Date: Fri, 4 Nov 1994 09:18:48 -0500
From: Logical Language Group <lojbab>
Message-Id: <199411041418.AA21608@access3.digex.net>
Subject: Re: context in Lojban
Cc: lojbab@access.digex.net
Status: RO
X-From-Space-Date: Fri Nov  4 09:45:09 1994
X-From-Space-Address: lojbab

My opinion is that the status quo for Lojban is that
LO is +veridical
LE is -veridical
There is secondary usage that LE is +specific, because specificity is implied
in having an in-mind object that is not necessarily veridical.  There has
been an example of metonymy used (like using "the elephant" for a picture of
an elephant) with LE, but the standard has been that metonymy should be
expressed with la'e/lu'e.  But even metonymy can be considered a kind of 
specificity; just to a lesser degree, so the association of non-veridicality
and specificity still seems relevant, if not formally part of the language
definition.

LO being usually contrasted with LE, it therefore has fallen on LO to reflect
non-specificity.  But a non-specific, non-veridical should not be expressed
with LO.  I think veridicality therefore remains pre-eminent, with the
words being neutral on specificity.  I believe that in the ideal case, though,
LO is making a minimal claim ONLY of veridicality, which implies nonspecificity
unless there is an expressed restriction.  I would be willing to back off
on the requirement of restriction before backing off on veridicality.

Note that Russian gets along just fine without a specific/non-specific
distinction.  It is something alien to English speakers, though I have
surprisingly never had trouble in Russian determining the level of
specificity when it was relevant from context.

On the other hand, LO would not exist in Lojban without the veridicality
criterion.  "lo" is our version of JCB's "lea" (= all of those in the set
that veridically are) which in Lojban is "ro lo".  It was my suggestion
to pc that "lo" be contrasted with "le" and thus usable with individuals
out of that veridical set if appropriate, ENABLING non-specific selection
from the veridical set to be the default, because "all broda" statements
really aren't that useful in language when people are really concerned with
truth values.  Hence the outer quantifier "su'o".

Having defined "lo" this way, I also quickly saw that two contradictory usages
of JCBs massification word ("lo" in TLI Loglan) corresponded to a specific
in-mind massed unit, and a massification of the entire set of things (i.e.
a massification of the veridical set), and these became "lei" and "loi"
respectively.  "le'i" and "lo'i" were added soon afterwards when we realized 
that JCB often confused talking about the members of the set wwith talking 
about the set itself.  By that time I was on a pattern binge as we were
trying to reassign all the cmavo, and "lai" and "la'i" were added for symmetry.
JCB had "lo'e" and the pattern of the other members of LE made me want
to fit it in, and I realized that we could have two lo'e/le'e with useful
meanings that corresponded to this pattern.

Now if people convince pc and Nick and others that this design is wrong
logically, then we may have to redefine things.  BUt I really would be
disinclined to make any such change without it being a true flaw in the 
language.  I would rather see "lo" become less useful than to muddy its
definition further.  After all, JCB's group gets by pretty much stuck with
only "le" rather than our panoply of LE.

lojbab