[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: indefinites - Lojbab's phonecon with Cowan (finally!)
- To: Veijo Vilva <veion@XIRON.PC.HELSINKI.FI>
- Subject: Re: indefinites - Lojbab's phonecon with Cowan (finally!)
- From: John Cowan <cowan@LOCKE.CCIL.ORG>
- Date: Mon, 10 Jul 1995 16:30:26 -0400
- In-reply-to: <199507081812.OAA00685@locke.ccil.org> from "jorge@PHYAST.PITT.EDU" at Jul 8, 95 02:06:05 pm
- Reply-to: John Cowan <cowan@LOCKE.CCIL.ORG>
- Sender: Lojban list <LOJBAN@CUVMB.BITNET>
la lojbab. joi la xorxes. cusku be di'e casnu
L> If we say
1) le ci remna cu pencu le ci gerku
L> we are talking about
L> exactly 3 men, and exactly 3 dogs, but 9 events of touching.
X> Agreed. Absolutely.
Quite.
L> I contend that this should also be equivalent to:
2) ci lo remna cu pencu ci lo gerku
L> which should also refer to exactly 3 men, 3 dogs, and 9 acts of touching,
L> although we have lost the definiteness.
X> This is, of course, one possibility. This is the view I was espousing
X> before And convinced me that the other possibility was more natural.
My "epiphany" was the realization that it is this kind of sentence which
says what the difference between "lo" and "da poi" was intended to be, and
makes sense of pc's strictures on descriptors vs. quantified variables.
I now hold that Example 2 differs from
3) ci da poi remna cu pencu ci de poi gerku
precisely because Example 3 means
(E!3x:remna(x)) (E!3y:gerku(y)) pencu(x,y)
whereas Example 2 does not. In Example 3, we have 3 men, 3 to 9 dogs, and
9 acts of touching, whereas in Example 2, only 3 dogs are possible.
X> Or are you saying that we will have different rules for outermost {ci}
X> when it is in front of {da} and in front of {lo}?
Just so. In essence, the rule I'm proposing that da-series sumti have nested
scope, but descriptors co-equal scope. This is a departure from what I have
have said before (lo != da poi), but I think it captures the 'feel' of "lo"
as an article better.
An open question is what happens in sentences that contain both descriptors
and variables. I think that all the descriptors effectively move to the
left end of the prenex: they have scope over the full sentence.
L> I basically agree, as does Nora, though our agreement is more specific to
L> the "lo" form of the above:
4) pa lo jubme cu se tuple vo lo tuple
5) re lo jubme cu se tuple bi lo tuple
X> Which contradicts what you just said above for the men and dogs case.
X> If we had nine events of "man touches dog" before, how come we don't
X> have 16 events of "table has leg" here?
I think Example 5 is bad, and Jorge is correct. Just because I had an
epiphany while talking to Lojbab, doesn't mean that he understands it. :-)
L> Nora and I have gone similar but varying directions in regards to what
L> to make of indefinites "re jubme" and "bi tuple". I think we both now
L> feel that equating them to the same thing with "lo" inserted may be too
L> simplistic.
X> Before deciding how {re jubme} behaves I think it would be better to
X> settle on how {re lo jubme} does. It doesn't make sense to explain
X> the first in terms of the second if we are not clear on the first.
X> I think the two possibilities are:
6) re lo jubme cu se tuple vo lo tuple
6.1) Two tables have four legs each.
6.2) Each of two tables are in relationship "se tuple" with
each of four legs.
X> Above you said the meaning was (2) for the man-dog case, then you gave
X> an example where it is something else altogether (a mass type of thing)
X> in the table-leg case. I think that (1) is the best choice.
I would rather express Example 6.1 with:
6.1.1) re da poi jubme cu se tuple vo da [poi tuple]
6.1.2) re lo jubme cu se tuple vo da
where the last two words are redundant, since clearly anything in the x2
place of "se tuple" needn't be qualified with "poi tuple".
Examples 6.1.1 and 6.1.2 do the same work if we understand "re lo jubme"
to be pushed to the left, since it is already all the way to the left.
X> If you want {re lo jubme cu se tuple bi lo tuple} to be the sensible
X> thing to say, then this would mean that {re lo jubme} is a mass of
X> two tables, and {bi lo tuple} a mass of eight legs. In that case,
X> {re lo jubme cu se tuple bi lo tuple} would mean the same as
X> {lu'o re lo jubme cu se tuple lu'o bi lo tuple}.
This won't do, because it fails to exclude the case of a three-legged
table and a five-legged table.
This argument makes me wonder whether "vo tuple" should be ambiguous
between "vo lo tuple" and "vo da poi tuple".
--
John Cowan cowan@ccil.org
e'osai ko sarji la lojban.