Re: THEORY: OO theories: was: Re: double negatives

Just to reassure you a bit, Goldberg's Construction Grammar basically
comes down to this:

A grammar is a structured inventory of constructions which encode
meanings.  (A "construction" here is pretty much any recognizable
pattern among human-producable sounds, and includes morphemes, words,
and all kinds of syntactic and morphological patterns; a "meaning"
here is pretty much anything conceptualizable by a human.)

Forms can be combined to encode meanings for which a single
conventional form does not happen to exist.

That's all there is to it.  One doesn't have to even so much as
assume the existence of nouns and verbs to use the construction
grammar framework, though I expect one usually would.

The "object orientation" thing only comes up in explicating what kind
of "structure" organizes the "inventory" of constructions that make up
a language -- specifically, Construction Grammar suggests that a given
form-meaning pair may have a number of more specialized variants, like
subclasses in OOP.

So don't worry yourself about theories that get in the way of
grammars...  If anything, this kind of theory just tries to establish
an explicit groundwork for something resembling the Dixonian "basic
theory" that has captured your heart. :)

BTW, I don't know what fieldwork Fillmore or Goldberg have done, but
Ronald Langacker, who works within a nearly identical framework he
calls "Cognitive Grammar," cut his teeth on Mesoamerican languages and
peppers his theoretical work with examples therefrom.


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:

> I was already a bit afraid that there might be object-oriented theories
> of language out. I think that what the world needs is more grammars
> (and more languages ;-)), not more theories...
>
> Boudewijn Rempt  | http://www.xs4all.nl/~bsarempt
>