Content-Type: text/plain;
  charset="iso-8859-1"
Reply-To: phma@oltronics.net
To: lojban@yahoogroups.com
Subject: Re: [lojban] Toward a {ce'u} record
Date: Sun, 19 Aug 2001 16:38:02 -0400
References: <E15YYvK-0000Ax-00@mercury.ccil.org>
In-Reply-To: <E15YYvK-0000Ax-00@mercury.ccil.org>
MIME-Version: 1.0
Message-Id: <0108191638020N.01556@neofelis>
Content-Transfer-Encoding: 8bit
Sender: phma@ixazon.dynip.com
From: Pierre Abbat <phma@oltronics.net>
Content-Length: 1172
Lines: 29

On Sunday 19 August 2001 16:11, John Cowan wrote:
> > So, the disagreement is about whether the {ce'u} must always be written
> > in and, if not, where the implicit one is.
> > 1) Every {ce'u} must be explicit. [...]
> > 2) Not so -- some {ce'u} may be implicit, so long as there is a rule for
> > identifying the place(s).
>
> Or 3) Not so -- some {ce'u} may be implicit, and it is up to the
> intelligence of the hearer/reader to figure out where they go.

I think both 2 and 3. In formal Lojban, {ce'u} fills the first unfilled 
place, counting {zo'e} as filling a place; in informal Lojban, it fills the 
first unfilled place that makes sense. I would also apply these rules to ke'a 
in relative clauses.

li'o

> > For this and general
> > reasons, I suggest that {ce'u}, like KOhA generally, be taken as having
> > implicit subscripts (starting with 0) assigned in left to right order.
>
> I think this convention is overkill, though of course I cannot consistently
> say it is outright wrong.

Would ce'uxipa and ce'uxire apply to two sumti such that the property is a 
relationship between them, or would they apply to different levels of nested 
ka?