Message-Id: <m0qG9yG-00001cC@xiron.pc.helsinki.fi>
Date:         Tue, 21 Jun 1994 18:50:16 BST
Reply-To:     i.alexander.bra0125@oasis.icl.co.uk
Sender:       Lojban list <LOJBAN%CUVMB.bitnet@FINHUTC.hut.fi>
From:         i.alexander.bra0125@OASIS.ICL.CO.UK
Subject:      Re: (kau) and (du'u) and (jei)
To:           Veijo Vilva <veion@XIRON.PC.HELSINKI.FI>
Content-Length: 2786
Lines: 76

la xorxes. spuda tu'a mi di'e

> > So we could simply use a parallel construction with {ka}.
> >
> >     lo'e kanba cu zmadu mi leka tu'okau da tuple de
> >     Goats have more legs than me.

> I liked it at first, but I think even this use conflicts with indirect
> questions.

>      mi djuno le du'u lo'e kanba mi zmadu le ka xokau da tuple de

> Could mean:

>   1- I know that goats have more legs than me.

>   2- I know the number of legs which beleg us (each of me and
>      the goats), I being in this property surpassed by the goats.
>      (Or something like that.)

We've always got this problem with nested constructs.

    mi cusku le se du'u do djuno le du'u dakau bebna

    1) I say that you know who is foolish
    2) I say who you know is foolish

and there's a workaround using subscripts - I think the above is (1)
and (2) is

    mi cusku le se du'u do djuno le du'u dakau xipa bebna

but I need to bring myself up-to-date with the latest version
of the abstraction paper.

la djan. spuda tu'a mi di'e

> > No, the property disambiguator is just plain {da}.

> That was my original idea.  However, I've since become convinced that the
> effort of finding a hitherto-unused variable may become too bothersome,
> in contexts where "da"s, "de"s, and "di"s are already flying about.
> So I adopted someone's suggestion of using "kau" in this additional manner.

Pity - that scuppers my idea of using {ka} with DA (argument) and
{kau} (result).  I'm not convinced about the shortage of DAs,
but let's not pursue this too far until I've reread the abstraction paper.

> > John Cowan threatened
> > to respond to the "properties" half, but to the best of my
> > knowledge never did.)

> Right, and mostly because it seems to me that you are correct,

If you thought I had a good idea, why did you make it impossible? :-)

> but the
> more I thought about the matter, the more muddled I got.  Someone needs
> to rethink the whole question of abstraction, preferably in conjunction
> with a close reading of my draft paper on the subject, which glosses over
> a great deal.

I don't promise to redesign the whole system :-), but I've started reading
the new version of the paper, and I'll get back to you.

>  The trouble is that we inherited the nu-ka-ni distinction
> from JCB, who had (has?) no other abstractors, and the remaining ones were
> added in a most ad-hoc fashion, sometimes with random changes -- thus
> "du'u" did not originally distinguish between proposition and text (now
> "sedu'u"), and in fact very early on wasn't NU at all -- it was the
> grammatical equivalent of LE+NU and was usable only with MEX sentences.

Yes, NU is a bit miscellaneous.  Unfortunately it's a bit late to
redesign from the ground up, even if we knew how. :(

co'o mi'e .i,n.