Question formal status of trees in H&GPSG

[Andrew Carnie]

On Fri, 27 Jun 2003, Ken Shan wrote:

> failure".  Anyway, returning to your question, which distinguishes
> between the following options for a notion of domination.
>
> On 2003-06-26T16:45:42-0700, Andrew Carnie wrote:
> > 	-X  *replace* Y and Z (as in  TG); meaning that the representation
> > 		contains only Y and Z; X is a historial artifact (or vice
> > 		versa, the representation contains only X, and Y and Z are
> > 		historical artifacts.
> > 	- or does X *contain* Y and Z (as in MP), there is one object in
> > 		the representation (X), which contains all the material
> > 		formerly in Y and Z. Y and Z no long exist except
> > 		derivationally
> > 	- or does it *represent* Y and Z (as in GB), X, Y and Z are all
> > 		identifiable objects in the representation (and
> > 		derivation). They are related through structural relations.
> > 	-  or something entirely different?
>
> To clarify what you mean by "replace", "contain", and "represent",
> let me ask you the following question: Can you give an example of a
> grammatical rule that would "make sense" (i.e., use only available
> operations on phrases) if X contains Y and Z, but not if X represents Y
> and Z, or vice versa?  It would clarify for me the difference between
> domination in MP and in GB.
>
> Cheers,
> 	Ken

Hi Ken,

It's become clear to me that what is a crucial issue in MP phrase
structure theory has no bearing in HPSG, and that's due to the fact that
HPSG is essentially representational, while MP is procedural. So my
question was really a non-issue.

For your information, though, the issue in MP has to do with structure
preservation. In a GB style approach -- at least in principle -- you can
go back and "mess with" dominees. That is given a category X, which
dominates Y and Z, X represents the complex Y and Z, but Y and Z still
have independent status in the structure: You can perform
operations on them, *independently* of the category that dominates them.
(ie. in a derivation. You take Y and Z; form X which has say the
unification of the features of Y and Z. Then you can go back and "move"
some of the features of Z out of the structure. An example of this would
be verb movement. Where X is a VP. By moving V you are essentially
removing the feature structure that gave the "V"-ness to the VP. And as
such V is only the head of the VP derivationally. By contrast in
MP, at least in the most recent version of it (which involves a
combination of Bare Phrase Structure theory and phase theory), the
category X *IS* the collocation of Y and Z (I use collocation here because
I want to avoid the term unification, because it isn't really unification
in the sense you guys use it, but it corresponds roughly to the operations
you describe as unify and embed). That is, ontologically speaking, once Y
and Z are merged to for X, Y and Z are no longer available as independent
objects, although they are, of course present within X. But they are not
available for operations independent of one another.

I do need to say one caveat: Many MP-linguists actually work as if
they are using the GB conception of things; so my description above is
not to be taken as "definitive" of the MP view of the world. I'm just
taking what CHomsky said in BPS seriously (a dangerous thing, I know).

In any case, many thanks to all of you who responded. It has not only been
helpful in understanding the ontology of HPSG, but also in helping me
firm up some vague thoughts I had about MP.

Best,

Andrew