From newshub.ccs.yorku.ca!ists!helios.physics.utoronto.ca!news-server.ecf!news-server.csri.toronto.edu!rpi!usc!cs.utexas.edu!uunet!psgrain!quagga!hippo!ucthpx!uctvax.uct.ac.za!trdgav01 Tue Apr  7 23:23:16 EDT 1992
Article 4823 of comp.ai.philosophy:
Path: newshub.ccs.yorku.ca!ists!helios.physics.utoronto.ca!news-server.ecf!news-server.csri.toronto.edu!rpi!usc!cs.utexas.edu!uunet!psgrain!quagga!hippo!ucthpx!uctvax.uct.ac.za!trdgav01
>From: trdgav01@uctvax.uct.ac.za (Gavan Tredoux)
Newsgroups: comp.ai.philosophy
Subject: Re: Definition of understanding
Message-ID: <1992Mar28.180811.201383@uctvax.uct.ac.za>
Date: 28 Mar 92 18:08:11 +0200
References: <6384@skye.ed.ac.uk> <1992Mar12.191404.1316@ccu.umanitoba.ca> <6421@skye.ed.ac.uk> <1992Mar18.062736.5270@ccu.umanitoba.ca>
Organization: University of Cape Town
Lines: 62

In article <1992Mar18.062736.5270@ccu.umanitoba.ca>, zirdum@ccu.umanitoba.ca (Antun Zirdum) writes:

Let's try a little experiment here. Seeing that most readers
probably don't understand the Afrikaans language native to
South Africa, we might be able to shed some light on what it
is to understand a new word - more complicated things
like grammar can follow.

Start with the word `verkrampte'. This is Greek to most, so
in order to explain it I might try to tanslate it into 
English, say as `conservative'. Mail this information to 
John Searle. Now suppose that Searle,
eager to impress his colleagues at work, starts using this new word
casually in conversation, eg. "Isn't he a bit verkrampte?"
in reply to an invitation to dinner with Rameses II. 
Hmm. Does Searle know what the word MEANS ! Well, he
is familiar we assume with the SYNTAX, in this case simply how
the word is spelt, and all we have given him is a simple association
or mapping to something else, an English word. His use of the word is
based solely on this association, and it is fair to accuse him of blind
manipulation of the word in the accordance with a set of rules, those
associated with the manipulation of `conservative'. That he doesn't understand
the word, by his own vague criteria, is evidenced by his insensitive
faux pas - one of his colleagues is Afrikaans and realizes that `verkrampte'
might also be translated as `hidebound', a description a mummy might take
exception to. 

Now how will Searle come to understand the word? His faux pas might be pointed
out and he would avoid the blunder in future. But unfortunately, `verkrampte'
is one of those simply untranslateable words with no direct equivalent in the
English language. Yet most second-language Afrikaans speakers in South
Africa would claim to know what the word means, to understand it, even though
they have come to it by what is nothing more than an extension of Searle's
route. They can make a better claim to understanding simply because they can
use it correctly in more situations and more contexts; all the time the word
remains no more than a string of letters and an expanding set of rules for its
use, there is no intrinsic and hidden MEANING accesible only to those who
TRULY UNDERSTAND, just a set of conventions. Note that people can make mistakes
in the use of a word, even if they are native speakers of the language
involved, and this often results in meaning shifts over time; there is no
`keeper of the key' here.

That computers can do this sort of thing just as well, in principle, should be
obvious; its just the degree of complexity, a difference of quantity that
appears to entail a difference of quality, that obscures this. The example
above was meant to demonstrate that the process of LEARNING unveils the
pretensions of UNDERSTANDING. To take the experiment further, I'll supply
another Afrikaans word and invite posters to supply a sentence containing it.
Of course I'll give you a clue to its `meaning' so that things speed up a
little.

The word is `oorbeligte'. A clue to its usage is an example
`The University of Stellenbosch contains a number of oorbeligte academics who
are bent on selling the nation out'

Shoot.

+----------------------------------------------------------------------------+
| Gavan Tredoux, Dept. Math. |  "Einstein and people like Einstein said that |
| University of Cape Town    |   the world was flat" T(uesday) Lobsang Rampa |
| Rondebosch 7700 RSA        |                                               |
+----------------------------------------------------------------------------+


