Newsgroups: comp.ai.philosophy
From: khaled@mardam.demon.co.uk (Khaled Mardam-Bey)
Path: cantaloupe.srv.cs.cmu.edu!das-news2.harvard.edu!news2.near.net!howland.reston.ans.net!news.sprintlink.net!demon!mardam.demon.co.uk!khaled
Subject: Re: Roger Penrose's fixed ideas)
References: <snodgras.27.2ED798A1@cts.com> <39posv$mr0@nnrp.ucs.ubc.ca> <CzFr3J.990@cogsci.ed.ac.uk <JMC.94Nov22011226@white.wisdom.weizmann.ac.il> <Czu5zD.Dto@festival.ed.ac.uk>
Distribution: world
Organization: Lack thereof.
Reply-To: khaled@mardam.demon.co.uk
X-Newsreader: Newswin Alpha 0.6
Lines:  51
Date: Sun, 27 Nov 1994 01:50:08 +0000
Message-ID: <750280541wnr@mardam.demon.co.uk>
Sender: usenet@demon.co.uk

In article: <snodgras.27.2ED798A1@cts.com>
snodgras@cts.com (John E. Snodgrass) writes:
> In article <Czu5zD.Dto@festival.ed.ac.uk> cam@castle.ed.ac.uk
> (Chris Malcolm) writes:
>       The essence of the Chinese Room argument is that the Chinese room does 
> not understand what it is doing from _its own_ viewpoint, not from the 
> viewpoint of an outsider. What is meant by the term "understanding" and what 
> is meant by the term "adding" or for that matter "throwing" or "waltzing" etc 
> are fundamentally different concepts -- not because we feel less "proud" of 
> one than another, but because of the self-referential nature of understanding 
> as opposed to the essentially mechanical activity of the others.
> 
>       When we understand something, we are not simply performing a sequence of
> actions, but instead are recognizing a relationship between ourselves and the 
> environment, between our goals and our actions. We do not learn (or 
> understand) in a vacuum: we do so in a context of positive and negative 
> results -- _from our own perspective_. The dominance of pleasure and pain in 
> determining what we learn (no matter how indirect it may become) is a 
> non-trivial difference between ourselves and machines. Our hardware (if you 
> like) is constructed on the principle of maintaining integrity by recognizing 
> threats to its own stability. 
> 

And that is precisely what makes the chinese room argument fallacious. 
For one thing, the lookup tables and the books that the person in the chinese
room must go through cannot possibly be static. They *must* be dynamic. Why?
Because of memory. If a question refers to a situation or a statement 
or sentence that was presented 10 questions ago, the books/tables must 
be able to reply to *implicit* and *remembered* information.

What if a submitted question refers to a question or story that was 
presented 100 questions ago? An idea that was discussed a 1000 
questions or statements or whole stories ago?

What Searle is asking us to accept is that there is in fact not only a
language *understanding* mechanism in those books and tables, which
can deduce implicit information and idiosyncracies of language, but
also a complete human being! The books must reflect experience, knowledge,
learning, previous statements that they made etc. so that they are coherent 
in future replies. Otherwise the books might as well be choosing answers
randomly, and in such a case the replies would NOT make sense to people
outside the room.

Once we see this, the next glaring hole is that the human is not NEEDED
to perform the task! A simple machine could easily perform the same tasks
of matching symbols, turning pages etc. Thus the human doesn't *HAVE* to
understand or be conscious of *anything*. On top of this, the human's
role in the system ends up being *minute* when compared to the vast amount
of information required for the process, so the human could in no way, even
if he/she was translating an *ENGLISH* sentence using this vast number of
rules, ever understand what was happening! (until the whole reply had been
spelt out ofcourse, but that's a little bit too late to experience
the consciousness of the process :-)

Therefore, Searle's whole argument crumbles. What he should have said 
is "Imagine that we have in these books and tables rules that reflect 
a whole human being, complete with experiences, knowledge, beliefs, 
character, and the ability to think, learn, and feel..." at which point
we would probably laugh and say, "Yes, that's very interesting... Now
imagine Pink Elephants..." :-)

Cheerio,
K.Mardam-Bey

=================================================================
MSc. Cognitive Science Student |
xhlec@wmin.ac.uk               | "With a gentleman I try to be a
http://www.wmin.ac.uk/~xhlec   |  gentleman and a half, and with
                               |  a fraud I try to be a fraud
Flames will be used to roast   |  and a half." -Otto von Bismark
marshmellows...                |
=================================================================

