Message-Id: <199511300250.VAA28977@locke.ccil.org>
Date:         Wed, 29 Nov 1995 21:40:17 -0500
Reply-To: Jorge Llambias <jorge@PHYAST.PITT.EDU>
Sender: Lojban list <LOJBAN@CUVMB.BITNET>
From: Jorge Llambias <jorge@PHYAST.PITT.EDU>
Subject:      Re: TECH: PROPOSED GRAMMAR CHANGE 38: lambda via new selma'o CEhU
To: John Cowan <cowan@LOCKE.CCIL.ORG>
Status: OR

la djan cusku di'e

> Granted that no number (quantitas) is involved, nonetheless \lambda(x) is
> parallel to \E x and \A x.

Parallel in what sense? Ex and Ax bind the variable, i.e. together with
an expression F(x) they give a proposition.

The way we use the lambda variable, the expression F(x) remains open.
"le ka F(x)" refers to a function of x, where x is not bound.

{da} could be used for this job if it wasn't for its automatic default
binding even in the absence of an explicit quantifier. {ce'u} would serve
to block the automatic binding, but having {ke'a}, which is also an
unbound variable in relative clauses, there is no need to get into
such complications. In "da poi F(x)" the role of {ke'a} is precisely
the same as that needed in "le ka F(x)".

>  However, Lojban Central is still restricting overloading
> "ke'a"; how would {le re do} reckon a solution in which there were two cmavo,
> one for relative clauses ("ke'a") and one for lambda abstraction?

I would prefer that solution over the pseudo-quantifier, but I hate to
see a new cmavo for something that already exists and is actually so
rare. I don't think it's overloading. In any case, what's the rush?
If we find in practice that {ke'a} is causing confusion, a new one
can be added, but I don't see that happening.

Jorge