Re: na`e

[Lee Daniel Crocker <lcrocker@CALWEB.COM>]

>>Chapter 15:
>>    But what exactly does na'e negate?  Does the negation include only
>>    the gismu klama, which is the entire selbri in this case, or does
>>    it include the le zarci as well?  In Lojban, the answer is
>>    unambiguously ``only the gismu''.  The cmavo na'e always applies
>>    only to what follows it.

> So that would mean that {lo mlatu ca'o na'e vreta lo stizu} means "the cat
> other-than-sits on the chair" (NOT that the cat sits other than on the
> chair).  The "other statement" that is true might be {my. ca'o citka lo
> stizu} or {my. ca'o cadzu lo stizu} etc.  Does that make sense?  If {na'e}
> negates what comes after it, that's the brivla here, not the following
> sumti as you seem to have interpreted.
>
> Sorry if this is wrong, or right but old news.

You're right about that part, but that's not the argument.  Yes, na'e
negates--in some sense to be agreed upon--only the word following.  So
"le mlatu na'e vreta le stizu" is "the cat other-than-sits-on the chair"

That still leaves issues:

(1) Does that bridi assert that the cat is not in fact sitting on the
    chair, in addition to asserting that some other relationship exists
    between them, or does it only assert the latter (very weak) predicate?

    Just because na'e is translated as "other than", I don't think we
    necessarily have to stick to its English meaning, expecially when it
    is ambiguous and not as useful.  As the example of "4 is other-than-
    the-sum-of 2 and 2" shows, English strictly asserts only the weak
    predicate, while vaguely implying (falsely, in this case) the strong.

    Since "the cat <something> the chair" doesn't really say anything
    useful--the cat could dislike the chair, it could eat the chair,
    it could be the chair--I think Lojban should choose the stronger
    and more useful interpretation here, even though it is too late to
    clarify this issue in the books.  That's what errata sheets are for.
    The cartoon at the start of that chapter implies this interpretation
    to me, anyway.  We don't imagine that she actually did wash the
    car, and just wanted to imply that she did something else to it as
    well.  Again, English only /implies/ that the car did not get
    washed, but we should choose to either explicitly assert it or
    explicitly refuse to assert it.  I think the former is more useful.

    So, "le mlatu na'e vreta le stizu" asserts two propositions: that
    "le mlatu cu vreta le stizu" is false, and that some bridi with a
    different brivla and the same sumti is true.  That means that the
    bridi "li vo na'e sumji li re li re" is false, even though its
    naive translation into English is true.  I don't have a problem
    with that--the translation just needs to be smarter.

    Now, it could be argued that even if we decide upon the weaker
    condition, the stronger can be easily asserted by using two
    bridi.  My brain can't at the moment figure out how to say the
    weaker predicate alone if we choose the stronger for na'e;
    something with broda and a prenex, maybe, like "the cat <X>
    the chair, where <X> != sits-on"; can someone more fluent help
    me out here?

    The same issue applies to na'e on one sumti: the stronger
    interpretation of "le mlatu cu vreta na'e le stizu" is that
    the cat is not sitting on the chair AND is sitting on something
    else.  The weaker interpretation only asserts that the cat sits
    on something else, and posibly the chair as well (perhaps it
    is sitting on a pillow on the chair, or else the tense used
    means the cat has sat or will sit on many things).

(2) Multiple na'e: What does "na'e le mlatu cu vreta na'e le stizu"
    assert?  Do they group left-to-right, or collapse onto the
    bridi?  I'm inclined to go with the simpler, weaker meaning here,
    because you can always break up a complex bridi into parts and
    references, but the simplicity of collapsing all the na'e in a
    bridi is useful in conversation.  In other words, I would choose
    that this asserts that (a) "le mlatu cu vreta le stizu" is false
    and (b) it is true for some different value of /any/ of the
    sumti--in other words, maybe the dog sits on the chair, or the cat
    sits on the shelf, or the mailman sits on the dog.  I think that's
    a useful and ordinary meaning.

    The alternative, if we choose the strong na'e and left-to-right
    grouping, gets convoluted: the bridi above asserts that (a) "le
    mlatu cu vreta na'e le stizu" is false--the cat can't sit on
    anything--and (b) it's true for some other value of the sumti--
    somebody /can/ sit on something other than the chair.  The "inner"
    na'e then asserts other things, which must then be compounded with
    the outer assertions.  I definitely like the weaker "collapsed"
    multiple na'e here.

(3) How does one negate all or only certain brivla in a tanru, and
    what does that assert about the whole bridi?

--
Lee Daniel Crocker <lee@piclab.com> <http://www.piclab.com/lcrocker.html>
"All inventions or works of authorship original to me, herein and past,
are placed irrevocably in the public domain, and may be used or modified
for any purpose, without permission, attribution, or notification."--LDC