[YG Conlang Archives] > [jboske group] > messages [Date Index] [Thread Index] >
Nick:
[quoting whole message as a reminder]
> Our problem is: we want to use Any-x in contexts where there is no
> embedded clause in Lojban. So in "I want to see a doctor, any doctor",
> you have {lenu}, hence: {mi djica lenu su'o da poi mikce zo'u: mi viska
> da}. But when you look for something, there's no {lenu} there. Thence,
> we made the x2 of sisku be leka
>
> So sisku is the kind of predicate that causes us problem: it's
> intensional, but its x2 is supposed to be an entity, not a predication.
> So where's the prenex go?
>
> http://www.tulane.edu/~forbes/pdf_files/ITV_Lims.pdf is a paper on
> such predicates in English, that I found when I googled for listings of
> intensional predicates
>
> The author wants to put prenexes anywhere the hell he pleases,
> including mid-sentence. He says even old man Montague ended up doing
> so. The Lojban grammar is fixed, though, so we can't do that
>
> So he's considering the alternative --- that underlyingly such preds do
> have clausal complements after all. Just as we have done for sisku. He
> terms this propositionalism, and dates it from Quine. He thinks
> propositionalism is crap
>
> He has a list of intensional transitives that maybe surprising; they
> surprised me:
>
> prefer, require, has an urge for, longs for
> look for, seek, hunt, rummage about for
> sketch, caricature, picture, write about, imagine (because you can
> describe imaginary things)
> hate, admire, idolize, worship (this one, I admit, I don't get)
>
> He accepts that "want X" is propositionalistic, in that it can be
> paraphrased as "want to have X", and that's why "Walter wants a bigger
> boat tomorrow" is ambiguous: {ca le bavlamdei} can attach to the inner
> clause or the outer:
>
> la .ualter djica lenu su'o da poi bramau bloti zo'u: .ubu ponse da ca
> le bavlamdei
>
> la .ualter djica ledu'u su'o da poi bramau bloti zo'u: .ubu ponse da
> kei ca le bavlamdei
>
> He contends that seeking, depiction and evaluation are not discernably
> propositional in English
>
> His evidence is syntactic. I don't give a turd about syntax, as you
> know. In fact, his evidence seems singularly underwhelming to me
> applied to Lojban. And even inside English. e.g. he thinks it's odd to
> say "Walter is seeking a camera by noon", meaning that when he wants it
> by as opposed to that's when he's starting the search by (so you can't
> attach a separate time to the outer bridi and the inner bridi, "seek
> *to find a camera by noon*". Eh, bullshit; that makes perfect sense to
> me
>
> The imagining verbs are harder to me to work out, and I agree that
> even if "I imagine X" is "I imagine X to exist, "I write about X" isn't
> necessarily "I write about X existing". He then dismisses the prospect
> of imagining {lenu xy co'e}, though I think he's over-hasty
>
> With evaluation, propositionalists usually claim they're extensional,
> not intensional, rather than that they're propositional. The author
> claims they're intensional, and not propositional. He claims 'fear'
> must be intensional, since fearing Clark Kent and fearing Superman are
> distinct states; he dismisses as unmotivated the counterargument that
> there is a propositional claim here as well, "Lex Luthor believes that
> he fears Superman"; but I don't see why it's unmotivated, since we're
> relying on what Lex knows about the world (and has lack of awareness of
> Supe's identity) anyway. The example that, if Allah = Jehovah, and a
> Muslim never goes into a synagogue, then there's no extension and no
> proposition, is even more bogus: If Allah = Jehovah, then the Muslim
> *is* worshipping Jehovah; she just doesn't realise it --- and realising
> it is itself a proposition that's part of the story
>
> And in particular, and others as fancy strikes, what do you make of it?
> And can Lojban cope with all four classes of preds talking ce'u?
Nick, I commend you for finding the paper. It is very helpful in
laying out the background to our debate.
Your summary is good and accurate. His syntactic evidence is based
on the premise that syntax and logical form are nondistinct. The
paradigm he is working on goes further and assumes without much
evidence that this logical form is universal. I don't accept that
it is universal, but I am sympathetic to the view that syntax =
logicosemantic form on a language-particular basis. However,
just because something works in such and such a way in the logico-
syntactic form of English does not mean that it should (or should
not) work that way in Lojban. Instead, we need to find an adequate
language-independent logical form for the meaning we are trying
to capture, and then implement it in Lojban.
So, returning to the paper, it doesn't really advance us beyond
the point we had already reached some months ago. At that time
we had considered the wants/needs class, the seeks class and
the depicts class, but not the fears class.
What we had come up with was:
* the propositionalist solution for wants/needs and seeks
* the propositionalist solution seems inadequate for depicts
* what we are now calling Unique
In recent discussion I had said that Unique neutralizes the
transparent/opaque distinction, so doesn't in itself give us
the opaque reading. But I was wrong. The transparent reading
of "I seek a unicorn" says that there is a unicorn (and would
use {lo}). The opaque reading doesn't say that there is
a unicorn, and nor does Unique. So Unique does the opaque
reading.
I would now query the idea that the propositionalist solution
fails for depicts and fears: basically the question hinges
on whether it is illict to construe {X gerku} as {X nu da
gerku}. If it is not illicit, then propositionalism works,
I think.
But whether or not propositionalism works, Unique seems to do
the job (as xorxes as always said). *However*, I am having
big second thoughts about Unique and the Excellent Solution,
which I will take up in another message. Suffice it to say
for the present purposes, that an intensional gadri should
work for the cases discussed in the paper.
--And.