Daniel Andreasson wrote:
>
> Jerry wrote:
>
> > by DeLancy. He claims that all case
> > is at root THEME-LOCATION, spatial; and I am going to try to express
> > these cases with an adverbialized format of my own that is similar. So
> > far I have found it to be true that one can express case with the just
> > the verbs "is_located" and "moves; (dis and pas from my vector space
> > grammar) combined with adverbs.
> > In this system there are only two verbs. They are modified with
> > tense particles and optionally detachable adverbial suffixes.
> > Free standing adverbs together with the "glueable" adverbial
> But to have only two verbs. Hmm... That seems a little
> too unorthodox for me. I don't think I'm capable of making sentences
> in that fashion (yet). A bit too abstract for me.
Another idea of Delancey is to look at the "diachronic" (over time)
development of grammatical features. Maybe in a proto-vector-pidgin,
the 2 ur-verbs get noun roots glued onto them, so "move eyes past xxx"
would become "moveyepas xxx" or something like that. It seems that NGL
is especially fluid in its word boundaries, so such things might happen.
I like Delancey's papers on localism because they help to analyze
complex natlang constructions into intuitively pleasing parts,
and then use the diachronic processes to rebuild them ...