[YG Conlang Archives] > [jboske group] > messages [Date Index] [Thread Index] >
Jordan:
> On Wed, Oct 02, 2002 at 05:01:37PM +0100, And Rosta wrote:
> > jordan:
> > #This wasn't an example of a bad usage of "le" with du'u. Remember,
> > #x1 of du'u is a predication (formed by the abstraction inside the
> > #du'u). As far as I can think it, there's no difference between
> > #ledu'u and lodu'u, because you just said the predication. No one
> > #can ask "which predicaton?" (sensibly).
> >
> > That's why {le} is inappropriate (except arguably for the unusual
> > reading "it (viz the proposition blahblah)").
> >
> > It's kind of like if I say "a (certain) nose of mine is big" or
> "look at a (certain)
> > sun" -- since I have only one nose, and there is only one sun, the
> > referent is obvious, yet the locutions imply that I have more than one nose
> > and that there is more than one sun.
>
> I suppose you're not giving me an example because you're trying to
> claim it is *always* bad to say "ledu'u"?
Sort of. I wasn't giving you an example, because I took it for
granted that it had been established to everybody's satisfaction
that for a given proposition P, {pa da du'u P}.
> I don't think that using "le" implies there are more than the su'opa
> to which is being refered. Using "le" means "su'opa le ro", and
> makes no claim about how big ro is.
I forget whether CLL actually says {le} means {su'o pa le}. The
dialect I speak definitely has it meaning {ro le}.
Anyway, it is certainly true that {le broda} makes no claim about
the cardinality of {le'i broda}. But -- usage aside -- I do think
it implies "it is not the case that the cardinality of le'i
broda is 1 and that the addressee can be expected to know this".
For categories known to be singleton, the specific/nonspecific
contrast is redundant. To use gadri that encode that contrast
therefore griceanly implicates that their use is nonredundant,
especially if there are other gadri that could be used that
don't encode that contrast.
> If anything, for the du'u
> clauses "le" is more proper than a different article because the
> speaker obviously has the preposition in mind (as they're about to
> say it).
For singleton categories, le, lo, lei, loi, lo'e, le'e will all
be equally true (setting aside the veridicality factor). But
all of them other than {lo'e} encode contrasts that are pertinent
only to nonsingleton categories.
> Note that this is entirely different from the often improperly used
> "le" found with "nu". (x1 of nu is the event, where x1 of du'u is
> the predication expressed in the du'u -- i.e. du'u is more or less
> a no-op which just is there for grammar purposes).
I do note that, and the difference is that whereas {le nu} is
sometimes simply plain wrong (given the speaker's intended
meaning), {le du'u} is, pe'i, always merely infelicitous.
--And.