* Friday, 2011-09-02 at 11:47 -0700 - ianek <janek37@gmail.com>:
> On Sep 2, 4:51 pm, Martin Bays <mb...@sdf.org> wrote:
> > Cute Russel for properties, which translates nicely to lojban:
> >
> > xu lo ka ce'u ce'u na ckaji ku vo'a ckaji
> >
> http://dag.github.com/cll/11/4/
>
> Quote:
> It is also possible to have more than one “ce'u” in a “ka”
> abstraction, which transforms it from a property abstraction into a
> relationship abstraction. Relationship abstractions “package up” a
> complex relationship for future use; such an abstraction can be
> translated back into a selbri by placing it in the x2 place of the
> selbri “bridi”, whose place structure is:
>
> “bridi”: x1 is a predicate relationship with relation
> x2 (abstraction) among arguments (sequence/set) x3
> End quote.
>
> So, {lo ka ce'u ce'u na ckaji} is not right.
Good catch.
> It should be something
> like {lo ka ce'u lo nei na ckaji} unless I'm wrong (not sure whether
> {nei} refers to the inner bridi). {vo'a} can't be used here, as it
> refers to the x1 of the outer bridi, ie. the full abstraction. {ri}
> can't be used also, as it doesn't refer to KOhA.
I don't see why {ri} shouldn't work. CLL only says "Certain sumti are
ignored by "ri"; specifically, most of the other cmavo of KOhA, [...]"
-- http://dag.github.com/cll/7/6/
{ce'u} could reasonably be one of the exceptions.
Otherwise, I guess I'd use {goi ko'a}, or use {ce'u xi pa} for both.
> This opens a subject which bugged me for a long time: how are
> reflexive relationships handled in Lojban? I've seen lujvo with
> {sevzi} to handle them, but this doesn't seem right in general, as the
> definition of {sevzi} suggests a concious being (unless I'm wrong), so
> it couldn't be used to make eg. {sezra'a} mean "x1 is self-
> referential".
How about {du} instead of {sevzi} for reflexives? So {dubra'a} for
"x1 pertains to x2=x1".
Martin
Attachment:
pgpy3ly55xcVl.pgp
Description: PGP signature