Dylan:
At 16:54 01/07/2002 +0100, you wrote:
>It strikes me that a significant difference between WG and TG is to do with
>the interraction between syntax and lexical items. In TG, the syntax is seen
>as a 'processing device' which takes static lexical items and operates on
>them; whereas is WG the lexical items themselves are actively involved in
the
>syntactic process.
## Yes, this is very similar to the rather vague picture that I have. In TG
the syntax provides a framework - the X-bar tree which defines 'positions'
or 'slots' - and the lexicon provides the passive fillers of those slots.
In WG it's the lexical items themselves that provide the slots for each
other. To introduce a political metaphor (which Chomsky would no doubt
hate), TG is a model for totalitarian State control: the State provides the
jobs, the words passively carry them out. In WG it's absolutely egalitarian
anarchy: each word defines its own needs and finds other individual words
to fulfill them. Most of the time it works smoothly, and we can live with
the occasional mess.

>

>I truly believe that the syntactic component is universal, sorry, but I do.
>However, it can be nowhere near as complex and powerful as TG suggests. To
my
>mind, syntac takes two units (lexical item or pre-generated structure) and
>joins them into a new (immutable) item.
## If this is all you mean by universal syntax, surely no-one would
disagree? Certainly no dependency grammarian.



Richard (= Dick) Hudson

Phonetics and Linguistics, University College London,
Gower Street, London WC1E  6BT.
+44(0)20 7679 3152; fax +44(0)20 7383 4108;
http://www.phon.ucl.ac.uk/home/dick/home.htm