Alas, what you say may be practically correct, but is logically wrong. All predicates must have all arguments in place. If we omit some at the surface level, they have to be restored uniquely at some level before interpretation. At the moment, Lojban does not have a way of doing this. As you note, following CLL, sometime there just isn't something for that slot -- which means, of course, that we have the predicate wrong, if not the other places. On the other hand, when there is something there, we have no clear way of saying what should go there: the obvious value (I suppose this is sorta {zo'e}) or the any value that makes the whole true (presumably {da}), but no rule as to which to pick (and {zo'e}'s use here does not fit well with its use in "definitions" of gadri).
From: Ian Johnson <blindbravado@gmail.com>
To: lojban@googlegroups.com
Sent: Sunday, June 9, 2013 7:53 PM
Subject: Re: [lojban] cmevla as a class of brivla
--
You received this message because you are subscribed to the Google Groups "lojban" group.
To unsubscribe from this group and stop receiving emails from it, send an email to lojban+unsubscribe@googlegroups.com.
To post to this group, send email to lojban@googlegroups.com.
Visit this group at http://groups.google.com/group/lojban?hl=en.
For more options, visit
https://groups.google.com/groups/opt_out.