Message-Id: <m0sfz6z-0000ZHC@xiron.pc.helsinki.fi>
Date:         Tue, 8 Aug 1995 08:29:51 -0700
Reply-To:     "John E. Clifford" <pcliffje@CRL.COM>
Sender:       Lojban list <LOJBAN@CUVMB.BITNET>
From:         "John E. Clifford" <pcliffje@CRL.COM>
Subject:      lu'a
To:         lojban list <lojban@cuvmb.cc.columbia.edu>
Content-Length: 2472
Lines: 39

I confess extreme psychic dissonance here.  Each time xorxes explains an
example, I get a good hypothesis about what is going on.  But then, when I
try that hypothesis on an earlier example (which I thought I understood at
the time) I find that it does not fit and that my old understanding does
not work on the new example.  So, I try another example with the same
result.  I am sure there is a single theory that fits all of xorxes'
cases, but I do not see it yet.  I also have no reason yet to think that
that theory, whatever it is, is or ought to be Lojban and rapidly
strengthening assurance that it IS not Lojban, whether it should be or
not.

        Xorxes has often made the point that the use for these operators
that my interpretation (and I think Lojbab's) gives is useless, since we
already have all the resulting descriptions in primitive form -- and
shorter.  While this is true of unabbreviated forms, it does not take
anaphorized references into account, nor the habit of shifting back and
forth between the various kinds of references to a kind of thing, which
constitutes the most common sort of ambiguity in English (or vagueness,
English plurals m ay refer to an undifferentiated plurality and pray that
context picks out the right application). What we do too often heedlessly
(but -- or because -- easily) in English, we can do consciously (but
almost as easily) in Lojban with this device: move quickly from talking
about the things under one view to talking about them under another,
regardless of what view we last used.  Starting from any lV('V) broda,
anaphorized by _by_, say, we can reach any other simply by prefixing the
correct adapter.  Even the seemingly redundant ones -- lu'a lo broda, for
example -- have their place, to remind us where we are in an array of
shifting types of reference.  In this scheme, I do miss the shift to the
average.  I suspect that its absence comes from the other use -- for
simplifying complex sentential compounds -- where the average plays no
role, but sequences (which are also welcome in the mode-of-reference use)
do.
        Still, I think that xorxes is onto something interesting, probably
even important.  So I will keep pounding away at understanding it.  The
more so because it seems to be involved -- by contrast at least -- with
understanding the already legitimated multiple descriptions like lo ci lo
broda.  Please stay tuned -- or, better, please join in and help clarify
all this.
pc>|83