Re: lo

[ucleaar <ucleaar@ucl.ac.uk>]

pc:
>         We also agree that _lo_ and its ilk are +veridical and
> -definite.  I argue that, both because it is a description and to fill a
> gap in the pattern, _lo_ and its ilk are +specific.  The +veridical is
> then essential, for without a known referent (-definite), the referent
> cannot be determined except through its properties.

I can only make sense of this if "-definite" means what  we have so
far been calling "-specific" and not what we have so far been
calling "-definite". I think there are examples of +definite
-specific beyond the "lo pa" cases we were talking about (Archduke
Ferdinand's assassin, etc.):
   fifty-one of the states of the USA elect senators.
      [NB this example relies crucially on my belief that there are
      51 states in the USA. If I am wrong about this, please change
      the figure to the correct number.]
This involves existential quantification over states of the USA
(or sets of states of the USA), so it is -specific. But you know what
is being referred to (Arkansas, Arizona, ... Wyoming, etc.), so it is
+definite.
Jorge would call this "+specific, +definite", but the
point of this message is to aver that LO is not necessarily -definite.
By "-definite" I understand "the speaker does not believe that the addressee
is in a position to (a) identify the referent [under one definition
of definiteness] or (b) select the referent out of the set being quantified
over [under another definition of definiteness]". Either way, +definite
means (the speaker thinks that) the addressee knows which broda out of
the set of all brodas is being talked about.

I've already acknowledged that "definite" is sometimes used (in linguistics)
to mean "+specific" (in Lojban-list sense), but unless we avoid that
confusing usage we are hopelessly lost.

-----
And