* Thursday, 2011-09-08 at 19:48 -0300 - Jorge Llambías <jjllambias@gmail.com>:
> On Thu, Sep 8, 2011 at 12:42 AM, Martin Bays <mbays@sdf.org> wrote:
> >
> > The ambiguity would be widerspread than in English, of course. You can't
> > say "someone loves everyone" and mean that the generic lover does, but
> > you could with {da prami ro de}. And if this generic lover is invoked by
> > innocuous phrases like {ro se mamta cu se prami}, it wouldn't be
> > a particularly exotic reading.
>
> I think it would be somewhat exotic but only because you would be
> mixing two levels of abstraction for the same things in the same
> sentence.
But only to the same extent you do in your analysis of {ro se mamta cu
se prami}, if I'm not misunderstanding that?
> But "su'o da prami ro de" does have a reading that "someone loves
> everyone" lacks, because Lojban doesn't distinguish someone/everyone
> from something/everything. "su'o da prami ro de" could be instantiated
> by "dogs love everything" (as opposed to, say, cats, that only love
> themselves). Then, since dogs love everything, then something does
> love everything, i.e. "su'o da prami ro de".
>
> In general, Lojban can be more vague than English. But that's not a
> bad thing, as long as we have the means to be more precise when we
> want or need to.
Agreed, as a general principle - though a binary ambiguity between
different logical forms is taking it too far, imho.
But how would you disambiguate to precisely say "someone loves
everyone"? Or, for that matter, "some dogs love every human"?
(You've just indicated that {su'o gerku cu prami ro remna} won't do,
since it could be intended to be witnessed by the generic 'dogs' (or the
generics "chihuauas" and "German shephards", for that matter))
> > You may be right that it's impractical to grammatically distinguish
> > between generics and mundanes (and perhaps also that there isn't
> > a coherent distinction between the two), though I'm not really convinced
> > yet. But even if so, introducing with zo'e these generics which
> > demonstratedly *aren't* necessary, because they don't really exist in
> > natural languages, seems unhealthy.
>
> But they do exist in natural languages! They are all over the place.
Not as widely over the place as your zo'e-within-universal analysis
would require, surely?
> mi nelci lo nu mi te vecnu zo'e goi ko'a .i ku'i ba bo mi no roi djuno
> lo du'u mi punji ra makau
> "I love buying stuff, but then I never know where to put it."
>
> What is that if not a generic?
How would you analyse this using generics? "Things I like to buy"?
"Things I buy when I buy things"? I don't see.
I'd have thought this was rather an example of this weird thing English
can do sometimes, whereby an anaphoric pronoun appears to cross
a quantifier boundary, resulting in complicated semantics of which
I wouldn't like to posit a general theory...
i.e. I think the English is roughly equivalent to, though more natural
than, "I love it when there are things which I buy, but after each such
buying I don't know where to put those things".
We could try to copy this semantics into lojban if we could figure out
general rules for it... but for now I'd rather just leave such
boundary-crossing uses of prosumti undefined.
(Sorry for going on about a problem you didn't even mean to raise...
getting prosumti working is something else I want to sort out)
Martin
Attachment:
pgp0bIL2y3mJz.pgp
Description: PGP signature