flat structures

[Shannon West]

Hi Catherine. Thanks for the reminder on the relative clause paper, I'd
forgotten that one.

>
> Shannon -- Williamson has configurational trees for Lakhota in her article
> on internal-headed relative clauses.  (Janis S. Williamson, An
> Indefiniteness Restriction for Relative Clauses in Lakhota, in E. Reuland
> and A.G.B. ter Meulen (eds), The Representation of Indefiniteness.  MIT
> Press 1987).   Also in her dissertation (Williamson 1984 Studies
> in Lakhota
> Grammar, UC San Diego) there are trees which look configurational to me,
> though she states in the introduction that Lakhota has "flat
> structure" and
> "nonconfigurational phrase structure rules".  Maybe I don't
> understand what
> "flat" means!!!   Or at least what W. means by it.

She's got a bunch of things called virtual structures, in which she needs
the VP, but says that there is no surface structure difference between
Subject and Object. I've been having fun with her work. :)

>  I've assumed configurational structures for both nominal phrases (DP) and
> clauses in various papers on Omaha-Ponca, though it's not clear to me just
> HOW configurational it is --  the type of trees David and I both presented
> in Boulder lst fall, with layers of functional heads, seem very plausible,
> but I'd hesitate to try to argue for e.g. a VP constituent.  Williamson
> does show a VP in some of her trees.

I love the functional categories, they work so well for so many things. I'm
not using a whole pile of them (I'm very carefully avoiding the issue), but
that's just because I don't need this thing turning into a 1000 page volume.

> What are your thoughts on the issue?  I'd love to know.  Or do we have to
> wait for your dissertation?

Heh. It's directly tied into the pronominal argument stuff I was talking
about last year. I'm trying very hard to make a case for DP arguments *and*
pronominal arguments, the explanation Randy favours. It makes the most sense
intuitively, but it's hard to get the theory to work. It would require that
sentences with 1st and 2nd person arguments be somehow non-configurational.
The verb would have to check features via movement. In the event of 3rd
person arguments, those occupy the normal configurational structure argument
positions (or possibly heads of discourse functional categories)

Something hit me a while back. It doesn't matter which way you look at it
(lexical arguments or pronominal arguments), 1st and 2nd person are
mandatorily pronominal arguments. In a lex.arg. theory, they're pro. In a
pro.arg. theory, they're affixes. I've pretty much satisfied myself that DPs
are in argument position (Binding Condition C, no apparent Wh-movement, and
whatnot), so it really is a split system. I just had to decide whether I
wanted the 1st and 2nd person arguments to be pro or affixes. And it makes
more sense to me, that they would be affixes. If they were pro, then the
affixes would be agreement, and it would be nice to remove that layer of
abstraction.

I say that I've "pretty much satisfied myself" with regard to the position
of DPs, because I just read something else that seemed wonderfully plausible
(Gotta both love and hate it when that happens). Russell and Reinholtz
(1999) wrote a fantastic article about configurationality and pronominal
arguments in Cree. Cree has long been touted as the posterboy for
non-configurational languages. Almost no one argues against the pronominal
arguments in it. But R&R say that just because the arguments are pronominal,
that doesn't necessarily make the DPs adjuncts. Instead, they put them into
configurationally  structured trees under the functional categories of Topic
and Focus. I'm trying to work out all the implications of that idea before I
jump on it, but it seems like a nifty alternative, especially since in
Nakota the only way to have an OSV structure is if the object is the focus.
But I've not got any of the details worked out.

My advisor just got back in town after a year's leave, so I'll have lots of
things to work out with her. I'll get back to you once I cement my ideas at
least a little. :)

All the best,
Shannon