Re: bridi, tanru and lujvo

[Logical Language Group <lojbab@ACCESS.DIGEX.NET>]

la .eduard. cusku di'e

> >Here's how to assert rigorously that Mon Repos is a doghouse for Spot:

[example snipped]

> Aha! Excellent. Although I suppose you meant ".ije de zdani da" on line 2.

Yes.

> Now can you translate that into a form R that has the explicit place
> structure (house) (dog) without the z place so that I can say <la mon.
> repos. R la spot.>, and mean exactly what you just said? In other words,
> can lojban express not just a proposition embodying the relation, but the
> relation itself without its arguments, ready for further functional
> operations?

Partly.  At the moment, a statement can be converted into a relation-abstract-
object by wrapping it in "le ka...kei" and inserting "ce'u" before each
logical variable.  However, there is no way to convert this object into
an actual predicate.  This was discussed some time ago under the heading of
"anti-ka", but without resolution.

> Fine as far as it goes. So I can say, informally, that I have some
> undefined relation in mind, and create a tanru for it. Or I can say <I
> define "this tanru" to-stand-for "that bridi" in-the-context-of this
> discourse.> (Is that a gismu? Translation, please?) That will satisfy most
> of the needs I had in mind.

There are pro-bridi in gismu form, "brod[aeiou]", which can be assigned to mean
whatever relations we find it useful to express.  In addition, it is possible
to create context-dependent lujvo for things like "your-coat", whose
denotation is dependent on the current binding of "you".

--
John Cowan              sharing account <lojbab@access.digex.net> for now
                e'osai ko sarji la lojban.