Date: Wed, 14 Dec 1994 01:15:29 -0500
From: Logical Language Group <lojbab>
Message-Id: <199412140615.AA14079@access1.digex.net>
To: jorge@PHYAST.PITT.EDU
Subject: Re: plural
Cc: lojbab@access.digex.net, lojban@cuvmb.cc.columbia.edu
Status: RO


>A more valid comparison would be with na/ja'a. If none is given explicitly,
>ja'a is the default one.

na go'i .i go'i .i go'i

No I did not contradict myself.  "na" carried over as implict to the go'i.
My son and I get into nago'i/ja'ago'i arguments all the time - one little
bit of Lojban he knows well.

On the other hand, i will agree with you that for the rest of the language,
we are not metaphysical neutral withh respect to negation.  We ARE by ultimate
assumption, neutral with respect to number and tense, and especially if you
insist that the quantifiers of "le" give an implict number specification
parallelling singular/plural, I am obligated to deny that the default is
absolute.

>> I'm not sure where the default quantifiers are on "lei" at the moment - Cowa
>> disagreed with me on what we have said before, I think.
>--More--
>
>I hope it ends up as {piroi}. The consequences of {pisu'o} are just too
>horrifying to even consider.  :)
>
>With {piroi}, {lei broda} is a singular term (singular meaning that it is
>neither universally nor existentially quantified, or rather it is both).
>Singular terms are very good because they commute with everything, you
>don't have to worry about the order of negation and everything else.
>{le broda} is often a singular term too, when it means {le pa broda}.
>
>With another quantifier, {lei broda} is no longer a singular term, and
>you have to be very careful with the order in which it appears with
>respect to non-singular terms.
>
>The same with {le broda} when it is {le su'ore broda}. The sentence has
>different meanings if you change its position with a non-singular term
>like {lo broda}.
>
>That's why I think {le [pa] broda} and {lei broda} are going to be used
>much more than the others. They are much less trouble.

I see your claim, and that to me is a strong reason not to equate "lo" with
a prenex form - it should be locally scoped, I think.  To me, as far as is
possible, I want the language to be absolutely symmetrical under SE conversion
unless there are explicit bound variables (or NAKUs) present.

But even ignoring the impliocations for "lo", I do not intuitively see the
problem with fractional quantifiers that I do with "Everybody loves somebody".
Since "loi" is certainly pisu'o, you seem to be arguing that 
roda prami [pisu'o] loi prenu and [pisu'o] loi prenu se prami roda are not
the same in meaning.  I'm not sure I see this, but again if so, i will protest
that it should not be and that these must have a fixed identical meaning.

I am not happy with your assertion that quantification defaults make all but
1 or 2 of the LE gadri unusable, and will insist strongly that the 
quantificational interpretation render these meanings fixed.  Cowan said that
the only problem with lo equating to a dapoi parallel was under negation.
YOu are claiming a far greater problem.

lojbab