Re: Cowan's summary: opacity and sumti-raising

[Jorge Llambias <jorge@PHYAST.PITT.EDU>]

> > Here I would use {lo'e}:
> >
> >         mi sisku lo'e xe fanva be la'o sy Diana sy bei la gliban
> >
> >
> > {lo'e broda} doesn't claim that {lo broda} exists, does it?
>
> I don't know that that has been settled.  But I find the idea of an
> archetype of a non-existent (in the appropriate universe of discourse) thing
> rather problematic.  What could be predicated of this {lo'e xe fanva}, other
> than what we say in the embedded place structures?

Lots of things. It can be looked for, needed, wanted, etc. Any predicate
that makes sense with an opaque reference.


Another problem of not making the x2 of sisku the object of the search is
that it makes it very difficult to single out the looked for thing.
For example, in a game of hide-and-seek, there is {le sisku}, and what
I'd like to call {le se sisku}, but I can't with sisku as it is now.

Jorge