X-Digest-Num: 117
Message-ID: <44114.117.652.959273824@eGroups.com>
Date: Fri, 16 Apr 1999 21:24:15 -0300
From: "=?iso-8859-1?Q?Jorge_J._Llamb=EDas?=" <jorge@intermedia.com.ar>
Subject: Re: di'e preti zo nu
X-Yahoo-Message-Num: 652
Content-Length: 1179
Lines: 37



coi djer 

It's good to hear from you again! 

>I agree as always that  "lo tanxe" means E(x)
>T(x) where the x referred to is the same in each form, ie the scope of
>x is the sentence.  So we have: mi nitcu E(x) T(x). We have put a full
>predication, a compound sentence, in a slot calling for a noun. It's
>not going to work.

It can be made to work using a nu abstraction, which
allows us to have a new prenex embedded in the x2 slot:

            mi nitcu le nu da poi tanxe zo'u mi ponse da
            "I need that for some x which is a box, I have x."

>Since I moved on from lojban I wrote a set of modals (need is a modal)
>for NGL. I got around the problem by requiring that the modal take a
>proposition in all cases as grammatical object, never a noun. 

Maybe it makes more sense to define {nitcu} as "x1 needs that 
proposition x2 (du'u) obtains" rather than "x1 needs that event 
x2 (nu) happens", but I don't know, sometimes I think it doesn't
really gets us anything to distinguish nu and du'u. 

But your point still holds: whether nu or du'u, we need a prenex
inside the x2 of nitcu in order to be able to use {lo tanxe}. 

co'o mi'e xorxes