From newshub.ccs.yorku.ca!ists!helios.physics.utoronto.ca!news-server.ecf!news-server.csri.toronto.edu!rpi!zaphod.mps.ohio-state.edu!pacific.mps.ohio-state.edu!linac!mp.cs.niu.edu!rickert Thu Jan 16 17:21:55 EST 1992
Article 2735 of comp.ai.philosophy:
Path: newshub.ccs.yorku.ca!ists!helios.physics.utoronto.ca!news-server.ecf!news-server.csri.toronto.edu!rpi!zaphod.mps.ohio-state.edu!pacific.mps.ohio-state.edu!linac!mp.cs.niu.edu!rickert
>From: rickert@mp.cs.niu.edu (Neil Rickert)
Newsgroups: comp.ai.philosophy
Subject: Understanding.  (was: Penrose on Man vs. Machine)
Message-ID: <1992Jan15.171155.2326@mp.cs.niu.edu>
Date: 15 Jan 92 17:11:55 GMT
References: <1992Jan13.230532.26592@gpu.utcs.utoronto.ca> <1992Jan14.040820.26868@unixg.ubc.ca> <375@tdatirv.UUCP>
Organization: Northern Illinois University
Lines: 69

In article <375@tdatirv.UUCP> sarima@tdatirv.UUCP (Stanley Friesen) writes:
>
>Quite true.  I see the call for a definition as a way of trying to resolve
>the issue.  Right now we are talking in circles at least in part because
>we do not have an adequate definition of such terms as 'understanding'.
>
>If we had:
>	1 - a definition of understanding that gave an *objective* criterion
>	    for determining if some entity had understanding,

 I have tried to formulate what we mean by "understanding".  I anticipate
most readers will disagree, but perhaps a discussion will clear the air.

Consider the following scenario.  In the best traditions of Science Fiction,
we assume that our alien visitors, while technologically advanced enough
to make it here through space, and while completely fluent in the English
language, yet are unfamiliar with our technology.

	A space craft arrives from the Andromeda galaxy.  Two visitors
	arrive on earth, and are given independent guides to show them
	around.

	When A-Andromeda sees a television set, he looks puzzled.  His
	guide asks him about it, and indicates in interest in understanding
	how it works.  The guide has some electronics books brought in.
	A-Andromeda studies these intensely.  Fortunately he is a speed
	reader.  After a while he admits that he does not fully understand,
	but asks to resume the tour.

	Later B-Andromeda in brought into the same room.  He turns the
	light switch off at the door.  Then turns it on again.  He repeats
	this sequence twice more.  Then he walks over to the television
	screen.  He turns that off and on twice.  His guide asks if he has
	any questions, but he indicates that everything is clear to him, so
	they continue the tour.

	Later, on the intergalactic cruiser, the two are taken to the
	debriefing room.  First it is A-Andromeda's turn.  He expresses
	confusion about the television.  He is questioned further, and
	is able to explain all the principles from Maxwell's equations to
	RF oscillators, intermediate frequencies, video bands.  But he is
	mystified by the idea of electron guns shooting at the screen.

	B-Andromeda is now debriefed.  The debriefer brings him directly
	to the question of the television, and what he understands of it.
	"Oh, it is all very simple," he says.  "I understand it completely.
	They use magic switches.  The switches near the door make it bright
	at night time, and the switch on the brown box creates pictures."

The point of all this?  Understanding ain't what we understand it to be.
"Understanding" has no generally agreed meaning.  It is something that
we personally interpret for ourselves, and nobody else can tell exactly
what we mean by it in any particular case.  Understanding a problem
generally means we no longer think or worry about it.  Such an
understanding could come from correct knowledge, from incorrect knowledge,
or from indifference.  But in most cases it probably arises from a complex
mixture of the three.

Educators routinely give examinations to test the degree of understanding
by their students.  In other words, they test in an effort to determine
whether the student meets the teacher's criteria for understanding.  It
should be evident that such testing methods could equally be used as
part of a Turing test.

-- 
=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=
  Neil W. Rickert, Computer Science               <rickert@cs.niu.edu>
  Northern Illinois Univ.
  DeKalb, IL 60115                                   +1-815-753-6940


