From newshub.ccs.yorku.ca!ists!helios.physics.utoronto.ca!news-server.csri.toronto.edu!rpi!zaphod.mps.ohio-state.edu!qt.cs.utexas.edu!yale.edu!yale!mintaka.lcs.mit.edu!spdcc!dirtydog.ima.isc.com!ispd-newsserver!psinntp!scylla!daryl Mon Mar  9 18:33:58 EST 1992
Article 4149 of comp.ai.philosophy:
Newsgroups: comp.ai.philosophy
Path: newshub.ccs.yorku.ca!ists!helios.physics.utoronto.ca!news-server.csri.toronto.edu!rpi!zaphod.mps.ohio-state.edu!qt.cs.utexas.edu!yale.edu!yale!mintaka.lcs.mit.edu!spdcc!dirtydog.ima.isc.com!ispd-newsserver!psinntp!scylla!daryl
>From: daryl@oracorp.com
Subject: Re: Semantic and Syntax
Message-ID: <1992Feb28.172808.1956@oracorp.com>
Organization: ORA Corporation
Date: Fri, 28 Feb 1992 17:28:08 GMT
Lines: 46

kimhock@csar.uucp (Ng Kim Hock) writes:

[speaking about syntactic rules in mathematics]

> Viewed from this angle, it would appear that a purely syntactic approach 
> could solve certain problems and in fact when a person is asked to prove 
> certain properties. I am wondering whether if a person who found the 
> solution is deem to have understood the problem
> 
> Alternatively, we could associate the problem to the real world and try to 
> imagine a quantity of something being added to another quantity to form a 
> larger quantity. This is a more a Physical approach and really shows the 
> semantic. 
> 
> On the other hand, one could argue that the physical semantic is only a 
> particular interpretation of the syntactic relations and that the syntax
> encapsulates something very general that can carry different semantics 
> depending on how we wish to interprete it.

I believe that your description of things is correct. However,
controversy arises over the difference between semantics of computer
programs and semantics of human beings. Let me give you the standard
argument for why semantics in human beings is different from semantics
in computer programs. (I don't happen to believe this argument,
although I do understand it.)

The main issue is the existence of some semantics which is "correct".
The thoughts of human beings have a preferred semantics; when I talk
about "cats", there may indeed be a consistent way to interpret what I
am saying so that I seem to be talking about arithmetic or chess. (I
posted an article once showing how you can, in principle, consistently
interpret any statement to be about any subject.) However, I would say
that such nonstandard interpretations are wrong, even if they fit the
syntax of what I am saying.

On the other hand, in the case of computer programs, there doesn't
seem to be any wrong interpretations; one interpretation seems just as
good as another. Given that there can be multiple interpretations that
are equally good, it doesn't make sense to say that the interpretation
is part of the computer program---the interpretation must be something
imposed on the program from outside, and the program itself is
completely without semantics.

Daryl McCullough
ORA Corp.
Ithaca, NY


