Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!uunet!mcsun!ukc!edcastle!hwcs!raza
From: raza@cs.hw.ac.uk (Z. Raza Hussain)
Newsgroups: comp.ai
Subject: systemic grammar and realization rules
Keywords: natural language, linguistics, Halliday's systemic theory.
Message-ID: <2340@odin.cs.hw.ac.uk>
Date: 20 Feb 91 11:35:51 GMT
Sender: root@cs.hw.ac.uk
Reply-To: raza@cs.hw.ac.uk (Z. Raza Hussain)
Organization: Computer Science, Heriot-Watt U., Scotland
Lines: 18


something which i've been trying to understand for a long time about 
systemic grammars is how the idea of 'classification' and 'systems' fit 
with the syntax of languages.  from this, i mean how is it possible to
use computational methods to derive systems and classify constituents ?
is this where the idea of using realization rules comes into the theory ?
(is it that realization rules help to convert the system description
into classificatiion description? if this is so, then my original
question becomes : how is it possible to get a functional description
from examining syntax ?)

i realise that this theory is developed from a non-computing point of
view (sociological viewpoint of linguistics as originally observed by 
Malanowski) and is difficult to implement.

thanks for any ideas,

raza


Brought to you by Super Global Mega Corp .com