WikiDiscuss

xorxes:
> --- And Rosta wrote:
> > I have been raising this as a gadri issue only because
> > in an ancestral version of xorlo it was treated as such
> > and because that was the only solution I was ever
> > satisfied was satisfactory. But I'm not trying to
> > argue that xorlo needs to be altered.
>
> The way I understand it, xorlo hasn't changed much in the
> aspect we are discussing (but it's possible I'm still
> missing something). The only significant change from
> the old version has been the interpretation of the inner
> quantifier with respect to the outer one: in the old
> version the inner quantifier was the numerosity of each
> value that the quantified variable takes, in the current
> version it is the total number of values the quantified
> variable takes.

The relevant change concerns what the outer quantifier
quantifies over; at the time that I tuned out, we seemed
to be in agreement that (i) there was a significant distinction
between quantifying over instances of a kind and quantifying
over subkinds of a kind, (ii) PA + gadri constructions
ought probably to be underspecified with regard to the
distinction, and (iii) there needed to be some way to
make the distinction, e.g. using LAhE. I'm saying this
just in order to jog your memory & explain where I'm coming
from; I'm not trying to resuscitate that old scheme and
undermine xorlo.

> > Suppose there are two statues of unicorns in front of me, and I
> > draw them, and say "ta pixra re pavyseljirna ku noi zasti ta":
> > that would be true, because the two drawees do exist in the
> > same world as ta, even though in ta's world they aren't
> > unicorns (-- they're statues). So we'd also want a way to
> > express whether the drawees are unicorns in ta's world.
>
> But that issue only arises if {ta pavyseljirna} has
> two different readings. If we can say of the statue
> that it is a unicorn, then the picture of the statue
> is a picture of a unicorn. If we can't say that the
> statue is a unicorn, then the picture of the statue
> is not a picture of a unicorn.
>
> CLL says:
>
> (The notion of a ``really existing, objectively defined bear'' raises
certain
> difficulties. Is a panda bear a ``real bear''? How about a teddy bear? In
> general, the answer is ``yes''. Lojban gismu are defined as broadly as
> possible, allowing tanru and lujvo to narrow down the definition. There
> probably are no necessary and sufficient conditions for defining what is
and
> what is not a bear that can be pinned down with complete precision: the
real
> world is fuzzy. In borderline cases, ``le'' may communicate better than
> ``lo''.)
>
> I would add that in different contexts the boundaries can
> be different. Once we decide whether in the given context
> a statue of a unicorn counts as a unicorn, we know whether
> the picture of the statue counts as a picture of a unicorn.

I think this is a red-herring. Even if a statue of Abraham Lincoln
is categorically not Abraham Lincoln, a picture of a statue of
Lincoln can be a picture of Lincoln, whereas a photo of statue
of Lincoln can't be a photo of Lincoln.

> > > {ta pixra lo re pavyseljirna poi ranmi danlu} "that's a picture
> > > of two unicorns which are mythological animals" and
> > > {ta pixra lo re pavyseljirna poi ca'a zasti le ma'a munje}
> > > "that's a picture of two unicorns that actually exist in
> > > our world".
> >
> > The first seems okay, but the second doesn't, for two
> > reasons. Firstly, one can have imaginary things that are defined
> > as existing in the real world: "lo ca'a zasti le ma'a munje"
> > needn't refer to something that exists in the real world;
>
> This part I'm not sure I understand. Wouldn't an imaginary
> (or fictional) thing be non-existent by definition?

Pe'i imaginary things exist by virtue of being imaginable, but
they don't exist in the same world as real things. But anyway,
my point is that Mr Dog exists but is not a dog in our
local world, and likewise Mr Real World Dog exists but is not
a dog in our local real world.

> > it might refer to "Mr Exister in the RW".
>
> Mr Exister in the RW has to exist in the RW, that's in fact all
> it needs to do.

In that case, we must understand different things by "Mr X".

> > Secondly, what
> > matters (with regard to disambiguation) is whether the
> > depictee exists in the same world as the depicter (and
> > whether the depictee has the property of being depicted
> > in the same world as it has the property of unicornhood).
>
> That would seem to require the marking of statements for
> world of application, rather than the marking of sumti.

That'd be one solution, yes.

> > I'm not saying that Lojban should handle this overtly in
> > terms of different worlds. I am talking about different worlds
> > only as a means of explicating the different readings we
> > get with intensional sumti-places. And I'm suggesting that
> > Lojbanists might want a robust way to express the readings
> > distinctly on occasion. Is that reasonable?
>
> How was the distinction handled in the ancestral version
> according to you?

{lo broda} denoted brodakind. Actual broda were instances of
{lo broda}. {PA lo broda} was neutral between quantifying over
instances of brodakind and over subkinds of brodakind, but
the distinction could be made e.g. by LAhE, pending appropriate
decisions about how LAhE worked. For certain sorts of
intensional readings of quantified sumti in intensional sumti
places, quantification would be over subkinds. For extensional
readings, quantification would be over instances.

As I have said, though, I don't intend to suggest a reversion
to that scheme.

--And.