Message-Id: <m0sS9AN-0000YjC@xiron.pc.helsinki.fi>
Date:         Sat, 1 Jul 1995 16:13:17 EDT
Reply-To:     jorge@PHYAST.PITT.EDU
Sender:       Lojban list <LOJBAN@CUVMB.BITNET>
From:         jorge@PHYAST.PITT.EDU
Subject:      Re: masses
To:           Veijo Vilva <veion@XIRON.PC.HELSINKI.FI>
Content-Length: 2731
Lines: 51

la lojbab cusku di'e

> Shall I go out a limb, probably incorrectly as to terminology, and
> suggest that "loi remna" in a given claim is +definite while being
> -specific; i.e. that in evaluating a given predication as true, you need
> to instantiate the mass in such a way that the relationship holds, and
> that this instantiation is constant/has scope over the whole of a
> complex predication.

This is a possible convention. It is completely against the way
scopes work for other quantifiers, but it is possible. I just don't
see the advantage of complicating things like that.

> >You don't accept then that {la djan e la maris cu pinxe loi djacu} means
> >something different from {loi djacu cu se pinxe la djan e la maris}?
> >
> >With the normal rule for scopes, the first one says that there is some
> >mass for each of John and Mary, not necessarily the same mass, such that
> >they drink it.  The second one says that there is a mass such that they
> >both drink it, the same one.
>
> Intellectually, I would prefer them to mean the same, probably the
> former because of the desire for "se broda" to be symmetrical and to be
> a transform of "broda" rather than a different selbri

Your rule above would mean they both mean the latter, not the former!
In any case, {se broda} is always the same selbri as {broda}, all that
changes is the order of the arguments. The way the scopes interact is
important to determine the meaning of the bridi. The order in which
the quantifiers appear is the basic rule to determine quantifier scope.
Therefore, the same selbri can give bridi with different meanings,
depending on the order in which the arguments appear.

> My inclination would be to make the rule that, in the absence of
> explicit prenex scope or scope implicit through the use of bound
> variables like "da", that scope be determined by sumti number (x1/x2/x3)
> order. Jumping ahead in these threads, I would accept that there might
> be desirable an afterthought capability to override this order to make a
> given sumti have broadest scope or least scope (or perhaps some
> intermediate value).  At this time of night this idea is mushing
> together with pc's "context leaper" xVV proposal(s) that may or may not
> have anything to do with it - the commonalty seems to be the desire to
> specify a scope other than that implicit in the stated syntax.

I think scope determined by sumti number is the worse possible choice.
Working out what a sentence means on the fly if the arguments are out
of the normal order would become a nightmare. Not to mention the big
difference that might arise between saying {se broda} and {selbroda}.
What would be the advantage of having scope determined by sumti number?

Jorge