actual universe is.
>
>Final conclusion:  Subjective scientific knowledge is sufficient.
>	Subective free will suffices for acquiring subjective
>	knowledge.  Objective knowledge is unattainable, and in any
>	case is useless.  What we normally refer to as "objective" is
>	really the shared subjectivity of the culture.

I have no argument with this (maybe I'm just not trying hard enough :-),
except for your unnecessary insertion of the term "free will" in the final
conclusion.  And even then, you say that free will *suffices*, not is
required, so you haven't made the point you claimed to make before.  I still
see no coherent meaning of the phrase "free will", and I reject what I took to
be your argument that because meaning is usage-based then there must be a
coherent meaning for the term; that certainly doesn't apply to "flowing
purply", a point that you failed to address when I raised it.

There seems to be an inconsistency between your "result" and your final
sentence; the scientific conclusion is not bogus, it is an accurate
description of the subjective universe that is defined by our collective
subjective experience.  The conclusion factors in the behavior of the demon,
even though the demon is invisible in our model.

The scientific method is a mechanism or computation that produces a
distillation of a set of events; the distillation is a model that, when
instantiated, yields events of or approximating the same class as the input
events.  If the set of events extends into the future, i.e. is induceable,
then the model yields, or predicts, the future events.  Any number or type of
demons or whathaveyou can be held responsible for producing the event stream.
As long as they play the game, act consistently, remain subject to induction,
then the scientific method will work.  Hume told us that we can't count on the
demons to be so orderly; they could turn into imps at any time.  But William
of Ockham told us that we can do no better than to assume that the demons will
hold a steady course.  Perhaps you would say that we might as well assume they
have no free will.
-- 
<J Q B>
