Martin Wynne wrote:
> Actually this isn't a case of corpora showing us the problem -
> their intuitions about the phrase should have told them this.
> All the corpus work is doing is to provide the evidence to
> back up the intuitions.
As a Bad Guy on this list (read: Generative Linguist), I had to chuckle. Of
course the real point is that the two approaches are complementary, but
still...
Mike Maxwell
Linguistic Data Consortium
maxwell@ldc.upenn.edu
This archive was generated by hypermail 2b29 : Fri Oct 17 2003 - 15:25:45 MET DST