Message-Id: <199309061500.AA05270@eli.CS.YALE.EDU>
Date:         Mon, 6 Sep 1993 15:58:59 +0100
Reply-To: Colin Fine <C.J.Fine@BRADFORD.AC.UK>
Sender: Lojban list <LOJBAN%CUVMB.bitnet@YaleVM.YCC.YALE.EDU>
From: Colin Fine <C.J.Fine@BRADFORD.AC.UK>
Subject:      Re: TECH: Mark Shoulson waiting for a taxi
To: Erik Rauch <rauch>
Status: RO

And asks about his scurrilous example
++++++++>
 Perhaps if it is the case that
  do ba speni lo dotco "You will marry some German or other"
but not
  do ba speni le dotco "You will marry a certain German"
And you also want to (in the spirit of your example) equate
Germany with Sodom, then there is quandary about the choice
between:
  do ba speni lo ganxygletu
  do ba speni le ganxygletu  [gismu from memory - may be wrong]
nn the one hand, I don't know who it is you will marry - all
I know is that they'll be German: so lo ganxygletu is preferable.
On the other hand, lo ganxygletu excludes Germans who aren't
ganxygletu, which is not what I want: there's no guarantee your
spouse will be a ganxygletu, but there is a gurantee they'll
be German. So le ganxygletu is preferable.

 I guess you, John & Lojbab would definitely use "le ganxygletu"
here, but I wonder what Colin thinks, since he has been taking
the same line as me so far.
>++++++++

Took me several tries to see what this was about, but I at
last get that there is a problem. The problem is I think in the
definition of 'le', which ASSUMES that if you are using a
non-veridical description, there must be a specific target
individual or group that you intend: this example may be
a counter example.
On the other hand it may be that this does still work - that
the specific collection of individuals you are describing as
ganxygletu is in fact ro dotpre - which seemed a non-specific
collection when we though of the first example.

Hm.  More thought needed.

        Colin