Path: utzoo!utgpu!jarvis.csri.toronto.edu!mailrus!uflorida!haven!aplcen!jhunix!ins_atge
From: ins_atge@jhunix.HCF.JHU.EDU (Thomas G Edwards)
Newsgroups: comp.ai
Subject: Re: Question on Chinese Room Argument
Summary: Understanding
Keywords: people understand
Message-ID: <1139@jhunix.HCF.JHU.EDU>
Date: 15 Mar 89 17:12:10 GMT
References: <Mar.4.00.45.52.1989.4435@elbereth.rutgers.edu> <9560@megaron.arizona.edu> <2568@crete.cs.glasgow.ac.uk>
Reply-To: ins_atge@jhunix.UUCP (Thomas G Edwards)
Organization: The Johns Hopkins University - HCF
Lines: 44

In article <2568@crete.cs.glasgow.ac.uk> gilbert@cs.glasgow.ac.uk (Gilbert Cockton) writes:
>The system passes the LTT (because Searle so defines the gedanken
>experiment), but it DOES NOT understand - certainly not in the sense of
>the way people use the word.

	It appears that "people" cannot use the word "understand" to
refer to cognitive associations of computers, only people, and 
occasionally animals.  Despite popular usage, there is no reason to
expect artifacts not to be able to do what we consider "understanding"
in humans.  But until we ourselves define what "understanding" is,
which the discussion on this group have failed to come to a conclusion
upon, we cannot prove anything concerning artifactual understanding.

>Everyone is in the grip of some ideology, but the systems' one is just
>plain silly if it attributes "understanding" to a system.  I am a
>holist, but I don't see how an attribute of a part can be transferred
>to the whole if it doesn't exist in the part.

   You say you cannot understand how an attribute of a part can be
transfered to the whole if it doesn't exist in the part.  This is
reasonable.  However, an attribute of _parts_ can be transfered to
a whole if it doesn't exist in any singular _part_.  (i.e. summing
a+b+c+d can be accomplished by a system of three parts, one which
sums a+b, another which sums c+d, and a third which sums the output
of the first two parts).

>  The interesting thing
>about systems is the attributes of the whole which CANNOT be attributes
>of the parts, not true here I'm afraid.

	You are saying the attributes of the whole CAN be the attributes
of the parts here...I am not sure I understand your concept here,
but if we assume you mean "here" to refer to "cognition" then you
say that the parts of cognition are capable of cognition, the
parts of understanding understand.  If we assume the human body to be
made up of parts (atoms and electrons), from the above assumption,
we are assuming that (atoms and electons) can understand.

	The conclusion I'd like to draw is that systems _typically_
have attributes which one would find very, very difficult to imply
from examining each part.  (examples...any dynamic system:  Julia Sets,
turbulence in air or fluids, time between faucet drips, etc.)

-Thomas Edwards