Message-Id: <199308190000.AA02564@eli.CS.YALE.EDU>
Date:         Thu, 19 Aug 1993 00:57:43 +0100
Reply-To: Matthew Faupel <matthew@CPDAPO.TELE.NOKIA.FI>
Sender: Lojban list <LOJBAN%CUVMB.bitnet@YaleVM.YCC.YALE.EDU>
From: Matthew Faupel <matthew@CPDAPO.TELE.NOKIA.FI>
Subject:      ZAhO: A view from the sidelines
To: Erik Rauch <rauch>
In-Reply-To:  Jorge LLambias's message of Wed, 18 Aug 1993 11:57:14 EDT
Status: O

Otherwise known as "fools rush in where angels fear to tread" :-)

I've been trying to follow the great ZAhO debate.  It's not been easy given
that it started before I joined the list and I barely know enough Lojban to
figure out what's being talked about (let alone cope with the terminology),
but I think I've sort of got it.  I'm listing what I think the points of
contention are, partly because no-one else has put them all in one place,
and partly because I want to make sure I've got them right before I throw in
my own tuppence-worth of opinion.  It might also help out some other
beginners; you never know!  Here they are as point, summary and comment.

1.  Are ba'o/pu'o the wrong way round?

S:  The argument being (I think) that at the moment (possibly illogically)
    "ba'o broda" implies "pu broda" instead of "ba broda" (and similarly for
    "pu'o").

C:  I didn't think that the assignments of pu'o and ba'o were illogical
    until I started reading this debate!  "mi pu broda" is "mi broda ca lenu
    mi pu'o cusku dei" which seems reasonable enough.  Another mnemonic is
    that pu/ca/ba are before/while/after relative to the speaker and
    pu'o/ca'o/ba'o are before/while/after relative to the event.

2.  How far back before the event does the period covered by pu'o extend?
    (and how far after for ba'o?)

S:  Theoretically, the period covered by pu'o extends infinitely before the
    event and ba'o infinitely after, however some people have been using
    them to suggest "about to" and "just after" i.e. marking a point in time
    shortly before the event has happened or just after it has happened.

C:  Apart from the fact that they're slightly more cumbersome, why can't
    "bazi co'a" and "puzi mo'u" be used to express these concepts?  This
    would leave pu'o and ba'o free from the need to be interpreted according
    to context.  Or is there some fatal flaw I'm not aware of here?

3.  ZAhO used as sumti tcita (things that add a qualification to a bridi?)
    have the effect of limiting the entire period of validity of the bridi
    (this behaviour differs from the effect of the other selma'o VA, PU, ZI
    and FAhA).

S:  This derives (so I believe) from the example in "Imaginary Journeys":

    mi morsi ba lenu mi jmive
    I am-dead in-the-future-of the-event-of I live

    mi morsi ba'o lenu mi jmive
    I am-dead in-the-aftermath-of the-event-of I live

    where the explanation states that in the first case, the "I am-dead" is
    asserted to be true at some point after "the-event-of I live" has become
    true (thus implying I could be dead while I'm alive, just so long as I'm
    alive first :-) whereas in the second "I am-dead" is asserted to be true
    only after "the-event-of I live" has been true and become not true.
    (See footnote).

    The upshot of this is that if the explanation is followed, ZAhO tcita
    mark the time of the beginning and the end of the bridi being true as
    having to be within the event phase of the qualifying sumti described by
    that tcita, whereas PU tcita et al. do not delimit both ends of the
    event in the same way.

C:  This interpretation seems to be limiting; the desired effect could be
    achieved without imposing the delimitation by:

    mi ca'o morsi ba'o lenu mi jmive

    (though I prefer "mi co'a morsi mo'u lenu mi jmive" to eliminate the
    philosophically worrying gap that otherwise would be possible between
    living and dying :-)  If a non-restrictive interpretation is used, a
    complementary ZAhO can always be used in the main bridi to impose the
    restriction, while the reverse is not always true.

Footnote 1:
    What's the event contour of an event that never happens?  Is it always
    pu'o?

Footnote 2:
    Incidentally I note that:

    mi broda PU mi ba brode -> mi baPU broda .i mi ba brode

    but

    mi broda ZAhO mi ba brode -> mi ba broda .i mi ba ZAhO brode

    (that second one's not really correct, but I hope you see what I'm
    getting at - it's getting late and I don't have the brainpower to figure
    out what it ought to be).  Is this just an odd feature, or does it have
    some real significance?  If so, what is that significance?

Caveat Lector: I'm a beginner.  Please don't get too upset if large chunks
of the above don't make sense; just try and point out the mistakes :-)


co'o mi'edoi matius.