* Thursday, 2011-10-06 at 17:21 -0400 - John E. Clifford <kali9putra@yahoo.com>:
> Which is why I added "or would have I intended if he had thought of
> it". So you are saying he might have intended a variable.
Ah, I was reading your "something" in "something I have in mind (or
would have, if I thought about it)" as being an ordinary something (or
somethings, presumably), i.e. just some element (possibly plural) of the
universe.
Having it be a variable is roughly right, yes. This can't be dealt with
at a textual level - simply substituting {da xi ci ze} for {zo'e} -
because {zo'e ro da broda} should be the same as {ro da zo'u zo'e da
broda}. Hence the "close-scoping (plural) existential with glorked
domain" suggestion made in this thread.
Do you still think that suggestion is wrong/bad in some way?
> I personally think it is {zi'o}, but that probably has problems
> too--though I can't think of one.
My main problem with that is just the icky ambiguity it would introduce.
In principle, {zi'o klama} is an entirely new 4-place predicate, whose
semantics are related to those of {klama} but not in any very
predictable way. So if an omitted place can be {zi'o}, understanding the
possible meanings of any expression would, in principle, involve
understanding many such zi'o-derived selbri.
Martin
> On Oct 6, 2011, at 15:35, Martin Bays <mbays@sdf.org> wrote:
>
> > * Thursday, 2011-10-06 at 11:10 -0700 - John E Clifford <kali9putra@yahoo.com>:
> >> Well, I am not sure just what the all the complications that MB and
> >> xorxes have stirred up are, but, so far as I can follow it, the
> >> position seems to be that {zo'e} stands for something I have in mind
> >> (or would have, if I thought about it). It is a constant (a different
> >> one at each occurrence -- an ongoing problem in Lojban), not
> >> a variable and not under any quantifier. Then {lo broda} is that with
> >> the additional information that the something is in some way, broda.
> >> So, it can refer to anything from a single broda to all brodas, past,
> >> present, future, and possible but not actualized. This referent can
> >> then be said to have a further property in a variety of ways, mainly
> >> to be grokked from context, since the ways to specify them are not ywr
> >> well-established. My understanding is that MB disagrees with this
> >> specification of {zo'e} and xorxes with the extrapolation to the
> >> referent, but juast why is hard to see.
> >
> > Because it doesn't seem to explain the behaviour of {zo'e} with respect
> > to negation and quantifiers - at least if we accept that an unfilled
> > place is implicitly filled with a {zo'e}, and if we don't use kinds.
> >
> > To dig out the old example, in
> > A: xu do pu klama lo zarci
> > B: mi na klama
> > , and assuming that there's only one market in question, B probably
> > intends to refer to that market by the implicit {zo'e} in {klama}'s x2.
> > But B is unlikely to mean only to mean that for some specific route,
> > B didn't go to the market by that route. B probably means that B didn't
> > go to the market via *any* route, or means of transport.
> >
> > Your explanation of {zo'e} seems not to deal with that.
Attachment:
pgpu286gm8bwM.pgp
Description: PGP signature