From newshub.ccs.yorku.ca!ists!helios.physics.utoronto.ca!news-server.csri.toronto.edu!rpi!zaphod.mps.ohio-state.edu!wupost!darwin.sura.net!mojo.eng.umd.edu!mimsy!harwood Mon Dec  9 10:48:23 EST 1991
Article 1900 of comp.ai.philosophy:
Path: newshub.ccs.yorku.ca!ists!helios.physics.utoronto.ca!news-server.csri.toronto.edu!rpi!zaphod.mps.ohio-state.edu!wupost!darwin.sura.net!mojo.eng.umd.edu!mimsy!harwood
>From: harwood@umiacs.umd.edu (David Harwood)
Newsgroups: comp.ai.philosophy
Subject: Re: Searle and the Rumpus Room
Message-ID: <44289@mimsy.umd.edu>
Date: 6 Dec 91 03:48:27 GMT
References: <YAMAUCHI.91Dec5040116@heron.cs.rochester.edu> <1991Dec5.191043.10565@psych.toronto.edu> <1991Dec5.210724.12480@cs.yale.edu>
Sender: news@mimsy.umd.edu
Organization: University of Maryland, College Park, MD 20742
Lines: 74

A bright young girl sits down in the rumpus room, and opens a
"WFF n Proof" game box, containing some instructions and dice with 
letters. It was given to her by an indifferent relative, a philosopher.
What should she do with it? What should she do?

If she has been reading the discussions of Searle, correspondents, and
critics, she knows enough to throw it away if it can't be swapped for 
something better.

In this story, we may assume that the bright girl in the rumpus room
understands English, but she hasn't heard of this game before. The game
box includes some instructions with rules, written in English, but with
some terms involving strings of letters (and possibly some other symbols)
which are * more-or-less * explained in English. However, she understands
the instructions (even the special symbols).

There are instructions and rules for 3 games. The first game is simply 
to roll the dice a few times and decide whether the string of letters 
is a WFF; the second game is to decide whether such a (WFF) string
is valid; the third game is to decide whether a such a (WFF) string is
"a theorem of Monopoly." (We'll have to wait to see what this means for
our story. All we can assume for now is that she's heard of Monopoly but
doesn't know how to play it, since her indifferent relative, the
philosopher, disdains to buy the game for her, nevermind play it with her.)

The little girl roles the dice and plays the game the best she can, 
according to the rules as she understands them, and does pretty well 
indeed. (Much better than her indifferent relative, the philosopher,
did at the same age, to be sure.)

Then her indifferent relative, the philosopher, arrives for supper with 
his friends, and proceeds to criticize her as having the IQ of a Coke 
machine (meanwhile he and his friends are trying to decide which forks, 
spoons, and knives to use).

What does this story prove?

Not a whole lot - I would venture to say - except that philosophers are
no fun.

Rather, what is proved by this story "more-or-less" depends on what 
were supposed to be instructions of the "WFF n Proof" games.

Let us suppose that the instructions allow the little girl:
(i) to correctly decide whether a string of letters is a WFF; this game
is pretty easy, if you can use paper and pencil.
(ii) to usually correctly decide whether a WFF string is valid; this game
pretty frustrating.
(iii) to usually correctly decide whether a WFF string is "a theorem of
Monopoly;" this game is very frustrating (especially since she's never
played the game).

Is she smarter than an augmented Coke machine (ACM)? Does she really,
really understand what the "WFF n Proof" games are all about?

Well, the philosophers publish disputes about this for years, in Brain 
and Behavioral Science. But in my opinion the problem is ill-posed.

That is:
------------
Do these instructions for the games, written in English but partly 
referring to strings of letters, interpret these strings by translation 
in English, or not? They don't have to, in order for the little girl
to play the games so far. But then she doesn't "understand" the games,
as being games of logic, neither does she learn anything about Monopoly.

In either case, the little girl plays the "WFF n Proof" games about
equally well, following the purely syntactical rules. But if there are
auxiliary rules of interpretation by translation, then she may indeed
"understand" the games depending on the completeness of these rules and
upon her prior ability to correctly interpret English translations
given by them. And accordingly she will learn something about Monopoly,
which her indifferent relative, the philosopher, disdained to teach her.



