From: Logical Language Group <lojbab>
Message-Id: <199412192236.AA13923@access2.digex.net>
Subject: Re: plural
To: lojban@cuvmb.cc.columbia.edu (Lojban List)
Date: Mon, 19 Dec 1994 17:36:08 -0500 (EST)
Cc: lojbab@access.digex.net (Logical Language Group)
In-Reply-To: <199412120803.AA08716@access3.digex.net> from "Logical Language Group" at Dec 12, 94 03:03:25 am
Mime-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Content-Length:       4302
Status: RO

la xorxes. cusku di'e

> >Suppose that there are five people in front of us, and I say:
> >
> >        ro le prenu cu citka lo plise
> >
> >This I will understand to mean "Each of the (5) people eats an apple."

la lojbab. cusku di'e

> Would it?  I'm not sure where "lo" ended up, but if it ended up as
> implicitly equivalent to "da poi", then the sentence may mean that they
> all ate the same apple.  Is this your intent?  (your English is
> ambiguous).  This seems no more sensible than your alternative below.
> It would seem likely to me, especially if the equivalence of "lo" and
> "da poi" were established, that "lo"/"loi" do clearly make a
> distributive/mass distinction.  Not so likely for "le"/"lei".

I believe that Jorge is almost right.  Each person eats one or more apples.
We don't know how many apples there are: there might be one (each person
eats on it, but doesn't consume it); there might be less than five, or five,
or more than five.  (I neglect the possibility that "le prenu" might not
mean "le mu prenu" in this case: some persons present might be implicitly
excluded from the statement.)  If "lo plise" becomes "pa [lo] plise", then
the number of apples is bounded: still could be one, could be as many as five.

> >But you are saying that it could also mean that the mass of five people,
> >which I'm calling {le prenu}, eats an apple.
> >
> >It could, but it makes very little sense.

I think I now have to concede that "le prenu" might mean "lei prenu", on
the grounds that the so-called "prenu" could really be a "prenu gunma",
a person-blob.  So "le" as catch-all is saved, by postulating the right
+inmind +veridical predicate underlying the +inmind -veridical "le" selbri.

> Given that you knew that there were 5 people and 5 apples before, and
> none of the apples remain, then "le prenu cu citka le plise" suggests
> that the apples were respectively eaten by the people, but to me makes
> no implication about which of the people ate which apple(s).

Now we have "le...le".  I think this has to mean that each of the 5 persons
eats (but does not consume) each of the five apples: this statement is
doubly distributive.  Because "le" is +specific, we can't have each person
eating five different apples, but "respectively" is impossible too, because
of the force of "le" = "ro le" = "each of those I have in mind".

One can escape this by applying the kludge I mentioned above, and believing
that "le plise" here = "le plise gunma" = "lei plise", but it's forced.

> Explicitly saying "ro le prenu cu citka le plise", which merely supplies
> the already implicit outer quantifier, pragmatically emphasizes TO ME
> that the people are intended to be considered as individuals, and I
> would understand that as saying that each person separately ate an apple
> (but if told later that 2 people had shared an apple while another ate 2
> apples, I might feel misled, but I was not told a falsehood).

Won't work, because the second "le" is as distributive as the first.

> Likewise "le mu prenu cu citka le mu plise" more strongly emphasizes the
> correspondence than the same sentence without the quantifiers.  But I
> think you would insist that I expand this to 25 sentences implying that
> each of the people ate each of the apples - a rather incomprehensible
> concept.  Is this what we want?  (Nora thinks so.  She notes that as
> long as only one sumti is to be interpreted distributively, there is
> seldom problem with expansion.)

I agree with Nora.

> To unambiguously state that they each ate separate apples, I think "le
> mu prenu cu citka le pa'a plise, using the "respectively" meaning of
> "pa'aku".

That works on a discursive level, perhaps.  You can say

	le mu prenu cu citka lo pa le mu plise
	each-of-the five persons eats one-of-the one-of the five apples

to which the gloss does no justice.  Even that may not work, I'm not sure.

> I think that Lojban makes number as invisible as tense as an obligatory
> category.  This does not mean that people will not make assumptions,
> correct or incorrect, about both tense and number, based on context.
> But assuming singular as a rule would be risky.

Assuming plural as a rule, however, is safe.

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