[Corpora-List] Is a complete grammar possible (beyond thecorpus itself)?

[Rob Freeman]

On 9/10/07, John F. Sowa <sowa at bestweb.net> wrote:
>
>
> I was not complaining about the conclusion, but about the argument.
> You were getting into issues of completeness and decidability, which
> only apply to formal languages.  Since NLs are not formal, you can't
> apply those notions to NLs.


The properties of formalizable systems are the properties of grammars, which
were the topic of this thread.

Attempts to describe natural language as a formal system have been uniformly
unsuccessful. One reaction has been to reject all formal analysis. This
seems to be your position.

My argument here has been that no-one has considered a third possibility:
that NL may be amenable to formal analysis (e.g. distributional analysis),
but simply incomplete when described as a formal system.

As far as I know this possibility has not been proven wrong, it has simply
not been considered. If you have evidence to the contrary I would love to
see it.

-Rob
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listserv.linguistlist.org/pipermail/corpora/attachments/20070910/814dc522/attachment.htm>
-------------- next part --------------
_______________________________________________
Corpora mailing list
Corpora at uib.no
http://mailman.uib.no/listinfo/corpora