Message-Id: <199511221902.OAA13172@locke.ccil.org>
Date:         Wed, 22 Nov 1995 13:41:50 -0500
Reply-To: Jorge Llambias <jorge@PHYAST.PITT.EDU>
Sender: Lojban list <LOJBAN@CUVMB.BITNET>
From: Jorge Llambias <jorge@PHYAST.PITT.EDU>
Subject:      Re: ke'a
To: John Cowan <cowan@LOCKE.CCIL.ORG>
Status: OR

Lojbab & Jorge & John:
> > > For the sake
> > > of saving a cmavo, the complications of resolution aren't worth the
> > > trouble.
> >
> > Exactly the same problems apply to {xe'u}. It also needs subscripts
> > for many arguments at the same level and for different levels of
> > embedding.
>
> No, the problems apply to "da", not "xe'u", since "xe'u" is just a quantifier.

Ok, I should have said: "The same problems apply to the resolution that
uses {xe'u}."

> And they don't apply, at least not with such force.

With practically the same force.

> There are already three
> da-series cmavo, which solves many problems all by itself before subscripting
> has to be brought in at all.

If you are arguing from the principist view, the problems are the same,
since you will need subscripts for the general n-argument case.

If you argue it from the pragmatic view, the problems are also the same:
there are none, because properties with several arguments are never needed,
and if they ever are, they can probably be left implicit.

The only way that the complications would apply with less force is if
two- and three-argument properties were frequent but much less frequent
than those with four or more arguments. Even in that case, only if they
are not used together with the normal use of da, de, di.

I still don't see a case for a new cmavo. I would have to see examples
of normal sounding things that one would want to use it for. Esoteric
examples the current language can already handle with due pain.

Jorge