[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [lojban] {le} in xorlo

Jorge Llambías, On 14/04/2010 15:31:

On Tue, Apr 13, 2010 at 11:07 PM, And Rosta <and.rosta@gmail.com> wrote:

I'm fine with "so'i li mu" being "many fives", coerced by the
quantification.
In English, one does get things like "my mother that bore me", "London that
I have luved in for so many years", which don't coerce a "many
mothers/Londons" interpretation. So restrictiveness needn't coerce plural
interp.


I Googled "my mother that bore me", and the first (of only six) hits was:

"My mother and I–and by "my mother" I mean always one of my two
mothers, for my mother that bore me was dead–"

and that seems like the most natural use for "my mother that bore me"
to me, but some of the other hits are not so clear that the
restriction is doing actual restrictive work.

The London example sounds odd to my non-native sense without a "the"
in front, and I would want to change "that" to "where" (unless it's a
vocative?), but I don't really want to dispute that there may be
restrictive clauses that don't do any actual restriction.


It's rare and stylistically marked, as you'd expect it to be.


Also, I can't think how the poi/noi distinction can apply when there is no
predicate (selbri) present either explicitly, or implicity as when a
quantifier is applied to a sumti.


With quantification the distinction is very stark:

   ro da poi xunre cu kukte
   Everything that is red is delicious.

   ro da noi xunre cu kukte
   Everything, which is red, is delicious.


Of course. Also the distinction should pertain to "lo broda poi/noi" too. "lo broda poi brode cu brodi" = "lo ge broda gi brode cu brodi"; "lo broda noi brode cu brodi" = (roughly) "lo broda cu ge brode gi brodi".



"PA lo gunma" is ordinary quantification over groups, while "PA loi"
is quantification over the members of the group.

Yes, I see. I can't decide if the problem goes away if "loi broda" is not
"lo gunma" but rather "zo'e noi ke'a gunma",


I'd say that "lo gunma" is just the same as "zo'e noi ke'a gunma".


since "PA zo'e noi ke'a gunma"
quantifies over members of zo'e.


which must be things that satisfy the x1 of gunma.


I guess it comes down to whether "zo'e noi
ke'a broda" necessarily means zo'e is a single broda (as required for "zo'e
noi ke'a gunma" to work as a solution)


No, that's not required. Many things can work as a solution working
together without that making them one thing. (That's essentially
McKay's argument.) But the kind of thing that satisfy the x1 of gunma
is the kind of thing that has members, as opposed to the kind of thing
that satisfy the x2 of gunma, which are typically many things working
together.


or whether it can mean zo'e is a
bunch of broda (as required, I'm fairly sure, by xorlo).


zo'e can be many broda without necessarily being one anything (whether
bunch or whatever).


This boggles my mind. I've looked at Ch 1 of McKay -- pdfs of it and the contents pages are on his webpages. I suspect I must be an exponent of the singularism that his Ch 2 attacks.


Actually, those are
both meanings one needs to be able to express. Maybe "zo'e noi pa gunma ne
ke'a"? (The thinking in this para is low quality, so feel free to ignore
it...)


The two meanings are easily expressible without any recourse to "loi":
"lo broda" vs. "lo gunma be lo broda".


I think the contrast I had in mind was "lo pa broda" versus "lo (su'o) broda". "loi" is "lo pa gunma be lo", but "zo'e noi ke'a gunma be lo" is probably "lo su'o gunma be lo".


The actual English example "came by bus" seems to me to demand a generic
reading (because that seems to be the effect of using _bus_ without an
article), but if we can use, say, "I will drink wine" as an example, then
the nongeneric reading can be specific or nonspecific, "Ex, x is wine: I
will drink x" being the nonspecific.


Right, but I think we don't need to commit to one of the two readings,
or perspectives, to get the meaning. In the case of the bus we may be
forced to by English:

  They came by bus, so they arrived earlier than those of us that came
by bike, even though (?it / their bus) had to stop for gas.

It seems to me that "it" can't be used there, because grammatically
"bus" doesn't have a nongeneric perspective available, but:

  They drank wine, and I only had water, so I will drive. Not that I
wouldn't have wanted to drink (it / the wine) too, but they had
finished it before I arrived.

In this case, it seems to me, "it" seems more acceptable (what are the
native intuitions?).


Your intuitions are (unsurprisingly) accurate. ("It" in the bus example is not impossible, but it would refer to the implied bus travelled on, not to the mode of transport referred to by "(by) bus". The problem with the example is not that bus is generic but that it wasn't the generic bus, qua mode of transport, that had to stop for gas.) In your wine example, the version with "it" is consistent with all three interpretations (generic, specific, existential).

I can see how a linguistic construction can be semantically ambiguous between these interpretations, but not how the interpretations can be conflated into one.


But in Lojban both perspectives remain always available, giving
something that would seem weird in English like:

  They came by bus, so they arrived earlier than those of us that came
by bike, even though bus had to stop for gas.

So "bus" can be generic and still stop for gas in a particular occasion.


English too: "She offered me tea or coffee; I chose coffee, and then spilt it"


What I'm trying to say (I think) is that the level of abstraction,
which is to some extent arbitrary, can set the stage in such a way
that the issue of specificity will be affected. But I know I'm not
saying anything very convincing about it at this point.

OK, I understand your point, I think. Translated into my terms, it is that
specific readings are nongeneric; so if something is viewed generically,
then perforce it's not specific.


Mmm... I think I'm saying the opposite, that generic readings are
always specific, but because of the arbitrary perspective of
genericity, they can often be taken as nongeneric nonspecific as well.
But I have to keep insisting that I'm not certain if that's exactly
what I'm saying. :)


I now understand what you mean. Some particular broda could be the generic broda, you're saying. Our difference was mainly terminological.

But if you think "lo broda" means "some particular broda, which may be the generic broda", then I see why you think you can do without e-gadri. (How to explicitly do generics, though?) Actually, I guess the sole difference between o-gadri and e-gadri might then be veridicality, which, mirabile dictu, might actually suddenly make pre-xorlo gadri usage mean approximately what the writers thought they were saying.


"zo'e noi ke'a broda" effectively gives you the generic reading, since
there's no quantification.


Yes, though I would want to say "it allows" the generic reading,
rather than "gives".


Yes.


"PA zo'e noi ke'a broda" quantifies over
tokens/members of the category.


Thereby forcing a nongeneric (or less generic) reading (and
quantification is always nonspecific).


"zo'e'e", with optional noi or no'oi, gives
specific, and "PA zo'e'e" quantifies over tokens/members of "zo'e'e". That
seems pretty straightforward... (Admittedly, not straightforward if the goal
is to paraphrase gadri using fairly common nonexperimental cmavo.)


Yes, I think if we have "zo'e'e" then we have "le" figured out, and
vice versa. But putting one in terms of the other doesn't really get
to the bottom of it (at least for me).
Maybe zo'e already is zo'e'e? 
On the one hand one can see a distinction between a reading of "I ate X" where X is some specific thing amd a reading where X is just whatever it was I ate -- the generic thing-eaten-by-me, but on the other hand one can see how the second reading can be treated as merely an instance of the first.

--And.


--
You received this message because you are subscribed to the Google Groups "lojban" group.
To post to this group, send email to lojban@googlegroups.com.
To unsubscribe from this group, send email to lojban+unsubscribe@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/lojban?hl=en.