From newshub.ccs.yorku.ca!ists!helios.physics.utoronto.ca!news-server.csri.toronto.edu!psych.toronto.edu!christo Tue Apr  7 23:24:33 EDT 1992
Article 4962 of comp.ai.philosophy:
Newsgroups: comp.ai.philosophy
Path: newshub.ccs.yorku.ca!ists!helios.physics.utoronto.ca!news-server.csri.toronto.edu!psych.toronto.edu!christo
>From: christo@psych.toronto.edu (Christopher Green)
Subject: Re: SHRDLU's mind
Organization: Department of Psychology, University of Toronto
References: <1992Apr6.023638.518@organpipe.uug.arizona.edu> <1992Apr6.182533.109@psych.toronto.edu> <1992Apr6.224129.7406@organpipe.uug.arizona.edu>
Message-ID: <1992Apr7.211232.6930@psych.toronto.edu>
Date: Tue, 7 Apr 1992 21:12:32 GMT

In article <1992Apr6.224129.7406@organpipe.uug.arizona.edu> bill@NSMA.AriZonA.EdU (Bill Skaggs) writes:
>Bill Skaggs:
>>  I think I'm a backer of strong AI, but I don't believe that
>>mind is an all-or-nothing concept.  Minds come in varying degrees
>>of sophistication.  
>
>Christopher Green:
>>Varying degrees of sophistication have no bearing on the question of
>>presence or absense.  A small mind is still a mind, just as a small ball
>>is still a ball.
>
>  Is a single molecule of rubber a ball?  How many molecules of
>rubber does it take to make a ball?
>
A single molecule of a rubber ball is not a ball, now is it. A small
rubber ball, no matter how small, is.

>  I believe that all natural concepts are fuzzy at the edges.  For
>prototypical exemplars presence and absence are clear, but there
>are always doubtful exemplars where presence or absence is a
>matter of degree.  

I believe this is a notion borrowed from Rosch, who borrowed it not
entirely accurately from Wittgenstein.  Moreover, I believe it is
fundamentally misguided, for reasons that I borrow from Stevan Harnad.
They go, roughly, like this: the simple fact that we have trouble
definitively classifying some objects does not imply that they are
not definitively classifiable. Nor does it imply that they have some
sort of graded classification. It simply means that we have trouble
classifying it.  When we say that we're not sure, e.g., whether a
preist is a bachelor, we are not saying, a la Rosch, that he is 80%
bachelor or some such nonsense.
>
>  I like to think of SHRDLU as having mind to a very small degree,
>and a thermostat as having an "atom" of mind.  A rock, because it
>contains no representation of the world and does not communicate
>or act purposefully, has no mind at all.
>
What you "like to think" has little bearing on the question. Surely
whether SHRDLU, thermostats, and rocks have minds is an empirical
question.

>BS:
>>The essential ingredients are a picture of
>>the world, the ability to use it to accomplish goals, and some
>>ability to communicate.  
>
>CG:
>
>I don't quite understand the description of the robot.  Does 
>the robot have some sort of representation of its position?  

Yes. It has a pointer that moves around the paper in a manner isomorphic
to its movements around the table (assuming both have been set correctly
from the outset). It stops *because* it gets to the edge of the paper.
This also *happens* to be when it gets to the edge of the table.

>Does
>it move according to some plan?  Or does it just move randomly
>around on the paper until it comes to the edge?  

I'm not sure, but it's irrelevant. If you want a plan (move left two
feet, forward two feet, right one foot, etc.) take one.

>The answers
>to these questions determine how much mind I would be willing
>to attribute to the robot (but very little in any case).

I still contend this is a yes/no question. And that the answer is "no".


-- 
Christopher D. Green                christo@psych.toronto.edu
Psychology Department               cgreen@lake.scar.utoronto.ca
University of Toronto
---------------------


