Re: LAhE = lo broda be ?

xorxes:
> la and cusku di'e
> 
> > xorxes:
> > > Can we say that for every LAhE there is some broda that
> > > can replace it so that {LAhE ko'a} = {lo broda be ko'a}?
> > 
> > PA1 LAhE ko'a = PA1 broda be ko'a
> 
> Right.
> 
> > > lu'i ko'a = lo selcmi be ko'a
> [...] 
> > > If this is so, then understanding the interaction of LAhEs and
> > > quantifiers is relatively easy and there are no additional
> > > decisions to make. 
> > 
> > Right.
> 
> But notice that this means {lu'i ko'a} is a set that includes ko'a
> but does not necessarily exclude anything else.

Ah, sorry. I wasn't meaning to comment on the actual choices of
predicates. I thought they were just candidate examples, since
that wasn't the main point of your message.

> If we want the set that has ko'a and only ko'a as a member, we
> can't use {selcmi}. If we use a broda that means "x1 has x2 as its 
> only member", then we can't use it for {lu'i ko'a e ko'e} to
> get {ko'a ce ko'e}...

Yes. I was thinking we'd want a predicate that works as I described
in my earlier message -- "includes and excludes everything except".
You have shown that this fails for disjunctive cases.

--And.