[YG Conlang Archives] > [jboske group] > messages [Date Index] [Thread Index] >
Nick:
> From: "And Rosta" <a.rosta@hidden.email>
> >> unquantified gadri: ambiguous between intension and extension ---
> >> between appearing in the prenex or not
> >> quantified: extensional
> >> LAhE3 LE gadri: intensional
> >
> > I still don't see how the "LAhE3 LE" works. If LE is quantified, then
> > it won't work. If it is unquantified, what does it mean (and if it
> > means Kind, then LAhE3 is redundant).
>
> If LE is quantified, LAhE3 will work if I say it work, because it will
> have an idiosyncratic interpretation rule. *Like all LAhE*.
These would fail the criterion of consistency (of which compositionality
is a subcase).
> To my mind, LAhE3 re mikce == LAhE lo mikce remei, and is an adequate
> rendering of "I'm looking for two doctors". It doesn't work? Why?
Because it works only if the sumti "re mikce" means something different
from what it means in other environment. That is inconsistent & to be
deprecated: it is the sort of malady the BF is supposed to remedy, not
cause.
> If I refuse to grant quantified lo == Kind (and I do), and we rather
> have unquantified lo be ambiguous,
Nobody wants quantifed lo == Kind: that would be nonsensical. As for
unquantified lo, it does not exist in CLL-Lojban, and there are no
other candidate meanings for unquantified lo, so there is no good
reason not to make it mean Kind.
> then obviously LAhE3 is not redundant: it is disambiguation.
If bare lo is ambiguous between quantified and unquantified, then
it should be sufficient, as a disambiguation, to mark explicitly
whether or not it is quantified.
> In any case, {lu'a lo broda} is legal and intelligible, even if
> redundant.
It's not redundant: "at least one member of at least one broda".
> So what? Remember, elegance has never been my criterion.
>
> >> The ambiguity of {lo mikce} sucks, but I think it politically
> >> necessary now. Those who would (a) exoticise Lojban (hello xod) or (b)
> >> are semantic formalists who wish to make up for lost time and
> >> emphasise
> >> intensionality wheresoever it roams should use LAhE3 and LAhE4 (= lo
> >> se
> >> ka?) whensoever possible, and let it be known that by default their
> >> unquantified {lo} is extensional.
> > ???? Is there a typo here? Extensionality arises from quantification.
> > So unquantified lo can't be extensional.
>
> CLL-lo is extensional, because of its assumed {su'o}. I refuse to be
> barred from glorking {su'o} before {lo}.
If this is all you mean by being extensional, then it is too trivial to
quibble with. CLL-intent is that the outer PA be glorked (and it never
occurs to CLL that null quantification might be possible: *that's* why
it is extensional). CLL-interpretation has been that su'o is elidable
before lo: but that doesn't make lo itself extensional in any way.
As for whether it would be legitimate to glork su'o before bare lo, I
discuss this on my XS wiki page. It's not my preference that lack of
quantification must be marked overtly, but it is not incompatible with
the rest of XS.
> I will allow you to also glork {tu'o}. That is as much as I will
> concede.
That's not a concession, because CLL already allows it.
> Because we're obviously getting tangled in terminology, by unquantified
> I only mean "absent an overt quantifier"; I am admitting that {lo} without
> an overt quantifier may not even have an implicit one, but I will not
> admit that it must not.
Okay. Better to use my terminology "bare lo" ('bare' borrowed from
linguistics). So anyway, what you describe is in fact the status quo
wrt CLL qua CLL-intent, and AFAIAC it is an acceptable compromise.
> > A compromise, which is consistent with CLL-intent, would be to make the
> > outer PA glorked. Glorking it as tu'o would give intensional. Glorking
> > it
> > as something else would give extensional.
>
> That's what I'm proposing.
Okay. And Bob wanted this too.
> >> but I'm convinced default tu'o actually breaks usage. If I
> >> say {lo mikce cu cpedu lo sodva}, I sure as hell ain't talking either
> >> Mr Doctor or Intension of Doctor.
> > I'll concede that, though I think it is clear that usage is also not
> > consistent with CLL.
>
> Usage doesn't know the time of day, though. Seriously.
Of course. So we can't take "this breaks usage" very seriously, given that
usage is intrinsically broken.
> >> Note that Kind is *not* necessary the solution to intensions. Kind is
> >> perfectly usable in extensional contexts --- the fish and chips
> >> example
> >> --- and there can equally be intensional, non-existent but distinct
> >> entities. We may seriously see And's Kind split into LAhE3 and LAhE4,
> >> Mr and Intension: Uniqueness is not necessarily the same as Intension.
> >> That's the way John was heading in February. So the work is nowhere
> >> near done.
> > Can you give examples? I can't remember what it is you have in mind
> > here.
>
> I'm looking for a particular Unicorn. Not LAhE3 at all: this Unicorn is
> not the Any Unicorn. But still manifestly intensional, or at the very
> least not in the external prenex.
That's "le LAhE-subkind (tu'o)lo unicorn".
> I eat the same fish and chips as you: certainly Mr Fish and Chips is
> involved, but I'm having a hard time crediting this is really intension
> as we know it.
I don't understand the example. The same kind of fish and chips? We
both ate the same meal? That'd be "su'o LAhE-subkind (tu'o)lo meal cu
se citka ro mi'o".
> I think Intension and Mr are orthogonal, and John, in
> sabotaging your intent to have "All Kinds exist, whether their
> referents do or not", is who has pushed me to think so. John is being
> Extensionalist, and that, I think, is what I'm defending.
I'm not opposed to making distinctions. Indeed, in the Great Debate I
myself, I seem to recall, was arguing for a distinction between a Kind
and its avatarage. But I haven't yet grasped the distinctions you
want to make.
> > I guess you mean "skicu fo" (describe x2 to x3 as x4). I myself am not
> > sure that anything but a ka abstraction would make sense there. Or
> > rather, I can see how only Kind or only ka could make sense, but not
> > both, and not anything else. So I'm perplexed at this being the lydian
> > stone.
>
> Then ask Jorge. Who used that very sentence of my own usage against me.
That usage would be legit (IMO) with bare lo construed as Kind & with
x4 of skicu being a Kind.
> It's because I too am starting to see a ka abstraction there that I
> added the proviso at the end. If this is not the Lydian stone, then
> someone come up with one (with an explicitly Unique lo that strikes all
> and sundry as commonsensical); because without that, I withdraw my
> concession. I don't think it will be hard to do so; but I need concrete
> examples.
I don't know what exactly you're asking for. Sumti that make sense *only*
with a Kind interpretation?
> > As I see XS (which I've now described on the wiki), it is largely
> > comprised of additions to CLL -- attributions of meaning to forms that
> > formerly had no meaning. The exception to this is the actual change to
> > the interpretation of PA in loPA broda. The new interpretation is
> > undeniably more useful (& the very unusefulness of the former meaning
> > accounts for why past usage has been so meagre) but also undeniably
> > at odds with CLL: it is the only part of the proposal that is in
> > conflict with CLL. Had this element not been crucial to the expressive
> > capability of XS, I would have removed it, so that it could be said
> > that there is no conflict at all between XS and CLL.
>
> We already have what you want mentioned in the lessons, as my
> (Johanninely blessed) extrapolation from CLL: ci lo re lo plise.
That's a bogus extrapolation -- it's not even an extrapolation. (Or am I
missing something?) So no, we don't already have what we want: if it's
mentioned in the lessons then the lessons have introduced innovations
considerably more radical than anything in XS. "ci lo re lo plise"
gives nonsense; "re lo ci lo plise" is "two of three apples", not "two
trios of apples".
However, if "ci lo re lo plise" is something people can actually agree
on as a change to CLL, then maybe I could rethink XS so as to build on
those innovations that conservatives have agreed to.
> If your only criterion is expressivity, I'd fain have you remove this
> violation of CLL. The parallelism with lePA does not compel me: I've
> learned X lo Y as another cornerstone of Lojban!
>
> I see we're having pc join in now too, though through wiki rather than
> here. So we'll have even more chaos and dissension. Peachy.
I answered pc on the wiki, but I am not planning to recap or rehash
jboske debates on the wiki, or on phpbb for that matter. I put XS on
the wiki for reference purposes for BF people who are debating gadri
on jboske.
--And.