[Corpora-List] ad-hoc generalization and meaning

[Rob Freeman]

Eric,

On 9/12/07, Eric Atwell <eric at comp.leeds.ac.uk> wrote:
>
> On Wed, 12 Sep 2007, Rob Freeman wrote:
>
> > Do I need to map from one set of rules to another? Can't I just
> interpret
> > the syntactic rules directly as some kind of logic? These are formal
> > systems, after all, just combinations of symbols. It might not be the
> logic
> > we are used to, but doesn't any combination of symbols define a "logic"
> of
> > its own?
>
> I've managed not to be drawn into this discussion but couldnt resist
> making one point:
>
> A logic isnt just a set of symbols (representing fact/knowledge),
> but also rules defining how to make inferences: how to infer new facts
> from known facts. The problem is that symbols/categories used as
> standard in grammar are not all the same as those used by logicians.
> For example in grammar, "and" and "or" are both Coordinating
> Conjunctions, but they are mapped onto very different logical functions.
>
> Some researchers have tried a "surface" approach to knowledge
> representation and reasoning with specific syntactic constructs,
> eg Silver N "Inferencing methods using systemic functional grammar"
> (PhD Leeds Univ 1995) dealt with polar and wh-questions; but noone has
> proposed a logic covering the whole gamut of grammatical constructs found
> in a Corpus, as far as I know - please correct me if I'm wrong.


I'm surprised you didn't mention cognitive grammar here. Isn't cognitive
grammar essentially an attempt to enumerate "a logic covering the whole
gamut of grammatical constructs found in a Corpus"? They try to equate
grammar with meaning, anyway.

-Rob
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listserv.linguistlist.org/pipermail/corpora/attachments/20070913/64f04093/attachment.htm>
-------------- next part --------------
_______________________________________________
Corpora mailing list
Corpora at uib.no
http://mailman.uib.no/listinfo/corpora