Message-Id: <199310141452.AA06016@eli.CS.YALE.EDU>
Date:         Thu, 14 Oct 1993 06:49:01 -0400
Reply-To: Logical Language Group <lojbab@ACCESS.DIGEX.NET>
Sender: Lojban list <LOJBAN%CUVMB.bitnet@YaleVM.YCC.YALE.EDU>
From: Logical Language Group <lojbab@ACCESS.DIGEX.NET>
Subject:      Re: TECH: Lean Lujvo and fat gismu
To: Erik Rauch <rauch>
Status: RO

On zi'o lujvo
I disagree a) because no convention has been adopted (at least I wasn't
consulted to the level of "convention" that I know of) and b. there
is an analysis that supports my place order, to wit
selzilklama reflects the tanru expansion se zei zi'o
zei klama
Since this does not look like any standard tanru g
rammar, you have to assume
something is left out.  Thus zi'o (which has sumti grammar) HAS  to be treate
d
as if it had some other grammar.  What grammar is somewhat arbitrary.  If it
is treated as a UI (a common fallback) it works.  If it is treated as
a tanru element (I guess as "me zi'o") then the grouping is (se *zi'o) klama
and you resolve the first two components as a place deletion with whatever
conventional interpretation you can agree on.  In this analysis, it seems
that there should be no reaarangement of terms since there is no suggestion
of rearrangement.

Since it is presumed that the place structure orders are desired to be in
a particular order as a norm, analyses that lead to nonstandard and not
inherently obvious orders are suspect.

I think the whole question of zi'o is still up in the air of course.  No
one has expressed satisfaction with it.  Indeed I think everyone dislikes it.
I cannot accept that there is a consensus on how it is to be used when no
one has agreed that it SHOULD be used.  To me it is an experimental cmavo
that we have assigned a real value because of the impending dictionary
publication and the need to have a minimum of such obvious experiments in
the dictionary.  But please don't cast its usage in concrete.

lojbab