Re: double negatives

Well, I think the situation here is that Object Oriented Programming
is intended to capture some aspects of human classification.

It does so -- in a limited way.  (Real human classification is
probably a bit more like an OOP programmer's acid-induced nightmare of
hideously complex mazes of multiple inheritance, all of which is
constantly changing in real-time in an ad hoc manner...)

Just to clarify, Fillmore & Goldberg's grammar uses the concept of
class heirarchy and inheritance from OOP because *in their opinion,
these aspects of OOP actually do usefully model aspects of human
classification and cognition*, and they want to include those in their
grammar.

Also because it gets good results.  Goldberg's analysis of the
caused-motion construction ("I sneezed the napkin off the table."
"They laughed the speaker off the stage") and Lakoff's analysis of the
existential "there is" construction in English work well, and work
well because of the analysis in terms of "inheritance" from other
constructions.

Ed Heil ------------------------------- edheil@postmark.net
"Facts are meaningless! You can use facts to prove anything
   that's even _remotely_ true!"           -- Homer Simpson

Boudewijn Rempt wrote:

> On Wed, 21 Jul 1999, Ed Heil wrote:
>
> > Boudewijn Rempt wrote:
> >
> > By Adele Goldberg, using the recent theory of Charles Fillmore.  She
> > assumes that a grammar consists of "core" constructions, and
> > derivative constructions based on those core constructions and on each
> > other, which "inherit" the characteristics of the constructions they
> > are derived from, unless explicitly specified otherwise.  Goldberg
> > specifically references object-oriented programming as the source for
> > this idea.
> >
>
> Now why do I get such an awful feeling of deja-vue? I'm very
> much reminded of the steam-machine theory of the human body...
>
> Boudewijn Rempt  | http://www.xs4all.nl/~bsarempt
>